international journal of business science and applied ... · the cognitive perspective in the...

Int. Journal of Business Science and Applied Management, Volume 7, Issue 1, 2012

Individual perception of different stimuli: Implications for

managers

María Valle Santos- Álvarez

Department of Business and Marketing, University of Valladolid

Avda. Valle de Esgueva, 6, 47011, Valladolid, Spain

Telephone: +34 983 183813

Email: [email protected]

María Teresa Garcia-Merino

Department of Business and Marketing, University of Valladolid


Telephone: +34 983 183813


Eleuterio Vallelado-Gonzalez

Department of Finance and Accounting, University of Valladolid


Telephone: +34 983 423387


Abstract

Managerial perception is the process by which managers form an image of the stimuli they receive. According

to research, perception is conditioned by the individual’s cognitive profile. But the different nature of incoming

stimuli suggests that it would be interesting to study whether the cognitive profile’s influence varies in the

presence of different stimuli. This paper analyses the effect of the cognitive profile on perception of differently-

structured stimuli. The results clearly show that the cognitive style, tolerance of ambiguity, and proactivity have

an effect. Specifically, they condition the recognition of stimuli, particularly when the stimuli are relatively

unstructured. The results also show that the cognitive variables have less influence in the interpretation stage.

Keywords: managerial perception, stimulus, structure, strength, cognitive profile

Int. Journal of Business Science and Applied Management / Business-and-Management.org

2

1 INTRODUCTION

The cognitive perspective in the management literature recognises managerial perception as the firm’s

“black box” that hides the key parameters determining decision-making from view (Sáez and Jiménez, 2004;

Yanes, 2004). Managers form a mental representation from the stimuli they receive from their competitive

context (external stimuli), and this becomes the foundation on which they adopt their strategic decisions.

Research also shows that managers differ individually in terms of how they perceive, acquire, interpret and use

information (see Walsh, 1995, Livengood and Reger, 2010, Tang et al, 2012, Barreto, 2012). Thus it would be

extremely useful to understand the factors that affect managers’ interpretations of stimuli (Kuvas and Kaufman,

2004), because these interpretations affect their knowledge about internal and contextual factors, and

consequently condition entrepreneurial actions and intentions (Wood, 1991; Miller, 1993; Thomas and Simerly,

1994; Shane, 2003; Baron, 2006; Ubcbasaran et al., 2008; Kaplan, 2008; Blume and Covin, 2011; Plambeck,

2011).

In this respect, research has stressed that managers’ perception is far from being an objective and rational

process. Rather, the process is influenced by aspects as diverse as the context (Santos and García, 2008), the

qualities of the manager, and even the nature of the stimulus (Dutton and Ashford, 1993, Gielnik, et al., 2011;

Plambeck, 2011). The empirical literature offers clear evidence that the organisational context – the type of firm

and its strategic orientation – influences managers’ interpretations of the stimuli coming from the environment

(Thomas and McDaniel, 1990; Ginsberg and Venkatraman, 1992; Thomas et al., 1993, 1994; Calori et al., 1994;

Sutcliffe, 1994; Denison et al., 1996; Chattopadhay et al., 1999; Nadkarni and Barr, 2008; Nicolau and Shane,

2009; Singh, 2010). At the same time, both theoretical and empirical literature recognises that managers’

personal characteristics and beliefs also condition the perception process (Daft and Weick, 1984; Schwenk,

1984; Milliken and Lant, 1991; Kautonen, 2008; Blume and Covin, 2011). The difficulty arises when it comes

to how to operationalise and analyse these individual characteristics that affect perception. Perhaps for this

reason researchers frequently use demographic characteristics (age, education, experience, etc.) to measure

individual characteristics (Schneider and DeMeyer, 1991; Thomas et al., 1993, 1994; Kautonen, 2008). But the

researchers often explain the effect of these demographic variables using arguments involving cognitive aspects

and the different managerial attitudes that affect the process of recognition and interpretation of the competitive

context (Hambrick and Mason, 1984; Gillingwater and Gillingwater, 2009). It would be more appropriate to

recognise that the individual characteristics that most affect the perception process are the cognitive traits.

The cognitive profile refers to the set of individual qualities related to the different forms of thought and

action managers can engage in, in other words their capacity to recognise and interpret information. After a

review of the available literature, we did not find any studies that have analysed the relationship between

managerial perception and cognitive traits.

As proposed in signal detection theory (Swets, 1992), a stimulus is a signal that could foster an idea or

knowledge in the receiver. In management literature it is common to introduce external stimuli as signals that

issue the context but that influence managers’ reactions in such a way that they condition business behaviour

(Miller, 1996; Green et al., 2008). Furthermore, entrepreneurship research emphasizes the key role of stimuli in

opportunity recognition processes (Shane, 2003, Baron, 2006; McMullen and Shepherd, 2006, Ucbasaran, et al.,

2008; Nicolay and Shane, 2009). Additionally, the research stresses the diversity and complexity of the stimuli

that managers receive (Mintzberg et al., 1976; Eggers and Kaplan, 2008; Phan et al 2009, Tang, et al., 2012) and

the varying amount of attention managers pay them. Research has shown that managers pay more attention to

stimuli linked to threats than to those linked to opportunities and linked with previous knowledge (Dutton and

Ashford, 1993, Shepherd and Detienne, 2005; Blume and Covin, 2011). Researchers have also shown that the

strength and clarity of the stimulus moderates the extent to which the receiver’s personality affects the

perception process. Thus the clearer the stimulus is, the less important the individual perception process (Fiske

and Taylor, 1991; Waller et al., 1995; Entrialgo et al., 2001). Other researchers offer features to classify the

stimuli into different categories. Some distinguish between opportunities and threats, while others stress

structure and strength (Haukedal, 1994; Haukedal and Gronhaug, 1994).

Thus the objective of the current paper is to study the effect of the manager’s cognitive traits on the process

of the perception of stimuli. The study also analyses whether the effect of the cognitive profile on perception

depending on the nature of the stimuli. The purpose of this paper is to illuminate the perception of diverse

stimuli. For this purpose, the following section looks at the literature on the perception process and the influence

of the cognitive profile. The stimuli are characterised according to two features (Haukedal, 1994; Haukedal and

Gronhaug, 1994): structure and strength. This leads to the formulation of the model that guides the study.

Section 3 offers the methodology. This work uses experimentation, a technique that is frequently used in

research on cognitive biases and managerial perception (Samuelson and Zeckhauser, 1988; Sterman, 1989;

Schwenk, 1995; Shout and Bolger 2002, Gielnik et al., 2011). The experiment involves business administration

students, who were considered representative of potential managers. These students were chosen for two main

reasons: they had a basic understanding of the workings of the capital market and takeover processes, and they

María Valle Santos- Álvarez, María Teresa Garcia-Merino and Eleuterio Vallelado-Gonzalez

3

had no previous managerial experience, so they would not be influenced by their beliefs (Ucbasaran et al., 2010;

Baron et al., 2012; Aguinis and Lawal 2012). Section 4 offers the main results of the study, and Section 5 the

discussion. The paper ends with some final remarks.

2 INDIVIDUAL PERCEPTION OF STIMULI: A REVIEW

Perception is the process by which the individual forms an image of the surrounding reality. In other words,

it is the cognitive representation resulting from the individual interpretation process (Haukedal, 1994). Two

distinct stages are evident in this process (Starbuck and Milliken, 1988; Santos and García, 2006): (1) the

recognition of the stimuli; and (2) their subsequent interpretation. In the first stage the individuals (managers,

for instance), faced with the stimuli coming from the context, select only those that they consider relevant (i.e.,

classify each stimulus as either a signal or noise). In the second stage the managers attribute meaning to those

stimuli they selected in the first stage. Thus the managers form a more or less reliable image of the original

stimulus. This final image can contain various biases –misperception- that distance managers’ perception from

the original signal. This is the case with omission biases, when relevant stimuli are ignored, or attribution biases,

when the meaning does not correspond to the real strategic importance of the stimulus. What the manager

registers then is not the real stimulus. Something is always added, taken away or changed when the stimulus

passes from the senses to the mind of the manager (Haukedal, 1994).

The effect of the cognitive profile on perception

This perception process is affected by receivers' individual characteristics, especially their cognitive traits.

The current researchers understand the cognitive traits or cognitive profile as the individual framework in which

information is processed and analysed. Specifically, the terms refer to the knowledge structures and mental

models that the individual uses to make assessments, judgements or decisions (Cools and Van den Breek, 2008).

The cognitive profile coincides, in essence, with what is sometimes called “cognitive schema”, defined as a

mental structure that directs information processing, guides attention and memory towards consistent schemata,

and fills gaps in the information (Gioia and Poole, 1984; Ericson, 2001; Baron et al, 2012). Defined like this,

individuals’ cognitive traits undoubtedly govern and condition their perception of stimuli. The central

hypothesis of this research follows:

H1: The individual cognitive profile conditions the perception of stimuli.

Accepting the influence of the cognitive profile on the perception process, the difficulty is in

operationalising these concepts. Researchers use several dimensions in the literature, and the current paper uses

the following three: cognitive style, tolerance of ambiguity, and proactivity.

Cognitive style refers to the individual information processing process1 (Hayes and Allinson, 1994; Sadler-

Smith and Badger, 1998; Cools and Van den Breek, 2008). Cognitive style affects how people look at their

context to gain information, how they analyse their environment, how they organise and interpret that

information, and how these interpretations guide their actions (Hayes and Allinson, 1998). Likewise, the

concept refers to the way individuals perceive stimuli and how they use that information to guide their

behaviour (Hayes and Allinson, 1998). The individual cognitive style can affect preferences for different types

of learning, knowledge, information processing and decision-making (Dubard et al., 2007). Thus the cognitive

style makes it possible for managers to see what they see but at the same time blinds them to other aspects.

Moreover, the cognitive style conditions the stimuli that the individual pays attention to and how that stimulus is

interpreted and understood (Hayes and Allinson, 1998, Dutta and Thornhill, 2008), so it conditions the two

stages of perception.

The literature offers different models of reference for the cognitive style. Authors have identified up to 54

items (Hayes and Allinson, 1994; Sadler-Smith, 1999; Armstrong, 2000; Sadler-Smith et al., 2000; Hodgkinson

and Sadler-Smith, 2003; Kozhervnikov, 2007; Cools and Van den Breek, 2008). Thus some authors have tried

to frame these items in a bipolar scale (Allinson and Hayes, 1996). Of all the scales in the literature, the current

authors use Allinson and Hayes’ (1996) cognitive style index (CSI). The CSI is a bipolar construct that classifies

individuals into two groups: intuitive versus analytical. The reason for this choice is that various studies find a

relation between the CSI and business behaviour (Allinson et al., 2000), making the indicator appropriate for the

current research. The analytical individual studies problems in detail and makes decisions on the basis of mental

reasoning. The intuitive individual makes decisions and deals with problems based on feeling (Allinson and

Hayes, 1996, Dutta and Thornhill, 2008).

1 For a more detailed explanation of the differences between cognitive style and other closely related

concepts – learning style, cognitive strategies, etc. – see Hayes and Allinson (1994).


4

Tolerance of ambiguity is the way in which the individual deals with an ambiguous situation (Furnham and

Ribchester, 1995; Cools and Van den Breek, 2008). The concept represents the person’s capacity to accept the

lack of information about the range of possible outcomes and their probabilities (Sherman, 1974; McNally et al.,

2009). Tolerance of ambiguity also refers to the individual’s capacity to make decisions in risky or highly

uncertain environments (Westerberg et al., 1997). Individuals with less tolerance of ambiguity perceive

ambiguous situations as threats (Sully de Luque and Sommer, 2000; Ling et al., 2005), while individuals with a

higher level of tolerance perceive such situations as non-threatening (Budner, 1962). Moreover, individuals with

less tolerance perceive a higher level of risk in ambiguous situations than those with a higher level of tolerance

(Conchar et al., 2004). Ambiguous situations are totally new situations, complex situations with a large number

of elements, or contradictory or ill-defined situations (Conchar et al., 2004).

Finally, proactivity refers to individual differences in how people seek to influence and change their

environments (Bateman and Crant, 1993; Cools and Van den Breek, 2008). Proactive individuals look for

opportunities, show initiative, do things, and persevere in their behaviour until they achieve the set goal

(Bateman and Crant, 1993). Proactive behaviours are anticipatory and anticipation depends on imagining future

outcomes. Proactivity helps individuals understand the information they receive (Crant, 2000; Grant and

Ashford, 2008). Proactive individuals also feel less conditioned by the situational forces of their environment

than less proactive individuals (Kickul and Gundry, 2002).

Stimuli: their diversity and perception

Managers are overloaded with information, since they receive a huge amount of diverse stimuli (Mintzberg

et al., 1976; Baron, 1998, 2004). The diversity of stimuli makes it necessary to use some criterion to classify

them into different categories. The literature offers a number of different typologies, for example the one that

stresses the content of the stimuli and distinguishes between opportunities and threats (see Dutton and Jackson,

1987, Kuvaas and Kaufman, 2004). But this classification is somewhat confusing, because depending on the

receiver a stimulus can be classified as either an opportunity or a threat (Ramaprasad and Mitroff, 1984; Kuvaas

and Kaufman, 2004). In this paper, the authors use the classification with the features “strength” and “structure”

(Haukedal, 1994; Haukedal and Gronhaug, 1994) to classify the stimuli. The first feature – strength –

distinguishes between easily recognisable stimuli and stimuli that are more difficult to recognise, discriminating

between stimuli in terms of their visibility. On the other hand, the second feature allows the researcher to

distinguish between well-structured stimuli and stimuli with an undefined, diffuse structure (Simon, 1973). This

feature is difficult to measure, and it is represented by three attributes (Kaufman, 1987): novelty, complexity,

and ambiguity. Novelty refers to the individual’s lack of experience or knowledge that can be directly applied to

the perceived stimulus. Complexity refers to the quantity of information contained in the signal. Finally,

ambiguity refers to situations in which the problem is to decide what the best options are, or which suit the

overall objective best. These are situations with various alternatives and the manager is unsure of which is the

best one. Using these two features: strength and structure, we can distinguish among four different types of

stimulus:

Table 1: Adopted from Haukedal and Gronhaug (1994: 357)

Strength: high Strength: low

Structure: high Type A: Messages Type C: Whispers

Structure: low Type B: Symptoms Type D: Itchers

Type A – “messages” – is surely the least problematic in terms of perception since it refers to stimuli that

are perfectly visible – so they catch the manager’s attention easily – and well-structured – so the managers know

what they mean and how important they are. The stimuli classified as “itchers” (D) are at the other extreme,

since they barely catch the manager’s attention because of their low visibility, and they are also difficult to

interpret. “Whispers” (C) are stimuli that are difficult to recognise but that have a clear, known structure. These

stimuli consequently have a problem of strength. Finally, “symptoms” (B) are strong stimuli but they are

relatively unstructured. Managers faced by these stimuli have the problem of finding the right interpretation and

determining the most appropriate actions.

Thus it can be deduced that type A stimuli are the clearest and type D stimuli the least clear, and the other

two categories are somewhere in between. Moreover, and as mentioned above, different stimuli present different

challenges to managers in their perception process. Thus while some stimuli generate problems of recognition,

others give rise to difficulties in the interpretation process, and the clearest lead to simpler perception processes.

Also, research recognises that the “force of the situation” moderates the extent to which the receiver’s

personality affects the perception process (Waller et al., 1995; Entrialgo et al., 2001). The force of the situation


5

refers to the clarity of the stimulus, which is composed of the strength and structure (Fiske and Taylor, 1991).

The clearer the stimulus (high strength and high structure), the less effect the receiver’s individual

characteristics have on the perception process. But in relatively unclear stimuli (low strength and low structure)

the cognitive variables decisively condition the perception process (Sutcliffe and Huber, 1998). The second

hypothesis of this paper follows:

H2: The influence of the cognitive profile on an individual’s perception of stimuli is weaker with clear

stimuli (high strength and high structure).

Comparing two stimuli with the same structure but different strength, the perception process depending on

the ease or difficulty in recognising the stimulus, rather than of interpreting the stimulus or deducing its strategic

implications. Thus the cognitive traits most involved in the process of recognition of the stimulus will be more

relevant than those linked to interpretation. Similarly, comparing two stimuli with the same strength but

different structures, the cognitive traits most involved in the process of interpretation will be more relevant.

On the other hand, the characterisation of the cognitive profile reveals that the three dimensions considered

have different effects in different information contexts. Thus proactive behaviour has more effect in uncertain

situations (Weick, 1979; Grant and Ashford, 2008). Equally, tolerance of ambiguity is also associated with

situations of uncertainty. With regard to the cognitive style, the analytical character seems to be more suited to

structured contexts with information available, while the intuitive character may be more appropriate in

uncertain situations. With all this, the third and final hypothesis of this research is as follows:

H3: Tolerance of ambiguity, proactivity and intuitive character have more effect on the perception of

unclear stimuli (low strength and low structure) than on the perception of clear stimuli.

This completes the analytical model upon which the current research is founded. This research will study

the effect of the cognitive profile on the process of the perception of different stimuli.

3 METHODOLOGY

Experiment design

The empirical analysis aims to analyse whether perception of stimuli is conditioned by the individual’s

cognitive profile, and above all, if the effect of the cognitive profile is stronger with unclear stimuli. For this

purpose, the authors compare the perception of two stimuli with the same strength (high) but different structure

(high vs. low). Of the three attributes determining the level of structure, novelty and complexity are kept

unchanged, and the authors compare the perception of two stimuli that differ only in their level of ambiguity.

They are consequently comparing a type A stimulus (message) and a type B stimulus (symptom). Recall that the

former has high strength and high structure, while the latter has high strength but low structure.

To carry out this study the authors needed to observe the behaviour of individuals in a controlled way, so

they used a laboratory experiment2 (Aguinis and Lawal, 2012). The sessions of the experiment were carried out

in our experimental laboratory. A total of 96 people participated in the experiment, all of them undergraduate

business administration students enrolled at the University of Valladolid (Spain), recruited through a public

announcement. These students were considered representative of potential managers. Incentives were used to

encourage the participants to behave realistically in response to the announcement of a takeover bid (Camerer

and Hogarth, 1999). Participants were remunerated according to the level of coherence of their decisions. The

remuneration was a fixed amount, from which sums were subtracted for each financially incoherent decision.

Specifically, two situations were penalised: (1) responses in which the participant predicted a rise (fall) in the

share price of the firm – bidder or target firm – and then declared their decision3 in the market to be to sell (buy)

those shares; and (2) responses in which the participant predicted a fall in the share price of the target firm.

The experiment involved presenting the participants – in writing – with an event that contained two stimuli.

The event was the announcement of a takeover bid (henceforth, TOB4) in six different scenarios. The event was

the same in each of the scenarios – the announcement of the TOB – but the presentation and source differed.

Thus, for example, some scenarios stated that a financial newspaper reports on the TOB, while others stated that

rumours about a possible bid are heard in the branch of a bank. In each scenario the announcement of the TOB

2 Experimentation is one of the most widely recommended techniques for research into cognitive biases

(Schwenk, 1995). This type of research frequently involves university students (Zalesny and Ford, 1990). 3 For a more detailed explanation, please contact the authors.

4 In general terms, a TOB refers to purchases in which firms acquire securities with voting rights from a

company: shares, convertible bonds, subscription rights, etc. Although bidders generally only want to acquire

part of the firm, TOB operations are considered typical of the market of control (Farinós and Fernández, 1999).


6

was described, and this gave rise to the two stimuli subsequently analysed: the stimuli referring to the bidder and

the target firm, respectively.

A TOB is a strategic decision that, as Baker et al. (2009) recognise, requires setting a value that is

sufficiently flexible to make the offer attractive to the shareholders and managers of the target firm but that at

the same time avoids giving the shareholders and managers of the bidding firm the feeling they are overpaying.

In these circumstances, the cognitive profile helps manage the complexity of the decision in an environment

with an information overload (Raghubir and Das, 1999). The choice of this event was also based on two other

reasons. First, the event would have barely any affective connotations for the participants, since they had had no

previous experience in the area; they would consequently not be influenced by their beliefs. Second, all the

participants had a basic understanding of how the capital market and takeover processes work, so the event was

familiar to them.

For each scenario, and for both the bidder and the target firm, the participants were asked to say whether

they thought the event stimulus would affect the share price or not. Specifically, they were asked whether, in

light of the TOB announcement, they believed that each firm’s share price would rise, fall or remain unchanged

in the short term (Appendix 1).

The financial literature is unanimous in saying that credible TOB announcements generate powerful

information, create value and produce abnormal profits for the shareholders of the target firm (Farinós and

Fernández, 1999), since they have a positive effect on the firm’s share price (Mitchell et al., 2004). The same

cannot be said for the firm that launches the takeover bid. Firms launch such operations for a wide range of

reasons, and so the effects on the bidder are equally diverse. The bidder’s share price sometimes rises after

announcing a TOB but sometimes falls.

These are consequently the two stimuli that will be used in this study: the TOB announcement for the

bidder firm and for the target firm, respectively. These two stimuli have the same degree of strength since they

appear in the same event. They also have the same degree of novelty (the participants have no previous

experience in TOB operations but understand how they work) and complexity (they receive the same

information). Nevertheless, the ambiguity of the two stimuli differs: the TOB announcement is less ambiguous

for the target firm than for the bidder. Thus the TOB announcement will be a type A stimulus for the target firm

but a type B stimulus for the bidder.

Measures design

The participants were also asked a series of questions to determine the three dimensions of their cognitive

profile: cognitive style (COG), tolerance of ambiguity (TA), and proactivity (PRO). The authors used Allinson

and Hayes’s (1996) cognitive style index (CSI)5 to measure cognitive style. This indicator consists of 38 items

that the participant must score on a three-point scale (“true”, “uncertain”, “false”), and measures how intuitive

or analytical the individual is. Thus the range of scores runs from 0 to 76. Individuals with an analytical style

obtain high scores on the CSI, while intuitive individuals get low scores. The Cronbach alpha for this scale is

0.857, so its internal consistency can be considered satisfactory6.

To measure tolerance of ambiguity, the authors used Acedo and Jones (2007) and included four statements,

which the participants had to score at 1-5 depending on their level of agreement. In this case, the reliability

indicator is acceptable (α = 0.83). For proactivity, the authors included a 10-item scale, each item scored at 1-5

depending on the degree of agreement, as in the previous scale. The Cronbach alpha in this case (0.72)

guarantees the reliability of this scale.

The authors built two individual indicators for the stimuli analysed: an indicator of the proportion of

stimuli classified as relevant (AFF), and another of the proportion of stimuli considered to have a favourable

effect on the firm’s share price (FAV). The first (AFF) reflects the number of times the individual says that the

stimulus received affects the share price in the six scenarios presented in the experiment. The second (FAV) is

calculated as the ratio of the number of times a favourable effect is recognised (the share price rises) to the

number of times the stimulus is considered to have an effect. This indicator varies in a range from 0 to 1. These

indicators were calculated separately for each of the stimuli analysed: AFF-A and AFF-B for recognition of the

stimuli and FAV-A and FAV-B for attribution of meaning to the stimuli recognised as relevant (favourable or

unfavourable) (Appendix 2).

5 The authors used this index with the kind permission and authorisation of its authors, Professors Allinson

and Hayes from the University of Leeds (UK). 6 This result is consistent with results of previous studies, which are in the range 0.78-0.9.


7

4 RESULTS

The authors analysed whether the participants’ cognitive profile affects their perception of the two stimuli

analysed. First, they analysed whether recognition of the stimuli depends on the individual’s cognitive profile.

For this purpose, the authors used a regression analysis. Table 2 shows the results of this analysis.

Table 2: Results of regression analysis

AFF-A AFF-B

R2

Adj. R2

F (Sig.)

0.072

0.042

2.39 (0.074)

0.088

0.059

2.97 (0.036)

Constant

COG

TA

PRO

1.14 (0.000)

-0.33 (0.010)

-0.022 (0.9)

-0.063 (0.74)

0.97 (0.00)

0.005 (0.977)

0.398 (0.08)

-0.65 (0.010)

Mean

SD

Wilcoxon signed rank test

Signed test

0.843

0.179

Z: -4.39a (0.00)

Z: -4.18a (0.00)

0.715

0.233

a: (AFF-B)-(AFF-A)

The results show that the level of recognition of stimuli is significantly higher for Stimulus A than for

Stimulus B. As might have been suspected, Stimulus B poses greater problems of visibility than Stimulus A, a

sure consequence of its higher ambiguity. The authors next analysed the explanatory power of each of the three

dimensions included in the cognitive profile. Some interesting points come out of the results. On the one hand,

the regression analyses carried out are statistically significant, thereby confirming that the recognition of the

stimuli analysed depends on the receiver’s cognitive profile. On the other hand, the results show considerable

differences in the explanatory power and in which dimensions of the cognitive profile are influential.

Specifically, the F statistic and the adjusted R2 indicate that the cognitive variables have greater explanatory

power in Stimulus B. In other words, the cognitive profile has an effect on the recognition of stimuli, above all

when the stimuli are relatively unclear.

With regard to the cognitive dimensions that are influential, the recognition of Stimulus A depends on the

cognitive style. Specifically, the results show that the greater the analytical orientation, the lower the level of

recognition of Stimulus A. With Stimulus B, the most significant dimensions are tolerance of ambiguity and

proactivity. Tolerance of ambiguity shows a positive effect on recognition of the stimulus, as would be

expected, while proactivity has a negative effect. This latter result means that proactive individuals may

recognise fewer stimuli as relevant because they anticipate that they are not going to be affected by them. It

should be recalled that proactivity has connotations of anticipation.

Having analysed the stimulus recognition stage, the authors then went on to the interpretation stage, in

other words, the stage in which the participants attribute meaning to the event. The authors analysed the

variables FAV-A and FAV-B for this purpose. The descriptive statistics of these variables show that the mean is

higher for Stimulus A than for Stimulus B (1.8 compared to 1.52). This means that participants recognise a

favourable effect more frequently for Stimulus A than for Stimulus B. The variability of the indicator is clearly

higher in the second case (Sd-A: 0.24; Sd-B: 0.29).

The authors then looked at the dependence relation between the stimulus interpretation variables and the

individual cognitive profile. They decided to make some regroupings. In the case of Stimulus A, they grouped

the sample into two categories: (a) the participants who always say that Stimulus A has a positive effect7; and

(b) the participants who say that the effect of the stimulus varies (sometimes positive, sometimes negative). The

ANOVA analysis shows that none of the cognitive profile dimensions can explain the differences evident in the

interpretation of Stimulus A (Table 3). Consequently, the authors observe that for Stimulus A the attribution of

meaning does not depend on the cognitive profile.

The process was similar for the interpretation process with Stimulus B. In this case the authors grouped the

sample into three categories8: (a) participants recognising a varied influence, but with the negative effect being

in the majority; (b) participants giving an equal number of positive and negative scores; and (c) participants who

7 Recall that Stimulus A refers to the TOB announcement and its effects for the target firm. In this case the

announcement, if credible, leads to a rise in the firm’s share price. 8 Stimulus B refers to the effect of the TOB announcement for the bidding firm, and its effects are more

uncertain.


8

recognised mainly positive effects. Table 3 shows the results of the ANOVA analysis with the cognitive profile

dimensions. The results show that the cognitive style has a significant effect on the meaning that participants

attribute to Stimulus B. Specifically, intuitive individuals attribute a positive effect to Stimulus B more

frequently than analytical individuals.

Table 3: Results of ANOVA

COG TA PRO

STIMULUS A F: 1.855 (0.176) F: 0.204 (0.653) F: 0.060 (0.808)

STIMULUS B

Levene statistic (Sig.)

F (Sig.)

1.2 (0.29)

3.22 (0.04)

1.13 (0.32)

0.92 (0.40)

1.64 (0.20)

0.29 (0.74)

0: 34 participants

1: 20 participants

2: 42 participants

Total: 96 participants

0: (0.74)

1: (0.74)

2: (0.67)

total: (0.71)

0: (0.61)

1: (0.62)

2: (0.59)

total: (0.60)

0: (0.76)

1: (0.77)

2: (0.75)

total: (0.76)

Finally, the authors analysed how the perception process develops with respect to the cognitive dimensions.

In other words, they examined whether differences exist in the perception process between intuitive and

analytical individuals, and whether the level of proactivity and tolerance of ambiguity have an effect on the

perception process. For this purpose, they split the sample into two groups depending on how subjects score on

COG: low (intuitive) or high (analytical). Then a difference of means analysis (ANOVA) was used to determine

if this grouping was associated with the differences evident in the perception process of the different participants

(AFF-A; AFF-B; FAV-A; FAV-B). The process was identical with the other two dimensions: tolerance of

ambiguity and proactivity. Table 4 shows the results of this analysis.

Table 4: Results of ANOVA

COG:

INTUITIVE/ANALYTICAL

TA:

- TA / + TA

PRO:

- PRO / + PRO

AFF-A 1: 0.89

2: 0.79

TOTAL: 0.84

LS: 6.73 (0.011)

F: 8.71 (0.004)

1: 0.83

2: 0.85

TOTAL: 0.84

LS: 0.26 (0.609)

F: 0.27 (0.602)

1: 0.833

2: 0.85

TOTAL: 0.84

LS: 0.004 (0.952)

F: 0.351 (0.555)

AFF-B 1: 0.70

2: 0.73

TOTAL: 0.71

LS: 0.00 (0.956)

F: 0.39 (0.53)

1: 0.68

2: 0.76

TOTAL: 0.71

LS: 0.42 (0.51)

F: 2.88 (0.093)

1: 0.75

2: 0.674

TOTAL: 0.715

LS: 7.52 (0.007)

F: 2.83 (0.096)

FAV-A 1: 1.77

2: 1.84

TOTAL: 1.80

LS: 0.22 (0.637)

F: 2.12 (0.14)

1: 1.80

2: 1.81

TOTAL: 1.80

LS: 0.103 (0.909)

F: 0.043 (0.83)

1: 0.183

2: 1.78

TOTAL: 1.80

LS: 3.54 (0.063)

F: 0.854 (0.358)

FAV-B 1: 1.59

2: 1.44

TOTAL: 1.52

LS: 1.21 (0.27)

F: 6.14 (0.015)

1: 1.52

2: 1.50

TOTAL: 1.52

LS: 1.21 (0.27)

F: 0.109 (0.74)

1: 1.51

2: 1.52

TOTAL: 1.52

LS: 0.66 (0.417)

F: 0.018 (0.89)

LS: Levene statistic of difference of variances (Sig.).

F: F statistic of difference of means test (Sig.).

The grouping with the variable COG gives rise to 2 groups (intuitive versus analytical individuals). The

results show that significant differences exist between the two groups in the recognition of Stimulus A (AFF-A),

in the difference in the level of recognition of the two stimuli considered (DIFF), and in the interpretation of

Stimulus B (FAV-B). The results show that analytical individuals recognise Stimulus A less than intuitive ones.

Finally, analytical participants recognise the favourable effect of Stimulus B less frequently than intuitive

individuals.

The grouping for tolerance of ambiguity is associated with significant differences only in the behaviour of

the variable that measures the recognition of Stimulus B. Specifically, and consistent with the above arguments,

the results show that participants with a higher TA recognise unclear stimuli as relevant signals more frequently.


9

Finally, the analysis shows that the level of proactivity is associated with significant differences in AFF-B.

Proactive participants recognise Stimulus B less frequently than non-proactive ones.

5 DISCUSSION

A number of interesting points can be made from the results of this analysis. First, and most importantly,

the results show that the cognitive profile is a determinant in the process of individual perception. Specifically,

in the experiment carried out here the cognitive dimensions are determinant mainly in the initial stage of the

perception process, in which the individual classifies the stimulus as either a signal (relevant stimulus) or noise

(irrelevant stimulus). These results confirm – at least partially – Hypothesis H1. On the other hand, the cognitive

profile has greater explanatory power in the perception process for Stimulus B than for Stimulus A. In the

recognition stage this difference is clearer when comparing the goodness of fit of the estimated regression

models. In other words, the results show that the cognitive profile is more relevant in explaining the recognition

stage of the stimuli, especially when the stimuli are relatively unstructured, and so unclear. This provides

support for Hypothesis H2, which postulates that the cognitive profile has a stronger effect given relatively

unclear stimuli (low strength and low structure).

Second, the results also indicate that recognition of Stimulus A is associated with the cognitive style. Thus,

intuitive individuals recognise this stimulus more frequently than analytical ones. As could be expected, the

attribution of meaning to Stimulus A is not associated with the participants’ cognitive qualities. It will be

recalled that Stimulus A refers to the effects on the share price of the target firm, and the financial literature

indicates that the likely effects are positive.

And third, recognition of the least clear stimulus (Stimulus B) is related to greater tolerance of ambiguity

and individuals with a low proactivity. Stimulus B’s interpretation stage is related to the participants’ cognitive

style, so that intuitive individuals recognise a favourable effect more frequently than analytical ones. This result

is also consistent with the situation. Stimulus B has some uncertain effects that depend on the bidding firm’s

motives in making its bid. The participants in the experiment have no information about these motives, so any

attribution of meaning can only be based on intuition.

With regard to H3, the results diverge slightly from what was expected. The authors postulated that

cognitive style would have more effect on the perception of Stimulus B, the less well-structured stimulus, than

on that of Stimulus A. The results from the interpretation stage confirmed this hypothesis because cognitive

style helps explain the interpretation of Stimulus B but not that of Stimulus A. But in the recognition stage the

relation is not confirmed because cognitive style explains the recognition of Stimulus A but not that of Stimulus

B. The effect of tolerance of ambiguity, as predicted, contributes to recognising relatively unstructured stimuli

(Stimulus B in the current analysis). Likewise, the previous arguments led the authors to postulate that

proactivity would surely have an active role in the perception of unclear stimuli. The results support this

argument and show that proactive individuals recognise this stimulus as relevant less frequently. Considering

the anticipatory nature of proactivity, this result shows that the most proactive participants anticipate that they

will not be affected by Stimulus B and so ignore it, in other words they classify the stimulus as irrelevant.

In short, the authors show that the cognitive dimensions have a significant effect on the process of the

perception of stimuli, above all in the initial recognition stage and in the presence of relatively unstructured

stimuli. Thus, these results give support to the idea that the perception process is dependent on the manager’s

cognitive traits and that the type of stimulus conditions the relevance of the cognitive dimensions on the

perception process.

6 FINAL REMARKS

The perception of stimuli can be seen as the “black box” that hides the secrets of the managerial decision-

making that marks out the path for firms. The process of managerial perception determines which stimuli

deserve a strategic response and attributes their meaning. Previous research has shown that the individual’s

characteristics condition this perception process. Many authors have studied the effect of the managers’

demographic characteristics, but it is their cognitive traits that are determinant in the process of recognising and

interpreting stimuli. Moreover, the incoming stimuli differ in nature, hence the interest in studying whether the

effect of the cognitive profile varies in the presence of different stimuli. Thus, in the current research the authors

have analysed the effect of the cognitive profile on the process of perception of differently-structured stimuli.

The stimulus was the share price of a firm involved in the announcement of a takeover bid. The cognitive profile

was measured using three dimensions: cognitive style, tolerance of ambiguity, and proactivity.

The results obtained strongly support the idea that the cognitive profile analysed here does have an effect

on the perception of stimuli. Specifically, the cognitive dimensions considered are determinant in the initial

stage of the recognition of stimuli and especially when the stimuli are relatively unstructured. But the results

also show that these cognitive dimensions have a weaker effect in the interpretation stage. With all this, the

results of this research point to the need to study the managerial perception process further in two directions.


10

Thus researchers need to refine their operationalization of the individual cognitive dimensions to analyse their

influence on the different stages of the perception process in more detail. At the same time, they should examine

the contextual features that would allow them to characterise and differentiate different stimuli to learn more

about the different associated perception processes. Finally, we recognise that interdisciplinary research is likely

to provide novel insight into exploring and conceptualizing organizational behaviour (Gillingwater and

Gillingwater, 2009). All this will mark the direction of future research.

Appendix 1

TYPE-A STIMULUS: TAKEOVER ANNOUNCEMENT – THE TARGET COMPANY

TYPE-B STIMULUS: TAKEOVER ANNOUNCEMENT – THE BID COMPANY

Appendix 2

VARIABLES

Stage I: Recognition of the stimuli

AFF-A= the proportion of stimuli-A classed as relevant:

The number of times the individual says that stimulus-A affects the share price in the six scenarios

presented in the experiment.

AFF-B= the proportion of stimuli-B classed as relevant

The number of times the individual says that stimulus-B affects the share price in the six scenarios

presented in the experiment.

Stage II: Interpretation of the stimuli:

FAV-A = the proportion of stimuli considered to have a favourable effect on the price of the target

company.

The ratio of the number of times a favourable effect is recognised (the share price rises) to the number of

times stimulus-A is considered to have an effect.

FAV-B = the proportion of stimuli considered to have a favourable effect on the price of the bid company.

The ratio of the number of times a favourable effect is recognised (the share price rises) to the number of

times stimulus-B is considered to have an effect.

Cognitive Dimensions:

Cognitive Style: CSI

Tolerance of ambiguity: TA

Proactivity: PRO

Stage I:

Does a takeover

announcement affect

the market price of the

target company?

Stage II:

What is the reaction of the

target company stock

market price?

YES

NO

Favourable

(the share prices rise)

Unfavourable

(the share prices have fallen)

Stage I:

Does a takeover

announcement affect the

market price of the bid

company?

Stage II:

What is the reaction of the

bid company stock market

price?

YES

NO

Favourable

(the share prices rise)

Unfavourable

(the share prices have fallen)


11

REFERENCES

Acedo FJ, Jones MV (2007): Speed of internationalization and entrepreneurial cognition: Insights and a

comparison between international new ventures, exporters and domestic firms. Journal of World Business,

42: 236-252.

Aguinis, H.; Lawal, S.O. (2012): Conducting field experiments using eLancing’s natural environment. Journal

of Business Venturing, 27 (4): 493-505.

Allinson CW, Chell E, Hayes J. (2000): Intuition and Entrepreneurial Behaviour. European Journal of Work

and Organizational Psychology, 9 (1): 31-43.

Allinson CW, Hayes J. (1996): The cognitive style index: a measure of intuition-analysis for organizational

research. Journal of Management Studies, 33 (1): 119-135.

Armstrong, S.J. (2000): The influence of individual cognitive style on performance in management education.

Education Psychology, 12 (3): 323-339.

Baker M, Pan X, Wurgler J. (2009): The psychology of pricing in mergers and acquisitions. Working papers,

Electronic copy available at: http://ssrn.com/abstract=1364152.

Baron RA. (1998): Cognitive mechanisms in entrepreneurship: why and when entrepreneurs think differently

than other people. Journal of Business Venturing, 13 (4): 275-294.

Baron, R.A. (2004): The cognitive perspective: a valuable tool for answering entrepreneurship's basic “why”

questions. Journal of Business Venturing, 19 (2): 221-239.

Baron, R.A. (2006): Opportunity recognition as pattern recognition: how entrepreneurs “connect the dots” to

identify new business opportunities. Academy of Management Perspectives, 20 (1): 104–119.

Baron, RA., Hmieleski, KM; Henry, R.A. (2012): Entrepreneurs’ dispositional positive affect: The potential

benefits – and potential costs- of behing up. Journal of Business Venturing, 27 (3): 310-324.

Barreto, I. (2012): Solving the entrepreneurial puzzle: the role of entrepreneurial interpretation in opportunity

formation and related process. Journal of Management Studies, 49 (2): 356-380.

Bateman TS, Crant JM (1993): The Proactive Component of Organizational Behavior: A Measure and

Correlates. Journal of Organizational Behavior, 14 (2): 103-118.

Blume, B.D; Covin, J.G. (2011): Attributions to intuition in the venture founding process: Do entrepreneurs

actually use intuition or just say that they do? Journal of Business Venturing, 26: 137-151.

Budner S. (1962): Intolerance of ambiguity as a personality variable. Journal of Personality, 30: 29-50.

Calori R, Johnson G, Sarnin P. (1994): CEOs cognitive maps and the scope of the organization. Strategic

Management Journal, 15: 437-457.

Camerer CF, Hogarth RM. (1999): The effects of financial incentives in experiments: a review and capital-

labor-production framework. Journal of Risk and Uncertainty: 7-42.

Chattopadhyay P, Glick WH, Miller CC, GP Huber (1999): ‘Determinants of executive beliefs: comparing

functional conditioning and social influence’, Strategic Management Journal, 20 (8): 763-789.

Cools E, Van den Breeck H. (2008): The Hunt for the Heffalump Continues: Can trait and cognitive

characteristics predict entrepreneurial orientation? Journal of Business Strategy, 18 (2): 23-41.

Conchar MP, Zinkhan GM, Peters C, Olavarrieta S. (2004): An integrated framework for the conceptualization

of consumers’ perceived-risk processing. Journal of Academy of Marketing Science, 32 (4): 418-436.

Crant, JM (2000): Proactive Behavior in Organizations. Journal of Management, 26 (3): 435-462.

Daft RL, Weick KE. (1984): Toward a model of organizations as interpretation systems. Academy of

Management Review, 9 (2): 284-295.

Denison DR, Dutton JE, Kahn JA, Hart SL. (1996): Organizational context and the interpretation of strategic

issues: A note on CEOs’ interpretations of foreign investment. Journal of Management Studies, 33 (4):

453-474.

Dubard S, Gerhardt MW, Richard J. (2007): The Role of Cognitive Style and Risk Preference on

Entrepreneurial, Journal of Leadership and Organizational Studies, 13 (4): 86-104.

Dutta, D.K.; Thornhill, S. (2008): The evolution of growth intentions: Toward a cognition-based model. Journal

of Business Venturing, 23 (3): 307-332.

Dutton JE, Ashford SJ. (1993): Selling issues to top management. The Academy of Management Review, 18 (3):

397-428.

Dutton JE, Jackson SE. (1987): Categorizing strategic issues: links to organizational action. Academy of


http://www.sciencedirect.com/science/article/pii/S0883902603000089




12

Eggers, J.P., Kaplan, S, (2008): Cognition and renewal: comparing CEO and organizational effects on

incumbent adaptation to technical change. Organization Science, 20 (2): 461–477.

Entrialgo M, Fernández E, Vázquez CJ. (2001): Los efectos de la turbulencia del entorno y del grado de

discrecionalidad percibido en el comportamiento del emprendedor. Revista Europea de Dirección y

Economía de la Empresa, 10 (1): 21-36.

Ericson T. (2001): Sensemaking in organizations – towards a conceptual framework for understanding strategic

change, Scandinavian Journal of Management, 17: 109-131.

Gielnik, M.M., Frese, M., Graf, J.M.; Kampschulte, A. (2011): Creativity in the opportunity identification

process and the moderating effect of diversity of information. Journal of Business Venturing, in press.

Green, K.M.; Covin, J.G.; Slevin, D.P. (2008): Exploring the relationship between strategic reactiveness and

entrepreneurial orientation: The role of structure–style fit. Journal of Business Venturing, 23: 356-383.

Farinós JE, Fernández M. (1999): La incidencia de una OPA sobre la actividad negociadora y la estimación del

riesgo sistemático de las empresas objetivo, Revista Española de Financiación y Contabilidad, 100: 381-

401.

Fiske ST, Taylor SE. (1991): Social cognition. McGraw Hill: New York.

Furnham A, Ribchester T. (1995): Tolerance of Ambiguity: A review of the concept, its measurement and

applications. Current Psychology, 14 (3): 179-200.

Gillingwater, D., Gillingwater, T.H. (2009): A neuroanatomical approach to exploring organizational

performance. International Journal of Business Science and Applied Management, 4 (3): 34-52.

Ginsberg A, Venkatraman N. (1992): Investing in new information technology: The role of competitive posture

and issue diagnosis. Strategic Management Journal, 13: 37-53.

Gioia D, Poole P.(1984), Scripts in Organizational Behavior, Academy Management Review, 9: 449-459.

Grant AM, Ashford SJ. (2008): The dynamics of proactivity at work. Research in Organizational Behavior, 28:

3-34

Hambrick DC, Mason PA. (1984): Upper Echelons: The organization as a reflection of its top managers,

Academy of Management Review, 9 (2): 193-206.

Hayes J, Allinson CW. (1994): Cognitive Style and its relevance for management practice. British Journal of

Management, 5: 53-71.

Hayes J, Allinson CW. (1998): Cognitive style and the theory and practice of individual and collective learning

in organization. Human Relations, 51 (7): 847-871.

Hodgkinson G.P. (1997): Cognitive inertia in a turbulent market: A comparative causal mapping study, Journal

of Management Studies, 34: 921 - 946.

Hodgkinson GP, Sadler-Smith E. (2003): Complex or Unitary? A Critique and Empirical Re-assessment of the

Allinson-Hayes Cognitive Style Index. Journal of Occupational and Organizational Psychology, 76 (2):

243-268.

Haukedal W. (1994): Categories of strategic stimuli: their implications for managers’ sense making of

organizational environments. Scandinavian Journal of Management, 10 (3): 267-279.

Haukedal W, Gronhaug K. (1994): Context-Specific Rationality in sense-making of strategic stimuli.

Scandinavian Journal of Management, 10 (4): 355-367.

Kaplan, S. (2008): Framing contests: strategy making under uncertainty. Organization Science, 19 (5): 729-752.

Kaufmann G. (1987): Reflections on the concept of a problem. In Bjorgen, I. (Ed.), Fundamental Problems in

Psychology (Bergen: Sigma, 1987).

Kautonen, T. (2008): Understanding the older entrepreneur: Comparing third age and prime age entrepreneurs in

Finland. Journal of Business Science and Applied Management, 3 (3): 3-13.

Kickul J, Gundry LK (2002): Prospecting for Strategic Advantage: The proactive Entrepreneurial Personality

and Small Firm Innovation, Journal of Small Business Management, 40 (2): 85-97.

Kozhevnikov M. (2007): Cognitive Styles in the Context of Modern Psychology: Toward and Integrated

Framework of Cognitive Style. Psychological Bulletin, 133 (3): 464-481.

Kuvaas B, Kaufman, (2004): Individual and organizational antecedents to strategic-issue interpretation.

Scandinavian Journal of Management, 20: 245-275.

Ling Y, Floyd SW, Baldridge DC. (2005): Toward a model of issue-selling by subsidiary managers in

multinational organizations. Journal of International Business Studies, 36: pp. 637-654.




13

Livengood, RS., Reger, RK (2010): That’s our turf! Identity domains and competitive dynamics. Academy of


McMullen, J.S., Shepherd, D.A., (2006): Entrepreneurial action and the role of uncertainty in the theory of the

entrepreneur. Academy of Management Review, 31 (1): 132–152.

McNally RC, Durmusoglu SS, Calantone RJ, Harmancioglu N. (2009): Exploring new product portfolio

management decisions: The role of managers’ dispositional traits. Industrial Marketing Management, 38:

127-143.

Miller, D. (1996): “Configurations revisited”. Strategic Management Journal 17: 502–512.

Miller KD. (1993): Industry and country effects on managers’ perception of environmental uncertainties.

Journal of International Business Studies, 24 (4): 693-714.

Milliken FJ, Lant TK. (1991): The impact of an organization’s recent performance history on strategic

persistence and change: The role of managerial interpretations. In Advances in Strategic Management,

Dutton J, Huff A, Shrivastava P (eds). JAI Press: Greenwich, CT; 7: 129-156.

Mintzberg H, Raisinghani D, Théorét A. (1976): The structure of unstructured decision process. Administrative

Science Quarterly, 21: 246-275.

Mitchell M, Pulvino T, Stafford E. (2004): Price pressure around mergers. Journal of Finance, 59 (1): 31-63.

Nicolaou, N.; Shane, S. (2009): Can genetic factors influence the likelihood of engaging in entrepreneurial

activity? Journal of Business Venturing, 24 (1): 1-22.

Nadkarni S, Barr P. (2008): Environmental context, managerial cognition, and strategic action: An integrated

view. Strategic Management Journal, 29: 1395-1427.

Phan, P.H., Wright, M., Ucbasaran, D., Tan, W.L., (2009): Corporate entrepreneurship: current research and

future directions. Journal of Business Venturing, 24 (3): 197–205.

Plambeck, N. (2011): “The development of new products: The role of firm context and managerial cognition”.

Journal of Business Venturing, in press.

Raghubir P, Das SR. (1999): A Case for Theory-Driven Experimental Enquiry. Financial Analysts Journal, 55

(6): 56-79.

Ramaprasad A, Mitroff II (1984): On formulating strategic problems. Academy of Management Review, 9: 597-

605.

Samuelson W, Zeckhauser RJ. (1988): Status quo bias in decision making. Journal of Risk and Uncertainty, 1

(1): 7-59.

Sadler-Smith, E. (1999): Intuition-analysis cognitive style and learning preferences of business and management

studies. Journal of Managerial Psychology, 14 (1): 26-38.

Sadler-Smith E, Badger B. (1998): Cognitive Style, Learning and Innovation. Technology Analysis and

Strategic Management, 10 (2): 247-265.

Sadler-Smith E, Spicer DP, Tsang F. (2000): Validity of the Cognitive Style Index: replication and extension.

British Journal of Management, 11: 175-181.

Sáez, Martínez FJ, Jiménez Moreno JJ. (2004): Una revisión de las aportaciones de la escuela cognitiva a la

dirección estratégica. Revista Europea de Dirección y Economía de la Empresa, 13 (2): 43-60.

Santos MV, García MT. (2006): Managers’ opinions: reality or fiction: a narrative approach. Management

Decision, 44 (6): 752-770.

Santos MV, García MT. (2008): Environmental uncertainty: the sides object of perception. Journal of

Administrative and Social Sciences, 12 (32): 65-74.

Schneider, S.C., DeMeyer, A. (1991): Interpreting and responding to strategic issues: The impact of national

culture. Strategic Management Journal, 12: 307-320.

Schwenk CR. (1984): Cognitive simplification process in strategic decision making. Strategic Management

Journal, 5 (2): 111-128.

Schwenk CR. (1995): Strategic decision-making. Journal of Management, 21 (3): 471-494.

Shane, S., (2003): A General Theory of Entrepreneurship: The Individual-Opportunity Nexus. Edward Elgar

Publishing, Cheltenham.

Shepherd, D.A., DeTienne, D.R., (2005). The impact of prior knowledge and financial reward on the

identification of opportunities. Entrepreneurship Theory and Practice 29 (1): 91–112.

Sherman, R. (1974): The psychological difference between ambiguity and risk. The Quarterly Journal of

Economics, 88: 166-169.


14

Shrout, P.E., Bolger, N. (2002): Mediation in experimental and nonexperimental studies: new procedures and

recommendations. Psychological Methods, 7 (4): 422–445.

Simon, H.A. (1973): The structure of ill-structured problems. Artificial Intelligence, 4: 181-201.

Singh, K. (2010): Developing human capital by linking emotional intelligence with personal competencies in

Indian business organizations. Journal of Business Science and Applied Management, 5 (2): 29-42.

Starbuck W, Milliken F. (1988): Executive perceptual filters: What they notice and how they make sense. In The

executive effect: Concepts and methods for studying top managers, Hambrick D (ed). JAI Press:

Greenwich, CT: 35-65.

Sterman, J. (1989): Modeling managerial behaviour: misperceptions of feedback in a dynamic decision making

experiment, Management Science, 35 (3): 321-339.

Sully de Luque MF, Sommer SM. (2000): The impact of culture on feedback-seeking behavior: an integrated

model and propositions. Academy of Management Review, 25 (4): 829-849.

Sutcliffe, K.M. (1994): What executives notice: Accurate perceptions in top management teams. Academy of

Management Journal, 37: 1360-1378.

Sutcliffe K, Huber GP. (1998): Firm and industry as determinants of executive perceptions of the environment.

Strategic Management Journal, 19: 793-807.

Swets, J. A. (1992): The science of choosing the right decision threshold in high-stakes diagnostics. American

Psychologist, 47: 522-532.

Tang, J.; Kacmar, K.M.; Busenitz, L. (2012): Entrepreneurial alertness in the pursuit of new opportunities.

Journal of Business Venturing, 27 (1): 77-94.

Thomas, J.B.; Clark, S.M.; Gioia, D.A. (1993): Strategic sensemaking and organizational performance:

Linkages among scanning, interpretation, action, and outcomes. Academy of Management Journal, 36:

239-270.

Thomas, JB; McDaniel, R.R. (1990): Interpreting strategic issues: Effects of strategy and the information-

processing structure of top maangement teams. Academy of Management Journal, 33: 286-306.

Thomas, J.B.; Shakster, L.J.; Mathieu, J.E. (1994): Antecedents to organizationl issue interpretation: The roles

of single-level, cross-level, and content cues. Academy of Management Journal, 37: 1252-1284.

Thomas, A., Simerly, R. (1994): The chief executive officer and corporate social performance: an

interdisciplinary examination. Journal of Business Ethics, 13 (12): 959-968.

Ucbasaran, D., Westhead, P., Wright, M., (2008): Opportunity identification and pursuit: does an entrepreneur's

human capital matter? Small Business Economics 30 (2): 153–173.

Ucbasaran, D., Westhead, P.; Wright, M. (2009): The extent and nature of opportunity identification by

experienced entrepreneurs. Journal of Business Venturing, 24: 99-115.

Ucbasaran, D., Westhead, P., Wright, M.; Flores, M. (2010): The nature of entrepreneurial experience, business

failure and comparative optimism. Journal of Business Venturing, 25: 541-555.

Waller, M.J., G.P. Huber; W.H. Glick. (1995): Functional Background as a Determinant of Executive Selective

Perception. The Academy of Management Journal 38 (4): 943-974.

Walsh, J.P. (1995): Managerial and organizational cognition: Notes from a trip down memory lane.

Organizational Science, 6 (3): 280-321.

Weick, K.E. (1979): The Social Psychology of Organizing, 2nd

Edition, Addison-Wesley, Menlo Park, CA.

Westerberg, M.; Sing, J.; Hackner, E. (1997): Does the CEO matter? An empirical study of small Swedish firms

operating in turbulent environments. Scandinavian Journal of Management, 13 (3): 251-270.

Wood, DJ. (1991): Corporate social performance revisited Academy of Management Review, 16 (4): 691-718.

Yanes Estévez, V. (2004): La aproximación cognoscitiva a la realidad empresarial y la aportación de Herbert

Simon. . Revista Europea de Dirección y Economía de la Empresa, 13 (2): 25-42.

Zalesny, M.D., Ford, J.K. (1990): Extending the social information processing perspective: New links to

attitudes, behaviors, and perceptions. Organizational Behavior and Human Decision Processes, 47: 205-

246.


(In)justice contexts and work satisfaction: The mediating role of

justice perceptions

Maria Rita Silva

Business Research Unit of ISCTE – Lisbon University Institute

Av. das Forças Armadas, Edifício ISCTE, 1649-026 Lisboa, Portugal

Telephone: +351 217903001


António Caetano

Business Research Unit of ISCTE – Lisbon University Institute

Av. das Forças Armadas, Edifício ISCTE, 1649-026 Lisboa, Portugal

Telephone: +351 217903001


Qin Zhou

The University of York, The York Management School

University of York Freboys Lane Heslington York YO10 5GD, United Kingdom

Telephone: +1904 325008


Abstract

This study explores the impact of the social context, namely (in)justice climate and target, in workers' justice

perceptions and satisfaction. Individual's justice judgments are expected to mediate the relationship of (in)justice

climate and target with work satisfaction. We found mediation effects of procedural justice in the relationship

between justice climate and satisfaction, and interactional justice in the relationship between injustice target and

satisfaction. Distributive justice does not affect the relationship between the (in)justice context and satisfaction.

Findings demonstrate the relevance of framing organizational justice in a socially contextualized perspective

since they seem to influence individual justice reactions and work attitudes. Using an experimental

methodology, it was possible to explore the role of seldom studied contextual variables.

Keywords: Organizational justice, social context, justice climate, justice target, work satisfaction, mediation

Acknowledgments: Part of this work was supported by the Foundation for Science and Technology, Portugal

[PhD Grant number SFR/BD/61417/2009] awarded to Maria Rita Silva and [PEst-OE/EGE/UI0315/2011] the

strategic project of Instituto Universitário de Lisboa (ISCTE-IUL)


16

1 INTRODUCTION

Organizational justice is an important domain in the study of organizations (e.g., Ambrose & Schminke,

2002; Chia, Foo, & Fang, 2006). In organizations, justice is an important feature of outcome distribution (i.e.,

distributive justice), decision making processes (i.e., procedural justice) and daily personal interactions (i.e.,

interactional justice). Justice judgments are related to several aspects of workers’ attitudes that are reflected in

their performance (e.g., Cohen-Charash & Spector, 2001; Colquitt, Conlon, Wesson, Porter, & Ng, 2001; Lam,

Schaubroeck, & Aryee, 2002; Nowakowski & Conlon, 2005). Consequently, it is important to understand the

mechanisms through which justice perceptions develop and are reflected in workers’ behaviours and attitudes.

Many authors (e.g., Inness, Barling & Turner, 2005) have stated the relevance of the social context to

understand workers’ justice-related reactions. Well-established theories, such as the relational (e.g., Lind &

Tyler, 1988) or the group engagement model (e.g., Blader, & Tayler, 2009) have suggested that fairness

appraisals are socially constructed, yet most of the studies that examine the influence of justice perceptions on

workers’ attitudes and behaviour focus on hierarchical decision-making or resource allocation situations

(Fassina, Jones, & Uggerslev, 2008). These studies adopt an “exchange perspective” of justice (Bies, 2005).

Specifically, this approach emphasizes an individualistic view of justice, focused on the transactions between an

agent, the one who acts fairly or unfairly, and a recipient or justice target, the one who makes the fairness

appraisal.

However, a more socially contextualized “encounter perspective” of justice (Bies, 2005) has been

neglected. Justice is conceptualized as a set of socially construed values that guarantee that self-interest and

opportunism do not interfere with the overall social fabric of societies and organizations (Folger & Cropanzano,

1998).

One way to bridge the gap between the conceptualization of justice as a social phenomenon and the highly

individualistic study of justice perspectives is to explore the impact of the immediate social context of justice

events. One aspect of the immediate social context that has been often overlooked in organizational justice

literature is the relationship between co-workers. Namely, how work colleagues’ perceptions and experiences

help shape individuals’ reactions to fairness.

The importance of lateral relationships – relationships with other individuals situated in the same stratum of

an organizational hierarchy and with whom one executes tasks and has routine interactions has increased as

flatter organizational structures and team-based work have become widespread. Indeed, in most jobs, workers

interact more frequently with co-workers than with their supervisors (Chiaburu and Harrison, 2008). Work

colleagues often comment on their experiences and perceptions regarding workplace justice, and so share a

sense-making activity that may shape other workers’ interpretation of organizational events and their subsequent

attitudes and reactions. Therefore, co-workers may be at least as important as other organizational agents, in

shaping individuals’ justice perceptions (e.g., Hung, Chi, & Lu, 2009).

This study intends to explore the way co-workers perceptions and experiences influence the development

of justice perceptions, and their effect on workers’ satisfaction. Specifically, we will examine the influence of

two social context variables: (1) Justice climate, which refers to the degree of justice a group considers it is

receiving; (2) Justice target, which refers to the direction of the (in)justice event, which can target the self or a

co-worker. In addition, it is expected that these justice related social factors will impact employees’ work

satisfaction through their influence on individuals’ justice appraisals. In other words, we investigate justice

perceptions as a potential mechanism linking the social context and employees’ work attitudes.

This study aims to contribute to the literature in two ways. First, this study highlights the relevance of the

social context to the development of individual perceptions of workplace justice and wok satisfaction. It

examines the impact of seldom studied aspects of the (in)justice context, namely injustice target and justice

climate on individual justice judgments, and it links them to worker’s satisfaction. Second, by employing an

experimental methodology, this study contributes to the clarification of the causality directions of these

relationships. Because most management research on organizational justice is cross-sectional there may be

doubts regarding how social context and justice judgments interact to influence work attitudes. Some authors

have considered justice climate as a moderator of the impact of individual justice on workers attitudes (Liao &

Rupp, 2005). We believe that justice climate and target have a direct impact on the development of justice

judgments, and that justice judgments mediate their impact on workers attitudes. Applying an experimental

approach can help to clarify this process as it more accurately illustrates causality.

Maria Rita Silva, António Caetano and Qin Zhou

17

2 LITETERATURE REVIEW

Injustice target as an antecedent of justice judgments

People’s perception of the world is built, not only through direct experience but also by social learning,

through the experience of others (e.g., Carroll & Bandura, 1987). A recent meta-analysis by Chiaburu and

Harrison (2008) has demonstrated that co-workers’ actions predict perceptual, attitudinal, and behavioural

outcomes of their colleagues, even when the influence of the direct leaders is accounted for. According to the

authors, co-workers affect how their colleagues perceive and shape their work roles, and how they form, retain,

and access work related attitudes (Chiaburu & Harrison, 2008).

The study of the impact of other person’s justice experiences on one’s own justice perception has focused

primarily on the procedural justice dimension. In two experimental studies, (Van Den Bos & Lind, 2001) show

that the treatment received by others can sometimes be such as potent consideration in procedural justice

judgments as is one’s own treatment. Colquitt (2004) has explored this effect in a field and in an experimental

study in team settings. The author found that individual members’ own procedural justice perceptions interacted

with other members procedural justice evaluations, such that higher levels of role performance occurred when

justice judgments were consistent within the team. (De Cremer & Van Hiel, 2006) based on the results from a

scenario experiment, a cross-sectional survey, and a laboratory experiment, have proposed that how fairly

another is treated influences one’s emotional and behavioural reactions and that the way other colleagues are

traded can be considered important organizational information in shaping one’s own feelings and actions.

Some research on the effects of other people’s justice experiences has been based on the deontonic

perspective, an aspect of fairness theory (e.g., Cropanzano, Goldman & Folger, 2003, Beugré, 2010). This

perspective focuses on the observer’s inherent tendency to notice and react to perceived violations of justice.

Some research has shown that third party observers tend to experience negative emotional responses, and even

to engage in reciprocating behaviours against the perpetrator of the perceived unfairness (Folger, Cropanzano &

Goldman, 2005). These reactions seem to occur regardless of any connection observers might have with the

victim and even in situations where retaliation results in the loss of personal benefits (Turillo, Folger, Lavelle,

Umprheress & Gee, 2002). As such, one might expect that co-workers’ experiences have a strong effect on

employees’ attitudes.

Despite the deontonic perspective assumptions, some research has demonstrated that there appear to be

fundamental differences between the reactions of individuals towards unfair events that target the self as

opposed to events that target other people. Lind, Kray, and Thompson (1998) have demonstrated that justice

perceptions are more negative when the self is deprived of voice than when another person is deprived. Beehr,

Nair, Gudanowski and Such (2004) have found that people tend to consider promotions based on performance

fair, while promotions based on personal characteristics are considered unfair. However, workers tend to

evaluate a promotion based on personal characteristics as more fair if it targets themselves, rather than a

colleague. More recently, Spencer and Rupp (2009) have shown that an individual’s emotional labour increases

as a result of customer unfairness directed towards themselves as well as their co-workers. However, emotional

labour was significantly higher when the injustice was directed at the self.

Accordingly, although we acknowledge the relevance of co-workers’ experiences to an individual’s justice

perceptions, we conservatively expect that faced with an injustice situation, justice perceptions will be more

negative when the injustice is directed at the self. We suppose that the different evaluations regarding one’s own

injustice experiences and those of co-workers may be due to three different reasons: attribution bias, memory

bias and just-world beliefs.

First, individuals may use different criteria for themselves and for others when evaluating the justice of

events. Recent studies on accountability and multifoci justice suggest that injustice events tend to be attributed

to a specific source (Rupp, McCance, Spencer, Sonntag, 2008). Consequent attitudes and behaviours will be

directed back at the perceived source of injustice. It is probable that, when faced with an injustice directed at the

self, individuals tend to be more inclined to consider the injustice as being caused more by another

organizational agent and less by the person being mistreated (i.e. themselves) than when the injustice is directed

at a co-worker.

Second, research concerning self-enhancing bias has shown that people perceive and remember events in a

way that favours their self-image (e.g., Sedikides, Gaertner, & Vevea, 2005). These distortion processes tend to

occur more in regard to one’s own experiences than in the analysis of another’s experiences.

Finally, according to the just-world theory (Lerner, 1980), individuals tend to believe that people are

rewarded or punished as a direct consequence of good or bad deeds, so people supposable get what they

deserve. This theory was been successfully applied to the organizational context (e.g., Bobocel & Hafer, 2007).

The belief in a just world is a “basic illusion”, useful because it leads individuals to perceive the world as a more

predictable and controllable place (Hafer, Bègue, Choma, & Dempsey, 2005). But numerous studies (e.g., Hafer

et. al., 2005) indicate that the belief in a just world contributes to adverse reactions when another person suffers


18

an injustice, such as avoidance, depreciation or condemnation of the victim, or the minimizing of the injustice

situation.

H1: Organizational justice perceptions will be more negative when the injustice is directed at the self, than

when it is directed to a co-worker.

Justice climate as an antecedent of justice judgments

The justice climate refers to shared cognition concerning the degree of justice the workers’ normally

receive (Mayer, Nishii, Schneider & Goldstein, 2007). The justice climate can be considered an antecedent of

individuals’ justice perceptions. In contrast to justice perceptions, which are related to the individual cognitions

of each employee, the justice climate, like the organizational climate, is located at the inter-individual level.

Repeated reception of input from co-workers leads to reciprocation and fosters climates that encourage the

individual to display more positive or negative attitudes and interpersonal actions (Chiaburu & Harrison, 2008).

Workers may not spontaneously conclude that a certain event is fair or unfair (Lamertz, 2002). On the

contrary, they may experience a period of ambiguity that motivates them to look for social information to help

make sense of situations through social comparison and interpersonal validation of reality. Additionally, they try

to influence, and are in turn influenced by the people with whom they interact (Lamertz, 2002; Meyer, 1994).

Thus, in order to evaluate the impact of an (un)fair event in workers’ justice perceptions and satisfaction one

should take into account the social context in which it occurs.

Previous studies (e.g., Colquitt, Noe, & Jackson, 2002; Liao & Rupp, 2005; Mayer, et.al., 2007) have

addressed specific facets of the justice climate, such as procedural and interactional dimensions. Those studies

revealed that the justice climate predicted justice perceptions and individual and team outcomes above and

beyond the effects of individual-level justice perceptions.

In the present study, distributive, procedural and interactional justice concerns were operationalized so that

the distinct effects of these dimensions on the different facets of justice perceptions could be evaluated.

Recently, there has been growing interest in holistic justice judgments (Lind, 2001). Ambrose and Schminke

(2009) have suggested that the focus on the effects of specific types of justice may not capture the depth and

richness of individuals’ justice experiences. The authors proposed an overall justice scale comprising two

subscales. The first assesses the individual personal justice experience; the second is more socially focused and

assesses the perception of how fairly employees, in general, feel they are treated by the organization. This

dimension is similar to the justice climates’ construct proposed here.

Workers do not always have access to the causes behind decisions that affect them (Johanson, 2000). As

so, they experience uncertainty concerning their relationship with the organization. In order to avoid exclusion

or social exploitation, workers look for clues that serve as heuristics to evaluate the reliability and degree of trust

that organizational authority figures deserve, as well as for clues to assess their own status as members of the

organization (Lind & Van den Boss, 2002; Thau, Bennett, Mitchell & Marrs 2009). Through social influence

processes, the justice climate may serve as a source of information to help evaluate the organization and

interpret events experienced by workers (Lamertz, 2002).

By acknowledging that the justice climate serves as an information source for the interpretation of events, it

is possible to conceive two alternative relationships between the justice climate and justice event judgments.

The literature on expectancy violation (e.g., Bell, Wiechmann & Ryan, 2006), and some justice theories, such as

the group-value model (e.g., Blader & Tayler, 2009) and the fairness theory (e.g., Folger & Cropanzano, 2001)

emphasize the comparison process individuals engage in when evaluating their justice experiences. According

to these theories, when the justice climate is high, individuals may judge unfair experiences more negatively.

When there is a difference between the unfairness of personal experiences and the justice climate within the

group, it contradicts expectations about what “might” and “should” have been, and may be perceived as a sign

that the individual is not as respected and valued as other group members. On the other hand, the justice climate

may have a direct influence on individual perceptions, leading to an assimilation of the groups’ opinion. In that

case, an injustice event will lead to less negative justice perceptions when the justice climate is high, than when

it is low.

Most studies indicate that shared opinions in the work environment concerning justice influence an

individual’s justice perceptions (Degoey, 2000). For example, Lind, Kray & Thompson (1998) have shown that

communication between workers concerning a supervisor’s personal treatment influenced individuals’ justice

perceptions. Jones and Skarlicki (2005) also observed that a discussion between workers concerning the justice

reputation of an authority figure affected the interpretation of his subsequent behaviour. In both cases

individuals tended to conform to the group’s opinion. In addition, Choi (2008) found that when employees’

perceptions of the organization as a whole are positive, they tend to react less negatively to unfair events. So we

expect that the justice climate will exert an influence on the interpretation of personal experiences and that faced

with an injustice situation, justice perceptions will be less negative when the justice climate is high.


19

H2: Organizational justice perceptions will be more positive when the justice climate is high, as opposed to

when it is low.

The (in)justice social context and work satisfaction

The social context in which injustice takes place has an effect on workers’ satisfaction. A work

environment perceived by the work group as fair will have a greater chance of satisfying workers and raising

morale (Forret & Love, 2007) than a climate perceived as unjust. We expect that in high justice climate

situations, participants will experience higher levels of satisfaction. On the other hand, based on attributional

bias, memory bias and the just-world theory, we expected that an injustice directed at the self would result in

lower levels of satisfaction than an injustice directed at a colleague. It is predicted that these social context

variables will have an indirect effect on satisfaction, so when justice judgments are statistically controlled, the

amount of variance in work satisfaction explained by the justice climate or the injustice target will be

significantly reduced.

The relationship between justice perceptions and work satisfaction has been one of the most studied in

organizational justice literature. In fact, the first theories concerning justice perceptions (e.g., Adams, 1965)

stated that the evaluation of a situation as fair would lead to satisfaction, while an unfair situation would lead to

psychological tension and feelings of anger. Since then, many studies have replicated this relationship, finding

that organizational justice perceptions are strong and consistent cognitive predictors of work satisfaction

(Cohen-Charash & Spector, 2001; Colquitt et. al., 2001; Tremblay & Roussel, 2001). These emotional outcomes

have been related to withdrawal behaviours such as employee absenteeism, turnover intentions, work alienation,

and self-medication with alcohol (Howard & Cordes, 2010). Some authors state that among the multiple factors

that affect work satisfaction, the most important organizational characteristic is justice perception (Clay-Warner,

Reynolds & Roman, 2005).

Different approaches have been proposed to understand the relative effects of each dimension of justice on

job satisfaction. Instrumental models argue that distributive justice is the most important predictor of satisfaction

(McFarlin & Sweeney, 1992). These models state that employees value justice because it allows them to predict

and control the outcomes they are likely to receive from organizations (Cropanzano, Bowen & Gilliland, 2007).

On the other hand, some models support a more relational view of justice (e.g., Tyler & Blader, 2000, 2003;

Blader & Tayler, 2009). They state that fair procedures and personal treatment provide identity-relevant

information, which confer respect and status and thus improve the image of the group and increase self-esteem.

This in turn, leads to a positive evaluation of the organization as a whole, and an increase in satisfaction. Thus

far, we have argued that social context influence people’s justice perceptions, which, in turn, contribute to work

satisfaction. Therefore, we propose:

H3: Justice perceptions will mediate the relationship of the (in)justice target and climate with satisfaction.

Figure 1 presents a simplified model of the expected relationship between variables.

Figure 1: Theoretical Model

This model is consistent with socially contextualized justice perspectives such as the group value and the

deontonic models. The social context in which the (in)justice takes place is of great importance.

We expect that when faced with an injustice situation directed at the self, the worker’s justice perceptions

will be more negative than when the injustice is directed at other colleagues. Even if deontonic concerns may

lead people to be sensitive to others’ mistreatments, the joint effect of attribution and memory bias, and just-

world beliefs lead to an emphasis of one’s own experience and a deemphasizing of other people injustice

Justice Perceptions:

Distributive

Procedural

Interactional

Injustice Target

Justice Climate

Work Satisfaction


20

experiences. On the other hand, we believe that justice climate will influence the interpretation of justice

experiences. Because individuals tended to conform to the group’s opinion (Jones & Skarlicki, 2005) we

believe that, contrary to group-value model’s (e.g., Blader & Tayler, 2009) and the fairness theory´s

assumptions (e.g., Folger & Cropanzano, 2001), justice perceptions will tend to be less negative when the justice

climate is high.

It is predicted that these social context variables will have an indirect effect on satisfaction, mediated by

justice perceptions. The relative importance of each justice dimension will be investigated, to compare

instrumental models’ (McFarlin & Sweeney, 1992; Cropanzano, Bowen & Gilliland, 2007) and relational

models’ predictions (e.g., Tyler & Blader, 2000, 2003; Blader & Tayler, 2009).

3 METHOD

Experimental design

To analyse the effect that the independent variables, justice climate and injustice target, have on the

mediating variable, organizational justice judgments (i.e., distributive, procedural, interactional), and their joint

effect on work satisfaction as the dependent variable, a 2 x 2 (justice climate: high vs. low x injustice target: self

vs. other) cross-factor design was used.

A scenario methodology (e.g., Lind & Tyler, 1988; Hafer, Bègue, Choma, & Dempsey, 2005) was used to

manipulate the independent variables. According to Lind and Tyler (1988, pp.47) “This method involves the

presentation of descriptions of scenarios, which respondents are asked to imagine happening to themselves or to

others, and the assessment of attitudes and beliefs with respect to the scenarios.” In the present study,

participants are invited to assume the point of view of an employee of a fictitious company, and evaluate what

their attitudes and behaviors would be in that situation.

Participants

A sample of 140 volunteers participated in this study. The participants were approached while waiting in a

government building with several governmental services divisions such as ID and Passport emission, Social

Security Services, Post Office, etc. and asked to complete the survey. Seven participants were excluded from the

sample for having correctly identified the purpose of the study in a control question. This left a valid sample of

133 individuals. Of that valid sample, 57.9% (n=77) were women. Ages varied between 17 and 65, with an

average age of 34 (SD=11.813). The majority of participants, 75% are employed and 44% of those had had

tenure for over 10 years (M=14.65; SD=11.59).

Procedure

The study was presented as an inquiry about “human resources management practices” in national

companies. Participants were given a text to read describing a company, designated company X. They were

asked to imagine they worked for company X and to answer the questions with that in mind.

At the beginning, the text describes the justice climate prevalent in company X. Depending on the

experimental situation, the justice climate is described as either high or low. Then, a specific injustice situation

taking place in this company is portrayed. Depending on the experimental situation, the target of the injustice

situation is presented as the self, as a worker of company X (target self), or as a co-worker (target co-worker).

After reading the text, participants are asked to turn the page and not to consult it again. Afterwards, they are

requested to position themselves on the scales from the point of view of an employee in company X.

Measures

A pre-test was conducted (n=11) to test the effectiveness of the independent variable manipulation, and that

the participants thoroughly understood the scales used. The pre-test was also useful to verify the level of the

scenario’s credibility, since “the key to valid scenario studies is to design the study to deal with situations the

respondents have experienced and understand” (Lind & Tyler, 1988, pp. 47). It was observed that participants in

the pre-test found the story told to be entirely believable, (M=4.36; SD=.667 in a 1 to 5 scale), some participants

even mentioned having had similar experiences in their professional life. Participants did not express any

difficulties in understanding the questions so there was no need to carry out further adjustment to the material.

The text displaying the experimental scenario started by requesting the participants to: “Imagine that you

have been working for some time now in company X”. The justice climate manipulation was adapted from the

procedural justice climate manipulation carried out by Aquino, Trips and Bies (2006). Aspects of distributive

and interactional justice were introduced to allow a manipulation of the different aspects of the justice climate.

“. . . they are known for their good (poor) treatment of employees. Employees feel they are treated with

(little) a lot of respect and consideration. (Often) Rewards are (not) distributed according to the effort each

employee makes. Besides, employees feel they are (never) involved in upper-level decision making. In fact, as


21

you and your colleagues have witnessed on numerous occasions, managers have always (rarely) considered

employees’ views before making major decisions, have always applied the rules consistently (frequently applied

the rules inconsistently) across employees, and make sure to base decisions (never ensure that decisions are

based) on accurate information. Even then, if employees disapprove of a decision, managers have shown a

willingness (not been willing) to reconsider.”

Then the manipulation of the target was presented:

“One day, your supervisor tells you and your colleagues that an element of the team, normally in charge of

a particular function is absent. And he says you (your co-worker Joseph) will have to replace him. You (He

tries) try to explain that you have (he has) never performed this task and you don’t (he doesn’t) think you have

(he has) the proper knowledge or skills needed to perform the task. But the supervisor does not reconsider,

saying he is available if you need (your co-worker needs) help.”

The description of the injustice situation was adapted from Kray and Lind (2002).

“(Later, your co-worker Joseph comments to you :) How frustrating! The tasks were difficult, the computer

repeatedly shut down, and to make matters worse, the supervisor didn’t help at all. You (I) sent messages after

each task asking for help because, besides not having sufficient knowledge or skills to perform the task, the

computer was off-line so much that you couldn’t work as hard as you wanted. Each time that you (I) sent a

message, you (I) got a response that made you (me) think your (my) supervisor hadn’t even read your (my)

message. Your (my) doubts and concerns were falling on deaf ears. In fact, on one occasion, the manager even

stated explicitly that your (my) message was unread. You were (I was) told your (my)‘‘excuses’’ weren’t good

enough and that they would not be considered when deciding on your (my) performance appraisal or on pay

bonuses. You think to yourself (I thought to myself): What did I do to deserve this kind of treatment? I try to

work hard...on top of it all, a promotion is going to be decided soon.”

To check the effectiveness of the manipulation of the justice climate a control question was introduced at

the end of the questionnaire: “In general, not taking into consideration the specific situation described in the

text, workers feel justly treated in company X”. In order to check whether participants recognized the situation

described as an injustice situation they were asked if: “The specific situation that happened to the worker in

company X is an injustice situation” and “The specific situation that happened to the worker in company X is

serious”. Finally, an open question, regarding the perceived aim of the questionnaire, was introduced to control

implicit justice theory biases. All items were measured on a 1- Totally disagree to 5- Totally agree Likert type

scale.

Organizational justice perceptions were measured using the justice scales adapted from Niehoff and

Moorman (1993) and Folger and Konovsky (1989). Four items assessed distributive justice (α=. 913) (e.g.,

“Overall, the rewards I receive here are quite fair”), three items assessed procedural justice perceptions (α=.

893) (e.g., “To make job decisions, my supervisor collects accurate and complete information”), seven items

assessed interactional justice (α=. 947) (e.g., “My supervisor treats me with kindness and consideration”).

Work Satisfaction was assessed through an item where participants showed how they would feel working

for company X by choosing from 1 - “Totally unsatisfied” to 5 - “Totally satisfied”. Many researchers (eg,

Hackman & Oldham, 1980, Quinn & Shepard, 1974) have adopted an additive approach in measuring job

satisfaction. Several items are used to separately measure satisfaction with different aspects of organizational

life, them a global satisfaction index is computed (Snipes Oswald, LaTour, & Armenakis, 2005). However,

some authors, such as Scarpello and Campbell (1983) believe that a single item satisfaction measure is more

credible than the sum of scales for each aspect of job satisfaction, since multi items scales may overlook aspects

of work that are important to the employee (eg., Ironson, Smith, Brannick, Gibson, & Paul, 1989; Scarpello &

Campbell, 1983; Wanous, Reichers, & Hudy, 1997). Furthermore, in a meta-analysis by Wanous and colleagues

(1997), the authors compared studies using multi-item scales, and studies using a single item of job satisfaction.

Results showed a high correlation, suggesting that job satisfaction can be operated through a single item. In

addition this study used a single item satisfaction measure because, given that the manipulation was conducted

through the presentation of a scenario, we believed participants would find it easier to assess what their general

levels of satisfaction would be, rather than to rate specific satisfaction dimensions.


22

4 RESULTS

Manipulation check and descriptive statistics

Participants agree that the situation described is an injustice situation (M=4.07; SD=.612). Additionally,

72% of the participants consider that the situation described is very serious (M=3.83; SD=.78).

To verify if there is a difference in the justice climate perception between the two situations, a t test for

independent samples was carried out. This revealed a significant difference (t(,414)=6.706; p≤.001) in the

expected direction. Those in the high justice climate situation tended to believe that, in general, the company

dealt with the workers more fairly (M=3.30, S.D= 1.301) than those in the low justice climate situation

(M=1.91, S.D= .91).

Table 1 presents the means, standard deviations and correlations of the variables considered in this study.

As can be observed, all variables are significantly correlated with each other except for: the justice climate and

procedural justice; the justice climate and interactional justice; and the justice climate and injustice target. All

correlations show positive relationships between variables except for the injustice target, which appears to have

a negative relationship with the remaining constructs. These magnitudes of correlations, ranging from relatively

high to medium, have often been found between justice dimensions (e.g., Colquitt et. al., 2001).

Table 1: Descriptive statistics

Measures M SD 1 2 3 4 5

1. Distributive Justice 2.55 1.00

2. Procedural Justice 2.35 1.10 .63**

3. Interactional Justice 2.11 .95 .44** .43**

4. Work Satisfaction 2.48 1.10 .47** .56** .46**

5.Justice Climate .49** .56** .17 .35**

6. Injustice Target -.27** -.10 -.25** -.17 -.022

Notes: **= p≤.001.

Hypotheses testing

Hypothesis 1 predicts that an injustice directed at the self will lead to more negative justice perceptions

than an injustice directed at a co-worker. Hypothesis 2 predicts that when the justice climate is high justice

perceptions will be less negative than when the justice climate is low. Multiple regression analyses were

conducted to assess the effect of these social context variables on justice perceptions. Because the independent

variables are a dichotomy they were recoded as dummy variables. Recoding was conducted in the following

way: justice climate 0= low; 1 =high; injustice target 0 =self; 1 =co-worker. The regression coefficients

represent the difference between the lower and higher values, that is, between low and high justice climate

situations, and between injustice directed at the self or at a co-worker. The results are summarized in table 2.

The model is significant (F(2.123)= 26.214; p≤ .001) in predicting distributive justice perception. The

(In)justice climate and target have an effect on distributive justice perception, which would explain 29% of its

variance. From the low to the high climate situation there is an increase in distributive justice perception.

Distributive justice perception tends to be more positive when the injustice targets a co-worker than when it

targets the self.

The effect of (in)justice climates explains 31% of the variance in procedural justice perceptions (F(2.126)=

29.911; p≤ .001). In the high justice climate situation, perceptions of procedural justice tend to be higher than in

the low climate situation. Although the relationship between the injustice target and procedural justice follows

the expected direction, it is not significant.

The model explains about 8% of the variance in interactional justice (F(2.121)= 6.137; p≤.05). Interactional

justice perception tends to be higher when the target is a co-worker than when it is the self, or when the justice

climate is high as opposed to when it is low. Although the relationship between the justice climate and

interactional justice follows the expected direction, it is not significant.


23

Table 2: Organizational Justice regressed on (in)Justice climate and target

Depended Variables Independent Variables βa R2a Δ R2

Distributive Justice Climate .48** .23 .23**

Target -.24* .29 .06*

Procedural Justice Climate .56** .31 .32**

Target -.07 .31 .00

Interactional Justice Climate .17 .08 .00

Target -.25* .08 .09*

Notes: *= p≤.05; **= p≤.001

These results show that distributive and interactional justice perceptions were more negative when the

injustice situation was directed at the self rather than towards a co-worker. So hypotheses 1 was supported for

distributive and interactional justice perceptions. Results also show that distributive and procedural justice

perceptions tend to be less negative in the high justice climate situation. Hypothesis 2 was supported for

distributive and procedural justice judgments.

Although the hypotheses did not predict any interaction between the (in) justice climate and target, we

conducted a multiple variance analysis to explore that possibility. No effects were found.

Hypothesis 3 predicts that organizational justice will mediate the relationship of the justice climate and the

injustice target with satisfaction. The proposed model involves multi mediators that have relatively high to

medium correlation levels. Taking into account that the sample size does not permit structural equation analysis,

we have chosen to follow Preacher and Hayes (2008) recommendations and employ bootstrap analysis. The

authors state this method has four major advantages over classic multiple linear analyses: (1) it tests the “total

indirect effect” of the independent variable on the dependent variable which makes it possible to determine

whether an overall mediation effect exists and to assess its magnitude; (2) it makes it possible to determine to

what extent specific mediator variables mediate the relationship between the independent variable and the

dependent variable, in the presence of other mediators in the model; (3) the likelihood of parameter bias due to

omitted variables is reduced; (4) it makes it possible to compare the relative magnitudes of the specific indirect

effects associated with which mediators within the model. Separate analyses were conducted for each

independent variable including all three justice dimensions as mediators, and satisfaction as a dependent

variable. The results are presented in tables 3 and 4.

Justice perceptions fully mediate the effect the justice climate has on satisfaction. The model explains 37%

of the variance in satisfaction levels (F(4.114)= 17.998; p≤ .001). Only procedural justice was a significant specific

indirect effect, mediating the relationship between the justice climate and work satisfaction. This means that,

within this model and relative to the other mediators, procedural justice explains the overall effect of climate on

satisfaction. Satisfaction tends to be higher when the justice climate is high because it promotes more positive

procedural justice judgments.

Table 3: Bootstrap Analysis: Justice Dimensions as mediators between Justice Climate and Satisfaction

Note— BCa. bias corrected and accelerated; 5.000 bootstrap samples.

Justice perceptions also fully mediate the effect the injustice target has on satisfaction. The model explains

37% of the variance in satisfaction levels (F(4.114)= 18.012; p≤ .001). One can observe that interactional justice is

the only mediator with a significant specific indirect effect. This means that within this model and relative to the

other mediators, interactional justice explains the overall effect of injustice targets on satisfaction. Satisfaction

tends to be lower when injustice is directed at the self, compared to when it is directed at a co-worker, because

injustice directed at the self-promotes more negative interactional justice perceptions.

Ba 95% CI

β P Lower Upper

Direct Effect

Justice Climate .80 .00

Indirect Effects

Justice Climate .16 .41

Total .64 .00 .32 .98

Distributive Justice .10 .34 -.10 .32

Procedural Justice .44 .00 .15 .78

Interactional Justice .10 .07 .01 .26

R2 .37


24

Table 4: Bootstrap Analysis: Justice dimensions as mediators between injustice target and satisfaction

Notes: BCa. bias corrected and accelerated; 5.000 bootstrap samples.

The results partially support hypothesis 3. The relationship between the injustice target and satisfaction is

totally mediated by interactional justice. The relationship between the justice climate and satisfaction is totally

mediated by procedural justice. However, no mediating effect of distributive justice was verified.

4 DISCUSSION

This study attempted to understand the influence of the (in) justice social context in the development of

justice judgments, and its consequences. It explored the effect of co-workers’ experiences (i.e., the injustice

target), and co-workers’ shared perceptions (i.e., the justice climate) on employees’ justice perceptions and job

related attitudes, namely job satisfaction.

Our findings indicate that injustice situations experienced by the self-generate more negative distributive

and interactional justice perceptions than injustice situations directed at other people. The injustice target has a

greater effect on interactional justice when compared to distributive justice. Contrary to our hypothesis, as far as

procedural justice judgments are concerned it makes no difference if the unfair event targets the self or a co-

worker. This may be due to the fact that procedural justice judgments are related to structural aspects of

organizational decision making processes. Even if the unfair procedures are experienced by a co-worker, they

may have a more general impact in situations that may affect individuals directly in the future.

Interactional justice judgments mediates the effect injustice target has on satisfaction. Injustices directed at

the self, generate more negative distributive and interactional justice perceptions than those directed at another.

Unlike distributive justice perceptions, which focus on specific outcomes, interactional justice focuses on the

more subjective quality of human interactions most likely to be influenced by attributional and memory biases.

Interactional justice relates directly to the image of supervisors (Colquitt et. al., 2001; Nowakowski & Conlon,

2005; Aryee, Chen, Sun, & Debrah, 2007), who are seen as the key agents of the organization. Thus perceptions

of interactional justice affect the extent to which individuals are satisfied with their employment relationship

(Lamertz, 2002), influencing job satisfaction.

When workers hear about an unfair event directed at a co-worker they consider the responsibility of both

parties involved. Influenced by just-world beliefs (e.g., Bobocel & Hafer, 2007) they attribute some, or most, of

the responsibility to the co-worker, and the event is evaluated as more fair. On the other hand, when the injustice

is directed at the self, workers are influenced by self-enhancing bias (e.g., Sedikides, Gaertner, & Vevea, 2005),

and assign all the responsibility for the transgression to the perpetrator, thus perceiving the interaction as less

fair. This leads to a decline in the supervisor’s image and the employment relationship, resulting in lower levels

of satisfaction. This effect indicates that, despite the deontonic perspective assumption that a perceived injustice

provokes negative emotional and behavioural reactions by third party observers (Cropanzano, Goldman &

Folger, 2003) these reactions are not as extreme as when the self is the target of the injustice.

Regarding the justice climate, findings indicate that justice plays a role as an antecedent of individuals’

justice perceptions. Facing the same injustice situation, individuals’ distributive and procedural justice

perceptions are more positive when the work group considers it is being treated fairly. The results contradict

assumptions based on the literature on expectancy violation (Bell et. al, 2006), and justice theories, such as the

group-value model (Tyler & Lind, 1992) and the fairness theory (Folger & Cropanzano, 1998, 2001). If

individuals experience unfair events in groups that believe they are generally treated fairly, individuals will

likely compare their experience with that of the group. Even so, instead of generating more negative justice

perceptions and dissatisfaction, the dissimilarity between what was experienced and the fairness people are

generally treated within the organization, may lead them to think of the incident as a one-off occurrence and to

give it less importance.

β 95% CI

β p Lower Upper

Direct Effect

Injustice Target -.41 .03

Indirect Effect

Injustice Target -.14 .40

Total -.27 .00 -.55 -.03

Distributive Justice -.06 .38 -.22 .07

Procedural Justice -.09 .24 -.31 .05

Interactional Justice -.12 .05 -.29 -.03

R2 .37


25

The justice climate has a stronger impact on procedural than on distributive justice judgments, and has no

effect on interactional justice perceptions. This may be related to the differences between these justice

dimensions. According to Greenberg (1993a), people are conceptually able to form justice judgments that

evaluate relatively stable aspects of the work environment, as well as events that happen at work on a day-to-day

basis. Procedural justice perceptions relate to structural aspects of organizational decision-making. Interactional

justice on the other hand, pertains to the quality of event-level interactions between employees and direct

supervisors. Therefore, procedural justice may have a more general impact on work groups than more situational

interactional justice. As such, individuals may rely more on socially influenced information processes, like the

justice climate, to make procedural justice evaluations. While interactional justice judgments are more prone to

particular event related attribution processes.

Procedural justice mediates the effect of the justice climate on satisfaction. When the justice climate is

high, procedural and distributive justice perspectives will be more positive. But unlike distributive justice, which

targets particular individual outcomes, the fairness of organizational procedures is more widespread, so an

individual may resort to group-level information for their evaluation. Satisfaction is said to occur not only as a

result of affective reactions but also as a consequence of calculative evaluation about the organization's fair

dealings in the employment relationship (Lamertz, 2002). Procedural justice is a good indicator of the fairness

of structural institutional organizational practices and as such, it contributes to the cognitive facet of satisfaction.

Because of its generalized character, reflected in the coherence norm, workers’ may rely on social information

sources, like the justice climate, to make procedural justice judgments, which are then reflected in satisfaction

levels.

No mediating effects of distributive justice were found. Distributive justice does not appear to play a major

role in the relationship between contextual social factors, like the (in) justice climate and target, and satisfaction.

This finding questions the assumptions of instrumental models of justice (e.g., Thibaut &Walker, 1975;

McFarlin & Sweeney, 1992). Satisfaction does not appear to be primarily influenced by distributive concerns.

On the other hand, results illustrate the importance a justice related social context has on satisfaction, and they

support the prediction of group value models (Tyler & Blader, 2000, 2003) that fair procedures and personal

treatment provide identity-relevant information influencing satisfaction.

Unlike distributive justice, procedural and interactional justice can serve dual roles: as outcomes in their

own right, and as sources of information about events (Barclay, Skarlic & Pugh, 2005) and social identity

(Blader & Tayler 2009). Individuals are sensitive to procedural and interactional justice because they signal the

degree to which members are valued and respected in the group and because transgressions contradict

expectations about the norms that regulate procedures and personal treatment. When individuals’ expectations

for socio-emotional outcomes are violated, they can experience emotional reactions regardless of whether their

economic expectations have been violated (Barclay, Skarlic and Pugh, 2005). Although violations of any type of

justice perspective can trigger attributions, procedural and interactional justice, carry attributional information

which can be used in the appraisal of the situation. As such, procedural and interactional justice judgments are

important to attitudes workers form based on social information.

Like most studies, the present study has its limitations. Some authors (e.g., Fraizer, Barron & Tix, 2004)

recommend that in the case of multiple mediator model analyses of structural equations, analysis should be used

in order to control the combined effect of the variable and the fit of the overall model. In this study that was not

possible due to the sample size. However, we followed Preacher and Hayes’ (2008) recommendations and

conducted bootstrap analysis. Although bootstrap analysis does not indicate the overall fit of the model, or

permit the simultaneous testing of both independent variables, nevertheless, it allows us to compare the effects

of multiple mediators.

Another limitation is the fact that participants of this study were presented with a description of an injustice

situation and not involved in a real life situation. Although we made certain that the described situation was

believable, differences may exist between the way people perceive they would behave and the way they actually

would. In part because “people are especially inaccurate in predicting how they will behave in a new situation”

(Lind & Tyler, 1988, pp. 47) and in part, because this method is particularly likely to be biased by social

desirability, since instead of acting spontaneously, the participants cognitively esteem how they would behave.

Future research could test our model in an organizational field study.

Our findings have some implications for human resources management. On the one hand they suggest that

the social context in which the (in)justice occurs has an effect, both in workers’ justice perceptions, and in their

satisfaction levels. Work environments normally characterized as fair have a buffering effect in the development

of workers’ negative justice perceptions. This suggests that organizations and their supervisor must strive to

maintain a high justice climate work context, if they wish to encourage higher satisfaction levels and more

positive justice perceptions. So, it is not only necessary to treat workers fairly, organizations must strive for

workers to have a shared perception that indeed, fair treatment is part of “the way things work around here”. On

the other hand, our findings indicate that, at least as far as procedural justice judgments are concerned, it makes

no difference if the unfair event targets the self or a co-worker. This indicates that organizations should strive to


26

be consistent in the way they treat their employees. Furthermore, it seems that procedural and interactional

justice play an important role in the relationship between the context in which the (in)justice occurs and workers

attitudes. Managers should be aware that even if work outputs are fairly distributed, the way policies are

designed, and implemented by supervisors, have an impact in workers’ satisfaction.

Justice is an important predictor of workers’ attitudes and behaviours. These findings are relevant to human

resources management practices, since they imply refocusing justice concerns so that they are not solely on

individuals but extended to groups. If co-workers’ experiences and perceptions have an impact on individual

justice perceptions, we can infer that practices aimed at dealing fairly with individuals will have less effect if

employees, as a whole, do not believe they are treated fairly. Especially because of the increasing relevance of

teams in the structuring of today’s organizations, the notion that injustices directed at co-workers may influence

other individuals’ justice judgments is well worth noting. Further research should be conducted on the effect co-

workers’ injustice experiences have on employees’ justice perceptions. Results indicate that a greater effort

should be made towards a finer integration of conceptual models of justice, which frame the development and

functioning of organizational justice perceptions in a socially contextualized perspective.

REFERENCES

Adams, J. (1965). Iniquity in social exchange. In L. Berkowitz (Ed.) Advances in Experimental Social

Psychology. New York: Academic, 2, 276-299.

Ambrose, M., & Schiminke, M. (2009). The role of overall justice judgments in organizational justice research:

A test of mediation. Journal of Applied Psychology, 94: 2, 491–50

Ambrose, M., & Schminke, M. (2002). Organization structure as a moderator of the relationship between

procedural justice, interactional justice, perceived organizational support, and supervisory trust. Journal of

Applied Psychology, 88, 295–305.

Aquino, K., Tripp, T., & Bies, R. (2006). Getting even or moving on? Power, procedural justice, and types of

offense as predictors of revenge, forgiveness, reconciliation, and avoidance in organizations. Journal of

Applied Psychology, 91:3, 653–668.

Aryee, S., Chen, Z., Sun, L., & Debrah, Y. (2007). Antecedents and outcomes of abusive supervision: Test of a

trickle-down model. Journal of Applied Psychology, 92: 1, 191-201. doi:10.1037/0021-9010.92.1.19.

Barclay, L., Skarlicki, D., & Pugh, S. (2005). Exploring the role of emotions in injustice perceptions and

retaliation. Journal of Applied Psychology, 90, 629-643.

Beehr, T., Nair V., Gudanowski, M., & Such, M. (2004). Perceptions of reasons for promotion of self and

others. Human Relations 57: 4, 413

Bell, B. S., Wiechmann, D., & Ryan, A. M. (2006). Consequences of organizational justice expectations in a

selection system. Journal of Applied Psychology, 91, 455-466.

Beugré, C. D. (2010). Resistance to Socialization into Organizational Corruption: A Model of Deontic Justice.

Journal of Business and Psychology. doi:10.1007/s10869-010-9176-3

Bies, R. (2005) Are procedural and interaccional justice conceptually distinct. . In J. Greenberg & J.Colquitt.

Handbok of Organizational Justice (pp.85-112) Mahwah: Lawrence Erlbaum.

Blader, S., & Tayler, T. (2009) Testing and extending the group engagement model: Linkages between social

identity, procedural justice, economic outcomes, and extrarole behavior. Journal of Applied Psychology,

Vol. 94, No. 2, 445–464

Bobocel, D., & Hafer, C. (2007) Justice motive theory and the study of justice in work organizations: A

conceptual integration. European Psychologist, 12: 4, 283–289.

Carroll, W., & Bandura, A. (1987) Translating cognition into action: the role of visual guidance in observational

learning. Journal of Motor Behavior. 19: 3, 385-98.

Chia, H., Fang, R.,, & Foo, M.D. (2006). Work place as communities. Who seeks, gives, and accepts help on

justice issues. Journal of Community Psychology, 34: 3, 363-377.

Chiaburu, D., & Harrison, D. (2008). Do Peers Make the Place? Conceptual Synthesis and Meta-Analysis of

Co-worker Effects on Perceptions, Attitudes, OCBs, and Performance. Journal of Applied Psychology, 93,

1082–1103

Clay-Warner, J., Reynolds, J., & Roman, P. (2005). Organizational justice and job satisfaction: A test of three

competing models. Social Justice Research, 18: 4, 391-409.

http://www.biomedexperts.com/Abstract.bme/14988054/Translating_cognition_into_action_the_role_of_visual_guidance_in_observational_learning

http://www.biomedexperts.com/Abstract.bme/14988054/Translating_cognition_into_action_the_role_of_visual_guidance_in_observational_learning


27

Cohen-Charash, Y., & Spector, E. (2001). The Role of Justice in Organizations: A Meta-Analysis.

Organizational Behavior and Human Decision Processes, 86, 278–321

Colquitt, J., Noe, R., & Jackson, C. (2002). Justice in Teams: Antecedents and Consequences of Procedural

Justice Climate. Personnel Psychology, 55, 83-110.

Colquitt, J. A. (2004). Does the justice of the one interact with the justice of the many?

Reactions to procedural justice in teams. Journal of Applied Psychology, 89: 4, 633–646.

Colquitt, J.; Conlon, D.; Wesson, M.; Porter, C., & Ng, K. (2001). Justice at the millennium: A meta-analytic

review of 25 years of organizational justice research. Journal of Applied Psychology, 86, 425-445

Cropanzano, R.; Bowen, D., & Gilliland, M. (2007). The Management of Organizational Justice. Academy of

Management Perspectives, 21, 34-48

Cropanzano, R., Goldman, B. M., & Folger, R. (2003). Deontic justice: The role of moral principles in

workplace fairness. Journal of Organizational Behavior, 24, 1019-1024.

De Cremer, D., and Van Hiel, A. (2006). Effects of another person's fair treatment on one's own emotions and

behaviors: The moderating role of how much the other

cares for you. Organizational Behavior and Human Decision Processes, 100:2, 231–249.

Degoey, P. (2000). Contagious justice: Exploring the social construction of justice in organizations. Research in

Organizational Behavior, 22, 51–102.

Fassina, N.; Jones, D., & Uggerslev, K. (2008). Meta-analytic tests of relationships between organizational

justice and citizenship behavior: Testing agent-system and shared-variance models. Journal of

Organizational Behavior 29, 805–828

Folger, R., & Cropanzano, R. (1998). Organizational justice and human resource management. Beverly Hills,

CA: Sage.

Folger R. & Cropanzano, R. (2001) Fairness theory: Justice as accountability. In J. Greenberg & R. Cropanzano

(Eds.) Advances in organizational justice (pp.1-55) Palo Alto CA: Standford Press.

Folger, R., Cropanzano, R., & Goldman, B. (2005). Justice, accountability, and moral sentiment: The deontic

response to “foul play” at work. In J. Greenberg & J. Colquitt (Eds.), Handbook of organizational justice

(pp. 215-245). Mahwah, NJ: Erlbaum.

Forret, M., & Love, M. (2008). Employee justice perceptions and coworker relationships. Leadership &

Organization Development Journal, 29, 248-260

Frazier, P.; Tix, A., & Barron, J. (2004). Testing Moderator and Mediator Effects in Counseling Psychology.

Research Journal of Counseling Psychology, 51, 115–134.

Greenberg, J. (1993a). Stealing in the name of justice: Informational and interpersonal moderators of theft

reactions to underpayment inequity. Organizational Behavior and Human Decision Processes, 54, 81–103

Hafer, C., Bègue, L., Choma, B. & Dempsey, J. (2005) Belief in a just world and commitment to long-term

deserved outcomes. Social Justice Research, 18, 429–444.

Howard, L. W., & Cordes, C. L. (2010). Flight from Unfairness: Effects of Perceived Injustice on Emotional

Exhaustion and Employee Withdrawal. Journal of Business and Psychology. doi:10.1007/s10869-010-

9158-5

Hung, T., Chi, N., and Lu, W. (2009). Exploring the Relationships Between Perceived Coworker Loafing and

Counterproductive Work Behaviors: The Mediating Role of a Revenge Motive. Journal of Business and

Psychology, 24: 3, 257-270. doi:10.1007/s10869-009-9104-6

Inness, M.; LeBlanc, M., & Barling, J. (2005), Psychosocial Predictors of Supervisor-, Peer-, Subordinate-, and

Service- Provider-Targeted Aggression. Journal of Applied Psychology, 93: 6, 1401–1411

Johanson, J. (2000) Intraorganizational influence. Management Communication Quarterly, 13, 393-425.

Jones, D., & Skarlicki, D.P. (2005). The effects of overhearing peers discuss an authority’s reputation for

fairness on reactions to subsequent treatment. Journal of Applied Psychology, 90, 363–372

Lam, S.; Schaubroeck, J., & Aryee, S. (2002) Relationship between organizational justice and employee work

outcomes: A cross national study; Journal of Organizational Behavior, 23, 1-18

Lamertz, K. (2002). The social construction of fairness: Social influence and sense making in organizations.

Journal of Organizational Behavior, 23, 19–37.

Lerner, M. J. (1980). The Belief in A Just World: A Fundamental Delusion. New York: Plenum.

Liao, H., & Rupp, D. E. (2005). The Impact of Justice Climate and Justice Orientation on Work Outcomes: A

Cross-Level Multifoci Framework. Journal of Applied Psychology, 90: 2, 242-256. doi:10.1037/0021-

9010.90.2.242


28

Lind , E., & Tyler (1988). The Social Psychology of Procedural Justice. New York : Plenum Press.

Lind, E., & Van den Boss, K (2002). When fairness works: toward a general theory of uncertainty management.

In B. Stawn & R, Kramer (Eds) Research in Organizational Behavior, 24 (pp. 181-223) Boston: Elsevier.

Lind, E. (2001) Fairness heuristic theory: Justice judgments as pivotal cognitions in organizational relations. In

J. Greenberg & R. Cropanzano (Eds.) Advances in Organizational Justice. (pp. 56-57) Stanford, CA:

Stanford University Press

Lind, E., Kray, L., & Thompson, L. (1998). The social construction of injustice: Fairness judgments in response

to own and others’ unfair treatment by authorities. Organizational Behavior and Human Decision

Processes, 75, 1–22.

Lind, A., Kray, L., & Thompson (2001). Primacy effects in justice judgments: Testing the predictions of

fairness heuristic theory. Organizational Behavior and Human Decision Processes, 85, 189-210.

Mayer, D., Nishii, L., Schneider, B., & Goldstein, H. (2007). The precursors and products of justice climates:

group leader antecedents; Personnel Psychology, 60, 929-963.

McFarlin, D., & Sweeney, P. (1992). Distributive and procedural justice as predictors of satisfaction with

personal and organizational outcomes. Academy of Management Journal, 35, 626-637

Meyer, G.W. (1994). Social information processing and social networks: A test of social influence mechanisms.

Human Relations, 47, 1013–1047.

Nowakowski, J., & Conlon, D. (2005). Organizational justice: looking back, looking forward, International

Journal of Conflict Management; 16, 4-29.

Preacher, K., & Hayes, A. (2008). Asymptotic and resampling strategies for assessing and comparing indirect

effects in multiple mediator models. Behavior Research Methods, 40, 879-891

Rupp, D., McCance, A., Spencer, S., & Sonntag, K (2008) Customer (in)justice and emotional labor: the role of

perspective taking, anger, and emotional regulation. Journal of Management, 28, 904-925

Sedikides, C., Gaertner, L., & Vevea, J. (2005). Pancultural self-enhancement reloaded: A Meta-Analytic Reply

to Heine (2005). Journal of Personality and Social Psychology, 89, 539-551.

Spencer, S. & Rupp, D. (2009). Angry, guilty, and conflicted: injustice toward coworkers heightens emotional

labor through cognitive and emotional mechanisms. Journal of Applied Psychology, 94, 429–444

Thau, S., Bennett, R., Mitchell, M., & Marrs, M. (2009). How management style moderates the relationship

between abusive supervision and workplace deviance: An uncertainty management theory perspective.

Organizational Behavior and Human Decision Processes, 108, 10879–92

Thibaut J. & Walker, L. (1975) Procedural justice: A psychological perspective. Hillsdale NG: Erlbaum.

Tremblay, M., & Roussel, P. (2001). Modelling the role of organizational justice: effects on satisfaction and

unionization propensity of Canadian. The International Journal of Human Resource Management, 12: 5,

717-737. doi:10.1080/713769672

Turillo, C., Folger, R., Lavelle, J., Umphress, E., & Gee, J. O. (2002). Is virtue its own reward? Self-sacrificial

decisions for the sake of fairness. Organizational Behavior and Human Decision Processes, 89, 839-865.

Tyler, T., & Blader, S. (2000). Cooperation in Groups: Procedural Justice, Social Identity and Behavioral

Engagement. Philadelphia: Psychology Press.

Tyler, T., & Blader, S. (2003). The Group Engagement Model: Procedural Justice, Social Identity, and

Cooperative Behavior. Personality and Social Psychology Review, 7, 349–361

Tyler, T.R. & Lind, E.A. (1992). A relational model of authority in groups. In M. Zanna (Ed.), Advances in

Experimental Social Psychology (Vol. 25, pp. 115-191). New York: Academic.

Tyler, T.R. (1988). What is procedural justice? Criteria used by citizens to assess the fairness of legal

procedures. Law and Society Review, 22, 103-135

Van Den Bos, K., & Lind, E. A. (2001). The psychology of own versus others’ treatment: Self-oriented and

other-oriented effects on perceptions of procedural justice. Personality and Social Psychology Bulletin, 27:

10, 1324-1333.


Transformational leadership and job satisfaction: The mediating

effects of perceptions of politics and market orientation in the

Japanese context

Takuma Kimura

Faculty of Lifelong Learning and Career Studies, Hosei University

2-17-1 Fujimi Chiyodaku Tokyo, Japan

Telephone: +81-3-3264-6602


Abstract

This paper investigates the causal relationship among transformational leadership, perceptions of organizational

politics, market orientation, and work-related outcome. In this study, we assumed that organization-level

perceptions of organizational politics and market orientation mediate the relationship between top

management’s transformational leadership and employees’ work-related outcomes and that perceptions of

organizational politics diminish market orientation. Data were collected from a sample of 200 employees

working in Japanese companies. To test the hypothesized correlations, we used structural equation modelling

using AMOS 16.0. As we hypothesized, both perceptions of politics and market orientation mediated the

relationship between transformational leadership and employees’ job satisfaction. However, contrary to our

expectation, perceptions of organizational politics were not significantly correlated with market orientation. This

study is the first empirical research of organizational politics using a Japanese sample. In future studies, we

should resolve methodological limitations of this study and develop theoretical frameworks that reflect cultural

difference among countries.

Keywords: transformational leadership, organizational politics, market orientation, job satisfaction

Acknowledgments: The author would like to thank the anonymous reviewers for their insightful comments and

suggestions to improve the quality of the paper. And the author wishes to thank Ryuichi Hirano and Miku

Yamagishi at Neo Marketing Inc. for their support in conducting this research.


30

1 INTRODUCTION

Since the 1960s, organizational politics has been a prominent issue in organizational management studies.

One of the most often cited definitions of organizational politics is Mintzberg’s which views it as “individual or

group behaviour that is informal, ostensibly parochial, typically divisive, and above all in a technical sense,

illegitimate—sanctioned neither by formal authority, accepted ideology, nor certified expertise (although it may

exploit any one of these)” (Mintzberg, 1983, p. 172). Some studies define organizational politics as a broader

and more general set of social behaviours (e.g., Pfeffer, 1981). However, for the purpose of this study, we

limited our focus to narrower and more specific definitions, such as Mintzberg’s which regards politics as

behaviours strategically designed to maximize short-term or long-term self-interest (c.f. Ferris, Russ & Fandt,

1989).

Even more recently, there has been a considerable number of theoretical studies and empirical research in

organizational politics (cf. Atinc, Darrat, Fuller, & Parker, 2010). Although some recent seminal studies used

samples from Chinese organizations (e.g., Wei, Chiang, & Wu, 2012; Wei, Liu, Chen & Wu, 2010), most

studies were carried out in the Western context. To the best of our knowledge there are no studies that

empirically examine organizational politics in Japanese companies.

However, organizational politics is a common issue in Japanese firms. Japanese firms widely exhibit

political behaviours, such as sub-optimization, power struggles, factional disputes, and tactics used by

individuals to wield influences, such as ingratiation. Despite the lack of academic research, a lot of Japanese

business people have pointed out the existence of organizational politics and its harmful effects. For example,

Tsujino (2010), in memoirs of his working life at Sony stated that the recent slump of Sony was partly due to the

spreading of an introvert attitude and a decline in market orientation caused by organizational politics.

Market orientation enhances business performance and leads to other positive work-related outcomes

(Jaworski & Kohli, 1993). Thus, if organizational politics detrimentally affects market orientation, firms would

be well advised to effectively manage politics in their organizations. One of the keys to promoting market

orientation while diminishing political behaviours and their detrimental effects is transformational leadership

(e.g., Avolio & Bass, 1991; Bass & Avolio, 1993). Previous empirical research has shown that transformational

leadership behaviours are negatively related to perceptions of organizational politics (POPs), and positively

related to market orientation (Vigoda-Gadot, 2007; Menguc, Auh, & Shih, 2007; Menguc & Auh, 2008).

However, the relationship between POPs and market orientation has not yet been empirically examined.

In this study, we investigated correlations among transformational leadership, organizational politics,

market orientation, and work-related outcomes. More specifically, we examined the correlation between

transformational leadership and job satisfaction, and the mediating effects of POPs and market orientation on

the relationship. We tested our model by using a sample of employees working in Japanese firms. Cultural

differences among countries may lead to differences in some aspects of organizational politics. However,

because there have been no academic studies of organizational politics in the Japanese context, and some

previous studies implied the applicability of the Western framework to the Japanese context, we relied on

theoretical framework of Western studies.

As described below, previous Western studies have already examined many of our hypotheses. In this

respect, our study is an application of a Western theory in the Japanese context. However, to the best of our

knowledge, our study is the first to integratively investigate correlations among the four focal constructs of

transformational leadership, POPs, market orientation, and job satisfaction. Thus, our study aimed to not only

replicate results of previous Western studies but to also investigate an unexamined research issue.

2 THEORY AND HYPOTHESES

2.1 Transformational Leadership

In leadership literatures, we often distinguish two types of leadership; transactional leadership and

transformational leadership. This distinction was first suggested by Burns (1978) and has been developed by

subsequent studies (e.g., Avolio & Bass, 1991; Bass & Avolio, 1993). Whereas transactional leaders seek to

satisfy the current needs of followers through transactions or exchanges mediated by contingent reward

behaviours, transformational leaders arouse heightened awareness and interests in the group or organization,

increase confidence, and move followers gradually from concerns for existence to concerns for achievement and

growth (Yammarino & Dubinsky, 1994).

Previous empirical research revealed that while transactional leadership behaviours are negatively related

to organizational performance, transformational leadership behaviours lead to higher organizational

performance (Howell & Avolio, 1993; MacKenzie, Podsakoff, & Rich, 2001; Parry, 2003). Although some

studies emphasized a ‘culture-specific’ perspective of leadership effectiveness (for review see Dickson, Den

Hartogo, & Mitchelson 2003), recent empirical research has found the effectiveness of transformational

leadership across cultures, supporting Bass’s (1997) ‘universal’ perspective (Madzar, 2005, Muenjohn &

Takuma Kimura

31

Armstrong, 2007; Wang, Law, Hackett, Wand, & Chen, 2005). Moreover, in a team-level analysis, Ishikawa

(2012) found a positive influence of transformational leadership on R&D team performance in Japanese firms.

Traditionally, researchers define and measure transformational leadership behaviours in various ways.

Reviewing influential studies of transformational leadership, Podsakoff, MacKenzie, Moorman, and Fetter

(1990) identified six key behaviours of transformational leaders: (1) identifying and articulating a vision, (2)

providing an appropriate model, (3) fostering the acceptance of group goals, (4) high performance expectations,

(5) providing individualized support, and (6) intellectual stimulation.

Podsakoff et al. (1990) demonstrated that there is a great deal of consensus among the researchers on some

of these behaviours, but not on others. For example, almost all of the reviewed literatures had established

“identifying and articulating a vision” as an important component of the transformational leadership process.

Similarly, more than half of the studies had identified “fostering the acceptance of group goals” and “providing

an appropriate model” as elements of transformational leadership. In contrast, concerning the other three

behaviours (i.e., high performance expectations, providing individualized support, and intellectual stimulation),

only a few studies suggested their importance as aspects of transformational leadership.

2.2 Perceptions of Organizational Politics (POPs)

The literature on organizational politics can be classified into three approaches: (1) studies on influence

tactics, conflict, and actual political behavior in organizations, (2) studies on POPs, (3) studies on political skills

and capacities of the self within the workplace (Drory & Vigoda-Gadot, 2010). Since the seminal work by Ferris

et al. (1989) proposed a model for the examination of employees’ POPs, researchers have conducted empirical

analyses mainly on the second approach. This line of research relied on Lewin’s (1936) proposition that

individuals respond based on perceptions of reality rather than on objective reality. Thus, in this approach,

organizational politics are conceived of as a state of mind rather than as an objective state (Gandz & Murray,

1980; Harris, Andrews, & Kacmar, 2007).

Theoretical and empirical studies of POPs have focused on the negative side of organizational politics.

Indeed, a lot of empirical research has shown that POPs bring about negative outcomes such as turnover

intentions, job stress, job dissatisfaction, and declines in organizational commitment (cf. Atinc et al., 2010;

Kacmar & Baron, 1999: Miller, Rutherford, & Kolodinsky, 2008). Therefore, reduction of employees’ POPs

should be viewed as an important issue both in the theory and practice of organizational management.

Political behaviours are likely to occur when a work environment is characterized by a high degree of

uncertainty or ambiguity (Ferris et al., 1989) which are generated when objective and specific criteria for

decision making are absent. Such uncertainty and ambiguity give organization members room to engage in

opportunistic behaviours and influence-related tactics. Even though an organization has some formal rules and

regulations, employees’ perceptions of the decision-making process can be political if the evaluation criteria for

decision making are obscure. Such a political environment will be viewed as unjust and unfair (Andrews &

Kacmar, 2001). Indeed, Andrews and Kacmar’s (2001) empirical research revealed a significant negative

relationship between POPs and organizational justice.

Vigoda-Gadot (2007) assumed that transformational leadership reduces POPs on the grounds that a

transformational leader offers a vision, a mission, and an operative plan for goal achievement. This, then,

eliminates ambiguity and professional uncertainty, and validates the feeling that it is possible to deal with

organizational challenges in a decent way based on justice and fairness. In their empirical analysis, Vigoda-

Gadot (2007) found a negative relationship between transformational leadership and POPs. Thus, we

hypothesized the following;

H1: Transformational leadership is negatively correlated with POPs.

As noted above, Ferris et al.’s (1989) POPs model assumes that POPs negatively influence work outcomes.

Indeed, a lot of empirical research has shown various detrimental effects of POPs on employees. Ferris et al.’s

(1989) model posits that organizational politics is a source of stress in the work environment, and thus POPs

induce negative work-related outcomes. Some researchers have argued that if the work environment is political,

employees’ investment into the organization (i.e., expenditure of effort to work in the organization) becomes

more risky (Cropanzano, Howes, Grandey, & Toth, 1997; Randall, Cropanzano, Bormann, & Birjulin, 1999). In

a political environment, rewards tend to be allocated based on informal power structures rather than on

contribution or efforts, and the rules may change from one day to the next. Because of this uncertainty,

individuals are less likely to be confident that their efforts will produce any outcomes beneficial to themselves.

Thus, individuals see their long-term contribution to such organizations as a risky investment, with the result

they are more likely to withdraw (Cropanzano et al., 1997).

Although POPs affect various kinds of work-related outcomes, we limited our focus to their effect on job

satisfaction. This was because of research implementation limitations we encountered. When decision-making is

governed by political considerations, employees usually view their work environments as unfair. In such unfair


32

environments, employees feel unhappiness and stress, and therefore reduced satisfaction with their job (Ferris et

al., 1989; Ferris, Frink, Galang, Zhou, Kacmar, & Howard, 1996; Poon, 2003). Indeed, a lot of empirical

researches found the negative relationship between POPs and job satisfaction (e.g., Cropanzano et al., 1997;

Ferris & Kacmar, 1992; Ferris et al., 1996; Harris et al., 2007; Poon, 2003; Vigoda-Gadot & Talmud, 2010).

The POPs model was established in the Western context (Ferris et al., 1989), and much empirical research

on the model was conducted in Western countries. However, in recent years some research has replicated the

antecedents and consequences of POPs suggested by Ferris et al.’s (1989, 2002) model in the Eastern context

(e.g., Huang, Chuang, & Lin, 2003; Liu, Liu & Wu, 2010; Poon, 2003; Poon, 2006). Thus, we assume that POPs

have negative effects on work-related outcomes in Japanese organizations as well. However, there has been no

empirical research in the Japanese context. According to Hofstede (2001), a notable feature of Japanese culture

is high uncertainty avoidance in which people tend to feel stressed by uncertain or unknown situations.

Therefore, Japanese employees may be likely to suffer a feeling of dissatisfaction when they perceive their

organization as a political environment, or in other words, work environment with high uncertainty. Thus, we

hypothesized that:

H2: POPs are negatively correlated with job satisfaction.

2.3 Market Orientation

Market orientation is defined as “an organization culture that most effectively and efficiently creates the

necessary behaviours for the creation of superior value for buyers.” (Narver & Slater, 1990, p.21). Narver,

Slater, and Tietje (1998) argued that market orientation must be understood as an organizational culture and not

merely a set of processes and activities separate from the it. According to Narver et al. (1998), the central

principle of market orientation is that every person in the organization understands that each individual and

function can, and must, continuously contribute skills and knowledge to create superior value for customers.

In addition to viewing market orientation as a kind of organizational culture, Narver et al. (1998) suggested

that top management plays a critical leadership role in achieving and maintaining successful changes in

organization’s culture. They argued that appropriate leadership is essential to create market orientation in an

organization. Jaworski and Kohli (1993) echoed these viewpoints. They argued that unless an organization gets

clear signals from top managers about the importance of being responsive to customer needs, the organization is

not likely to be market oriented. Indeed, their empirical analysis showed that the greater top management’s

emphasis on market orientation, the greater the market orientation in the organization.

To reinforce market orientation, top management should articulate their vision and propagate it among

members so as to integrate the interests and values of various individuals and/or departments/groups within the

organization, and enable members to prioritize goals and objectives of the organization. We can regard these

leadership behaviours as essential parts of transformational leadership. Specifically, these leadership behaviours

are equivalent to “identifying and articulating a vision” and “fostering the acceptance of group goals”

(Podsakoff et al., 1990). Empirical studies have supported these arguments. Menguc et al. (2007) and Menguc

and Auh (2008) directly examined the relationship between transformational leadership and market orientation,

and found a significant and positive relationship. Therefore:

H3: Transformational leadership is positively correlated with market orientation.

Political behaviours are unsanctioned attempts at using influence to promote self-interest at the expense of

organizational goals (Ferris & Judge, 1991; Ferris & Kacmar, 1992; Ferris et al., 1996). Thus, in a political

organization, individuals and/or departments/groups tend to prioritize their own interests, and pay less attention

to the interests of other groups and/or that of the overall organization. They are prone to engage in political

behaviours in an attempt to receive favourable treatment and acquire more resources. These behaviours distort

resource allocation leading to detrimental effects on organizational performance.

To be market oriented, an organization requires cross-functional customer-value creation processes and

activities. In other words, each member, function, department, and/or group has to cooperate to create desired

customer value (Narver et al., 1998). However, in a political organization, because of self-serving behaviours of

each group or member, cross-functional customer-value creation processes and activities might be unlikely to

emanate. Thus, a political climate might impede the development of market orientation. Organizational politics

reflects a process of power struggles among conflicting individuals and groups attempting to further their own

self-serving goals mainly in decision-making (cf. Drory, 1993). If such power struggles escalate, employees or

groups are likely to compete with co-workers or other groups within their organization rather than with outside

competitors. This predisposes an organization to being “self-oriented” rather than market oriented, and therefore

less likely to gain a competitive advantage.

Nwanko, Owusu-Frimpong, and Ekwulugo (2004) examined the relationship between organizational

climate and market orientation and concluded that to make a market orientation program more sustainable, an

Takuma Kimura

33

organization needs to create a supportive climate. Nye and Witt (1993) argued that employees who perceive

their work environment as political may be less likely to view their organization as supportive because they

cannot believe that the organization is truly interested in their welfare. Previous empirical research has

confirmed their argument, finding a negative relationship between perceptions of organizational support and

POPs (Nye & Witt, 1993; Harris, Harris, & Harvey, 2007).

On the basis of these arguments and findings, we can assume there is a negative association between POPs

and market orientation. Indeed, empirical research has shown that the extent of political behaviour by

organizational members is negatively associated with the degree of market orientation (Harris & Piercy, 1999).

Thus, we hypothesized:

H4: POPs are negatively correlated with market orientation.

Empirical research has shown that market orientation leads to positive work-related outcomes, such as job

satisfaction, trust in management (Ruekert, 1992), and organizational commitment (Jaworski & Kohli, 1993).

Empirical research has also supported a positive relationship between market orientation and organizational

performance, such as ROA, sales growth, new product success, customer retention, market share, subjective

business performance (Greenley, 1995; Jaworski & Kohli, 1993; Kirca, Jayachandran, & Bearden, 2005; Narver

& Slater, 1990; Narver & Slater, 1993; Ruekert, 1992; Slater & Narver, 1994). Although contextual factors

significantly affect the strength of the link between market orientation and performance, the link itself

universally exists across various contexts (Ellis, 2006). Therefore, we can assume the same association between

market orientation and its outcome in the Japanese context as in the Western context. As noted above, because

of the restrictions in research implementation, we limited our focus to job satisfaction. Thus, we suggested the

following hypothesis:

H5: Market orientation is positively correlated with job satisfaction.

Summarizing the above discussion, we suggest the model presented in Figure 1. This model proposes that

POPs and market orientation mediate the association between transformational leadership and job satisfaction,

and that POPs negatively affect market orientation. As noted above, previous Western studies examined our

hypotheses and most of their results support the hypothesized correlations. However, no empirical studies have

tested these correlations in the Japanese context. Thus, this study examined whether the findings in the Western

context could be replicated in the Japanese context. Furthermore, no empirical study ―including those in the

Western context― has examined the integrative mediating model of these four focal constructs. Therefore, our

study aimed to not only replicate results of previous Western studies but to also investigate an unexamined

research issue.

Figure 1: The Research Model

Transformational

Leadership

Perceptions of

Organizational

Politics

Market

Orientation

Job Satisfaction

H1

H3

H4

H5

H2

－

＋＋

－

－


34

3 METHOD

3.1 Sample and Procedure

The sample consisted of 200 full-time employees working in Japanese companies. Participants were

selected from registered members of an online research program named iResearch administered by Neo

Marketing Inc., a private Japanese research company. In this program, the company invites people who wish to

participate in research and registers them as members. The company conducts various kinds of research on

requests from firms, universities, or individual researchers. Registered members who become respondents

include full-time company employees, those who are self-employed, public servants, part-time workers. From

those registered, we selected people who satisfied the requirements of our survey (i.e. full-time employees,

tenure at their organization of ≥2 years, in sales, general affairs, research and development, accounting, finance,

or planning. These requirements will be explained below), and sent them written requests by e-mail. The survey

was conducted by an online form questionnaire accessed via a URL written in the request e-mail.

We sent screening questionnaires to 13,625 randomly selected members, all of whom were company

employees. This screening questionnaire was intended to sort out members who met the research requirements.

It included questions regarding job, organization tenure, and job category. A total of 5,657 responses were

obtained, representing a response rate of 41.5%. Among them 303 respondents met the research requirements

and were then sent the main questionnaires. In this main questionnaire we collected the data regarding our four

focal constructs (i.e., transformational leadership, POPs, market orientation, and job satisfaction). We obtained

200 responses (66.0% of 303 qualified members). To ensure anonymity and confidentiality, response data were

automatically collected and converted into comma separated values. In this collection and conversion process,

participants’ names or identifying information were not recorded.

All of the respondents were Japanese and 60% were male. Ages of the respondents ranged from 25 to 68

years with a mean age of 38 years. Judging from the results of the Japanese Labour Force Survey conducted by

The Statistics Bureau and the Director-General for Policy Planning of Japan, these percentages and

distributions of gender and age nearly corresponded to that of all employees in Japan.

Because of concerns about response burden, tenure at organization was measured categorically with

responses being grouped in four blocks (i.e. 2 years and more to less than 3 years, 3 years and more to less than

4 years, 4 years and more to less than 5 years, 5 years and more). We only included respondents with 2 years

and more at their organization because a certain degree of time is necessary for an employee to understand

subtle features of their work environment, especially organizational politics. Although lacking a scientific

rationale, as a rule of thumb in Japan it is said that it takes about 3 years for new employees to be well informed

about their workplace. However, limiting respondents’ tenure to 3 years and more might have resulted in too-

small a sample. Thus, we limited respondents’ tenure to 2 years and more. We also limited respondents to full-

time employees, omitting part-time workers. In Japanese firms, full-time and part-time employees generally

work under different human resource management systems even if they are in a same workplace. A lot of part-

time workers are housewives and students. Therefore, there are large differences between full-time employees

and part-time workers in social interactions and attitudes at work (Kimura, 2011). In this study, we focus on

full-time employees and limited the scope of our research to them.

The sample included various job categories: sixty-seven (33.5%) in sales, forty-nine (24.5%) in general

affairs, forty-four (22.0%) in research and development, twenty-nine (14.5%) in accounting and finance, and

eleven (5.5%) in planning. Generally, in Japanese work settings, interactions among members are required, with

power struggles and conflicts a common and inevitable outcome in these jobs. Therefore, they are appropriate

for investigating organizational politics. Although this occupation classification is popular in Japan on a

practical level, it does not fully correspond to that of Japanese government surveys. Giving priority to

respondents’ convenience, we adopted this type of popular classification. Thus, we could not rigorously

compare the distribution of job categories in our sample with that of the entire Japanese labour force.

Nonetheless, by comparison with the results of the Japanese Labour Force survey, we would say that the job

distribution of our sample was not very different from that of the national labour force.

3.2 Measures

All items were measured on a five-point Likert scale. The anchors were “1 = strongly disagree” to “5 =

strongly agree”. The items in the scales were simply averaged to create an overall mean for each variable (i.e.,

not weighted by using the loadings of factor analysis from the construct).

Transformational leadership behaviours: Top management’s transformational leadership was measured by

nine items from Podsakoff et al.’s (1990) scale. Of the nine items, five items reflect behaviours of “articulating a

vision”. Sample items include “Our top manager has a clear understanding of where we are going.” and “Our

top manager paints an interesting picture of the future for our group.” The remaining four items reflect

behaviours of “fostering the acceptance of group goals”. Sample items include “Our top manager fosters

Takuma Kimura

35

collaboration among work groups.” and “Our top manager gets the group to work together for the same goal.”

Cronbach’s alpha for scores on this scale was 0.94.

Perceptions of organizational politics: Ferris et al. (1989) noted that political behaviour can occur in

multiple levels, namely, individual, group, and organization. Recently, some theorists have emphasized the

importance of applying a multi-level perspective to empirical analyses of organizational politics (e.g. Darr &

Johns, 2004; Fedor, Maslyn, Farmer, & Bettenhausen, 2008; Hochwarter, Kacmar, Perrewé, & Diane, 2003;

Maslyn & Fedor, 1998). The larger organization can be nonpolitical while an employee experiences high levels

of politics in his or her immediate work group (Maslyn & Fedor, 1998). Indeed, Maslyn and Fedor (1998) and

Hochwarter et al. (2003) found that employees distinguish between political behaviours occurring within their

own work groups versus those in the larger organization.

Since our focus was at the organization-level, in the analysis we focused on organization-level variables,

namely, top management’s leadership behaviours, and organization-wide market orientation. Therefore, as for

organizational politics, we also focused on organization-level POPs. We measured POPs with Maslyn and

Fedor’s (1998) four-item scale of organization-level POPs. Sample items from this scale include “There has

always been an influential department in this organization that no one ever crosses.” and “I have seen changes

made in policies here that only serve the purposes of a few individuals, not the work unit or the organization.”

Cronbach’s alpha for scores on this scale was 0.61.

Market orientation: Market orientation was measured by an eight-item scale developed by Farrell and

Oczkowski (1998). Sample items from this scale include “We monitor our level of commitment and orientation

to serving customers’ needs.” and “Our strategy for competitive advantage is based on our understanding of

customer needs.” Cronbach’s alpha for scores on this scale was 0.89.

Job Satisfaction: Job satisfaction was measured by a two-item scale developed by Hackman and Oldham

(1975). A sample item from this scale is “Generally speaking, I am very satisfied with this job.” Cronbach’s

alpha for scores on this scale was 0.82.

3.3 Factor analysis

We performed exploratory factor analysis (EFA) on these items by the principal factor method with

varimax rotation and extracted predicted factors (i.e., transformational leadership, POPs, market orientation, and

job satisfaction). Since two items of POPs, “Rewards come only to those who work hard in this organization”

(reverse scored) and “Pay and promotion decisions are consistent with existing organizational policies” (reverse

scored) demonstrated low factor loadings (i.e., lower than 0.50), we deleted these items from the analysis below.

Next, we performed four separate confirmatory factor analyses (CFAs) to assess the discriminant validity

of these constructs. CFAs were conducted on the proposed four-factor model and three alternative models.

Then, we compare the fitness of the four-factor model with those of alternative models. We used chi-square

value, Tucker-Levis index (TLI), the comparative fit index (CFI), and the root mean square error of

approximation (RMSEA) as comparative criteria. Table 1 shows the CFAs results.

Table 1: Results of Confirmatory Factor Analyses

chi-square df p TLI CFI RMSEA

4-factor model 20.415 14 .118 .982 .991 .048

3-factor model 1 33.751 17 .009 .961 .976 .070

3-factor model 2 118.735 17 .000 .765 .857 .173

1-factor model 183.937 20 .000 .678 .770 .203

Note: 3-factor model 1: transformational leadership and market orientation are combined

3-factor model 2: transformational leadership and job satisfaction are combined

The CFA results provided support for the four-factor model indicating the distinctiveness of the four

constructs. The chi-square value for the four-factor model (chi-square=20.415, df=14, p=0.12) was lower than

those for the three-factor model 1 (combining transformational leadership and market orientation; chi-

square=33.751, df=17, p<0.01), the three-factor model 2 (combining transformational leadership and job

satisfaction; chi-square=118.735, df=17, p<0.01), and the one-factor model (chi-square=183.937, df=20,

p<0.01).


36

Furthermore, the fit indices indicated a better fit for the four-factor model (TLI=0.982, CFI=0.991,

RMSEA=0.048) relative to the three-factor model 1 (TLI=0.961, CFI=0.976, RMSEA=0.070), the three-factor

model 2 (TLI=0.765, CFI=0.857, RMSEA=0.173), and the one-factor model (TLI=0.678, CFI=0.770,

RMSEA=0.203). Results of CFAs showed that the four-factor model is more appropriate than alternative

models. Thus, we proceeded to test the four-factor model.

3.4 Statistical analysis

We used structural equation modelling (SEM) methods for data analysis using AMOS 16.0. We used

maximum-likelihood estimation methods. We assessed the goodness-of-fit of our model by absolute and relative

indices (Jöreskog & Sörbom, 1993). The absolute goodness-of-fit indices calculated were (1) the chi-square

goodness-of-fit statistic, (2) the root mean square error of approximation (RMSEA), (3) the goodness-of-fit

index (GFI), and (4) the adjusted goodness-of-fit index (AGFI). The relative goodness-of-fit indices computed

were (1) the normed fit index (NFI), (2) the comparative fit index (CFI), and (3) the incremental fit index (IFI).

4 RESULTS

4.1 Descriptive statistics and correlations

Table 2 shows descriptive statistics and correlations among the study variables. This table demonstrates

that the correlations between the research variables were in the expected directions. Top management’s

transformational leadership was negatively correlated with organization-level POPs (r=-0.205, p<0.01) and

positively correlated with market orientation (r=0.755, p<0.01).

Organization-level POPs was negatively correlated with market orientation (r=-0.172, p<0.05), and job

satisfaction (r=-0.201, p<0.01). Market orientation was positively correlated with job satisfaction (r=0.499,

p<0.01).

Table 2: Means, Standard Deviations, and Correlations

Mean S.D. 1 2 3 4

1. Transformational

leadership 2.790 .917

2. Perceptions of

Organizational Politics 3.118 .992 -.205 (***)

3. Market Orientation 2.866 .773 .755 (***) -.172(**)

4. Job Satisfaction 2.915 1.001 .436 (***) -.201(***) .499(***)

Note: N=200; *p<0.10, **p<0.05, ***p<0.01

4.2 Hypotheses Testing

Figure 2 shows the SEM results of the structural model. Each numerical value shown in Figure 2 is a

standardized coefficient. The resulting fit indices indicated an acceptable fit of the model as all fit indices

showed a good fit. As for the absolute goodness-of-fit indices, chi-square=1.640 (p=0.20), RMSEA=0.057,

GFI=0.996, AGFI=0.959. As for the relative goodness-of-fit indices, NFI=0.993, CFI=0.997, IFI=0.997.

Figure 2 shows that top management’s transformational leadership was negatively correlated with

organization-level POPs (r=-0.205, p<0.01). Organization-level POPs was negatively correlated with job

satisfaction (r=-0.119, p<0.10). Thus, hypothesis 1 and hypothesis 2 were supported..

Top management’s transformational leadership was positively correlated with market orientation (r=0.751,

p<0.01). Market orientation was positively correlated with job satisfaction (r=0.478, p<0.01). Thus, hypothesis 3

and hypothesis 5 were supported. However, organization-level POPs was not correlated with market orientation

(r=0.018, p=0.71). Therefore, hypothesis 4 was not supported.

Finally, to test whether the mediation of POPs and market orientation was ‘partial’ or ‘full’, we set a

comparative ‘partial’ model which assumed partial mediation: the direct effects of transformational leadership

on job satisfaction. In this comparative model’s setting, we added an arrow from transformational leadership to

job satisfaction, and deleted the arrow from POPs to market orientation. The resulting fit indices indicated that

the fitness of the hypothesized model was better than that of the ‘partial’ model. As for the absolute goodness-

of-fit indices of the ‘partial’ model, chi-square=8.370 (p<0.01), RMSEA=0.192, GFI=0.980, AGFI=0.798. As

for the relative goodness-of-fit indices, NFI=0.971, CFI=0.974, IFI=0.974.

Takuma Kimura

37

Our SEM results supported many of the hypotheses suggested in this study. Taken together, these results

demonstrate that organization-level POPs as well as market orientation may mediate the relationship between

top management’s transformational leadership and job satisfaction. To put it another way, top management’s

transformational leadership is positively correlated with job satisfaction through reducing POPs and enhancing

market orientation.

Figure 2: Results of the Proposed Research Model (Standardized Coefficients)

Transformational

Leadership

Perceptions of

Organizational

Politics

Market

Orientation

Job Satisfaction

-.205***

.751***

-.018

.478***

-.119*

Note: N=200; *p<0.10, **p<0.05, ***p<0.01

chi-square=1.640 (p=0.20), RMSEA=0.057, GFI=0.996, AGFI=0.959, NFI=0.993, CFI=0.997, IFI=0.997

5 DISCUSSION

5.1 Findings

Our SEM results support many of our hypotheses. In line with our expectations, top management’s

transformational leadership had a significant influence on job satisfaction through organization-level POPs and

market orientation. As stated by Podsakoff et al. (1990), essential components of transformational leadership

include identifying and articulating a vision, and fostering the acceptance of group goals. These functions serve

to make employees concentrate their attention on their company’s goals and to reduce unproductive conflicts.

Thus, top management’s transformational leadership can serve to enhance organization’s market orientation and

lower the level of organizational politics.

Since organizational politics represents certain unique domains of organizational culture and climate

(Shaker, 1987), it may be that we need to examine POPs in consideration of the national culture. However, since

there has been neither theoretical nor empirical research on POPs in the Japanese setting, we developed a

correlational model based on Western literatures. Our findings are in line with previous empirical research in the

Western context (e.g., Vigoda-Gadot, 2007). We thus assumed a similar theoretical basis and expected similar

correlations to be found for various aspects of POPs in the Japanese context as has been reported for that in the

Western context.

Unexpectedly, organization-level POPs were not significantly correlated with market orientation. Although

a variety of theoretical reasons may account for this result, one possible explanation is that our measure of

organization-level POPs did not reflect the all the aspects of organizational politics in Japanese firms. Based on

the EFA results, we used only two items of Maslyn and Fedor’s (1998) POPs measure. Kacmar and Ferris

(1991) argued that POPs consist of three dimensions―“general political behaviour”, “go along to get ahead”,

and “pay and promotion”―and much recent empirical researches has adopted this view. Thus, the two items

used in this study may not have been enough to measure whole aspects of organizational politics in the

workplace. This may be one of the reasons why POPs was not correlated with market orientation.

The EFA result that two items showed low levels of factor loading might reflect work environment features

in Japanese organizations. These two deleted are matters of distributions and procedures of rewards such as pay

and promotion. In Japanese firms, pay and promotion decisions are sometimes distorted not only by

organizational politics but also by “seniority-based evaluation”. Since the 1990s, many Japanese firms have

discarded the “seniority-based personnel system” and introduced a “performance-based personnel system”

(Keizer 2010). However, attitudes of seniority orientation still exist in many workplaces in Japan. Thus, we can


38

infer that when Japanese read the sentences of these two items, some of them may have interpreted them as

items concerning seniority-based systems and not as having to do with political climate. That may be the reason

why these two items were not integrated into the factor of POPs.

5.2 Implications for Theory and Future Research

This study makes some important contributions. First, we explored correlations among transformational

leadership, POPs, market orientation, and work-related outcomes (i.e. job satisfaction). While previous

empirical research has examined the relationship between two (some three) of these variables, the present study

is the first to comprehensively explore the relationship among all of these variables.

Second, our study used a Japanese sample. Although a lot of empirical research on POPs has been

conducted over the last several decades, most of it was in the Western context. As far as we know, this is the

first empirical research on POPs using Japanese sample, and confirming, at least in part, the replicability of the

POPs model (Ferris et al., 1989) in the Japanese context. Although many of our hypotheses replicated those

made in the Western context, few studies have examined the mediating effects of POPs and market orientation

as examined in our study. In future studies, we should examine our mediating model in various contexts other

than a Japanese one.

As Ferris et al. (1996) stated, politics are inherent in the very contextual fabric of organizations. This may

be true for Japanese organizations as well because our data indicated that a considerable number of our

respondents showed relatively high POPs scores: on the average score of five-point items of POPs, 31.5% of

respondents had scores ≥3.5, 18.5% had ≥4.0, and 17.0% had ≥4.5. However, it is possible that general features

of organizational politics reflect the traits of national culture. According to Hofstede (2001), Japanese

organizations are more collectivistic than Western organizations. Thus, it may be that political behaviours in

Japanese organizations are more group or clique-centred and less individual-centred than in Western

organizations. Moreover, Japanese people have a tendency to avoid uncertainty compared with Western people.

Therefore, Japanese may be more likely to feel stress when in uncertain environments such as political

workplaces. That is, it is possible that POPs have more detrimental effects on work-related outcomes in Japan.

Although these cultural traits of organizational politics in Japan are only speculative, this conjecture can provide

an avenue of future research in Japanese organizations.

Another important contribution of this study is the multi-level perspective of the analysis. Focusing on

organization-level variables, we were able to explore correlations of top management’s leadership with

organization-level politics and market orientation. Future studies could use this approach at a group level.

5.3 Implications for Practice/Management

Our findings have useful practical implications for Japanese firms. Although organizational politics and

market orientation have received a lot of attention in practical fields, few empirical studies have examined the

relationship between these constructs and leadership in the Japanese context.

The results of our study suggest that top management’s transformational leadership is correlated with and

may lead to the reduction of POPs and the enhancement of market orientation. In the Western context, a lot of

previous empirical research has confirmed the impact of organizational politics and market orientation on a

firm’s performance. Our study suggests the importance of transformational leadership for Japanese firms

suffering from the prevalence of organizational politics and a shortage of market orientation.

5.4 Limitations

This study has several limitations. First, our analysis was based on the same source of cross sectional data

and could thus be affected by common-method variance. Because we only examined correlations, this study did

not look at causal relationships among the variables. To investigate causality, a ‘predictive’ research design in

which independent variables are measured before dependent variables is optimal. However, we relied upon a

‘contemporaneous’ design in which the independent variable, the mediating variables, and the dependent

variable were measured contemporaneously. Future research should use a longitudinal and multi-source

approach to examine questions of causality.

Second, the validity of our POPs scale is questionable. To measure POPs, recent empirical research

generally used Kacmar and Carlson’s (1997) fifteen-item, empirically validated scale (e.g., Miller & Nicols,

2008; Vigoda-Gadot & Talmud, 2010; Zetller & Hilbig 2010). Because of concerns about organization-level

POPs, we used Maslyn and Fedor’s (1998) four-item scale. However, this scale showed a 0.61 of Cronbach’s

alpha which is a little lower than desired level (Henson, 2001). Furthermore, two of them were deleted in the

analysis because of their low levels of factor loading. Thus, it is possible this scale did not sufficiently measure

organization-level POPs.

Third, we did not analyse all aspects of transformational leadership. Namely, our analysis did not include

some aspects of transformational leadership such as individualized consideration, inspirational motivation, and

intellectual stimulation (Avolio, Waldman, & Yammarino, 1991). These factors might contribute to reduce

Takuma Kimura

39

political behaviours and to strengthen market orientation by building a collective identity and mutual trust

among subunits. Finally, as for work-related outcomes, we only examined job satisfaction. Because of some

restrictions in the research implementation process―mainly, a concern for response burden―we did not include

scales of employees’ performance. Future studies should measure employees’ or organizations’ performance

and include them in research models.

6 CONCLUSIONS

Despite several limitations, this study makes some contributions by finding correlations among

transformational leadership, POPs, market orientation, and work-related outcomes. This is the first study

applying POPs to empirical research in the Japanese context, and our findings did not contradict results of

previous Western studies. Therefore, the theoretical framework of POPs generated in the Western context might

be applicable to the Japanese context.

However, our research indicated the possibility that the POPs measure should be adjusted to fit more with

the Japanese context. Furthermore, because of the characteristics of Japanese culture and the climate of Japanese

firms, in Japanese organizations it is possible that POPs have different antecedents and/or outcomes than that of

Western organizations. Differences in the mediators and/or moderators of the effects of POPs may also exist.

Future studies should consider cultural differences in more depth and reflect them in the research framework.

REFERENCES

Andrews, M. C., and Kacmar, K. M. (2001), “Discriminating among organizational politics, justice, and

support”, Journal of Organizational Behavior, Vol. 22 Issue 4, pp. 347-366.

Atinc, G., Darrat, M., Fuller, B. and Parker, B. W. (2010), “Perceptions of Organizational Politics: A Meta-

analysis of Theoretical Antecedents” Journal of Managerial Issues, Vol. 22 No. 4, pp. 494-513.

Avolio, B. J. and Bass, B.M. (1991), The Full-Range of Leadership Development, Center for Leadership

Studies, Binghamton, New York.

Avolio, B. J., Waldman, D. A., and Yammarino, F. J. (1991). “The four I’s of transformational leadership”,

Journal of European Industrial Training, Vol. 15 No. 4, pp. 9-16.

Bass, B. M. (1985), Leadership and performance beyond expectations. New York: Free Press.

Bass, B. M. and Avolio, B. J. (1993), “Transformational leadership theory: a response to critiques”, in

Chemmers, M. M. and Ammons, R. (Eds), Leadership and Research: Perspectives and Direction,

California Academic Press. Los Angeles, CA, pp. 49-80.

Burns, J. M. (1978), Leadership, Harper and Row, New York.

Cropanzano. R., Howes, J. C., Grandey, A. A. and Toth, P. (1997), “The relationship of organizational politics

and support to work behaviors, attitudes, and stress”, Journal of Organizational Behavior, Vol. 18, No. 2,

pp. 159-180.

Darr, W. and Johns, G. (2004), “Political decision-making climates: Theoretical processes and multi-level

antecedents”, Human Relations, Vol. 57, No. 2, pp. 169-200.

Dickson, M. W., Den Hartogo, D. N. and Mitchelson, J. K. (2003), “Research on leadership in a cross-cultural

context: Making progress, and raising new questions” The Leadership Quarterly, Vol.14 No.6, pp.729-768.

Drory, A. (1993), “Perceived Political Climate and Job Attitudes”, Organization Studies, Vol. 14 No. 1, pp. 59-

71.

Drory, A. and Vigoda-Gadot, E. (2010), “Organizational politics and human resource management: A typology

and the Israeli experience”, Human Resource Management Review, Vol. 20 No. 3, pp. 194–202.

Ellis, P. D. (2006), “Market orientation and performance: A meta-analysis and cross-national comparisons”,

Journal of Management Studies, Vol.43 No.5, pp.1089-1107.

Farrell, M. A. and Oczkowski, E. (1997), “An Analysis of the MKTOR and MARKOR Measures of Market

Orientation: An Australian Perspective”, Marketing Bulletin, Vol. 8 No. 3, pp. 30-40.

Fedor, D., Maslyn, J., Farmer, S. and Bettenhausen, K. (2008), “The Contribution of Positive Politics to the

Prediction of Employee Reactions”, Journal of Applied Social Psychology, Vol. 38 No. 1, pp. 76-96.

Ferris, G.R., Frink, D.D., Galang, M.C., Zhou, J., Kacmar, K.M. and Howard, J. E. (1996), “Perceptions of

organizational politics: predictions, stress-related implications, and outcomes”, Human Relations, Vol. 49

No. 2, pp. 233-266.

Ferris, G. R. and Judge, T. A. (1991), “Personnel/human resources management: A political influence

perspective”, Journal of Management, Vol. 17 No.2, pp. 447-488.


40

Ferris, G. R. and Kacmar, K. M. (1992), “Perceptions of organizational politics”, Journal of Management, Vol.

18 No. 1, pp. 93-116.

Ferris, G. R., Russ, G. S. and Fandt, P. M. (1989), “Politics in Organizations”, in Giacalone, R.A. and

Rosenfield, P. (Eds), Impression Management in the Organization. Hillsdale, NJ: Lawrence Eribaum, pp.

143-170.

Gandz, J. and Murray, V. (1980), “The experience of workplace politics”, Academy of Management Journal,

Vol. 23 No. 2, pp. 237-251.

Greenley, G. (1995), “Market orientation and company performance: empirical evidence”, British Journal of

Management, Vol. 6 No. 1, pp. 1-13.

Hackman, J. R. and Oldham, G. R. (1975), “Development of the Job Diagnostic Survey”, Journal of Applied

Psychology, Vol. 60 No. 2, pp. 159- 170.

Harris, K. J., Andrews, M. C. and Kacmar, K. M. (2007) “The Moderating Effects of Justice on the Relationship

Between Organizational Politics and Workplace Attitudes”, Journal of Business and Psychology, Vol. 22

Issue 2, pp. 135–144

Harris, R. B., Harris, K. J. and Harvey, P. (2007) “A Test of Competing Models of the Relationships Among

Perceptions of Organizational Politics, Perceived Organizational Support, and Individual Outcomes”, The

Journal of Social Psychology, Vol. 147 No. 6, pp. 631-655.

Harris, L. C. and Piercy, N. F. (1999), “Management behavior and barriers to market orientation in retailing

companies”, Journal of Services Marketing, Vol. 13 Issue 2, pp. 113-131.

Henson, R. K. (2001), “Understanding internal consistency reliability estimates: A conceptual primer on

coefficient alpha”, Measurement and Evaluation in Counseling and Development, Vol. 34 No. 3, pp. 177-

189.

Hochwarter, W. A., Kacmar, C., Perrewé, P. L. and Diane, J. (2003), “Perceived organizational support as a

mediator of the relationship between politics perceptions and work outcomes”, Journal of Vocational

Behavior, Vol. 63 No. 3, pp. 438-456.

Hofstede, G. (2001), Culture’s consequences: Comparing values, behaviors, institutions, and organizations

across nations (2nd ed.), Newbury Park, CA: Sage.

Howell, J. M. and Avolio, B. J. (1993). “Transformational leadership, transactional leadership, locus of control,

and support for innovation: Key predictors of consolidated business unit performance”, Journal of Applied

Psychology, Vol. 78 No. 6, pp. 891-902.

Huang, I., Chuang, C. J. and Lin H. (2003), “The Role of Burnout in the Relationship Between Perceptions of

Organizational Politics and Turnover Intentions Public Personnel Management”, Vol.32 No.4, pp.519-532.

Ishikawa, J. (2012), “Leadership and performance in Japanese R&D teams”, Asia Pacific Business Review, Vol.

18, No. 2, pp.241-258.

Jaworski, B. J. and Kohli, A. K. (1993), “Market orientation: antecedents and consequences”, Journal of

Marketing, Vol. 57 No. 3, pp. 53-70.

Jöreskog, K. G. and Sörbom, D. (1993), Lisrel 8: Structural equations modeling with the SIMPLIS command

language. Chicago: Scientific Software International.

Kacmar, K. M. and Baron, R. A. (1999), “Organizational Politics: The State of the Field, Links to Related

Processes, and an Agenda for Future Research”, in Ferris, G.R. (Ed.), Research in Personnel and Human

Resources Management. JAI Press, Greenwich, CT, pp. 1-39.

Kacmar, K. M. and Carlson, D. S. (1997), “Further Validation of the Perceptions of Politics Scale (POPS): A

Multiple Sample Investigation”, Journal of Management, Vol. 23 No. 5, pp. 627-658.

Kacmar, K. M., and Ferris, G. R. (1991), “Perceptions of organizational politics scale (POPS): Development and

construct validation”, Educational and Psychological Measurement, Vol. 51 No. 1, pp. 193-205.

Keizer, A. B. (2010) Changes in Japanese Employment Practices, Routledge: London & New York.

Kimura, T. (2011) “Hiseikiroudousya no kujosyori to syokuba deno komyunikeisyon” Hiseikiroudousya no

‘Hatsugen’ no kakudai to kyaria appu. Japanese Trade Union Confederation Research Institute for

Advancement of Living Standards. pp.56-79.

Kirca, A. H., Jayachandran, S. and Bearden, W. O. (2005), “Market orientation: A meta-analytic review and

assessment of its antecedents and impact on performance”, Journal of Marketing, Vol. 69 No. 2, pp. 24-41.

Klenke-Hamel, K. E. and Mathieu, J. E. (1990), “Role Strains, Tension, and Job Satisfaction Influences on

Employees’ Propensity to Leave: A Multi-Sample Replication and Extension”, Human Relations, Vol. 43

No. 8, pp. 791-807.

Takuma Kimura

41

Lewin, K. (1936), Principles of topological psychology. New York: McGraw-Hill.

Liu, Y., Liu, J. and Wu, L. (2010), “Growth Are You Willing and Able? Roles of Motivation, Power, and

Politics in Career”, Journal of Management, Vol. 36 No. 6, ember, pp.1432-1460.

MacKenzie, S. B., Podsakoff, P. M. and Rich, G. A. (2001), “Transformational and transactional leadership and

salesperson performance”, Journal of Academy of Marketing Science, Vol. 29 No. 2, pp. 115-134.

Maslyn, J. and Fedor, D. B. (1998), “Perceptions of politics: does measuring different foci matter?”, Journal of

Applied Psychology, Vol. 84 No. 4, pp. 645-653.

Madzar, S. (2005), “Subordinates’ information inquiry in uncertain times: a cross-cultural consideration of

leadership style effect”, International Journal of Cross-Cultural Management, Vol.5 No.3, pp.255-273.

Menguc, B. and Auh, S. (2008), “Conflict, leadership, and market orientation”, International Journal of

Research in Marketing, Vol. 25 Issue 1, pp. 34-45.

Menguc, B., Auh, S. and Shih, E. (2007), “Transformational leadership and market orientation: Implications for

the implementation of competitive strategies and business unit performance”, Journal of Business

Research, Vol. 60 No. 4, pp. 314-321.

Miller, B. K. and Nicols, K. M. (2008), “Politics and Justice: A Mediated Moderation Model”, Journal of

Managerial Issues, Vol. 10 No. 2, pp. 214-237.

Miller, B. K., Rutherford, M. A. and Kolodinsky, R. W. (2008), “Perceptions of Organizational Politics: A

Meta-analysis of Outcomes”, Journal of Business and Psychology, Vol. 22 No. 3, pp. 209-222.

Mintzberg, H. (1983), Power in and around organizations, Englewood Cliffs, NJ: Prentice-Hall.

Muenjohn, N. and Armstrong, A. (2007), “Transformational Leadership: The Influence of Culture on the

Leadership Behaviours of Expatriate Managers”, International Journal of Business and Information, Vol.2

No.2, pp.265-283.

Narver, J. C. and Slater, S. F. (1990), “The effect of a market orientation on business profitability”, Journal of

Marketing, Vol. 54 No. 4, pp. 20-35.

Narver, J. C. and Slater, S. F. (1993), “Market orientation and customer service: The implications for business

performance”, European Advances in Consumer Research, Vol. 1, pp. 317-321.

Narver, J. C., Slater, S. F. and Tietje, B. (1998), “Creating a market orientation”, Journal of Market Focused

Management, Vol. 2 No. 3, pp. 241-255.

Nwanko, S., Owusu-Frimpong, N. and Ekwulugo, F. (2004), “The effects of organisational climate on market

orientation: evidence from the facilities management industry” Journal of Services Marketing, Vol. 18

Issue. 2, pp. 122-132.

Nye, L. G. and Witt, L. A. (1993), “Dimensionality and construct validity of the perceptions of organizational

politics scale (POPS).” Educational and Psychological Measurement, Vol. 53 No. 3, pp. 821-829.

Parry, K. W. (2003), “Leadership, culture and performance: the case of the New Zealand public sector”, Journal

of Change Management, Vol. 3 No. 4, pp. 376-399.

Pfeffer, J. (1981), Power in Organizations, Boston, MA: Pittman.

Podsakoff, P. M., MacKenzie., S. B., Moorman, R. H. and Fetter, R. (1990), “Transformational Leader

Behaviors and Their Effects on Followers’ Trust in Leader, Satisfaction, and Organizational Citizenship

Behaviors”, Leadership Quarterly, Vol. 1 No. 2, pp. 107-142.

Poon, J. M. L. (2003). “Situational antecedents and outcomes of organizational politics perceptions”, Journal of

Managerial Psychology, Vol. 18 Issue 2, pp. 138-155.

Poon, J. M. L. (2006), “Trust-in-supervisor and helping coworkers: moderating effect of perceived politics”,

Journal of Managerial Psychology, Vol. 21 No. 6, pp.518-532.

Randall, M. L., Cropanzano, R., Bormann, C. A. and Birjulin, A. (1999), “Organizational politics and

organizational support as predictors of work attitudes, job performance, and organizational citizenship

behavior”, Journal of Organizational Behavior, Vol. 20 Issue 2, pp. 159-174.

Ruekert, R.W. (1992), “Developing a market orientation: an organizational strategy perspective”, International

Journal of Research in Marketing, Vol. 9 Issue 3, pp. 225-245.

Shaker, Z. (1987), “Organizational politics and the strategic process”, Journal of Business Ethics, Vol. 6 Issue 7,

pp. 579-587.

Slater, S. F. and Narver, J. C. (1994), “Does Competitive Environment Moderate the Market Orientation-

Performance Relationship?”, Journal of Marketing, Vol. 58 No. 1, pp. 46-55.

Tsujino, K. (2010), Google de hitsuyouna kotoha, minnna Sony ga oshietekureta, Shinchosya.


42

Vigoda-Gadot, E. (2007), “Leadership style, organizational politics, and employees’ performance: An empirical

examination of two competing models”, Personnel Review, Vol. 36, No. 5, pp. 661-683.

Vigoda-Gadot, E. and Talmud, I. (2010), “Organizational Politics and Job Outcomes: The Moderating Effect of

Trust and Social Support”, Journal of Applied Social Psychology, Vol. 40 No. 11, pp. 2829-2861.

Walton, R. E. and Dutton, J. M. (1969), “The Management of Interdepartmental Conflict: A Model and Review”

Administrative Science Quarterly, Vol. 14 No. 1, pp. 73-84.

Wang, H., Law, K. S., Hackett, R. D., Wang, D. and Chen, Z. X. (2005), “Leader–member exchange as a

mediator of the relationship between transformational leadership and followers’ performance and

organizational citizenship behavior”, Academy of Management Journal, Vol.48 No.3, pp.420-432.

Wei, L-Q., Chiang, F. F. T. and Wu, L-Z. (2012), “Developing and Utilizing Network Resources: Roles of

Political Skill”, Journal of Management Studies, Vol. 49 No. 2, pp. 381-402.

Wei, L-Q., Liu, J., Chen, Y-Y. and Wu, L-Z. (2010), “Political Skill, Supervisor–Subordinate Guanxi and

Career Prospects in Chinese Firms”, Journal of Management Studies, Vol. 47 No. 3, pp. 437-454.

Yammarino, F. J. and Dubinsky, A. J. (1994), “Transformational leadership theory: using levels of analysis to

determine boundary condition”, Personnel Psychology, Vol. 47 No. 4, pp. 787-811.

Zettler, I. and Hilbig, B. E. (2010), “Honesty-Humility and a Person-Situation Interaction at Work”, European

Journal of Personality, Vol. 24 Issue 7, pp. 569-582.


Developing a performance measurement system for public

research centres

Deborah Agostino

Dipartimento di Ingegneria Gestionale, Politecnico di Milano

Piazza Leonardo da Vinci, 32 - 20133 Milano, Italy

Telephone: +39 02.2399.4073


Marika Arena



Telephone: +39 02.2399.4070


Giovanni Azzone



Telephone: +39 02.2399.3981


Martina Dal Molin



Telephone: +39 02.2399.4044


Cristina Masella



Telephone: +39 02.2399.4081


Abstract

This study aims at developing a performance measurement system (PMS) for research and development (R&D)

activities carried out by public research centres. Public research institutions are characterized by multiple

stakeholders with different needs, and the management of R&D activities requires balancing the multiple goals

of different stakeholders. This characteristic is a key issue in the process of construction of the PMS. Empirical

evidence is provided by an Italian public research centre, where the researchers carried out a project aimed to

develop a PMS following action research principles. This project gave the possibility to researchers to interact

with different stakeholders and integrate their different information needs in a comprehensive set of key

performance indicators (KPIs). As a result, multidimensional framework for measuring R&D performance in a

public research centre is proposed and a set of Key Performance Indicators is developed, suggesting

implications for academics and practitioners.

Keywords: performance measurement, public research centres, stakeholders, accountability, decision-making


44

1 INTRODUCTION

In recent years public research systems in several industrialized countries, underwent an intense

transformation process due to the increasing organizational complexity of research & development (R&D)

activities and the hit of the worldwide financial crisis. Leading research institutions claim that many complex

problems of the society demand innovative solutions which combine knowledge from different scientific areas

that can be achieved through interdisciplinary research (National Academies, 2005). To achieve this, very often,

research centres carry out interdisciplinary R&D projects, which require collaborative and sometimes informal

behaviours (Cross, Borgatti and Parker, 2002; Allen, James, & Gamlen, 2007) and new forms of organization

and management (Welter, Jooß, Richert, Jeschke, & Brecher, 2012). At the same time, the financial crisis has

reduced the overall government spending and specifically the budget for research centres (e.g. Arena

&Arnaboldi, 2013). This reduction of funding has increased the competition between research institutions, while

at the same time they are required to demonstrate their value for money. To face these challenges, research

institutions need to consciously manage their core processes, the creation and development of their knowledge

assets (Rowley, 2000), and consistently redesign their support processes and managerial instruments (e.g.

Arena, Arnaboldi, Azzone & Carlucci, 2009; Arena, Arnaboldi & Azzone, 2010a;Welter et al. 2012).

In this context, our paper focuses on one specific managerial instrument - i.e. performance measurement

system (PMS). In the last decades, measuring R&D performance has become a fundamental concern for R&D

managers and executives. Scholars and practitioners have already recognized the relevance of performance

measurement in R&D in relationship to different purposes (e.g. Chiesa, Frattini, Lazzarotti&Manzini, 2009;

Kulatunga, Amaratunga&Haigh, 2011). A PMS can be useful for motivating researchers, evaluating R&D

projects’ profitability, supporting decision-making, communicating the centre’s results to external

constituencies, and stimulating organizational learning.

However, most of the current research about PMS in R&D focuses on private sector organizations (e.g.

Bremser&Barsky, 2004; Chiesa et al., 2009) and these results are not easily applicable to Government-funded

research centres, since they overlook some key characteristics of these organizations. In industrial firms, R&D

activities are primarily financed by the company itself and they represent one of the activities in their value

chain. In this context, the company does not have the need of searching for research funding, given that it is self

financed and research outputs represent the input for further processes of the firm’s value chain. The research

results are incorporated into products and, in the end, brought to the market and sold by the company,

‘hopefully’ resulting in an increase in the revenue and profits of the firm, and contributing amortize the R&D

investment. On the contrary, in Government-funded research centres, such as public research institutions, the

research activity represents the core mission of the organization and the R&D activity needs to be financed by

external parties. This structure has two main implications. On the one hand, the role played by the research

activity is central, because the output of the research represents the end itself. Research institutes contribute to

the early stages of the innovation process of various customers within the national innovation system and serve

thus as an important research infrastructure, (Shenker, 2001; Leitner & Warden, 2004). On the other hand, the

need for searching funding is associated with the pressure of demonstrating the ability of generating research

outputs that provide value for the society. The output of public research centres is expected to have a positive

impact on the wider society; their principal aim is to disseminate the research results and to have a return in

terms of scientific-technological progress, driving the national strength (Senker, 2001; Coccia, 2004). However,

once the funding are received, mainly by the public government (Senker, 2001; Coccia, 2004), the research

centre has to be accountable on how it spend public money. In this sense, Coccia (2005) argued that research

institutions are dependent on the government for carrying out their activities. As consequence of these

specificities, public research centres are subject to different and contrasting stakeholders pressures that are wider

compared to private sector organizations (Dixit, 2002; Coccia, 2004; Arnaboldi, Azzone & Savoldelli, 2004).

Public government, the wider society, the researcher, the administrative manager but also private partners, have

different objectives and their alignment can be extremely complex, especially in the current R&D landscape

characterized by reduced finance and higher competition between institutions.

To our best knowledge, relatively a few contributions have dealt with the issue of R&D performance

measurement in the public sector (e.g. Coccia, 2001a; 2004; Leitner & Warden, 2004; Secundo, Margherita, Elia

& Passiante, 2010) and a comprehensive framework for measuring performance of public research institutes is

still missing, setting the motivation for this work. This paper has the objective to develop a performance

measurement system (PMS) for a public research centre. The centre has to deal with the informative needs of a

diversified range of stakeholders that require accountability in relationship to the centre’s activities and, in

particular, the use of funding, but also aim to encourage collaboration between different research units and

support decision making processes. The PMS has been developed through action research, whereby the

researchers interacted with different stakeholders and integrated their different information needs to a

comprehensive set of key performance indicators (KPIs). Interviews and meetings with scientific and

Deborah Agostino, Marika Arena, Giovanni Azzone, Martina Dal Molin and Cristina Masella

45

administrative directors over a time horizon of ten months were useful for designing a dashboard for public

research institutions that integrates the information needs of a plurality of stakeholders.

The paper is organized as follows. The next section outlines the state of the art on the issue of performance

measurement for R&D activities. The third section details the methodology used for data collection and

analysis, describing the phases of the research project. The fourth section presents the PMS framework and the

set of indicators that have been developed. Finally, the last section clarifies contributions and limitations of this

research.

R&D performance measurement in research centres

Various studies on public research centres show a growing interest in R&D performance measurement

(Luwel, Noyons & Moed, 1999; Senker, 2001;Coccia, 2004; Coccia & Rolfo, 2007). Originally, particular

attention has been given to the use of bibliometric and technometric indicators to evaluate public research centre

performances (e.g. Narin & Hamilton, 1996; Luwel et al., 1999; Noyons, Moed & Luwel, 1999). Bibliometric

involves the quantitative analysis of the scientific and technological literature. It relies on the assumption that

the frequency with which a paper or a patent is cited is a measure of the impact of the paper or patent. The most

common bibliometric indicators are literature indicators – i.e. indicators that measure the scientific performance

based on the number of publications and the count of citations in the scientific literature, and, obviously, the

higher the better. Similar considerations are applicable to patents too. Those patents which are most highly cited

and most science linked are also the patents that tend to be most heavily licensed, and are therefore making the

largest contribution to the economy (Narin & Hamilton, 1996).However, bibliometric indicators provide a

partial and non-systemic picture of the performance of a research centre, because they overlook some relevant

elements such as the impact of the research activity on the society and they ignore the role played by

government funding (Coccia, 2004).

Moving from these considerations, Coccia proposed a weighted index to provide a synthetic measurement

of public research’s activities - Relev Model I and Relev Model II - (Coccia2001b; 2004, 2005). The Relev

Model is based on an input-output model, where inputs consist in public funds, personnel payrolls and cost of

labour, and outputs are represented by self-financing deriving from activities of technology transfer, training

(e.g. number of degree students, PhD students), teaching (measured as the number of courses held by

researchers), international and domestic publications, international and domestic conference proceedings. Based

on a weighted input – output ratio, the Relev model attempts to synthesize the performance of public research

centres in a unique score. This score has the final objective to support external accountability, providing

evidence of the value for money generated by these institutions. Indeed, these studies generally adopt one

specific perspective. They reflect the aim of governments to define metrics that could be used to assess the

research centres performances in order to facilitate the identification of the most and the least productive

laboratories, and to support policy decisions on the level and the direction of the public funding of research

(Coccia, 2004). These kinds of metrics, instead, are less useful to support internal decision making and report a

centre’s performances to a broader range of stakeholders.

To respond to this need, a few studies started to focus on the use of key performance indicators for

assessing R&D results in public sector research centres (e.g. Leitner & Warden, 2004; Chu, Lin, Hsiung & Liu,

2006; Secundo et al., 2010). This stream of research is based on Intellectual Capital reporting models (e.g.

Stewart, 1997; Edvinsson & Malone, 1997). At a general extent, the Intellectual Capital encompasses three

components: human capital, structural capital, and relational capital. The human capital refers to the

characteristics of the entire organization’s staff and management. The structural capital refers to organizational

infrastructure. The relational capital refers to the establishment, development and maintenance of relationships

with external stakeholders. The combination and integration of these forms of capital determines the entity’s

results that include both economic performances and other intangible results, such as research- and society-

oriented results (Leitner & Warden, 2004). These approaches to performance measurement provide a more

complete picture of the resources available to research centres, however they seem to overlook the performances

of the centres in the transformation process of inputs in outputs (e.g. they overlook issues such as efficiency and

productivity).

Finally, further evidence in relationship to R&D performance measurement is provided by the private

sector literature, where the topic has been investigated from two different perspectives. In particular, one stream

of research focuses on the choice of the performance dimensions and the performance indicators that are best

suited to the characteristics of R&D. Some of these works consist in the application of well-known PMS

frameworks, such as Balanced scorecard, Skandia navigator, Intangible asset monitor, to R&D activities (e.g.

Kerssens-Van Drongelen & Bilderbeek, 1999; Bremser & Barsky, 2004). Other works consist in the

development of ad-hoc frameworks and set of indicators to assess the performance of R&D (e.g. Pawar &

Driva, 1999; Kim& Oh, 2002; Marr, Schiuma & Neely, 2004; Mettänen, 2005). These papers can provide a

reference point for the identification of performance indicators for R&D, though they are not tailored to the

specific characteristics of public sector organizations. The second stream of research, instead, focuses on the


46

strategic design of the PMS in R&D firms and addresses the relationship between the characteristics of the

PMS, the objective of the PMS and contextual variables, such as type of R&D activity, industry and company’s

size (e.g. Kerssens-van Drongelen & Cook, 1997; Chiesa & Frattini, 2007; Chiesa et al. 2009). These papers

highlight how the PMS can be configured in a different way on the basis of the objectives that it is supposed to

pursue (e.g. diagnostic, motivation and coordination), hence suggesting to pose particular attention to the

different objectives of the stakeholders of a research centre.

The above review has highlighted that there has been widespread dispute among researchers and

practitioners about which measures are more suitable in assessing the performance of research organizations,

and, at present, this debate is still open (Elmquist & Le Masson, 2009; Whelan, Teigland, Donnellan & Golden,

2010; Dumay & Rooney, 2011).

2 RESEARCH METHOD

Choice of the research method

This paper reports the results of a project aimed at developing a PMS for an Italian technological research

centre. To carry out this goal we adopted action research (AR) as research methodology. Action research has

been selected given its aim “to contribute both to the practical concerns of people in an immediate problematic

situation and to the goals of social science by joint collaboration within a mutually acceptable ethical

framework” (Rapoport, 1970: 499). On the one hand, this methodology gives the possibility to focus on a

problematic practical situation, the need to introduce a PMS to deal with the current complex research

environment that represents at the same time a relevant academic concern. On the other hand, the collaborative

nature of the interaction between the researchers and the employees of the research centre allowed capturing the

specific requirements from each stakeholder supporting the definition of the most appropriate set of KPIs.

Specifically, we conducted AR drawing on the seminal work of Susman & Evered (1978). According to the

authors, AR is seen as a cyclical process articulated in five phases: diagnosing, action planning, action taking,

evaluating, and specifying learning (see Figure 1). At the centre of the cycle there is a client system that is the

social system in which the members face problems to be solved by action research. Different techniques were

used for data collection and analysis, including in-depth interviewing, direct observation and documental

analysis (data were drown from the records, memos, and reports that the client system routinely produces). The

interviews were recorded and transcribed, though the empirical material was not codified, but instead analyzed

textually, with each author highlighting emergent themes. Overall 64 interviews were performed and three

official plenary meetings took place; the process was highly participative (Reason & Bradbury, 2001) and

different stakeholders involved continued to provide comments about the researchers’ proposals.


47

Figure 1: The action research cycle (Susman and Evered, 1978)

ACTION PLANNING

Considering alternative

courses of action for

solving a problem

ACTION TAKING

Select a course of action

EVALUATING

Studying the

consequences of action

SPECIFIYING LEARNING

Identifying general f indings

DIAGNOSING

Identifying or def ining a

problem

Development

of a client

system

inf rastructure

The client system infrastructure

The client system is a public research centre in Italy, active in different fields of technological

development. The centre currently employs more than 1,000 people, with more than 900 researchers. The centre

is governed by an executive committee that defines the centre’s long term strategies and is in charge of its

management. It is characterized by a dual structure: the scientific director is responsible for all the research

activities of the centre; the administrative director, on the other hand, is responsible for the administrative

structure. The distinctive characteristic of the centre is represented by interdisciplinary R&D activities that cover

several fields, ranging from robotics, neuroscience, energy, to smart materials and drug discovery. The

interdisciplinary nature of the research activity is then reflected in the organizational structure that consists of 12

central research units, 11 research centres located at the premises of, and in collaboration with, other research

institutions both in Italy and abroad, and 3 facilities that provide support services to the research units (e.g.

animal facility).

At the beginning of the project, a working group was constituted within the client system, including the

Head of the Management Control Office and a member of its staff, the Head of the ICT Office, and two internal

consultants. The working group interacted with the researchers in all the phases of the AR cycles. In addition,

different employees of the centre were involved as informants and prospect users of the PMS (see Table1).


48

Table1: The client system

Participants and informants Organizational unit Role in the project

Scientific Director Central administration Initiator and user

Administrative Director Central administration Initiator and user

Head of the Management Control Office Central administration Head of the client working group

Management and Control Office Staff Central administration Member of the client working group

Head of ICT Office Administrative office Member of the client working group

Consultant 1 External Member of the client working group

Consultant 2 External Member of the client working group

Head of RU 1 Research unit Informant and user










Head of RC 1 Research centre Informant and user



Senior Scientist 1 Research unit Informant and user

Senior Scientist 2 Research unit Informant

Senior Scientist 3 Research unit Informant

Senior Scientist 4 Research facility Informant

Technician 1 Research facility Informant

Technician 2 Research unit Informant

Technician 3 Research unit Informant

Team Leader 1 Research unit Informant

Team Leader 2 Research unit Informant

Head of RF 1 Research facility Informant and user



Head of Research Office Support office Informant and user

Head of Project Office Support office Informant and user

Head of Technology Transfer Office Support office Informant and user

Head of Human Resource Office Administrative office Informant

Head of Engineering office Administrative office Informant

Head of Health and Safety Administrative office Informant

Head of Purchasing Office Administrative office Informant

Head of Legal Office Administrative office Informant

Head of Internal Audit Office Administrative office Informant

Head of Accounting Department Administrative office Informant


49

The AR Cycles

In this section we outline the phases of the two AR cycles, posing particular attention to specifying the

sources used for data collection in each step of the research.

AR Cycle I

The diagnosing phase started with a literature review concerning performance measurement in research

organizations, a series of project meetings and a brainstorming session on the role of the performance

measurement system in the specific setting. The outcome of this phase was the identification of the process-

oriented model (Pollanen, 2005) as a starting point for the development of the PMS and the acknowledgement

of the existence of different information needs in relationship to the PMS from a plurality of stakeholders.

Action planning consisted in mapping data and information needed to support the development of the PMS. A

list of informants and relevant documentation was prepared by action researchers in collaboration with a

working group of the research centre. In addition a list of critical issues (interview protocol) to be discussed with

the informants was prepared. In the action taking phase, 38 interviews were carried out with the heads of

scientific units and researchers of the centre, and also the administrative personnel. The analysis was

complemented with public documents and confidential reports, entering in detail the scientific activity carried

out by each department. Based on these data, a prototype of the PMS model and a preliminary set of KPIs were

developed. In the evaluating phase the proposed model and the preliminary set of KPIs were presented in two

plenary meetings, one involving the scientific director, the administrative director and a selected number of

other internal officers and one involving the heads of the research units. Feedback, comments and suggestions

for improvements were collected. Finally, specifying learning consisted in summing up the learning outcomes of

the AR cycle. All the comments and feedbacks deriving from the evaluation phase were integrated. The

outcome of this phase was twofold: the validation of the overall PMS model and an ‘explosion’ in the number of

KPIs due the receipt of many proposals from the participants. Overall 42 indicators and several variations to

each of them were suggested, posing the basis for the second AR cycle, aimed to the refinement of the set of

indicators. The following table outlines the techniques used for data collection and analysis in different steps of

the AR cycle, the role of the researchers and the client working group and the output produced (Table 2).

Table 2: AR Cycle I

Techniques Role of the researchers Role of the client

working group

Output

Diagnosing

phase

Literature review,

Project meetings,

Brainstorming meetings

Challenge the

informants to identify

emerging issues

Contribute to the

discussion

Definition of the role of

the PMS in the research

centre

Action

planning

Literature review

Project meetings

Define the agenda and

design data collection

tools

Support the researchers

in identifying potential

informants and provide

feed-backs and

suggestions

Design of the data

collection tools to

support the preparation

of the PMS prototype

Action

taking

Literature review

Interviews

Project meetings

Perform interview and

data analysis

Contribute to the

discussion and

participate in data

analysis

Development of the

PMS overall model and

preliminary set of

indicators

Evaluating Plenary meetings Present the PMS

prototype

Contribute to the

discussion

Presentation of the PMS

prototype in two plenary

meetings

Specifying

learning

Internal meetings Collect and analyse

feed-backs and

suggestions

Support researchers in

interpreting feed-backs

and suggestions

Integration of feed-backs

and suggestions


50

AR Cycle II

The diagnosing phase started with a structured analysis of the comments received concerning the set of

indicators and the acknowledgement that the proposals received needed to be organized and selected. Action

planning consisted in the definition of shared selection criteria to reduce the number of the indicators, still

ensuring that the information needs of different stakeholders were met. To this aim, the following selection

criteria were adopted: relevance, measurability, cost and timeliness (Lynch and Cross, 1991; Neely et al., 2003).

In addition, to support the selection of the KPIs, a second round of 26 interviews was planned with selected

informants. Action taking consisted in carrying out the interviews with an application of the selection criteria to

define the new set of KPIs. For each KPI, an information protocol was prepared (see Table 4). It defines in

operational terms how the indicator should be computed, the unit of measure, the level of detail, the relevance of

the indicator, the frequency of data collection and the owner of these data (see also Arena & Azzone, 2010).

Table 3: The information protocol

KPI Number of Patents

Definition Number of patents distinguished by category

Computation procedure Number of patents of the year distinguished by the following categories:

First filing

Extension abroad

Abandoned

Granted

Data source: Patents database

Data calculated over the following time horizon: year

Unit of measure Units

Level of detail Scientific structure

Research unit

Research team

Relevance of the indicator This KPI evaluated the output of the research centre and its related research units in

terms of patents, with respect to the different phases of a licensing process (first filing,

extension abroad, abandoned and granted)

Frequency of data collection Year

Owner of the measure Technology Transfer Office

In the evaluating phase the results were presented in a new plenary meeting involving the scientific and

administrative directors. Feed-backs we received mainly dealt with some specifications in the information

protocols in the indicators, leading to new revisions. Hence, after this second round of evaluation the final set of

KPIs was proposed suggesting practical directions for the research institute, but also general guidelines at the

academic level (see Table 4).


51

Table 4: AR Cycle II

Techniques Role of the researchers Role of the client

working group

Output

Diagnosing

phase

Project meetings Identify key issues based

on the output of the first

AR cycle

Contribute to the

discussion

Identification of the key

issues

Action

planning

Literature review,

Project meetings

Define the selection

criteria and design data

collection tools

Support the researchers

in identifying potential

informants and provide

feed-backs and

suggestions

Design of the selection

criteria and data

collection tools to

support the PMS

revision

Action

taking

Project meetings,

Interviews

Perform interview and

data analysis

Participate in data

analysis and support the

researchers in the

definition of the

information protocols.

Development of the final

set of indicators based

on the selection criteria

Evaluating Plenary meetings Present the final set of

KPIs

Contribute to the

discussion

Final presentation

Specifying

learning

Project meetings Collect and analyse

feed-backs and

suggestions

Practical directions for

the research institute, but

also general guidelines

at the academic level

3 RESULTS

Results are organized in three sections. First, we introduce the overall PMS model; then, we outline

different performance dimensions; finally, we present the complete list of KPIs.

The PMS model

The starting point for the development of the PMS model is the production system process of a R&D

organization. It can be represented through a process-oriented model(Brown & Svenson, 1998; Pollanen, 2005),

in which research inputs are processed and transformed into output, such as publications or patents. Accordingly

to this general model, performance measures can be defined on the basis of three interrelated elements: input,

output and outcome. Input refers to the amount of resources used in performing a certain activity; output refers

to the result of a transformation process; outcome refers to the long-term impact of the output on the external

environment(Lettieri & Masella, 2009). Based on the three above elements, three performance dimensions can

be identified, namely effectiveness, efficiency and impact (Azzone, 2008). Effectiveness refers to the output

characteristics, both quantitatively and qualitatively; efficiency refers to the ratio between output and input;

impact is a measure of the outcome and it refers to the long-term effects of the output on the external context.

Interaction with research and administrative personnel was useful in recognizing the importance of the

aforementioned dimensions. The quality of the output, the impact of the output on the society and the amount of

resources required for generating the output emerged as crucial aspects for successfully manage the research

centre. However, two further aspects were highlighted as central and they were consequently added to the

traditional model: risk and network. The risk is associated with the uncertainty that characterizes the research

activity. It is widely acknowledged that the context in which research institutes operate is characterized by

continuous changes and high variability (Leitner & Warden, 2004). When discussing with the administrative

and the research director, a question continuously emerged was the following: do we have enough resources, in

terms of quantity and quality, in order to deal with the unpredictability of the events, such as a change in

technology or a budget reduction? The importance of posing the attention to the quantity and quality of

resources to deal with uncertainty, led us to complement the traditional model with the risk dimension. The

network dimension is specifically related to the research activity. The heads of research units underlined several

times the importance of working collaboratively in order to achieve results. Publications as well as projects or

patens are rarely the result of a single researcher; rather, they derive from the joint working of more teams,

belonging to the same department, but also to different departments or different research institutions. This

reason justifies the choice of adding a further dimension to the model, the network, in order to capture the ability

of working collaboratively and output of this collaboration.

The final model we obtained (Figure 2) is characterized by five dimensions, suggesting that the holistic

management of a public research centre requires to consider the quantity and the quality of the output

(effectiveness), the impact of the output on the society (impact), the ability in transforming input into output

(efficiency), the quantity and the quality of the available resources (risk) and the ability and effects of working

collaboratively (network). Following, each dimension will be discussed separately, posing the attention on the

different relative importance given to stakeholders.


52

Figure 2: The proposed PMS model

RESEARCH

CENTERINPUTS OUTPUTS

Effectiveness

Publications

Projects

Patents & licences

Technology transfer

Human resources

Financial resources

Equipments

COMMUNITYOUTCOME

Impact Efficiency

Risk

Network

The performance dimensions

Effectiveness

The effectiveness dimension allows to evaluate the output of the R&D activities defining the level of

achievement of research objectives (Garcia-Valderrama and Mulero-Valdigorri, 2005). This aspect is related to

a key issue that is what is the output of a research centre and how to measure it (Steiner and Nixon, 1997). In the

case at hand, the output is represented by publications, funded projects, patents and technology transfer

activities. Publications and projects were the ‘immediate answer’ when the informants were asked about the

output of the research centre. This answer is in line with extant studies on research institute that recognized that

publication, books and reports represent the explicit transfer of knowledge of research bodies (e.g. Coccia,

2001b). On the other hand, patents and technology transfer activities emerged in a more ‘fragmented’ way.

Some scientific directors and researchers were fully aware of the relevance of these activities, as emerges from

the following quotation:

“Technology transfer is a key pillar in the mission of our institution, otherwise,

what are we here for?” (Head of RU 8).

Whilst in the other case, patenting was seen somehow as a secondary activity that can be explained

considering that the institute was at that time measuring outputs in terms of number of scientific publications

and funding obtained through competitive projects. The Head of the Technology Transfer Office pointed out:

“Technology Transfer activity is a green field. It needs to be organized and then

evaluated because in the early years of the research centre, the attention has been

attracted by scientific publications only. Yet we are growing and we have now a

portfolio of 68 patents and several commercial projects. They need to be monitored

and managed” (Head of Technology Transfer Office)

Based on these considerations, seven KPIs were defined to measure the research centre effectiveness: four

of them aim to evaluate the quantity of the output and the other three the quality (see Table 5 for the details of

the KPIs).

It is interesting to highlight that there was a polarization in the perceived usefulness of different aspects of

effectiveness depending on the personnel qualification. The administrative personnel and the executive

committee were mainly interested in the quantity of the output generated in terms of ‘how much money’, ‘how

much patents’, ‘how much publications’ (several informants). They considered these KPIs particularly useful to

support external accountability; in fact they have been included in the annual report and have been published on

the website of the research institute. Instead, the scientific personnel considered these measures too ‘generic’


53

and posed particular attention to the quality of the output of research activities, highlighting the importance to

account for ‘good publications’, ‘impact on the scientific community’ and ‘participation in breakthrough

innovation projects’ (several informants). From this perspective, the Impact Factor (IF) and Citation Index (CI)

distribution were evaluated particularly useful to support motivation and decision making processes of research

units. Finally, the scientific director underlined the importance of the integration of the two types of indicators,

for a decision making purpose, to have an overall idea of the productivity of each RU. At the same time, he also

recognized the limitations of these figures: the impossibility to compare them between RUs because their

average value varies widely from one research area to another one.

Efficiency

The efficiency dimension evaluates the amount of resources used to generate the output, (i.e. publications,

funded projects, patents, and technology transfer activities). It represents a relevant aspect in publicly funded

services as governments are interested in receiving feedback on how public resources are used (Coccia 2001a;

Moreno & Tadepalli, 2002; Lettieri, Masella & Nocco, 2009).

The diversity of the output of research centres requires to define different measures in order to assess the

value of the resources to generate a unit of the output, either a publication, a patent or a financed project, leading

to the identification of four KPIs (see Table 2).

The relative importance of these measures was different according to the stakeholder and its organizational

position. The administrative director and the scientific director were the stakeholders most interested in

efficiency indicators. They strongly underlined the need to keep under control how many resources the centre

employs for carrying out research activities and the time spent doing this.

“Monitoring costs is a key issue for us. We have to keep out costs under control and

we need to be able to show how much efficient we are. […] From time to time,

someone stands up and claim they we spend too much. We need to be able to prove

that it is not true, we spend money, but we produce more” (Administrative director)

In order to satisfy his requirements, a synthetic efficiency measure to evaluate the overall ability of the

administrative staff in supporting research activities was included: the ratio between the external budget and the

number of FTE (Full Time Equivalent). The fundamental resource of the institute is represented by people, and

this measure gives the possibility to evaluate the cost for managing a unit of personnel. The executive

committee considered efficiency relevant too, though from a different perspective. The committee poses

particular attention on two specific performance indicators, cost of a scientific publication and incidence of the

external budget on the internal budget, both considered useful in supporting the external accountability. These

numbers are already included in the annual report of the institute and they are explicitly depicted with the

purpose to increase the awareness of external stakeholders about the research activity. The heads of the research

units and the researchers, on the other hand, downplayed the importance of efficiency measures, following the

idea that ‘research is what counts’ no matter how much it costs. Some of them even argued that measuring the

‘cost per output’ is not relevant for the management of a department, ‘what they need to know is the budget’.

Impact

The impact dimension measures the outcome, i.e. the effects of the output of the research activity on the

external environment. This aspect has been highlighted several times as crucial for the research centre. It

represents the mission of the institute:

“The Foundation has the scope to promote the technological development of the

country and the technological education, in line with the scientific and technological

national policies with the final aim to support the development of the productive

Italian system” (Foundation Statute, 2012: 3).

The importance of measuring the outcome is closely connected with the need of being externally

accountable, providing information to external stakeholders on how public money is used. The administrative

director was clear on this point since the kick-off meeting of the project:

“The philanthropic mission of our Institute is to generate know how and

employment for our country. We are receiving public money and we have, not only

the duty to account about how we are using public funding, but also we want to

demonstrate that we are using this money to support the technological development

of our country”(Administrative Director).


54

However, the selection of the most appropriate impact indicator required to balance the trade-off between

the relevance of measuring the outcome and the cost for collecting these data. In fact, the measurement of this

aspect is far from being straightforward given the intangibility of the output and the existence of external factors

that makes it difficult to isolate the impact of research activities on the territorial environment. The KPIs finally

selected were related to cash flows from patents, licenses and technology transfer (see Table 2). They are a

proxy of the outcome because they give the possibility to account for the first order effect of a research activity,

which are leading indicators of second orders effect about impacts on the environment.

The scientific and the administrative directors of the centre were particularly keen on starting measuring

these aspects, whilst part of the administrative and research staff were somehow worried. The motivation at the

basis of this situation was that at present that the cash flows generated by these activities are still modest and

they feared that numbers could lead to wrong interpretations. Directors of research units did not perceived the

importance of measuring impact, being more attracted by KPIs about the quantity and the quality of the

publications.

“Patents, for my department, represent a cost. They do not generate income; this

information can be useful at the overall institute level. It takes some years for a

patent before gene rating money; hence, it is not relevant to manage my activity.

I’ve already seen the result of my research!” (Head of SF3)

Risk

The risk dimension is associated with the need to deal with the uncertainty and the variability of the

research context, as stated by the scientific director:

“Research trends change every year. On the basis of what happens outside, I need to

reconfigure the research activity inside here. The organizational structure should be

flexible and follow these changes accordingly. I need to be able to modify timely

the structure of scientific units; acquire new competencies, understand what

happens if I close a department and open a new one” (Scientific Director).

In particular, the emphasis given to flexibility and adaptability drove the choice of posing particular

attention to how the centre can monitor its resources in order to ensure its ability to respond to unexpected

events. In highly uncertain contexts, resources and the capability to reconfigure them are conceived as the basic

mean through which an organization can face risks (Arena, Arnaboldi & Azzone, 2010b; Arena, Arnaboldi &

Azzone, 2011; Lettieri, 2009; Lettieri, Masella & Radaelli, 2009). Hence, monitoring the ‘state of resources’ can

provide relevant information about how the organization could face a transformation. Obviously, this also poses

the problem of identifying which are the key resources, i.e. which are the resources critical for the success of

research institute. In the case at hand, four main elements were identified: human resources, financial resources,

scientific equipment and reputation; and seven indicators were defined to monitor these elements (see Table 5).

Once again, a different level of awareness and sensitivity of the stakeholders towards different issues

became visible. In particular, the scientific director was very attentive to read the implications of any variation

in different types of resources, going beyond numbers.

[…]“I want to monitor the age of scientific personnel because it provides very

useful information to me. I want a young research centre with a high turnover and

many PhD: by monitoring the age of the staff I can account for this aspect avoiding

the risk of increasing the average age” […]“The trend of PhD students for me is a

signal. If their number falls down, it is an alarm bell to me. First, they are a

resource, they are our primary row material. Second, if they stop coming here, they

going somewhere else. This means that other institutions are more attractive than

us. Why? I mean, it’s something to keep under control” (Scientific Director).

On the other hand, the administrative director and the executive committee appeared more focused on

financial resources and reputation (in particular in relationship to the public opinion) that was sees as a

considerable source of risk.

“We are on the newspaper almost every day. Public opinion poses particular

attention on the way in which we are employing public money. If we do not justify

how the public funding is used, we are immediately accused of wasting financial

resources that could be used for doing something else” (Administrative Director).


55

Finally, directors of RUs showed more interest for the scientific reputation as it gives the possibility to

understand how individual researchers, not only the centre as a whole, are known and considered in a specific

field of research.

Network

The network dimension refers to the ability of the research personnel working together with researchers

from internal teams or from other external institutes. This aspect finds a rationale in the organizational changes

that are affecting the research activities. Collaboration is nowadays considered an essential requirement (Saez et

al., 2002) to survive in a research landscape characterized by inter-disciplinary activities (National Academies,

2005). The network dimension accounts for this collaborative behaviour considering the structure of

relationships between actors, the quantity and the quality of the joint output and the cost of collaboration. This

aspect is emerging in the public sector literature as a relevant performance dimension to evaluate collaboration

(e.g. Provan & Milward, 2001), while it has received less attention in the R&D field (e.g. Allen at al., 2007).

Three KPIs were included in the set of indicators, focusing on the quantity and the quality of joint publications

(see Table 2). Even though collaborative activities relate also to patents and research projects, no network KPIs

were associated with these typology of output. With respect to patents, the decision was to not include indicators

because of the limited amount of patents compared with publications. Considering projects, measurement

difficulties were identified, mainly related to the complexity of accounting joint internal projects. The main

concern associated with this aspect was the following:

“The majority of financed projects is carried out in a collaborative way; it is

therefore not useful to know the level of collaborations as it is by definition that we

are collaborating. This is not true for internal research projects, but they are usually

carried out by the single research team”. (Head of RC 3).

Again, the importance of evaluating the network dimension was different depending on the type of

stakeholder. Collaborative activities emerged several times during interviews and meeting with the scientific

personnel and the scientific director.

“We are working several times with people from other research centres. This is

associated with a greater effort in terms of coordination but better results because of

different competences (Senior scientist 1).”

On the contrary, they were never mentioned by the administrative staff and the executive board.


56

The final set of indicators

Table 5 outlines the final set of indicators related to the five performance dimensions previously introduced

and highlights the priorities assigned to each indicator by different stakeholders.

Table 5: The set of indicators and stakeholders priorities

Dimension KPI

Stakeholders

Admin.

Director

Scientific

Director

Executive

committee

Head of

RU

Effectiveness

Number of scientific publications X X X

Number of patents X X X

Number of technology transfer activities X X X

Stock of competitive projects

X X X

IF distribution

X

X

CI distribution

X

X

Success rate of competitive projects

X X

Outcome Revenue generated by patents, licenses and spin-off X X X

Revenue generated by technology transfer activities X X X

Efficiency

Cost of a scientific publication X

X

Cost to manage and maintain a patent X X

External budget/Internal budget X X X

Internal budget/FTE X

Risk

Scientific reputation X X X X

Press coverage X X X

Capex and Opex X

Employees (category, nationality, age, degree, genre)

X X

Turnover

X X

Organizational climate

X

X

Saturation of equipment

X

X

Network

Number of joint publications

X

X

IF of joint publications

X

X

CI of joint publications

X

X

During the second phase of the action research cycle, the cost and validity of the proposed set of KPIs was

verified, providing practical guidelines to the research centre for the implementation of these measures. Results

of this phase vary according with the performance dimension.

KPIs related to effectiveness are diversified in terms of cost of measures. Most of them can be easily

calculated because they base on data that are already collected for different purposes. The most relevant

exceptions are the Impact factor and Citation index distribution. These indicators, though considered very

important for the assessment of the quality of the research by the scientific director, require ad hoc data

collection and further elaboration in order to ‘clean’ data. Effectiveness indicators are particularly relevant to

have an overall idea of how the centre is performing and to assess the performances of each RU with respect to

its research areas; however, they require some cautions in relationship to data interpretation, since different RUs

cannot be easily compared with each other based on these numbers (e.g. bibliometric indicators change

significantly in different research fields).

KPIs in the outcome dimension aim to provide a proxy of the impact of the research activity on the society.

Data about cash flows are already monitored for accounting purposes. However, they provide only a partial


57

information about the impact of the research activity on the society. Hence, it could be interesting to

complement them through qualitative information to have a more comprehensive idea of research outcomes.

KPIs related to the efficiency can be quite easily calculated because the institute already monitors cost

information very precisely. However, the interpretation of these data can be controversial. In particular, this is

the case of the cost per scientific publication that answer to the need of external accountability. Similar to

effectiveness indicators, it varies significantly across different research units depending on the specific research

field (equipment, materials, and other research instruments significantly vary in different research areas,

resulting in different costs per publication).

KPIs concerning the risk dimension are characterized by some differences in term of measurement costs.

Capex, opex, turnover and data about employees can be determined easily, instead this is not the case for

indicators about the reputation, saturation of key equipment and the organizational climate that entail ad hoc

data collection. Reputation and saturation of key equipment capture two aspects that are particularly relevant for

the research centre, but, at present, they cannot be calculated automatically, and the computation ask a great

effort to collect data and to elaborate them; in the next future, these indicators could benefit from the

exploitation of some supporting tools to collect data routinely. The organizational climate, on the other hand, is

assessed through a survey yearly and it suffers from the limitations that are typical of similar approaches (e.g.

risk of bias, limited timeliness).

Finally, KPIs in the network dimension exploit the same data base of those related to effectiveness, and

share their strengths and weaknesses. In addition, they provide an overall picture of the extent to which

researchers of different RUs collaborate, without entering into details of the contribution of different parts to the

research outputs.

4 CONCLUSION

This paper aimed at constructing a performance measurement system (PMS) for a public research centre

characterized by the presence of a diverse set of stakeholders with multiple objectives, ranging from making

decisions, motivating researchers to demonstrating external accountability. The PMS has been developed

through action research with respect to a specific context, an Italian technological research centre, where the

researchers recursively interacted with different stakeholders in the construction of the PMS. Based on the state

of the art literature and data collected from the interviews, a multidimensional framework for measuring R&D

performance was proposed and a set of KPIs was developed. The traditional process-oriented model (Pollanen,

2005) that encompasses the performance dimensions of effectiveness, efficiency and impact, was complemented

with two further dimensions of risk and network. The former aims to monitor the uncertainty that characterizes

the research activity, and the latter allows to capture the ability of working collaboratively. The introduction of

these two dimensions moves from the idea that the PMS should be aligned with the characteristics of the

organizations in which it is used and the trends of development they should deal with (e.g. Chiesa et al.,

2009).The set of KPIs puts together the information needs of different stakeholders that showed a definite

interest in relationship to a certain subset of measures, often disregarding the others. The integration of different

information needs in a comprehensive set of indicators aimed at providing a holistic picture of the centre

performances to all the relevant actors, raising their attention on the existence of potential trade-offs between

different measures.

This study contributes to the extant literature in two different ways. First, models, frameworks and

methodologies for measuring R&D performances have mostly focused at the firm level, with an economic or

strategic focus (Secundo et al., 2010). We deployed them in the context of a public research institute, where

attention of researchers, so far, concentrated on the one specific stakeholder – the government – that is mainly

interested in an overall assessment of research centres’ performances in relationship to funding allocation.

Second, we tried and integrated the concepts for performance measurement and performance management of

interdisciplinary collaborative research, which depicts new challenges for promoters and science managers in

the research environment of the twenty-first century (Agostino & Arnaboldi, 2012; Arena & Azzone, 2005;

Arnaboldi & Azzone, 2010).

Since the research was conducted in a specific organization, the possibility of generalizing the results to

similar institutions is a key issue. Hence, it is important to distinguish between what is general in scope and

what is case-specific. First, the multi-dimensional framework proposed aims to be general in scope. The use of

the process oriented model as a starting point and the integration of the two dimensions of risk and network aim

to make the PMS model more coherent with the pressures that public research organizations are currently

experiencing in different countries. Focus on efficiency and effectiveness, call for interdisciplinary research and

growing uncertainty are key challenges for these institutions. Moreover, risk and network represent two trends

of development in relationship to performance measurement in different types of organizations (e.g. Provan and

Milward, 2001), though, in most of cases these performances are not formally integrated in the PMS (e.g. Arena

et al., 2011). This paper represents a possible way to explicitly integrate these aspects in the PMS, opening the

path to future research to further adapt the proposed model to different contexts. The second result of general


58

validity is the approach followed to develop the set of KPIs. The choice of the KPIs was guided by the priorities

of different stakeholders. Their requirements and the purpose of use in relationship to different information were

mapped and cross-checked in order to build a set of indicators that would be comprehensive and transversal to

the whole organization. The set of indicators, on the other hand, is case specific and reflects the characteristics

of the research centre under investigation, even though it could provide source of inspiration for similar

organizations.

Finally we address the limitations of this work. The data were collected through a qualitative and

collaborative methodology, action research; hence the results suffer from the limitations that are typical of this

research methodology. In particular, action research is situational, hence, many of the relationships between

people, events, and things are a function of the situation as relevant actors currently define it and they change as

the definition of the situation changes (Susman & Evered, 1978). Accordingly, while the collaboration between

the researcher and the institute staff led to the identification of the KPIs, it may have also resulted in a degree of

bias. However, this methodology gave us the possibility of having lively, detailed and involved discussions with

members of the organizations. Our ideas about performance dimensions and KPIs were critically challenged in

order to satisfy their specific requirements. Moreover, the active participation of users and the integration of

their different informational requirement contributed to reducing resistance to the project and promoting the

development of the PMS. An important field of future research remains the applicability of the proposed model

and methodology to other public research organizations in order to verify its generalizability in different

institutions operating in different countries.

REFERENCES

Agostino, D., & Arnaboldi, M. (2012). Design issues in Balanced Scorecards: the “what” and “how” of control.

European Management Journal, 30(4), 327-339.

Allen, J., James, A.D., &Gamlen, P. (2007). Formal versus informal knowledge networks in R&D: a case study

using social network analysis. R&D Management, 37(3), 179-196.

Arena, M., &Arnaboldi, M. (2013). How dealing with Spending Reviews? The case of Italy. Public Money and

Management, forthcoming.

Arena, M., Arnaboldi, M., & Azzone, G. (2010a). Student perceptions and central administrative services: The

case of higher education in Italy. Studies in Higher Education, 35 (8), 941-959.

Arena, M., Arnaboldi, M., & Azzone, G. (2010b). The organizational dynamics of Enterprise Risk Management.

Accounting, Organizations and Society,35(7), 659-675.

Arena, M., Arnaboldi, M., & Azzone, G. (2011). Is enterprise risk management real? Journal of Risk Research,

14(7), 779-797.

Arena, M., Arnaboldi, M., Azzone, G., & Carlucci, P. (2009). Developing a performance measurement system

for University Central administrative services. Higher Education Quarterly, 63(3), 237-263.

Arena, M., &Azzone, G. (2005). ABC, Balanced Scorecard, EVA™: An empirical study on the adoption of

innovative management accounting techniques. International Journal of Accounting, Auditing and

Performance Evaluation, 2(3),206-225.

Arena, M., &Azzone, G. (2010). Process based approach to select key sustainability indicators for steel

companies. Ironmaking and Steelmaking, 37(6), 437-444.

Arnaboldi, M.,& Azzone, G. (2010).Constructing performance measurement in the public sector. Critical

Perspectives on Accounting, 21(4),266-282.

Arnaboldi, M., Azzone, G., &Savoldelli, A. (2004).Managing a public sector project: The case of the Italian

Treasury Ministry. International Journal of Project Management, 22(3), 213-223.

Azzone., G. (2008). Controllo di gestione nelle amministrazioni pubbliche. Decisioni e accountability per

ministeri, enti locali e università. Milano:ETAS.

Bremser W.G., &Brasky, N.P. (2004).Utilising the Balanced Scorecard.R&D Management, 34(3), 229-238.

Brown, M.G.&Svenson, R.A. (1988).Measuring R&D productivity. Research-Technology Management, July–

August,11–15.

ChiesaV.,& Frattini F. (2007). Exploring the differences in performance measurement between research and

development: evidence from a multiple case study.R&D Management, 37(4), 283-301.

Chiesa, V., Frattini, F., Lazzarotti V., & Manzini R. (2009). Performance measurement of research and

development activities. European Journal of Innovation Management,12(1), 25-61.


59

Chu, P.Y., Lin, Y.L., Hsiung, H.H.,& Liu, T.Y. (2006). Intellectual Capital: an empirical study of ITRI.

Technological Forecasting & Social Change, 73, 886-902.

Coccia, M. (2001a). Scientific research performance within public institutes. Scientometrics, 65(3), 307-321.

Coccia, M. (2001b). A basic model for evaluating R&D performance: theory and application in Italy. R&D

Management, 31(4), 453-464.

Coccia, M. (2004). New models for measuring the R&D performance and identifying the productivity of public

research institutes. R&D Management, 34(3), 267-280.

Coccia, M.(2005). Scientometric model for the assessment of the scientific research performance within the

public institutes. Scientometrics,65(3), 297–311.

Coccia, M., &Rolfo, S. (2007). How research policy changes can affect the organization productivity of public

research institutes. Journal of Comparative Policy Analysis, Research and Practices, 9(3), 215-233.

Coughlan, P. &Coghlan, D. (2002). Action Research for operations management. International Journal of

Operations & Production Management, 22(2), 220-240.

Cross, R., Borgatti, S.P.& Parker, A. (2002). Making invisible work visible: using social network analysis to

support strategic collaboration. California Management Review, 44(2), 25–46.

Dumay J.,& Rooney, J. (2011). Measuring for managing? An IC practice case study. Journal of Intellectual

Capital, 12(3), 344-355.

Edvisson, L., & Malone, M.S. (1997). Intellectual Capital: Realizing Your Company’s True Value by Findings

its Hidden Brainpower. New York, NY:HarperBusiness.

Elmquist M., & Le Masson, P. (2009). The value of a “failed” R&D project: an emerging evaluation framework

for building innovative capabilities. R&D Management, 39(2), 136-152.

Garcia-Valderrama, T.,&Mulero-Mendigorri, E. (2005). Content validation of a measure of R&D effectiveness.

R&D Management, 35(3), 311-331.

Hult, M., &Lennung, S. (1980). Towards a definition of action research: a note and a bibliography. Journal of

Management Studies, 17, 241-250.

Kerssen-van Dronelen, I.C. &Bilderbeek, J. (1999). R&D performance measurement: more than choosing a set

of metrics. R&D Management, 29, 35-46.

Kim, B.,& Oh, H. (2002). An effective R&D PMS: survey of Korean R&D researchers. International Journal of

Management Science, 30, 19–31.

KulatungaU., Amaratunga D., &Haigh R. (2011). Structured approach to measure performance in construction

research and development. Performance measurement system development. International Journal of

Productivity and Performance Management, 60 (3), 289-310.

Lepori, B., van den Besselaar, P.,Dinges, M.,Potì, B.,Reale, E.,Slipsaeter, S.,Thèves, J., & van der Meulen, B.

(2007). Comparing the evolution of national research policies: what patterns of change? Science and Public

Policy, 34(6), 372–388.

Leitner, K.H., & Warden, C. (2004). Managing and reporting knowledge-based resources and processes in

research organization: specifics, lessons learned and perspectives. Management Accounting Research, 15,

33-51.

Lettieri, E., (2009). Uncertainty inclusion in budgeting technology adoption at a hospital level: Evidence from a

multiple case study. Health Policy, 93(2):128-136.

Lettieri, E., Masella, C., (2009). Priority Setting for Technology Adoption at a Hospital Level: Relevant Issues

from the Literature. Health Policy, 90(1):81-88.

Lettieri, E., Masella, C., Nocco, U., (2008).Budgeting And Health Technology Assessment: First Evidence

Obtained From Proposal Forms Used To Submit The Adoption Of New Technology. International Journal

of Technology Assessment in Health Care, 24(4):502-510.

Lettieri, E., Masella C., Radaelli, G., (2009).Disaster Management: findings from a systematic review. Disaster

Prevention and Management, 18(2):117-136.

Luwel, M., Noyons, C.M. &Moed, F. (1999).Bibliometric assessment of research performance in Flanders:

policy background and implications. R&D Management, 29(2), 33–141.

Lynch, R.L. & Cross, K.F. (1991). Measure Up ± the Essential Guide to Measuring Business Performance.

Mandarin, London.

Marr, B., Schiuma, G. & Neely, A. (2004). Intellectual capital: defining key performance indicators for

organisational knowledge assets. Business Process Management Journal, 10(5), 551-69.


60

Mettänen, P. (2005). Design and implementation of a performance measurement system for a research

organization. Production Planning & Control, 16(2), 178-88.

Moreno, A.A. & Tadepalli, R. (2002). Assessing academic department efficiency at a public university.

Managerial and Decision Economics, 23, 385-397.

Narin, F., & Hamilton, K.S. (1996).Bibliometric performance measures.Scientometrics, 36(3), 293–310.

National Academies (2005).Engineering and Public Policy. Facilitating Interdisciplinary Research.

Washington, DC:National Academies Press.

Neely, A., Marr, B., Roos, G., Pike, S. & Gupta, O. (2003). Towards the third generation of performance

measurement. Controlling, 15(3/4), 129-135.

Noyons, E.C.M., Moed, H.F. ,&Luwel, M. (1999). Combining mapping and citation analysis for evaluative

bibliometric purposes: A bibliometric study. Journal of the American Society for Information Science,

50(2), 115–131.

Pawar, K.S., &Driva, H. (1999). Performance measurement for product design and development in a

manufacturing environment. International Journal of Production Economics, 60-61, 61–68.

Pollanen, R.M. (2005). Performance measurement in municipalities Empirical evidence in Canadian context.

International Journal of Public Sector Management, 18(1), 4-24.

Provan, K.G.,&Milward, H.B. (2001). Do networks really work? A framework for evaluating publicsector

organizational networks.Public Administration Review, 61, 413-423.

Rapoport, R. (1970). Three Dilemmas of Action Research. Human Relations, 23(6), 499-513.

Rowley, J. (2000). Is higher education ready for knowledge management? The International Journal of

Educational Management, 14(7), 325-33.

Reason, P., & Bradbury, H. (Eds.). (2001). Handbook of Action Research: Participative inquiry and practice.

London : Sage Publications

Saez, C.B., Marco, T.G. &Arribas, E.H. (2002). Collaboration in R&D with universities and researchcentres: an

empirical study of Spanish firms. R&D Management, 32, 321–341.

Secundo G., Margherita A., Elia G., &Passiante G. (2010). Intangible assets in higher education and research:

mission, performance or both? Journal of Intellectual Capital, 11(2), 140-157.

Senker, J. (2001). Changing organization of public research in Europe. Implications for benchmarking human

resources in RTD, Paper prepared for Human Resources. In:RTD session of the contribution of European

socio-economic research to the benchmarking of RTD policies in Europe, Brussels, March 15-16.

Steiner, A. & Nixon, B. (1997). Productivity and performance measurement in R&D. International Journal of

Technology Management, 13(5/6), 486-496.

Stewart, T.A. (1997). Intellectual Capital: The New Wealth of Organizations. New York: Doubleday/Currency.

Susman, G.I. &Evered, R.D. (1978). An Assessment of the Scientific Merits of Action Research. Administrative

Science Quarterly, 23(4), 582-603.

Whelan E., Teigland R., Donnellan B.,& Golden W. (2010). How Internet technologies impact information

flows in R&D: reconsidering the technological gatekeeper. R&D Management, 40(4), 400-413.

Welter, F., Jooß, C., Richert, A., Jeschke, S. &Brecher, C. (2012). Organisation and Management of Integrative

Research. In: C. Brecher (Ed.) Integrative Production Technology for High-Wage Countries. Berlin

Heidelberg: Springer-Verlag, pp. 64–73.

Yin, R.K. (1994). Case Study Research: Design and Methods. Thousand Oaks: Sage Publications.

international journal of business science and applied ... · the cognitive perspective in the...

Documents