the moderating effect of procedural justice on the

415

Supriyadi—The Moderating Effect of Procedural Justice on the Effectiveness of the BSC....

Gadjah Mada International Journal of BusinessSeptember-December 2010, Vol. 12, No. 3, pp. 415–434

Supriyadi*Faculty of Economics and Business, Universitas Gadjah Mada, Yogyakarta, Indonesia

This study extends prior studies on the effectiveness of theBalanced Scorecard (BSC) to improve managerial performancedone by Lau and Mosser (2008) and Lau and Sholihin (2005).Specifically, the study empirically tests the moderating effects ofprocedural justice on the relationship between the financial andnonfinancial dimensions of BSC and managerial performance. Italso tests the impact of organizational commitment on performance.Based on survey data from 76 respondents, the results indicate thatperceived procedural justice in the use financial and nonfinancialdimensions of the BSC is associated with managers’ organizationalcommitment. It further finds that organizational commitment ispositively related to performance. The study extends the literatureby providing empirical evidence about the moderating effect ofprocedural justice on the relationship between the financial andnonfinancial dimensions of BSC and organizational commitment.

THE MODERATING EFFECT OFPROCEDURAL JUSTICE ON THE

EFFECTIVENESS OF THE BALANCEDSCORECARD IN IMPROVING MANAGERIAL

PERFORMANCE THROUGHORGANIZATIONAL COMMITMENT

Keywords: balanced scorecard; organizational commitment; financial measures;managerial performance; moderating effect; nonfinancial measures;procedural justice

* I thank the editor and two anonymous reviewers for their helpful comments and suggestions

416

Gadjah Mada International Journal of Business, September-December 2010, Vol. 12, No. 3

Introduction

Since Kaplan and Norton (1992)introduced the balanced scorecard(BSC), more companies have replacedtheir traditional financial performancemeasures with the BSC. Prior studiesfound that 44 percent of organizationsin the US (Rigby 2001) and 35 percentof large US firms (Marr et al. 2004) usethe BSC. Silk (1998 in Lipe and Salterio2000) indicated that the BSC has beenwidely used by around 60 percent ofUS Fortune 1000 firms. Kaplan andNorton (2001) also report that the BSChas been adopted by not only profit-oriented organizations, but also non-profit organizations and governmententities. Moreover the results of anunpublished survey conducted by theresearcher to part-time MBA studentswho took a Management Control Sys-tem class at Universitas Gadjah Madain 2007 indicated that 68 percent of therespondents report that their firms haveemployed the BSC in various forms tomeasure business unit managers’ per-formance.

The purposes of this study is toexamine the moderating effect of pro-cedural justice on the effectiveness ofemploying the financial and nonfinan-cial dimensions of BSC as managers’performance measures in improvingmanagerial performance. The studyextends Lau and Sholihin (2005) studyand completes Lau and Moser’s study(2008) by evaluating the moderatingeffect of procedural justices on therelationship between the BSC mea-sures and organizational commitment.

Procedural justice is expected tostrengthen the effect of employing theBSC measures on managers’ commit-ment. It also examines the impact ofcommitment on performance. The firstobjective of this study is to completeLau and Moser’s (2008) study by em-ploying the four perspectives of theBSC measure in judging managerialperformance. Using the structural equa-tion analysis, they map the direct andindirect relationships between nonfi-nancial measures (three of the fourBSC performance perspectives): pro-cedural fairness, organizational com-mitment, and managerial performance.They found the effect of nonfinancialmeasures on managerial performanceis indirect via organizational commit-ment and procedural justice.

Lau and Sholihin (2005) docu-mented that the four BSC perspectivessignificantly affect managers’ behav-ior. The use of the BSC perspectives(financial and non-financial) indirectlyaffects job satisfaction through fair-ness of the measures and trust betweensupervisor and supervisee. This indi-rect relationship is strengthened by theinsignificant direct relationship be-tween the BSC measures and job satis-faction. However, Lau and Buckland(2001) and Lau and Mosser (2008)found inconsistent evidence regardingthe effects of financial and nonfinan-cial measures on procedural justice.Lau and Buckland (2001) documentedthat financial measures represent ob-jective and truthful measures and thatthey are perceived more fair than non-financial measures, which represent

417


subjective and bias measures. On theother hand, Lau and Mosser (2008)indicated that nonfinancial measuresrepresent complete and accurate mea-sures and, that why they are perceivedmore fair than financial measures,which represent narrow and rigid mea-sures.

Furthermore, since evaluatingmanagerial performance is a process,then the fairness in evaluation proce-dures will be perceived through thelength of the evaluation process andthat is why the fairness of the proce-dures might be gradually perceivedduring the process, not as a conse-quence of the process. Hence, the fair-ness in evaluation process is not medi-ating but moderating the relationshipbetween the BSC measures andmanager’s job satisfaction, as ex-plained by the contingency theory(Fiedler 1983). The second objectiveof this study is to address this issue byevaluating the moderating effect ofprocedural justice on the effectivenessof the BSC performance measures.

Finally, the last objective of thestudy is to investigate the relationshipsof the variables of interests using mul-tiple regression approach since thetheory (contingency theory) and logi-cal reasons as basis for modeling therelationships between variables exist.Lau and Sholihin (2005) and Lau andMoser (2008) nicely map the one-wayrelationship among variables of finan-cial measures, nonfinancial measures,procedural fairness, organizationalcommitment, and managerial perfor-

mance using structural equation analy-sis. They find significant relationshipsamong variables, but in fact their rela-tionships model among variables iscomplicated and is difficult to inter-pret to the real business conditions.

The study conducts surveys using76 part-time MBA students at GadjahMada University as a sample. First, itpredicts that procedural justice willstrengthen the positive relationshipbetween nonfinancial (financial) mea-sures and organizational commitment.Second, it also expects that organiza-tional commitment will positively af-fect managerial performance. The re-sults of the study show that managers’perceptions on the procedural justiceof the use of the financial and nonfi-nancial dimensions of BSC signifi-cantly strengthen the relationship be-tween nonfinancial measures and or-ganizational commitment (supportingLau and Moser 2008), and betweenfinancial measures and organizationalcommitment (supporting Lau andSholihin 2005). Finally, the study alsofinds that the relationship between or-ganizational commitment and mana-gerial performance is positive.

The remaining paper is organizedas follows. The next section describesthe theoretical background and hy-pothesis development in addressingthe research issues. This is followedby an explanation of the researchmethod employed and presentation ofthe results. The last section providesthe conclusions and limitations of thestudy.

418


Literature Review andHypothesis Development

The Balance Scorecard

The BSC system provides a com-plete view of business unit perfor-mance based on a set of financial andnon-financial measures, namely finan-cial, customer relationships, internalbusiness processes, and learning andgrowth performances. Kaplan andNorton (1996a) include a diverse setof performance measures in order tocompletely derive the business unitstrategy into a cause-and-effects rela-tionship between the strategy and theperformance measurement. However,some researchers have identified con-ditional aspects that may influence theway evaluators in considering the BSCmeasures when they assess and judgemanager’s performance in implement-ing business unit strategies (Liedtka etal. 2008). For instance, since the BSCmeasures usually reflect similar mea-sures across business units (commonmeasures) and specific measures toeach business unit (unique measure),Slovic and McPhillamy (1974) sug-gest that decision makers (evaluators)will focus on common measures andunderuse of unique measures. Lipeand Salterio (2000) using experimen-tal design conclude that most of theparticipants only consider commonmeasures, ignoring unique measureswhen they compare performance oftwo different business units.

Some factors have been indicatedto affect evaluators’ judgmental bias

in assessing managers’ performanceusing the BSC. Kaplan and Norton(1996b) suggest that the lack of align-ment between compensation and theBSC performance measurement sys-tems may affect the effectiveness ofimplementing the BSC. Lipe andSalterio’s (2000) experimental re-search concludes that general humancognitive tendencies are the main causeof the judgmental bias, while Liedtkaet al. (2008) indicate that evaluators’individual characteristics - ambiguityintolerance - influence the BSC-basedevaluation bias. These various resultssuggest that further research is neededto modify or clarify the BSC-basedperformance evaluation procedures(Liedtka et al. 2008).

In the development of the BSCperformance measures, Kaplan andNorton (1996a) suggest to use a phi-losophy of cause-and-effect relation-ship when deriving outcome measuresfrom strategies. Specifically, the fourperspectives of the BSC are insepa-rable performance measures with thelogic flow that internal business per-formance will affect learning andgrowth, that learning and growth per-formance will affect the customer, andfinally that customer performance willaffect financial performance. Hence,when managers’ performance will beassessed using the BSC measures, theywill consider the four perspectives ofthe BSC as important measures. Thisis the key success of using the BSC asa performance measurement system(Anthony and Govindarajan 2005).

419


Moderating Effect of ProceduralJustice

Procedural justice1 refers to thefairness of the managerial performancemeasurement process used to judge,evaluate, and reward subordinates’performance (McFarlin and Sweeney1992; Lind and Tyler in Lau andSholihin 2005; and Greenberg 1987).Procedural justice theory is expandedfrom the equity theory for resourceallocation process (Greenberg 1987).Similarly, Laventhal (1980) suggeststhat fairness of the judgmental processtoward decision-making will be re-flected into the procedural justice ofthe decision process. That is, proce-dural fairness refers to fairness of theprocesses that leads to a decision out-come. The rules of procedural justicessuggested by Laventhal (1980) includeconsistency, representativeness of in-terested parties, evaluativeness, accu-rate information, non-bias decisionmaker, and conformity to moral val-ues.

Earlier studies documented thatthere are various ways to relate theeffects of procedural justice on mana-gerial performance (Libby et al. 2004;Wentzel 2002, and Libby 1999; amongothers). Folger (1987) documented thateach of the organizational justices, e.g.procedural justice and distributive jus-tice, is independently related to atti-tude toward the decision and the orga-

nization. On the other hand, Korsgaardand Roberson (1995) found that re-gardless of the fairness of the decision,procedural justice will promote posi-tive behavior of the related parties.Based on a meta-analysis of twenty-five years of organizational justice re-searches, Colquitt et al. (2001) docu-mented that procedural justice hadmoderate correlations with system-ref-erenced evaluation of authority. Thisresult indicates that procedural justicewithin managerial performance evalu-ation consistently affects managerialperformance. However, the effect isvarious between two ends of direct vs.indirect effects.

In the context of the BSC, thecombination of financial and non-fi-nancial performance measures is torelate between short-term operationalcontrol and the long-term vision andstrategy of the business (Kaplan andNorton 1992, 1996a, 2001). The BSCemploys a multi-dimensional approachby including both financial and non-financial measures in order to com-bine the more traditional control meth-ods (financial measures) with the newfirm’s emerging success-leading-indi-cators (non-financial measures) (Otley1999). Amir and Lev (1996) concludedthat the non-financial measure willincrease the value-relevant of the fi-nancial measures related to the valueof the subject telephone-company.Therefore, the study employs the two

1 This study uses the term “justice” and “fairness” interchangeably. Some prior studies also usedboth of the terms in their studies (Linquist 1995; Lau and Sholihin 2005; among others). In thisstudy, however, the term “procedural justice” is used when it refers to variable under examination.

420


perspectives, financial and non-finan-cial measures in implementing the BSCsince both perspectives might appro-priately not be used separately.

It has been documented that theuse of the BCS in evaluating manage-rial performance is closely related tothe managers’ perception on the fair-ness of performance evaluation pro-cess. Referring to Laventhal’s proce-dural justice rules (1980), the finan-cial measures relate to the consistencyrule since the measures are consis-tently applied across time and personsor divisions, while the non-financialmeasures relate to the representative-ness rule since they represent the val-ues, concerns and perspectives of eachdivision. Furthermore, both the finan-cial and non-financial measures alsoaccommodate the rules of evaluative-ness, accurate information, non-biasdecision maker, and conformity tomoral values.

However, managers may perceiveprocedural justice differently betweennonfinancial and financial measuressince both measures have differentchcharacteristics. Lau and Buckland(2001) show that managers perceivefinancial measures as more fair thannonfinancial measures because finan-cial measures are objective and truth-ful, while nonfinancial measures aresubjective and biased. On the otherhand, Lau and Mosser (2008) indicatemanagers perceive nonfinancial mea-sures as more fair than financial mea-sures since nonfinancial measures arecomplete and accaccurate, while fi-nancial measures are narrow and rigid.

One possible reason explaining thisinconsistent evidence is the impor-tance of the process in evaluating per-formance. Since measuring managers’performance using the BSC is a pro-cess, then the fairness in evaluationprocedures will be perceived throughthe length of the evaluation process.Hence, managers will gradually per-ceive procedural justice during theprocess, not as a consequence of theprocess.

The study proposes that nonfi-nancial measures are associated withorganizational commitment and theeffect is strengthened when managersperceive the evaluation process as fair.Lau and Moser (2008) identify thethree-relationships between nonfinan-cial measures, procedural fairness, andorganizational commitment. They findthat nonfinancial measures positivelyand directly affect organizational com-mitment, while they also positivelyand indirectly affect organizationalcommitment through procedural fair-ness. Fisher (1992) documents thatsince nonfinancial measures arebroader terms, long-term dimensions,and tailor-made to conditions, theycan result in a variety of possible mea-sures. Consequently, they may affectuser judgments in assessing perfor-mance. Therefore, managers’ percep-tion on how objectively nonfinancialmeasures are being used to evaluatetheir performance is expected to affecteffectiveness of using nonfinancialmeasures to improve managers’ orga-nizational commitment. The study pro-poses the first hypothesis as follows.

421


Ha1

: Managers’ perception on proce-dural fairness of performanceevaluation criteria strengthensthe relationship between nonfi-nancial criteria and managers’organizational commitment

Kaplan and Norton (1996a) arguethat the financial measures are able tosummarize the readily measurable andimportant economic consequences ofthe actions of division managers. Fur-thermore, the financial measures willreflect the ability of managers to imple-ment its division strategy that will con-tribute to the bottom-line of the com-pany goal, profits (Kaplan and Mackey1992 and Kaplan and Norton 1996b).Hence, the study proposes that theprocedural justice is expected to posi-tively improve the effectiveness ofusing financial measures to improvemanagers’ organizational commit-ment. The study states the second hy-pothesis as follows.

Ha2

: Managers’ perception on proce-dural fairness of performanceevaluation criteria strengthensthe relationship between finan-cial criteria and managers’ or-ganizational commitment.

Some earlier researches in be-havioral aspects of performance mea-surement systems and participativebudgeting have consistently docu-mented that organizational commit-ment mediates the relationship betweensome behavioral effort variables andorganizational performance (Lau andMoser 2008; Lau and Sholihin 2005;Porter et al. 1974; Mowday et al. 1979;among others). Organizational com-

mitment is the strength of employees’involvement in a particular organiza-tion. The concepts of organizationalcommitment can be referred to as thecondition with which employees willstay with the organization because theyneed to, because they are obligated to,or because they want to (Jaros et al.1993). The third concept that is alsocalled affective commitment is the mostadopted concept, and hence, the studyuses the third concept as suggested byCohen (1993) and Cohen (1987).

Meyer at al. (1989) found a posi-tive causal relationship between orga-nizational commitment and organiza-tional performance. They differenti-ate organizational commitment intoaffective and continuance commit-ment. Affective commitment refers toemotional attachment to, identifica-tion with, and involvement in the orga-nization, while continuance commit-ment refers to perceived costs associ-ated with leaving the company. Theydocumented that affective commitmentcorrelated positively and continuancecommitment correlated negatively withjob performance. Accordingly, thestudy proposes the third hypothesis asfollows.

Ha3

: Managers’ organizational com-mitment is positively related tomanagers’ managerial perfor-mance

Figure 1 displays the researchmodel used to examine the relation-ship among variables of interests in thestudy. It proposes that procedural jus-tice will be moderating the effects ofusing the BSC (nonfinancial and fi-

422


nancial measures) on organizationalcommitment. Finally, the organiza-tional commitment, in turn, will affectmanagerial performance.

Methodology

Respondents

The hypotheses are tested usingprimary data collected from part-timeMBA students at Universitas GadjahMada, Indonesia. There are 76 stu-dents in three classes of managementcontrol systems subject (MCS) in theacademic years of 2007/2008 and 2008/2009 who participate in the study. Theselection of students in MCS classes isto assure that the respondents havebasic knowledge of and are familiarwith the topic and being asked aboutthe use of BSC in the performancemeasurement system. The researcherdistributes questionnaires to the re-spondents at the last session of theclass meeting. The researchers put thequestionnaires in the front of the classand announce to the students that there

is a study asking the students who arewilling to voluntarily participate in thestudy related to the BSC performancemeasurement system to take a ques-tionnaire and fill it out before the classbegins. It is also announced that it willtake around 15-20 minutes to fill outthe questionnaire, and that the studydoes not have any relationship with theMCS class.

The questionnaire is accompaniedby a cover letter describing the objec-tive of the study and assurance of re-sponse confidentiality. After the coverletter, the second page of the question-naire asks the respondent to write abrief description of demography data,followed by a description of a mission,strategy, and the plan of using the BSCby the business unit to measure effec-tiveness of strategy implementation,questions relating to measuring vari-ables in interests on the third to sixthpages, consecutively. Seventy nine stu-dents from 108 listed students (70.37%)in the three MCS-classes participate inthe study. Three responses are excludedfrom further analysis due to incom-

Figure 1. Relationship between Nonfinancial measures, Financial Measures,Organizational Commitment, and Managerial Performance

423


plete or incorrect responses, and henceleaving 76 usable responses, of which57 percent of them are men and 43percent are women. The non-responsebias among different times in threedifferent classes is tested based onmean comparisons for the four vari-ables used in the study (Hair 2006).The results show that there are nosignificant differences among meansfrom three MCS classes.

Table 1 depicts the distribution ofdemography data. The average age ofrespondents is 33 years with an aver-age working experiences of ten years,the shortest working experience of fouryears and the longest working experi-ence of fourteen years. Most of therespondents (94%) are involved in themanagerial function of monitoring,evaluating, and/or supervising subor-dinates in their current positions. Thiscondition indicates that most of therespondents have enough experiencesfunctioning as a manager in their com-panies. Finally, 89 percent of respon-

dents work in the service industry (59%in general and 30% in financial ser-vices) and 11 percent work in the manu-facturing industry.

Variables Measurement

The BSC Measures

The purpose of the study is toexamine individual behavior responsesto the four perspectives of performancemeasures used in the BSC, namelylearning and growth, internal business,customer, and financial perspectives.The study employs four items of per-formance measures for each perspec-tive (sixteen items for all perspectives)used by Lipe and Salterio (2000) withview wording modifications. However,the study selects only a set of the BSCmeasures for one division (ReadWeardivision) used by Lipe and Salterio’s(2000) study, since it is more appropri-ate with Indonesian business environ-ments. They use two sets of the BSCmeasures for two different divisions

Table 1. Distribution of Demography Data for Respondents

Man Woman Total

Number of Repondents 43 (57%) 33 (43%) 76 (100%)

Average Age 34 years 30 years 33 years

Working Experience 10 years 9 years 10 years

Company Background:

Manufacturing 6 (8%) 2 (3%) 8 (11%)

Services 28 (37%) 17 (22%) 45 (59%)

Financial 9 (12%) 14 (18%) 23 (30%)

Holding Managerial Function 40 (53%) 31 (41%) 71 (94%)

424


because their subjects are asked tocompare the performance of two divi-sions using the BSC measures.

The study selected Lipe andSalterio’s (2000) instrument based onthree reasons. First, as suggested byKaplan and Norton (1996), Lipe andSalterio’s instrument has been specifi-cally developed based on division strat-egy introduced in their experiment.Second, they decided to select sixteenitems of measures based on the consid-eration that individual cognitive ca-pacity to appropriately process cues ofinformation is between fifteen totwenty cues. Moreover, Lipe andSalterio (2000) and Liedtka et al. (2007)indicate that measuring individual per-formance based on too few variablesmay not represent the real conditions;however, employing too many vari-ables may complicate the process. Thefinal reason is that Lipe and Salterio(2000) have done a pilot testing fortheir instrument to groups of MBA andaccounting students resulting in validrepresentative of the BSC categories.

The items for learning and growthperspectives are hours of employeetraining per employee, employee sug-gestions per employee, average tenureof sales personnel, and stores comput-erizing. The four items used in internalbusiness perspective include returnsto suppliers, average major brandnames/store, average markdowns, andaverage orders filled within one week.The items used in customer perspec-tive are repeat sales, returns by cus-tomer as percentage of sales, customersatisfaction rating, and captured cus-

tomer. Finally, the four items for fi-nancial perspectives include return onsales, new store sales, sales growth,and revenue per sales visit.

The questionnaire asks the respon-dents, as a manager of the division, torank the importance of each item of theBSC performance measures used bytheir superior to assess his/her divi-sion performance, using a seven-pointLikert scale for each item, in whichscale 1 indicates the least importanceand scale 7 indicates the most impor-tance. Similar to Lipo and Salterio(2000), the first part of the instrumentdescribes the company mission state-ment, introduces the managers of thebusiness unit, describes the businessunit’s strategy, and presents the BSCfor the business unit. This scheme isdone to focus the respondents on theimportance of the BSC in assessingstrategy implementation. Furthermore,bringing respondents’ mindsets to un-derstand business unit strategy beforeassessing the BSC performance mea-sures will result in more appropriatejudgment relating to the importanceand fairness of the measures. How-ever, while Lipe and Salterio (2000)ask their subject to indicate businessunit performance on the scale of 0 to100 points, this study asks the respon-dent to indicate the importance of theBSC measures using a 7-point Likertscale.

The results of factor analysis withvarimax rotation show that 16 itemsload to four perspectives with eigen-value greater than 1 for all perspec-tives, namely eigenvalue for learning

425


and growth perspective is 2.705, inter-nal business is 2.094, customer is 2.015,and financial is 1.687; and those ex-plain 16.96, 13.09, 12.50, and 10.55percent of variances for each measure,respectively. These indicate that the

four perspectives significantly corre-spond to the BSC measurement per-spectives. Additionally, all four itemsalso significantly load to their respec-tive perspectives with loading factorsranging from 0.511 to 0.798.

Table 2. Factor Analysis and Eigenvalue for all BSC Perspectives

Measure Learning Internaland Growth Business Customer Financial

Process

Hours of employee training 0.566 per employee

Employee suggestions 0.511per employee

Average tenure of sales personnel 0.607

Stores computerizing 0.775

Returns to suppliers 0.514

Average major brand names per store 0.777

Average markdowns 0.558

Average orders filled within one week 0.626

Repeat sales 0.553

Customer satisfaction rating 0.593

Return by customer as a percent 0.798 of sales

Captured customer 0.776

Return on sales 0.604

New store sales 0.593

Sales growth 0.651

Revenue per sales visit 0.597 Eigenvalue 2.705 2.094 2.015 1.687

Percentage of variance explained 16.964 13.090 12.501 10.547

Cronbach alpha 7.617 7.617 7.617 6.738

426


In the analysis, the nonfinancialperspective scores (customers, inter-nal business, and learning and growthperspectives) are combined into onevariable. The data used for regressionanalysis (hypothesis tests) is the aver-age scores of raw data for each vari-able of interest, except for a manage-rial performance variable (see expla-nation under sub-heading of Manage-rial Performance). The average of non-financial performance is the mean of16 items’ scores of nonfinancial per-formance measures, the average scoreof financial performance is the meanof the four items’ scores of financialperformance measures, the averagescore of procedural justice is the meanof the four items’ scores of proceduraljustice measures, and the average scoreof organizational commitment is themean of the nine items’ score of orga-nizational commitment measures.Table 2 presents the results of factoranalysis and eigenvalue for validitytest of the instruments and Table 3provides the descriptive statistics of

all variables used in the regressionanalysis.

Procedural Justice

Using the BSC measures to judgeoverall managers’ performance willinvolve subjective perceptions of theuser on “the meaning” of the BSCmeasures. This condition is likely toaffect the ethical (justice) perceptionof managers being evaluated using theBSC. Two forms of justice may beinvolved in the process, namely dis-tributive justice and procedural jus-tice. The concepts of distributive jus-tice are developed based on the equityprinciple. It states that distributive jus-tice sustains when the allocation of theoutputs (benefits and/or costs) withina group is proportional to the contribu-tion of group members. On the otherhand, the concepts of procedural jus-tice rely on the fairness of the rules andprocesses involved in the activity. Thestudy relates to the justice perceptionof using the BSC (as a rule) to evaluatemanagers’ performance (as a process).

Table 3. Descriptive Statistics of Variables Used in Hypothesis Testing

Variable Actual TheoreticalStd. Range Range

Mean Dev. Min Max Min Max

Nonfinancial measure 5.22 0.23 4.83 5.67 1.00 7.00

Financial measure 5.59 0.28 5.00 6.00 1.00 7.00

Procedural justice 3.50 0.20 3.00 3.75 1.00 5.00

Organizational commitment 6.82 0.08 6.67 6.89 1.00 7.00

Managerial performance 5.58 0.78 4.00 7.00 1.00 7.00

427


Hence, the interest of the study is theconcepts of procedural justice.

Following Lau and Moser (2008)and Lau and Sholihin (2005), the studyemploys McFarlin and Sweeney’s(1992) instrument to measure manag-ers’ perception of procedural justice.The questionnaires ask subjects to rat-ing the procedural justice of perfor-mance evaluation process using a 5-point Likert scale (1 has the least im-portance and 5 has the most impor-tance) on four items used in the perfor-mance evaluation process. The fouritems include evaluating employee per-

formance, determining employee pro-motions, determining salary increases,and communicating performance feed-back. Table 4 presents the results ofthe factor analysis of the four proce-dural justice items. The rotated factorloadings for these items of proceduraljustice range from 0.559 to 0.759 andload on a single factor with an eigen-value of 1.679 that explains 41.923percent of the variances. The meanscore is 3.50 with the standard devia-tion of 0.20 (see Table 3), and Cronbachalpha of 0.717 indicating valid mea-sures.

Table 4. Factor Analysis and Eigenvalue for Procedural Fairness andOrganizational Commitment

Items Organizational Procedural FairnessCommitment

OC1 0.651

OC2 0.571

OC3 0.520

OC4 0.773

OC5 0.718

OC6 0.903

OC7 0.894

OC8 0.725

OC9 0.820

PJ1 0.661

PJ2 0.675

PJ3 0.759

PJ4 0.559

Eigenvalue 1.718 1.679

Percentage variance explained 19.084 41.923

Cronbach alpha 0.936 0.717

428


Organizational Commitment

Similar to Lau and Moser (2008),the study employs Mowday et al.’s(1974) instrument to measure organi-zational commitment. The question-naires ask subjects to rate their agree-ment on nine statements using a 7-point Likert scale (1 is strongly dis-agree and 7 is strongly agree) such as“I tell my friends that this organizationis a great place to work for” and “I amproud to tell others that I am part of thisorganization.” Table 4 presents theresults of the factor analysis of the nineorganizational commitment items. Therotated factor loadings for these itemsrange from 0.520 to 0.903 and load ona single factor with an eigenvalue of1.718 that explains 19.084 percent ofthe variances. The mean score is 6.82with the standard deviation of 0.08(see Table 3), and Cronbach alpha ofo.936 indicating valid measures.

Managerial Performance

The study employs Moahoney’s(1963) managerial performance instru-ment that has been widely used in priorstudies (Lau and Moser 2008;Lindquist 1995; and Brownell 1982;among others). The questionnaires asksubjects to rate nine-dimensions ofareas of managerial performance us-ing a 7-point Likert scale (1 is stronglydisagree and 7 is strongly agree). Thesedimensions are planning, coordinat-ing, evaluating, investigating, super-vising, staffing, negotiating, represent-ing, and overall performance.

As suggested by Brownell (1982)and also used by Lau and Moser (2008),

the study uses the overall performancescore as a measure of manager perfor-mance. Hence, to provide assurancethat the measure reflects the eight di-mension scores, the researcher regressthe eight performance dimensions (asindependent variables) on overall score(as dependent variables). The resultsshow that the eight performance vari-ables can explain around 58.8 percentof the variance in the overall perfor-mance variable, and this exceeds theminimum benchmark of 55 percentsuggested by Mahoney et al. (1963).The mean score for managerial perfor-mance is 5.58 with the standard devia-tion of 0.78 (see Table 3)

Hypothesis Tests

This study expects that proceduraljustice will strengthen the relationshipbetween the BSC measures (financialand nonfinancial) and the organiza-tional commitment. It is also expectedthat the organizational commitmentwill positively and significantly affectmanagerial performance. Regressionanalyses are used to test the hypoth-eses. Equation 1 depicts these tworegression models.

Equation 1. The Multiple andSimple RegressionModels

OC = 0 +

1 NM +

2 FM +

3 PJ

+ 4 NM*PJ+

5 FM*PJ +

1

MP = 0 +

1 OC +

2

where:

OC= organization commitment

429


NM= nonfinancial measure

FM= financial measure

PJ= procedural justice

MP = managerial performance

The first hypothesis is supportedwhen

4 is significant with a positive

sign. Similarly, the second hypothesisis supported when

5 is significant

with a positive sign, while the lasthypothesis is supported when

1 is also

significant with a positive sign.

Results of Hypothesis Tests

The study runs the regressionmodel using SPSS version 15.0. Table5 presents the results of running themoderating regression model based on76 observation data. The descriptivestatistics for the data used in the re-gression models are presented in Table3.

The results of regressing the mod-erating model indicate that the modelis reliably used to test the hypotheseswith the goodness of fit F-value of9.788 and significant at p-value of0.000. All independent variables in themodel can explain 36.9 percent varia-tions in the organizational commit-ment (adjusted R2 = 0.369). Further-more, all independent variables are

also significant at a= 1 percent (exceptvariable Nonfinancial Measure * Pro-cedural Justice at a= 5%). Interest-ingly, all the coefficient relationshipsbetween the main individual variablesof nonfinancial measure, financialmeasure, and procedural justice andorganizational commitment have sig-nificant negative signs, while earlierstudies document positive signs. Whena one-tail t-test isused with the expec-tation of positive direction, then thenegative coefficient will be consid-ered as insignificant. These results re-flect that procedural justice is neitherrelated to the BSC measures nor orga-nizational commitment, but it inter-acts with the BSC measures to modifythe relationship form of the BSC mea-sures and organizational commitment(Sharma et al. 1981). Thus, the proce-dural justice can function as a puremoderator variable in the relationship.2

Table 5 also shows that the coef-ficient of (Nonfinancial Measure *Procedural Justice) is 0.986 with t-value of 2.348 and at a of 5 percentindicating strong support to the firsthypothesis. As expected proceduraljustice significantly strengthens thepositive effect of using the nonfinan-cial measure when assessing manage-

2 The moderating model is robust since there are insignificant direct relationship betweenorganizational commitment and nonfinancial measures (coefficient 0.021, t-value 0.564, and p-value 0.574); organizational commitment and financial measures (coefficient -0.030, t-value -0.942, and p-value 0.349); organizational commitment and procedural justice (coefficient 0.180, t-value 0.416, and p-value 0.678); and nonfinancial measures and procedural justice (coefficient -0.001, t-value -0.002, and p-value 0.998). The direct relationship between finanacial measures andprocedural justice is significant, however, the sign is negative (coefficient -0.256, t-value -3.275,and p-value 0.002). This negative direct relationship is inconsistent with the finding in earlierstudies.

430


Table 6. The Regression Results of the Managerial Performance (MP) Model

MP = 0 +

1 OC +

2

Variable Beta Std. t-value p-valueError

Constant -48.533 5.183 -9.363 0.000

Organizational Commitment 7.978 0.760 10.498 0.000

R2 0.598

Adjusted R2 0.593

F-value 110.214 (p-value: 0.000)

Table 5. The Results of the Organizational Commitment (OC) ModeratingRegression Model

OC = 0 +

1 NM +

2 FM +

3 PJ +

4 NM*PJ+

5 FM*PJ +

1

Variable Beta Std. t-value p-valueError

Constant 34.808 8.188 4.251 0.000

Nonfinancial Measure -3.966 1.458 -2.721 0.008

Financial Measure -2.205 0.699 -3.155 0.002

Procedural Justice -7.258 2.337 -3.105 0.003

Nonfinancial Measure * Procedural Justice 0.986 0.420 2.348 0.022

Financial Measure * Procedural Justice 0.616 0.199 3.096 0.003

R2 0.411

Adjusted R2 0.369

F-value 9.788 (p-value: 0.000)

rial performance through organiza-tional commitment. Similarly, the re-sults also significantly support the sec-ond hypothesis when the coefficient of(Financial Measure * Procedural Jus-tice) is 0.616 with t-value of 3.096 andat the a of 1 percent. As expectedprocedural justice significantly

strengthened the positive effect of us-ing the financial measure when assess-ing managerial performance throughorganizational commitment. Theseresults support earlier studies done by,among others, Lau and Mosser (2008)and Lau and Sholihin (2005). Hence,introducing procedural justice into the

431


process of managerial performancemeasurement systems will improve theeffectiveness of implementing the BCSas a part of performance evaluationsystem.

Table 6 presents the simple re-gression results to test the third hy-pothesis that managers’ organizationalcommitment is positively related tomanagers’ managerial performance.The results of the regression indicatethat the model is reliably used to testthe hypotheses with the goodness of fitF-value of 110.214 and significant atp-value of 0.000. The independentvariable in the model can explain 59.3percent variations in the managerialperformance (adjusted R2 = 0.593). Asexpected the coefficient of organiza-tional commitment (7.978) was posi-tively significant with t-value of 10.498and p-value of 0.000 supporting hy-pothesis three.

Conclusion, Discussion, andLimitation

The aim of this study is to exam-ine the moderating effects of proce-dural justice on the effectiveness ofthe BSC performance measures inimproving managerial performancethrough organizational commitment.The study remodels the mediating ef-fect of procedural justice on the rela-tionship between the BSC measureand managerial performance found inprior studies, into the moderating ef-fect based on the contingency theory.

The results of this study supportprevious studies in organizational and

managerial accounting areas on theeffect of procedural fairness on indi-vidual manager or employee behavior(Wentzel 2002; Libby 1999; Magneret al. 1995; Lindquist 1995; andMcFarlin and Sweeney 1992; amongothers). However, this study extendsprior studies by remodeling the role ofprocedural justice and providing em-pirical evidence. While the earlier stud-ies indicate the role of procedural jus-tice as a mediating variable on therelationship between management per-formance measure and managerial per-formance, this study proposes the pro-cedural justice as moderating variableon the relationship. Empirical evidencestrongly supports the proposition inthis study that procedural justice has apure moderating role.

There are some cautions to inter-pret the results of the study due to thefollowing limitations: First, there is aninherent embedded limitation of usinga survey method especially related tothe possibility of missing appropriatecontext and of lacking respondents’knowledge. However, the use of anempirically tested valid instrument andexperienced respondents in this studyare expected to mitigate these draw-backs. Future studies may considerthese drawbacks by employing othermethods, such as field or laboratoryexperiments. Secondly, the study doesnot separate the dimensions of the non-financial measures of the BSC intothree perspectives, learning andgrowth, internal business, and customerperspectives, as developed by Kaplanand Norton (1992). Combining the

432


three perspectives into one dimensionmay reduce the appropriateness of themeasure when, in fact, respondentsinterpret the measures differently. Fu-

ture studies can deal with this limita-tion by separating the nonfinancialmeasures into three dimensional mea-sures.

ReferencesAmir, A., and B. Lev. 1996. Value relevance of nonfinancial information: The wireless

communications industry. Journal of Accounting and Economics 22: 3-30.

Anthony, R. N., and V. Govindarajan. 2005. Management Control Systems (11th Ed.)Chicago, IL: Irwin.

Brownell, P. 1982. The role of accounting data in performance evaluation, budgetaryparticipation, and organizational effectiveness. Journal of Accounting Research 20(1) Spring: 12-27.

Cohen, R. L. 1987. Distributive justice: Theory and research. Social Justice Research 1(1): 19-40.

Cohen, A. 1993. Organizational commitment and turnover: A meta analysis. Academy ofManagement Journal 36 (5): 1140-1157.

Colquitt, J. A., D. E. Conlon, M. J. Wesson, C. O. L. H. Porter, and K. Yee Ng. 2001. Justiceat the Millennium: A meta-analytic review of 25 years of organizational justiceresearch. Journal of Applied Psychology 86 (3): 425-445.

Fiedler, F. E. 1983. ‘Assessing the validity of fiedler’s contingency model of leadershipeffectiveness: A closer look at Strube and Garcia.’ Psychological Bulletin 93: 404–8.

Fisher, J. 1992. Use of nonfinancial performance measures. Journal of Cost Management(Spring): 31-38.

Folger. 1987. Distributive and procedural justice in the workplace. Social JusticeResearch 1 (2): 143-160.

Greenberg, J. 1987. A taxonomy of organisational justice theories. Academy of Manage-ment Review 12 (1): 9-22.

Hair, J. F., R. L. Tatham, R. E. Anderson, and W. Black. 2006. Multivariate Data Analysis.New Jersey: Prentice Hall.

Jaros, S., J. Jermier, J. Koehler, and T. Sincich. 1993. Effects of Continuance, affective andmoral commitment on the withdrawal process: An evaluation of eight structuralequation models. Academy of Management Journal 36: 951-995.

Kaplan, S. E., and J. T. Mackey. 1992. An examination of the association betweenorganizational design factors and the use of accounting information for managerialperformance evaluation. Journal of Management Accounting Research 4: 116-130.

Kaplan, R. S., and D. P. Norton. 1992. The balanced scorecard – measures that driveperformance. Harvard Business Review 70 (1): 71-79.

433


———. 1993. Putting the balanced scorecard to work. Harvard Business Review 71 (5):134-42.

———. 1996a. Using the balanced scorecard as strategic management system. HarvardBusiness Review 74 (1): 75-85.

———. 1996b. The Balanced Scorecard: Translating Strategy into Action. Boston, MA:Harvard Business School Press.

———. 2001. Transforming the balanced scorecard from performance measurement tostrategic management: Part 1. Accounting Horizons 15 (1): 87-104.

Korsgaard, M. A., and L. Roberson. 1995. Procedural justice in performance evaluation:The role of instrumental and non-instrumental voice in performance appraisaldiscussions. Journal of Management 21 (4): 657-669.

Lau, C. M., and A. Moser. 2008. Behavior Effects of Nonfinancial Performance Measures:The Role of Procedural Fairness. Behavioral Research in Accounting 20 (2): 55-71.

Lau, C. M., and M. Sholihin, 2005. Financial and nonfinancial performance measures:How do they affect job satisfaction? The British Accounting Review 37: 389-413.

Lau, C. M., and E. Lim. 2002. The intervening effects of participation on the relationshipbetween procedural justice and managerial performance. British Accounting Review34: 55-78.

Lau, C. M., and C. Buckland, 2001. Budgeting-role of trust and participation: a researchnote. ABACUS 37: 369-388.

Lau, C. M., L. C. Low, and I. Eggleton. 1995. The impact of reliance on accountingperformance measures on job-related tension and managerial performance: Addi-tional evidence. Accounting, Organizations and Society 20 (5): 359-381.

Laventhal, G. S., J. Karuza Jr., and W. R. Fry, 1980. Beyond fairness: A theory ofallocation preferences. In G. Mikula, Justice and Social Interaction: 167-218. Bern,Switzerland: Hans Huber Publishers.

Libby, T., 1999. The influence of voice and explanation on performance in a participativebudgeting setting. Accounting, Organizations and Society 24 (2): 125-137.

Libby, T., S. E. Salterio, and A. Webb. 2004. The balanced scorecard: The effects ofassurance and process accountability on managerial judgment. The AccountingReview 79 (4): 1075-1094.

Liedtka, S. L., B. K. Church, and M. R. Ray. 2008. Performance Variability, AmbiguityIntolerance, and Balanced Scorecard-Based Performance Assessments. BehavioralResearch in Accounting 20 (2): 73-88.

Lindquist, T. 1995. Fairness as antecedent to participative budgeting: Examining theeffects of distributive justice, procedural justice and referent cognitions of satisfac-tion and performance. Journal of Management Accounting Research 7: 122-147.

Lipe, M. G., and S. E. Salterio. 2000. The balanced scorecard: Judgmental effects ofcommon and unique performance measures. The Accounting Review 75 (3): 283-298.

———. 2002. A note on the judgmental effects of the balanced scorecard’s informationorganization. Accounting, Organizations and Society 27 (6): 531-540.

434


Mahoney, T., T. Jerdee, and S. Carroll. 1963. Development of Managerial Performance:A Research Approach. Cincinnati, OH: South-Western Publishing Company.

Marr, B., Neely, A., Bourne, M., Franco, M., Wilcox, M., Adams, C., Mason, S., &Kennerley, M. (2004). Business performance measurement - What is the state of usein large US Firms? Proceedings of PMA 2004 Conference. UK Edinburgh, July 2004.

McFarlin , D., and P. Sweeney. 1992. Distributive and procedural justice as predictors ofsatisfaction with personal and organizational outcomes. Academy of ManagementJournal 35 (3): 626-637.

Mowday, R. T., R. M. Steers, and L. W. Porter. 1979. The measurement of organizationalcommitment. Journal of Vocational Behavior 14 (2): 224-247.

Otley, D. T. 1999. Performance management: A framework for management controlsystems research. Management Accounting Research 10 (4): 363-382.

Porter, L. W., R. M. Steers, R. T. Mowday, and P. V. Boulian. 1974. Organizationalcommitment, job satisfaction, and turnover among psychiatric technicians. Journalof Applied Psychology 39 (5): 603-609.

Rigby, D. (2001). Management Tools and Techniques: A Survey. California ManagementReview 43 (2), 139-160.

Sharma, S., R. M. Durand, and O. Gur-Arie. 1981. Identification and Analysis ofModerator Variables. Journal of Marketing Research (August): 291-300.

Slovic, P., and D. MacPhillamy, 1974. Dimensional commensurability and cue utilisationin comparative judgment. Organisational Behavior and Human Performance 11:172-194.

Wentzel, K. 2002. The influence of fairness perceptions and goal commitment onmanagers’ performance in a budget setting. Behavioral Research in Accounting 14:247-271.

the moderating effect of procedural justice on the

Documents