J. Eng. Technol. Manage. 33 (2014) 1–17

Contents lists available at ScienceDirect

Journal of Engineering andTechnology Management

journal homepage: www.elsevier.com/locate/jengtecman

The antecedents of collective creative efficacy for

information system development teams

Hsiu-Hua Cheng a,1, Heng-Li Yang b,*a Department of IM, Chaoyang University of Technology, 168, Jifeng E. Rd., Wufeng District, Taichung 41349,Taiwan, ROCb Department of MIS, National Chengchi University, 64, Sec. 2, Zhinan Rd., Wenshan District, Taipei 11605,Taiwan, ROC

A R T I C L E I N F O

JEL classification:

M15

D83

Keywords:

Information system development

Collective creative efficacy

Knowledge integration capability

Achievement motivation

A B S T R A C T

Improvement processes for information system development, such

as creativity, have seldom been addressed. Based on social cognitive

theory, this study explores the influence and moderating effect of

collective creative efficacy (CCE) on software process improvement.

The partial least square method is applied to analyze data from 61

development teams. Analytical results indicate: (1) team knowl-

edge, achievement motivation, and knowledge integration cap-

ability positively influence CCE. (2) Interpersonal interaction

enhances the relationship between team knowledge and CCE. (3)

Project complexity weakens the relationship between team knowl-

edge and CCE, but strengthens that between achievement motiva-

tion and CCE. Theoretical and practical implications are discussed.

� 2013 Elsevier B.V. All rights reserved.

Introduction

Although software projects are typically well founded, they have had a high failure rate. Accordingto the Standish Group International (Jiang et al., 2004), about 15% of projects never delivered a finalproduct, costing US$67 billion per year. Failure to meet deadlines is endemic in the informationtechnology industry (Parolia et al., 2007).

Software process improvement can reduce the likelihood of software project failure. Manysoftware companies rely on continuous improvements to their processes to increase project

* Corresponding author. Tel.: +886 2 29399091x81005; fax: +886 2 29393754.

E-mail addresses: hhcheng@cyut.edu.tw (H.-H. Cheng), yanh@nccu.edu.tw (H.-L. Yang).1 Tel.: +886 4 23323000x7834; fax: +886 4 23742337.

0923-4748/$ – see front matter � 2013 Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.jengtecman.2013.12.001

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–172

performance (Allison and Merali, 2007; Galinac, 2009; Sun and Liu, 2010). During softwaredevelopment, an information system development (ISD) team generally proposes new ideas forsoftware processes or for improving existing software processes. However, software processimprovement has seldom been discussed from the perspective of creativity. This study assumessoftware process improvement is creativity for an ISD team.

Research has identified creative self-efficacy as a key influence on creativity (Tierney and Farmer,2002). Previous studies also indicated the relationship between efficacy and creativity at theindividual level. However, although teams are central to organizations, previous researchers onlyindicated that collective efficacy is positively related to team performance (Akgun et al., 2007); theyseldom discussed the relationship between self-efficacy and creativity at the team level (i.e., collectiveefficacy and collective creativity). Cheng and Yang (2011) combined literature on collective efficacyand creativity to propose collective creative efficacy (CCE) at the team level. In the ISD context,software process improvement may require that team members develop and share ideas, and mayrequire team creativity. This study (1) explores the antecedents of CCE from the perspective ofinformation accumulation, and (2) investigates the moderating effects of team and environmentalvariables on the relationships between antecedents and CCE.

Literature and hypotheses

Creativity

Oldham and Cummings (1996) defined creative outcomes as new, original, suitable and usefulideas or processes. Guilford (1984) argues that team creativity as an outcome of divergent teamthinking. Divergent and convergent thinking can help teams to be creative; that is divergent thinkingcan help teams develop new ideas, while convergent thinking can enable teams to integrate ideas asteam outcomes. According to Kurtzberg and Amabile (2001), information and interpersonalinteraction contribute to team creativity. Oldham and Cummings (1996) defined team creativityas creative synthesis, meaning that teams develop and share ideas and provide encouragement toproduce creative results. In ISD teams, members cooperate on various project-related matters andintegrate individual outcomes or ideas into systems. Thus, this study also defines team creativity as acreative synthesis, in which ISD team members propose new and useful ideas for improving softwaredevelopment procedures, suggest novel software processes, or integrate useful modules intoinformation systems.

Collective efficacy

Collective efficacy is the belief of a team in the abilities of its members to successfully completetasks (Bandura, 1997; Gibson, 1999, 2001); that is, it is a shared belief in the ability of a group toachieve specific goals (Bandura, 1997). Gibson and Earley (2007) indicated that collective efficacy is acognitive phenomenon, a belief in a general context. To discuss team creativity in depth, Cheng andYang (2011) developed collective creative efficacy (CCE), a new construct, and defined it as the sharedbelief that combining the abilities of team members will result in creative ideas.

Gibson (2001) asserted that collective efficacy is the product of group-level cognitive processes.Gibson and Earley (2007) indicated that collective efficacy is a cognitive phenomenon. A groupdeveloped collective efficacy through four collective cognitive processes (Gibson, 2001): (1)accumulation is the process through which teams acquire information and knowledge for developingcollective beliefs; (2) interaction allows team members exchange information and knowledge viainteractions; (3) examination is the process of team members cooperating to negotiate, interpret, andevaluate knowledge and information to form a collective efficacy; and (4) accommodation occurswhen team members select behaviors according to information processed during previous phases.

Based on the collective cognitive processes proposed by Gibson (2001), Gibson and Earley (2007)indicated that team members can combine their information and experiences to form collectiveefficacy by internal communication. This study integrates the collective cognitive processes of Gibson(2001), and Gibson and Earley (2007) and creativity theory by Amabile (1983) to generate the

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–17 3

proposed research model. This study explored the information sources for accumulating collectiveefficacy from the perspective of creativity literature. Thus, this study proposes that team knowledge,achievement motivation and integration capability are the primary antecedents of CCE. According tothe interaction and examination processes, this study proposes interpersonal interaction is amoderator of the relationship between team knowledge and CCE. Further, the study also proposes thatproject complexity is a moderator of the relationships between these three antecedents and CCE.

CCE and its antecedents

CCE is a result of collective cognition. Cognitive complexity is a characteristic of group cognition.Houghton et al. (2009) indicated that cognitive complexity refers to the breadth and integration ofmental models. Tetlock et al. (1989) indicated that cognitive complexity is composed of twodimensions: differentiation and integration. Differentiation is ‘‘the number of evaluatively distinctdimensions of a problem that an individual takes into account’’; and integration is ‘‘the developmentof complex connections among differentiated characteristics’’ (Tetlock et al., 1989, p. 635).

Several studies addressed group cognitive complexity. Curseu et al. (2007) examined the effects ofgroup variety and disparity on group cognitive complexity and discussed the moderating effect ofinteractions quality. Group cognitive complexity was deemed as one construct (Curseu et al., 2007).However, Houghton et al. (2009) mentioned that group cognitive complexity comprises differentia-tion and integration; their study investigated the relationship between team cognitive complexity andperformance. Similarly, this study also addressed these two components of group cognitivecomplexity: group differentiation and group integration. The former refers to team knowledge, andthe latter is synonymous with knowledge integration capability.

Gibson and Earley (2007) showed that team member characteristics are a key aspect of informationaccumulation during collective cognitive processes. Gibson and Earley argued that when teammembers are aware of the information and knowledge they possess, team certainty will increase,helping them achieve task objectives. In this study, team knowledge is similar to group differentiation,a component of cognitive complexity (Tetlock et al., 1989). Group differentiation is a function of thenumber of differentiated perspectives, ideas, or opinions (Choi and Coen, 2009).

Full and rich knowledge can help teams develop fresh and creative perspectives (Amabile, 1997).Un (2010) also noted that knowledge is essential to innovation. Teams that possess rich knowledgeand skills related to tasks can be creative (Dunbar, 1997). When a team has work-related knowledge,team members can often find solutions from different perspectives (Paulus, 2000). Tierney and Farmer(2002) identified knowledge as important for creative self-efficacy. Yang and Cheng (2009) alsoindicated that information system developers with considerable information technology knowledgeand skills are likely to feel confident that they can be creative at work.

According to this literature, this study defined team knowledge as the degree to which a projectteam possesses varied knowledge such as system analysis and design knowledge, programmingknowledge, project management knowledge, and domain knowledge (Barki et al., 2001). When teammembers perceive that their team has rich task knowledge, they are confident to that they will findsolutions from various perspectives and develop new ideas to complete their tasks or improvesoftware processes. This study thus proposes Hypothesis 1 as follows:

Hypothesis 1. Team knowledge positively influences CCE.

According to the accumulation phase of collective cognition by Gibson and Earley (2007), teammember characteristics form a context for acquiring and filtering information related to CCEdevelopment. Achievement motivation is defined as the strength of individual’s desire to excel, tosucceed at difficult tasks, and to complete them better than others (Chen, 2008). The study adoptsChen’s definition of achievement motivation. Zander and Forward (1968) demonstrated that teammembers with high achievement motivation are interested in team success. Furthermore, Chen et al.(2002) determined that average team member achievement motivation significantly increasescollective efficacy. Chen (2008) later indicated that people with high achievement motivation desiresuccess and to execute difficult tasks. Helmreich and Spence (1978a) demonstrated that individualswith high achievement motivation prefer to confront challenging tasks.

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–174

Modifying software processes and methods is a challenging task. When ISD team membersperceived that other team members have high achievement motivation, they believe that they favorconfronting challenge and work to complete tasks during software development. Thus, when an ISDteam contains many members with high achievement motivation, this achievement motivationincreases the team’s creativity development. This study thus proposes Hypothesis 2 as follows:

Hypothesis 2. Achievement motivation of all team members positively influences CCE.

Gibson and Earley (2007) indicated that team work process characteristic is one source of collectivebelief in team ability. Lester et al. (2002) also found that collaboration knowledge processes influencejudgments of collective efficacy. When team members flexibly integrate individual knowledge toaccomplish team tasks, they realize that their team can develop its creativity (i.e., collective belief inteam ability). In this study, knowledge integration capability resembles group integration of cognitivecomplexity (Tetlock et al., 1989). Group integration is defined as the extent to which conceptualconnections exist among differentiated perspectives within a group (Choi and Coen, 2009).

Amabile (1997) noted that creative thinking skills are necessary for individual/team creativity.Amabile (1997) argued that creativity skills include a cognitive style favorable to developing newperspective for existing problems, applying techniques for the exploration of new cognitive pathways,and developing a working style that underlines persistent, energetic pursuit of one’s work. Accordingto knowledge integration theory by Grant (1996), an organization with good knowledge integrationcapability can combine new and existing knowledge or recombine existing knowledge to producecreativity. Knowledge integration capability is therefore key to creativity.

Bhandar et al. (2007) defined knowledge integration as the process of combining, applying, andabsorbing different knowledge types. Furthermore, Tiwana et al. (2003) defined knowledgeintegration as a process via which organizations absorb external knowledge and combine thisexternal knowledge, internal techniques, and domain knowledge. Some researchers extendedorganizational knowledge integration capability to team knowledge integration capability, andasserted that it exists when teams have the ability to combine individual ideas and information tocreate products through teamwork (Okhuysen and Eisenhardt, 2002). Tiwana and McLean (2005)asserted that knowledge integration synthesizes individual expertise at the project level. Kleinsmannet al. (2010) indicated that knowledge integration is important in collaborative new productdevelopment. Houghton et al. (2009) argued that group integration ability influences performance. Inthis study, team knowledge integration capability is to the degree of capability a project team has totransform team members knowledge to be team outcomes.

Good knowledge integration capability for ISD teams means that team members can obtain ideasand knowledge within the team and combine them into team outcomes. Teams typically perceive CCEas high as their knowledge integration capability increases. This study thus proposes Hypothesis 3 asfollows:

Hypothesis 3. Knowledge integration capability positively influences CCE.

Interpersonal interaction among team members and CCE

Team members develop collective efficacy based on the meaning associated with information(Gibson, 2003). Hargadon (1999) noted that social interaction drives individuals to recognizeknowledge diversity. Team members can know which knowledge members have and use thisknowledge to resolve problems that other members confront. According to Leenders et al. (2003),team creativity requires a moderate frequency of communication because existing knowledge isdisseminated through interactions among employees with expertise in different areas.

Curseu et al. (2007) explored the effect of group interaction processes on group cognitivecomplexity. They argued that the quality of interpersonal interactions within a group is key to theintegration of individual knowledge structures (individual cognition) into a group knowledgestructure (group cognition). Furthermore, their empirical study indicated that the relationshipbetween average individual cognitive complexity and group cognitive complexity is moderated by thequality of interpersonal interactions.

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–17 5

In this study, team member interaction is the extent of closeness for working relationships, frequencyof communications and reciprocity among project team members. When working interaction is focusedon accessing information and knowledge for task completion, it might include debates, discussions,conflict, negotiations, and collaboration. As described, close interactions can assist team members inunderstanding the knowledge of others and in using this knowledge to develop different approaches toovercoming ISD problems. That is, close interactions can help team members increase their confidence inthe fact that team products will be creative. This study thus proposes Hypothesis 4 as follows:

Hypothesis 4. Team member interaction strengthens the positive relationship between team knowledgeand CCE.

ISD project complexity and CCE

Researchers have demonstrated that task characteristics moderate assessments of efficacy belief(Gibson and Earley, 2007) because task characteristics are closely related to task execution ability(Goodman et al., 1990).

Gibson and Earley (2007) argued that task complexity moderates the relationship betweenperceived abilities of team members and collective efficacy. This moderating effect occurs becausetask complexity complicates assessments of the likelihood of that tasks will be completed andreducing judgments of efficacy. At the team level, task complexity exists when many very difficultsub-tasks exist and team members have difficulty coordinating their efforts to complete team tasks.Thus, the complexity of team-level tasks makes it difficult for team members to predict whichoutcomes teams will produce.

Baccarini (1996) defined project complexity based on the number of project elements and their inter-relationships. The complexity of project structures and environmental dynamics define projectcomplexity according to Williams (1999). Xia and Lee (2005) asserted that ISD project complexity iscomprised of four components: structural organization complexity; structural IT complexity; dynamicorganization complexity; and dynamic IT complexity. Structural organization complexity and structuralIT complexity denote the multiplicity and interdependency of organizational and technologicalelements of an ISD project. End users, managers, and external contractors are typical organizationalelements. Technological elements often include technology platforms, software environments, otherintegrated systems, and data processing requirements. Additionally, dynamic organization complexityand dynamic IT complexity refer to the rate and pattern of changes in an ISD project organization and ITenvironments. Based on the work by Xia and Lee (2005), this study describes the complexity of ISDproject as the strength of the complicated correlations among project organizational and technologicalelements and the change degree of rate and pattern of these elements.

When facing numerous techniques and user units as well as dynamic techniques and user units,analyzing and forecasting the relationship between project inputs and outcomes is difficult for an ISDproject. This study thus proposes Hypothesis 5 as follows:

Hypothesis 5. Project complexity weakens the positive relationship between team knowledge and CCE.

Chen (2008) indicated that individuals with high achievement motivation enjoy challenges andovercome difficulties, and to complete such tasks better than others. Complex projects provideopportunities to people with high achievement motivation to experience success. According toNicholls (1984), the relationship between achievement motivation and task selection can be modeledfrom a capability perspective. When people are involved in tasks, they typically try to master tasks andincrease their capabilities. Completing easy tasks does not require outstanding capabilities; and byselecting easy tasks, people avoid revealing that they have poor capabilities. Helmreich and Spence(1978b) indicated that personal achievement comprises mastery, work, competition, and a lack ofpersonal concern for the opinions of others. Chen (2008) indicated that personal achievement includesperseverance, competition, and difficulty control. Referring to the work by Helmreich and Spence(1978b) and Chen (2008), achievement motivation in this study is focused on personal achievementcomprising mastery, work, and competition.

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–176

Task preferences of individuals generally vary according to their degree of achievementmotivation. Complex projects can satisfy individuals with high achievement motivation. Teammembers believe that their teams are capable of developing better ideas when projects are complexand team members have high achievement motivation. Conversely, team members with highachievement motivation can have difficulty developing their creativity. Complex projects may providean appropriate environment to increase the influence of achievement motivation on CCE. This studythus proposes Hypothesis 6 as follows:

Hypothesis 6. Project complexity increases the strength of the positive relationship between teammember achievement motivation and CCE.

Project teams with good knowledge integration capability can expect to achieve easily teamoutcomes to simple tasks. A complex project implies that numerous techniques and user units areinvolved, such that predicting the relationships among project elements is difficult for individuals (Xiaand Lee, 2005). Teams with high knowledge integration capability may still not achieve certainoutcomes for complex projects. This study thus proposes Hypothesis 7 as follows:

Hypothesis 7. Project complexity reduces the strength of the positive relationship between knowledgeintegration capability and CCE.

Based on the above reasoning, this study explores the antecedents and moderators of CCE of ISDteams. Fig. 1 shows the proposed research model.

Compared our model with the literature

This study integrated collective cognitive processes identified by Gibson (2001), and Gibson andEarley (2007) and creativity theory by Amabile (1983, 1997) to develop the antecedents of CCE, asshown in Fig. 1.

Amabile (1997) indicated that expertise, creative thinking, and task motivation are antecedents ofcreativity. The differences between this model and that by Amabile are as follows.

First, expertise can be viewed as the set of cognitive pathways that may be followed to solve a givenproblem or complete a task (Amabile, 1997). Thus, individuals with expertise can typically focus onthe important aspects of problems. This study adopted the cognitive perspective by Gibson and Earley(2007) to propose that team knowledge is an antecedent variable of CCE. Gibson and Earley (2007)argued that when team members are aware of the information and knowledge they possess and those

[(Fig._1)TD$FIG]

H1

H2

H3

H4

H5 H7 H6

Knowledge

integration

capability

Collective

creative

efficacy

Project

complexity

Interpersonal

interaction

Team knowledge

Achievement

motivation

Fig. 1. Research model.

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–17 7

of every other team member, team certainty increases, helping them achieve task objectives. Thus,this study argues team knowledge to be the antecedent of CCE.

Second, Amabile (1997) asserted that individual/team creativity skills include a cognitive stylefavorable to taking new perspective on problems, applying techniques to explore new cognitivepathways, and a working style conductive to persistent, energetic pursuit of one’s work. However, shedid not specify how to convert the skills of individuals into team creativity skills. In contrast withAmabile’s model, Gibson and Earley (2007) indicated that the characteristic of team work process is asource of collective belief in team ability. Lester et al. (2002) also found that collaborative knowledgeprocesses influence assessments of collective efficacy. When team members can integrate individualknowledge to accomplish tasks, they realize that their team has the ability to develop its creativity. Inthis study, knowledge integration capability refers to a collective belief in team integration ability.

Third, task motivation determines the extent to which a person will fully apply his expertise andcreative thinking skills to develop his creativity (Amabile, 1997). This study focused on teammembers’ intrinsic motivation instead of task motivation. According to the accumulation phase ofcollective cognition identified by Gibson and Earley (2007), team member characteristics form acontext for acquiring and filtering information related to CCE development. In the collective cognitiveprocess, when team members perceive that they have high achievement motivation, they realize thattheir team has a desire to develop its creativity. Thus, this study proposes that achievementmotivation is a key factor of CCE.

Additionally, during the interaction process, information and knowledge are exchanged; andduring the examination/negotiation/interpretation process, information and knowledge are evaluated(Gibson, 2001; Gibson and Earley, 2007). Group interactions (debates, negotiations, discussions) mayshape and reshape the content and structure of a group’s cognitive map (Curseu et al., 2007). Thus, thisstudy proposes interpersonal interaction moderates the relationship between team knowledge andCCE. Finally, considering the context of information system development, this study suggests thatproject complexity, another moderator, is an important environment variable.

Methods

Investigation procedures

Questionnaire data were collected and used for model testing. This study reached some ISD teams andpleaded for cooperation through personal relationships with or third-party introduction to the teammanagers. The ISD project teams were selected as the research participants. Three inclusion criteria wereapplied: (1) study participants must be on ISD project teams; (2) study participants were on teams with3–8 members (team size is related to the number of member interactions, this study did not survey largeteams); (3) teams had to exist for two months (the development of interpersonal relationships takes time.

This study enrolled 62 ISD teams with 322 members from 26 software companies in Taiwan whichdevelop custom information systems. After obtaining agreement from ISD teams, questionnaires weresent to project managers, who then distributed questionnaires to all of their information systemdevelopers. Each participant completed a self-reported questionnaire, which gathered data for teamknowledge, achievement motivation, knowledge integration capability, interpersonal interaction,project complexity, and collective creative efficacy. Participants were required to seal the completedquestionnaires themselves before returning them to their managers. All members of the 62 teams(including 322 members) completed and returned their questionnaires, for a valid response rate of 100%.

Of all 322 respondents, 72% were male and 28% were female. In total, 68% had a college degree. Theprojects included customizing enterprise resource planning systems, developing portal platforms,designing information systems to solve business problems, and migrating/upgrading systemplatforms. Of all projects, 28% had schedules exceeding 12 months.

Measures

First, this study reviewed literature to develop construct measures for the questionnaire.Second, three teachers and two software industry experts with ten years of work experience

Table 1The sources and reliability of the original questionnaire.

Construct Definition Source Reliability

of original

questionnaire

Team

knowledge

The degree to which the project

team possesses various knowledge,

such as system analysis and design

knowledge, programming

knowledge, project management

knowledge and domain knowledge

Items were drawn from the

measures of four variables of risk

exposure (‘‘lack of development

expertise’’, ‘‘lack of expertise with

applications’’, ‘‘lack of expertise

with task’’, and ‘‘lack of general

expertise’’) (Barki et al., 1993, 2001)

and then modified

>0.75

Achievement

motivation

The need to be outstanding and

successful, to execute difficult tasks,

and to excel in relation to others

Items were drawn from Chen (2008) >0.6

Knowledge

integration

capability

The degree of capability a project

team has to transform team

members knowledge to be team

outcomes

Items were drawn from Tiwana and

McLean (2005)

0.95

Interpersonal

interaction

The extent of closeness of working

relationships, frequency of

communications and reciprocity

among project team members

Items were drawn from Yang and

Cheng (2009). In addition, this study

adopted weighted-ties formulation,

which was proposed by Wasserman

and Faust (1994), to calculate

interpersonal interaction

0.94

Project

complexity

The strength of the complicated

correlations among project

organizational and technological

elements and the change degree of

rate and pattern of these elements

Items were drawn from Xia and Lee

(2005) and then modified

>0.65

Collective

creative

efficacy

The shared belief of team members

in combining their abilities to

develop creative work ideas

Items were drawn from Cheng and

Yang (2011)

0.92

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–178

were invited to verify the face and content validity of the questionnaire. Third, questionnaireitems were modified based on their recommendations. Table 1 lists construct definitions andsources.

Analysis and results

Aggregation tests

This study justified its response aggregation via test inter-rater agreement (rwg) and appliedANOVA to test whether between-group variance was sufficient for team-level modeling.

Testing for team-level effects requires aggregation by group member scores of team knowledge,achievement motivation, knowledge integration capability, CCE, and project complexity. This studytested the suitability of such aggregation for examining between-group differences and within-groupagreement for these measures (Goodman et al., 1990). According to Amason (1996), one-way ANOVAcan test between-group differences. Analytical results indicate that between-group variances for thefive constructs significantly exceed within-group variances.

This study estimated rwg using a method developed by James et al. (1993), which assess within-group consistency using ratings on a common scale. When rwg exceeds 0.7, within-group agreementexists (George, 1990). In this study, analytical results show that rwg of project complexity of one teamwas 0.6, lower than 0.7. Data from this team were thus deleted. Five constructs for remaining teamdata (61 teams composed by 316 members) were suitable for aggregation by averaging scores forgroup members.

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–17 9

Reliability and validity of the measurement

Since all participants completed the self-reported questionnaire, common method bias wasassessed. Common method bias can be tested via the Harmon’s one-factor test (George, 1990). Sixfactors (team knowledge, achievement, knowledge integration capability, interpersonal interaction,project complexity and CCE) with eigenvalues exceeding 1 were extracted, accounting for 73.71% ofvariance. Since no single factor emerged, and no single general factor accounted most variance,common method bias did not exist.

The research hypotheses were tested using partial least squares (PLS). PLS is a component-basedapproach that assesses construct reliability and validity and estimates the relationships amongconstructs (Wold, 1982). Thus, PLS can accommodate different variable type, as well as direct, indirect,and moderating effects (Wold, 1982), such that latent constructs to be modeled as formative orreflective indicators (data in this study are both types), and it makes minimal demands onmeasurement scales, sample size, and residual distributions (Chin, 1998b). Using ordinary leastsquares as its estimation technique, PLS performs an iterative set of factor analyses and appliesbootstrap approach to assess the significance (t values) of paths (Chin et al., 2003).

Before validating measurements, each construct must be judged to determine whether it belongsto the formative scale or reflective scale. According to Jarvis et al. (2003), the scale judgment ofconstructs is based on four principles: (1) cause–effect among constructs and indicators; (2) exchangeamong indicators; (3) covariance among indicators; and (4) theories of constructs. According toprinciples, team knowledge and project complexity are measured on the formative scale; otherconstructs are measured on the reflective scale.

ISD teams typically need various knowledge types, including system development knowledge anddomain knowledge. Additionally, not all indicators of team knowledge can be exchanged. Projectcomplexity is also a formative scale. Project complexity is influenced by many factors, such as numberof techniques elements, number of user units, the dynamic of techniques and the dynamic of userunits. For example, project complexity indicators included the dynamic of user requirements and thenumber of technology platforms. These two indicators are independent because a low covarianceexists between them.

This study applied factor analysis to analyze the communality of team knowledge and projectcomplexity. Analytical results indicate that team knowledge includes two factors: developmentknowledge and domain knowledge. Project complexity includes three factors: system requirementcomplexity; information technology complexity; and customer organization dynamic.

Reliability, internal consistency and validity of the measures were assessed. According to Chin(1998a), when estimating a measurement model, a formative scale must assess indicator weightsand a reflective scale must assess indicator factor loading. This study assessed internal validity basedon indicator loadings. Hair et al. (1998) noted that indicator loadings must exceed 0.3 which indicatesacceptability. Furthermore, weights and loadings testing demonstrate consistency amongindicators. This study assessed convergent validity based on average variance extracted (AVE).Fornell and Larcker (1981) indicated that an AVE score exceeding 0.5 indicates acceptability. Table 2lists weights, loadings, t-values for all indicators and the AVE of each construct. All path loadings ofreflective indicators exceeded 0.3 (p<0.05), and all path weights of formative indicators aresignificant. Thus, all indicators exhibit validity and all AVE value for constructs exceed theacceptability threshold.

Construct internal reliability should exceed 0.7 (Nunnally, 1978). Bagozzi and Yi (1988)recommended a threshold of 0.7 for composite reliability. The alpha value of tie-strength ofinterpersonal interaction is 0.93. All the other alpha coefficients exceed 0.7, as does the value forcomposite reliability of reflective constructs (Nunnally, 1978; Bagozzi and Yi, 1988). All constructs arethus reliable. Additionally, all AVE values exceed 0.5. That is, measurements exhibit convergentvalidity (Fornell and Larcker, 1981).

Finally, instrument discriminant validity was verified by deriving the square root of AVE (Fornelland Larcker, 1981). The outcome listed in Table 3 confirms that all constructs have discriminantvalidity: the square root of AVE for each construct exceeds that of the correlations involving theconstruct (Chin, 1998a).

Table 2The AVE, reliability, loading/weight and t value.

Construct and indicator Scale Cronbach’s alpha CR AVE Loading/

weight

t-value

Team knowledge Formative scale 0.86 – –

TK1: development knowledge (system analysis and design knowledge, programming knowledge,

project management knowledge)

0.82 4.14*

TK2: domain knowledge 0.21 2.01*

Achievement motivation Reflective scale 0.80 0.86 0.51

M1: doing challenging and difficult job 0.79 5.39*

M2: liking to work in competitive environment 0.63 2.98*

M3: enduring temporary sacrifice for long-term rewards 0.54 2.99*

M4: working hard till mastering it 0.80 5.20*

M5: insisting on a job 0.80 5.07*

M6: generating satisfaction while finishing a job perfectly 0.64 3.70*

Knowledge integration capability Reflective scale 0.91 0.94 0.80

KI1: synthesizing and integrating individual expertise into team project 0.93 43.22*

KI2: spanning several areas of expertise 0.88 25.01*

KI3: understand how different pieces of this project fit together 0.82 14.34*

KI4: blending new project knowledge with related existing knowledge 0.93 59.97*

Project complexity Formative scale 0.86 – –

PC1: the complexity of system requirements 0.15 2.22*

PC2: the complexity of information technology 0.32 2.48*

PC3: the dynamic of customer organization 0.74 3.55*

Collective creative efficacy Reflective scale 0.97 0.97 0.80

CCE1: will achieve most team goals creatively that we set 0.91 38.46*

CCE2: will complete difficult tasks creatively when our team confronts them 0.92 41.80*

CCE3: will be able to adopt creative way to obtain results which are important to our team 0.92 35.64*

CCE4: Will succeed to achieve creative efforts when we make up our minds. 0.85 18.94*

CCE5: will overcome many challenges creatively 0.93 39.83*

CCE6: will complete many different tasks creatively 0.91 31.44*

CCE7: Will have more creativity to finish most tasks than other teams 0.86 17.69*

CCE8: will accomplish tasks creatively even when our team faces difficulty 0.86 20.49*

Note: the constructs of formative scale do not have composite reliability (CR) and AVE.* p<0.05.

Table 3Correlations among constructs.

Construct Means SD (1) (2) (3) (4) (5) (6)

Team knowledge (1) 3.84 0.12 –

Achievement motivation (2) 3.52 0.13 0.24 0.71

Knowledge integration capability (3) 3.75 0.17 0.40 0.35 0.89

Interpersonal interaction (4) 0.72 0.01 0.09 0.09 �0.04 –

Project complexity (5) 3.23 0.21 �0.08 0.08 �0.03 �0.13 –

Collective creative efficacy (6) 3.58 0.19 0.46 0.40 0.60 �0.04 �0.17 0.89

Note: the diagonal elements represent the square root of AVE.

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–1710

Furthermore, this study examined criteria validity of CCE. Gibson (2001) indicated that beliefcognition influences team behavior. Previous research (Bandura, 1997) also demonstrated thatcollective efficacy has the great influence on performance. This study thus gathered data for processperformance and process creativity performance from each ISD team. Because objective data werehard to obtain, subjective data from team leaders were used. All team leaders completed anotherperformance questionnaire. Referencing to items developed by Henderson and Lee (1992),questionnaire items for assessing performance were designed. The reliability of process performancein Henderson and Lee (1992) exceeded 0.7. In this study, the alpha coefficient of process performanceis 0.72, and the value for composite reliability is 0.80. Items for process creativity performance were

Table 4The results of PLS.

Path b Coefficient T-value Hypothesis

Team knowledge!CCE 0.43 3.22* H1 is supported

Achievement motivation!CCE 0.48 2.07* H2 is supported

Knowledge integration capability!CCE 0.61 2.65* H3 is supported

Interpersonal interaction� team knowledge!CCE 0.35 3.14* H4 is supported

Project complexity�team knowledge!CCE �0.42 3.18* H5 is supported

Project complexity�achievement motivation!CCE 0.80 2.83* H6 is supported

Project complexity�knowledge integration capability!CCE �0.38 2.08* H7 is supported

* p<0.05.

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–17 11

developed in reference to items by Zhou and George (2001). The reliability of process creativityperformance in the study by Zhou and George (2001) was 0.96. In this study, the alpha coefficient ofprocess creativity performance is 0.77, while the value for composite reliability is 0.83. The coefficientfor the correlation between CCE and process performance is 0.33 (p<0.05), and that the correlationbetween CCE and process creativity performance is 0.34 (p<0.05). Thus, criteria validity of CCE issatisfactory. That is, a team with high CCE would have high process performance and processcreativity performance on project.

The testing on the structural model

The proposed hypotheses were tested with PLS as shown in Table 4. As shown in Fig. 2, teamknowledge, achievement motivation, and knowledge integration capability directly effect on CCE;this, H1 (b=0.43, p<0.05), H2 (b=0.48, p<0.05), and H3 (b=0.61, p<0.05) are supported. Knowledgeintegration capability has the strongest influence on CCE.

Beyond the direct effect, analytical results indicate that interpersonal interaction moderates therelationship between team knowledge and CCE, and that project complexity moderates the effects onCCE of team knowledge, achievement motivation and knowledge integration capability. Thus, H4(b=0.35, p<0.05), H5 (b=�0.42, p<0.05), H6 (b=0.80, p<0.05) and H7 (b=�0.38, p<0.05) aresupported.

Tests comparing R2 values for main and interaction effects used Cohen’s f2 (Cohen, 1988). A Cohen’sf2 value less than 0.02 indicates a small effect size; a value of 0.15 indicates a medium effect size; a[(Fig._2)TD$FIG]

0.43

0.48

0.61

0.35

-0.42 0.80 -0.38

(R2=0.73)

Team

knowledge

Knowledge

integration

capability

Achievement

motivation

Collective

creative efficacy

Interpersonal

interaction

Project

complexity

Fig. 2. PLS results for the proposed research model.

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–1712

value exceeding 0.35 indicates a large effect size. Significant direct effects explain 56% of variance,while moderating effects explain 73% of variance. In this study, the Cohen’s f2 moderating size reachesthe large size effect, supporting the proposed moderating effects.

Discussions

CCE and antecedents

Analytical results indicate that team knowledge, achievement motivation and knowledgeintegration capability markedly influence CCE for ISD teams.

Team knowledge is positively related to CCE, meaning that team knowledge positively influencesthe judgment on the ISD team CCE. Team knowledge, a stable factor, is one key contributor to efficacy(Gist and Mitchell, 1992) and a key influence on creative performance (Dunbar, 1997). Gibson (2001)described the processes for developing group efficacy based on group dynamics. These processes wereaccumulation of information, interaction, and examination. Gibson and Earley (2007) indicatedgroups often engage in ongoing information gathering, which affects the development of beliefs,perspectives, orientations, and decisions in the process of accumulation. In the process ofaccumulation, group members typically gather and understand the knowledge that their teamshave and thereby develop collective efficacy. In this study, ISD requires system analysis and designknowledge, programming knowledge, project management knowledge, and domain knowledge. Theconfidence of ISD team members increases when they perceive their team members have relativelycomplete development knowledge. Thus, Hypothesis 1 is supported.

Achievement motivation strongly increases CCE. Creativity literature (Amabile, 1997) positionsintrinsic motivation as a driver of creativity development. Gibson and Earley (2007) asserted that teammember characteristics are important to the accumulation phase of collective cognition for CCEdevelopment. In this study, for the achievement process and to ensure product quality, ISD memberswith high achievement motivation likely pay attention to project tasks and work hard to completethese tasks. For example, programmers with high achievement motivation generally actively seekways to rewrite programs to increase system efficiency. When an ISD team contains many memberswith high achievement motivation, team members can observe that their teammates have a strongdesire to complete tasks perfectly. Consequentially, members would increase the strength of theirassessment of CCE about how to improve software processes. Thus, Hypothesis 2 is supported.

Analytical results demonstrate that knowledge integration capability is the most importantantecedent of CCE. Kurtzberg and Amabile (2001) showed that team creativity is a synthesis of teammembers thinking, sharing, and developing ideas collectively to produce an outcome. Otherresearchers (Gibson and Earley, 2007; Lester et al., 2002) emphasized the importance of collaborationprocess to the development of collective efficacy. Team members of ISD teams must cooperate on aproject while performing different tasks. The performance of an ISD teams then depends on how itsynthesizes and integrates ideas and products for each member into its projects. When teams caneffectively combine member expertise at the team level, team members perceive teams as adept atfinding different solutions to increase CCE. Thus, Hypothesis 3 is supported.

Moderating effects of interpersonal interaction

The main effect of team knowledge on CCE is 0.43; the moderating effect of interpersonalinteraction on the relationship between team knowledge and CCE is 0.35. Restated, given increasinginterpersonal interaction, the influence of team knowledge on CCE becomes 0.78 (0.43 plus 0.35). Thisstudy demonstrates that members of teams with high interpersonal interactions communicatedfrequently, helping team members understand the previous work experience of other members inrelation to solution development. Thus, team members who have high-quality of interpersonalinteraction have a high degree of confidence in their team’s creativity.

The interpersonal interaction value range was from 0 to 1, and mean value was 0.72. To explore themoderating effect of interpersonal interaction on CCE, with the mean score for interaction as thethreshold, this study divided the teams into two groups: high interaction (defined as their interaction

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–17 13

higher than 0.72) and low interaction (defined as their interaction lower than 0.72). There are 25 and36 teams, respectively. The impact of team knowledge on CCE was tested for the 36 teams (207members) with low interaction. In these teams, team members have difficulty identifying theknowledge of other members. Analytical results indicate that team knowledge does not significantlyinfluence CCE for teams with low levels of interaction. That is, when team members have distantworking relationships, low communication frequency and poor reciprocity, they may not be aware ofthe knowledge of other members, or may make wrong judgments regarding that knowledge. Thus,interpersonal interaction increases the influence of team knowledge on CCE. Thus, Hypothesis 4 issupported. Such findings are consistent with arguments by Curseu et al. (2007), indicating that groupswith a high average individual cognitive complexity have the highest cognitive complexity as a grouponly when the quality of their interactions is high.

Moderating effects of project complexity

The main effect of team knowledge on CCE is 0.43; moreover, the moderating effect of projectcomplexity on the relationship between team knowledge and CCE is �0.42. In other words, as projectcomplexity increases, the impact of team knowledge on CCE decreases to only 0.01 (0.43 plus �0.42).

The project complexity had ranges at 1–5, with a mean of 3.23. To explore the moderating effect ofproject complexity on CCE, with the mean of complexity as the threshold, teams were divided into twogroups: high project complexity (defined as complexity higher than 3.23) and low project complexity(defined as their complexity lower than 3.23). There are 30 and 31 teams, respectively. The impact ofteam knowledge on CCE was tested for the 30 teams (145 members) with high project complexity.Analytical results indicate that for the influence of team knowledge on CCE for teams with high projectcomplexity reduced to a very small influence. In such cases, development knowledge does notinfluence CCE; and domain knowledge has only a small impact (0.14) on CCE. A team typically hasconfidence in its ability due to a belief that it can produce an expected outcome following team inputand process (Gibson, 2001). However, Gist and Mitchell (1992) suggested that complexity iscontextual, potentially influencing efficacy judgments. According to Cervone and Peake (1986), whenindividuals perceived tasks as complex, they decrease their confidence regarding their efficacy.Additionally, analytical results in this study demonstrate that the three complexity dimensions(complexity of system requirements, complexity of information technology and the dynamic ofcustomer organization) had high scores. Particularly, the dynamic of customer organization, whichincludes the rate and pattern of changes in user information needs, business processes, andorganizational structures, was scored highest. The dynamic of customer organization implies that thestructure of customer organization changes frequently, related business processes are frequentlymodified, and users must adjust work procedures accordingly. In this situation, ISD team members areconfronted with an uncertain, unexpected, and un-analyzable development environment. Con-sequentially, team members generally perceive their teams as having incomplete knowledge fordeveloping systems and cannot expect outcomes to be creative. Thus, Hypothesis 5 is supported.Project complexity weakens (i.e., has negative impact on) the positive relationship between teamknowledge and CCE.

The main effect of achievement motivation on CCE is 0.48, and the moderating effect of projectcomplexity on the relationship between achievement motivation and CCE is 0.80. Restated, whenproject complexity increases, the influence of achievement motivation on CCE increases to 1.28 (0.48plus 0.80). Analytical results show that project complexity significantly moderate the relationshipbetween achievement motivation and CCE. Team members with high achievement motivation likechallenges (Chen, 2008; Nicholls, 1984). Those individuals generally try to behave beyond themselvesand increase their confidence of developing creativity in the context of complex projects. Creativityliterature (Oldham and Cummings, 1996; Shalley and Gilson, 2004) revealed that people have manyopportunities to develop their creativity while working on challenging tasks. However, projects withlow level complexity do not provide opportunities for individuals with high achievement motivationto compete with others to satisfy their achievement needs. Furthermore, this study tested the impactof the direct effect of achievement motivation on CCE using the 31 teams (171 members) who workedfor low complex projects. Analytical results show that achievement motivation does not significantly

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–1714

influence CCE in teams working for low complex projects. Restated, when scores for the complexity ofsystem requirements, complexity of information technology, and dynamic of customer organizationare low, individuals with high achievement motivation typically perceive that few opportunities existto meet their achievement desires. In projects with low complexity, achievement motivation is notimportant in developing CCE judgments. Thus, Hypothesis 6 is supported.

The main effect of knowledge integration capability on CCE is 0.61; moreover, the moderatingeffect of project complexity on the relationship between knowledge integration capability and CCE is�0.38. That is, as project complexity increases, the impact of knowledge integration capability on CCEdecreases to 0.23 (0.61 plus�0.38). Furthermore, the impact of the integration capability effect on CCEwas also separately tested for the 30 teams (145 members) who worked for highly complex projects.Analytical results indicate that for teams working for highly complex projects, the influence ofknowledge integration capability on CCE would be 0.21.

When executing complex projects, teams need rich and diverse knowledge, including knowledge ofprogramming languages, knowledge of different platforms, and expertise in diverse domains.However, for complicated projects, the correlations among these diverse and heterogeneous types ofknowledge are difficult to identify. In such a situation, ISD team members perceive that they havedifficulty coordinating knowledge, and knowledge integration efficacy, scope, and flexibility reduce.Thus, when ISD projects are highly complex, knowledge integration has only a small influence on CCE.Thus, Hypothesis 7 is supported. Project complexity reduces (i.e., has negative impact on) the strengthof the positive relationship between knowledge integration capability and CCE.

Conclusions

Implications

This study explored the antecedents of CCE in the ISD context and investigated the moderatingeffects of team and environmental variables on the relationships among antecedents and CCE. Severalmanagerial implications exist.

1. T

eam composition. Watson. et al. (1993) indicated that diversity groups affect outcomes. Previousstudies identified the two important types of diversity: (1) diversity on observable or readilydetectable attributes including race, ethnic background, age, or gender; and (2) diversity withrespect to less visible or underlying attributes including education, technical abilities, functionalbackground, tenure in an organization, or socioeconomic background, personality characteristics orvalues (Tsui et al., 1992; Jackson et al., 1995; Milliken and Martins, 1996). Thus, project leadersshould staff ISD project teams with employees with complementary knowledge and skills that arerequired for project tasks. Additionally, team member achievement motivation should beconsidered when organizing an ISD project team. Teams should have several members with highachievement motivation because they benefit to ISD project teams in two important ways. First,these members pursue success to increase their opportunities for creativity. Second, such memberscan transfer information and thereby enhance their team members’ confidence in achieving teamsuccess.

2. S

haping team member interactions and the communication environment. Wegner (1987, 1995)proposed the theory of transactive memory, suggesting that people in continuing interpersonalrelationships often develop a specialized division of labor with respect to encoding, storage, andretrieval of information from different substantive domains. Group members can identify groupexperts and how to access this expertise through communicative processes. Based on transactivememory theory, project leaders should plan formal or informal activities that increaseopportunities for member interaction, communication and collaboration. Thus, through interac-tion, team members can identify their own knowledge, and knowledge they lack.

3. A

nalytical results indicate that project complexity weakens the positive influence of teamknowledge on CCE, and that of knowledge integration capability on CCE. The main reason is thatproject complexity prevents team members from characterizing the relationship between teaminput (team knowledge and knowledge integration capability) and team outcome. In this context,

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–17 15

vicarious experience, verbal persuasion and environmental resources can strengthen therelationship between team input and output (Bandura, 1997; Cohen, 1988). Furthermore, projectleaders should share information regarding the successful development of creativity in relation toother complex projects, encourage members to improve software processes, and provide additionalresources for implementing projects to increase confidence in creativity development.

4. A

nalytical results demonstrate that project complexity strengthens the positive influence ofachievement motivation on CCE. When making personnel assignments, task adjustments, orrecruiting new staff, project leaders should attempt to achieve a fit between the achievementmotivation of members and the complexity of assigned projects.

Limitations and future research directions

This study has several limitations. Data were gathered during a specific period. Longitudinalstudies are needed. Furthermore, hypotheses testing was performed with data from ISD project teammembers. However, the characteristics of ISD project-based teams differ from those of ISD product-based (software package) teams. Thus, over-generalization must be avoided.

Member interaction is an important factor when exploring collective efficacy. The ISD projectmembers communicate based on their functional role and clearly defined procedures for developinginformation systems. Both vertical and horizontal communication can assist in interaction among ISDteam members. The forms and closeness of interactions influence the exchange and explanation ofinformation among members (Gibson, 2001; Gibson and Earley, 2007) and influence creativitydevelopment (Chen et al., 2008). Future works can consider the relationship between interactionpatterns and CCE.

Other antecedents of CCE may exist. This study explores only the influence of achievementmotivation on CCE. However, other individual motivations, such as work motivation and learningmotivation, may contribute to collective creativity. Additionally, personal achievement and projectgoal achievement may be prioritized differently. Future studies can consider the influences of differentmotivations on CCE. Regarding the capability component, except for knowledge integration capability,many other team capabilities may influence creativity. For example, team learning capability and theability to handle conflict may be antecedents of CCE. Future studies can explore the influences ofdifferent team capabilities on CCE.

Social cognitive theory asserts that environmental resources influence efficacy judgments(Bandura, 1997). When organizations support innovation or team members observe creative behaviorin other teams, teams have increased CCE. Thus, future work can consider the influence ofenvironment on CCE.

Hackman (1987) used three criteria to assess team effectiveness. The first deals with actual groupoutput, such as productive output, meaning output that is acceptable to those who receive it or reviewit. The second focuses on a group’s state, such as viability, which is the capability of members to worktogether on subsequent team tasks. The third is the impact of group experience on individualmembers, such as satisfaction, meaning that group members’ needs are satisfied by groupexperiences. While verifying the criteria validity, this study only considered actual output (processperformance and process creativity performance); that is, the relationship between CCE and viability/satisfaction is not addressed. Thus, future work may verify the relationship between viability and CCEor between CCE and satisfaction.

References

Akgun, A.E., Keskin, H., Byrne, J., Imamoglu, S.Z., 2007. Antecedents and consequences of team potency in software developmentprojects. Information & Management 44 (7) 646–656.

Allison, I., Merali, Y., 2007. Software process improvement as emergent change: a structurational analysis. Information andSoftware Technology 49, 668–681.

Amabile, T., 1983. The social psychology of creativity: a componential conceptualization. Journal of Personality and SocialPsychology 45, 357–377.

Amabile, T.M., 1997. Motivating creativity in organizations: on doing what you love and loving what you do. CaliforniaManagement Review 40 (1) 39–58.

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–1716

Amason, A.C., 1996. Distinguishing the effects of functional and dysfunctional conflict on strategic decision making: resolving aparadox for top management teams. Academy of Management Journal 39, 123–148.

Baccarini, D., 1996. The concept of project complexity – a review. International Journal of Project Management 14 (4) 201–204.Bagozzi, R.P., Yi, Y., 1988. On the evaluation of structural equation models. Academy of Marketing Science 16 (1) 74–94.Bandura, A., 1997. Self-Efficacy: The Exercise Control. Freeman, New York.Barki, H., Rivard, S., Talbot, J., 1993. Toward an assessment of software development risk. Journal of Management Information

Systems 10 (2) 202–223.Barki, H., Rivard, S., Talbot, J., 2001. An integrative contingency model of software project risk management. Journal of

Management Information Systems 17 (4) 37–69.Bhandar, M., Pan, S.L., Tan, B.C.Y., 2007. Towards understanding the roles of social capital in knowledge integration: a case study

of a collaborative information systems project. Journal of the American Society for Information Science and Technology 58(2) 263–274.

Cervone, D., Peake, P.K., 1986. Anchoring, efficacy, and action: the influence of judgmental heuristic on self-efficacy judgementsand behavior. Journal of Personality and Social Psychology 50, 492–501.

Chen, G., Webber, S.S., Bliese, P.D., Mathieu, J.E., Payne, S.C., Born, D.H., Zaccaro, S.J., 2002. Simultaneous examination of theantecedents and consequences of efficacy beliefs at multiple levels of analysis. Human Performance 15 (4) 381–409.

Chen, L.H., 2008. Job satisfaction among information system (IS) personal. Computers in Human Behavior 24, 105–118.Chen, M.H., Chang, Y.C., Hung, S.C., 2008. Social capital and creativity in R&D project teams. R&D Management 38 (1) 21–34.Cheng, H.H., Yang, H.L., 2011. Student team projects in information systems development: measuring collective creative

efficacy. Australasian Journal of Educational Technology 27 (6) 881–895.Chin, W.W., 1998a. Issues and opinion on structural equation modeling. MIS Quarterly 22 (1) vii–xvi.Chin, W.W., 1998b. The partial least squares approach to structural equation modeling. In: Marcoulides, G.A. (Ed.), Modern

Methods for Business Research. Lawrence Eribaum Associates, Mahwah, NJ, pp. 295–336.Chin, W.W., Marcolin, B.L., Newsted, P.R., 2003. A partial least squares latent variable modeling approach for measuring

interaction effects: results from a Monte Carlo simulation study and an electronic-mail emotion/adoption study. Informa-tion Systems Research 14 (2) 189–217.

Cohen, J., 1988. Statistical Power Analysis for the Behavioral Sciences, 2nd ed. Lawrence Erlbaum, Hillsdale, NJ.Choi, K., Coen, C., 2009. The emergence of group cognitive complexity and its effects on group performance. In: Academy of

Management Annual Meeting Proceedings. pp. 1–5.Curseu, P.L., Schruijer, S., Boros, S., 2007. The effects of groups’ variety and disparity on groups’ cognitive complexity. Group

Dynamics: Theory, Research, and Practice 11 (3) 187–206.Dunbar, K., 1997. How scientists think: on-line creativity and conceptual change in science. In: Ward, T.B., Smith, S.M., Vaid, J.

(Eds.), Creative Thought: An Investigation of Conceptual Structures and Processes. American Psychological Association,Washington, DC, pp. 461–493.

Fornell, C., Larcker, D., 1981. Evaluating structural equation models with unobservable variables and measurement error.Journal of Marketing Research 18, 39–50.

Galinac, T., 2009. Empirical evaluation of selected best practices in implementation of software process improvement.Information and Software Technology 51, 1151–1364.

George, J., 1990. Personality, affect, and behavior in groups. Journal of Applied Psychology 75, 107–116.Gibson, C.B., 1999. Do they do what they believe they can? Group efficacy and group effectiveness across tasks and cultures.

Academy of Management Journal 42, 138–152.Gibson, C.B., 2001. Me and us: differential relationships among goal setting, training, efficacy, and effectiveness at the individual

and team level. Journal of Organizational Behavior 22, 789–808.Gibson, C.B., 2003. The efficacy advantage: factors related to the formation of group efficacy. Journal of Applied Social

Psychology 33 (10) 2153–2186.Gibson, C.B., Earley, P.C., 2007. Collective cognition in action: accumulation, interaction, examination, and accommodation in

the development and operation of group efficacy beliefs in the workplace. Academy of Management Review 32 (2) 438–458.

Gist, M.E., Mitchell, T.R., 1992. Self-efficacy: a theoretical analysis of its determinants and malleability. Academy of Manage-ment Review 17 (2) 183–211.

Goodman, P.S., Ravlin, E.C., Schminke, M., 1990. Understanding groups in organizations. In: Cummings, L.L., Staw, B.M. (Eds.),Leadership, Participation, and Group Behavior. JAI Press, Greenwich, pp. 323–385.

Grant, R.M., 1996. Prospering in dynamically-competitive environments: organizational capability as knowledge integration.Organization Science 7 (4) 375–387.

Guilford, J.P., 1984. Varieties of divergent production. Journal of Creative Behavior 18, 1–10.Hackman, J.R., 1987. The design of work teams. In: Lorsch, J.W. (Ed.), Handbook of Organizational Behavior. Prentice Hall,

Englewood Cliffs, NJ, pp. 315–342.Hair, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 1998. Multivariate Data Analysis, 5th ed. Prentice-Hall, Inc., New Jersey, pp.

1998.Hargadon, A.B., 1999. Group cognition and creativity in organizations. Research on Managing Groups and Teams 2, 137–155.Helmreich, R.L., Spence, J.T., 1978a. Achievement motivation and scientific attainment. Personality and Social Psychology

Bulletin 4, 222–226.Helmreich, R.L., Spence, J.T., 1978b. The work and family orientation questionnaire: an objective instrument to assess

components of achievement motivation and attitudes toward family and career. JSAS Catalog of Selected Documents inPsychology 8 (35) 21–35.

Henderson, J.C., Lee, S., 1992. Managing I/S design teams: a control theories perspective. Management Science 38 (6) 757–776.Houghton, S.M., Stewart, A.C., Barr, P.S., 2009. Cognitive complexity of the top management team: the impact of team

differentiation and integration processes on firm performance. In: Rahim, M.A. (Ed.), Current Topics in Management,vol. 14. Transaction Publishers, New Brunswick, NJ.

H.-H. Cheng, H.-L. Yang / Journal of Engineering and Technology Management 33 (2014) 1–17 17

Jackson, S.E., May, K.E., Whitney, K., 1995. Understanding the dynamics of diversity in decision making teams. In: Guzzo, R.A.,Salas, E. (Eds.), Team Effectiveness and Decision Making in Organizations. Jossey-Bass, San Francisco, pp. 204–261.

James, L.R., Demaree, R.G., Wolf, G., 1993. rwg: an assessment of within-group interrater agreement. Journal of AppliedPsychology 78, 306–309.

Jarvis, C.B., Mackenzie, S.B., Podsakoff, P.M., 2003. A critical review of construct indicators and measurement model mis-specification in marketing and consumer research. Journal of Consumer Research 30 (2) 199–218.

Jiang, J.J., Klein, G., Hwang, G., Huang, J., Hung, S., 2004. An exploration of the relationship between software developmentprocess maturity and project performance. Information & Management 41 (3) 279–288.

Kleinsmann, M., Buijs, J., Valkenburg, R., 2010. Understanding the complexity of knowledge integration in collaborative newproduct development teams: a case study. Journal of Engineering and Technology Management 27 (1–2) 20–32.

Kurtzberg, T.R., Amabile, T.M., 2001. From Guilford to creative synergy: opening the black box of team level creativity. CreativityResearch Journal 13, 285–294.

Leenders, R.T., van Engelen, J., Kratzer, J., 2003. Virtuality, communication, and new product team creativity: a social networkperspective. Journal of Engineering and Technology Management 20, 69–92.

Lester, S.W., Meglino, B.M., Korsgaard, M.A., 2002. The antecedents and consequences of group potency: a longitudinalinvestigation of newly formed work groups. Academy of Management Journal 45, 352–368.

Milliken, F.J., Martins, L.L., 1996. Searching for common threads: understanding the multiple effects of diversity in organiza-tional groups. Academy of Management Review 21 (2) 402–433.

Nicholls, J.G., 1984. Achievement motivation: conceptions of ability, subjective experience, task choice, and performance.Psychological Review 91 (3) 328–346.

Nunnally, J.C., 1978. Psychometric Theory. McGraw-Hill, New York.Okhuysen, G., Eisenhardt, K., 2002. Integrating knowledge in groups: how formal interventions enable flexibility. Organization

Science 13 (4) 370–386.Oldham, G.R., Cummings, A., 1996. Employee creativity: personal and contextual factors at work. Academy of Management

Journal 39, 607–634.Parolia, N., Goodman, S., Li, Y., Jiang, J., 2007. Mediators between coordination and IS project performance. Information &

Management 44 (7) 635–645.Paulus, P.B., 2000. Groups, teams and creativity: the creative potential of idea generating groups. Applied Psychology: An

International Review 49, 237–262.Shalley, C.E., Gilson, L.L., 2004. What leaders need to know: a review of social and contextual factors that can foster or hinder

creativity. Leadership Quarterly 15, 33–53.Sun, Y., Liu, X.F., 2010. Business-oriented software process improvement based on CMMI using QFD. Information and Software

Technology 52, 79–91.Tetlock, P.E., Skitka, L., Boettger, R., 1989. Social and cognitive strategies for coping with accountability: conformity, complexity,

and bolstering. Journal of Personality and Social Psychology 57 (4) 632–640.Tierney, P.T., Farmer, S.M., 2002. Creative self-efficacy: its potential antecedents and relationship to creative performance.

Academy of Management Journal 45 (6) 1137–1148.Tiwana, A., Bharadwaj, A., Sambamurthy, V., 2003. The antecedents of information systems development capability in firms: a

knowledge integration perspective. In: Twenty-Fourth International Conference on Information Systems.Tiwana, A., McLean, E., 2005. Expertise integration and creativity information systems development. Journal of Management

Information Systems 22 (1) 13–43.Tsui, A.S., Egan, T.D., O’Reilly, C.A., 1992. Being different: relational demography and organizational attachment. Administrative

Science Quarterly 37, 549–579.Un, C.A., 2010. An empirical multi-level analysis for achieving balance between incremental and radical innovations. Journal of

Engineering and Technology Management 27 (1–2) 1–19.Wasserman, S., Faust, K., 1994. Social Network Analysis: Methods and Applications. Cambridge University Press, Cambridge.Watson, W.E., Kumar, K., Michaelsen, L.K., 1993. Cultural diversity’s impact on interaction process and performance: comparing

homogeneous and diverse task groups. Academy of Management Journal 36, 590–602.Wegner, D.M., 1987. Transactive memory: a contemporary analysis of the group mind. In: Mullen, B., Goethals, G.R. (Eds.),

Theories of Group Behavior. Springer-Verlag, New York, pp. 185–208.Wegner, D.M., 1995. A computer network model of human transactive memory. Social Cognition 13 (3) 319–339.Williams, T.M., 1999. The need for new paradigms for complex projects. International Journal of Project Management 17 (5)

269–273.Wold, H., 1982. Systems under indirect observation using PLS. In: Fornell, C. (Ed.), A Second Generation of Multivariate Analysis.

Praeger, New York, pp. 325–347.Xia, W., Lee, G., 2005. Complexity of information systems development projects: conceptualization and measurement

development. Journal of Management Information Systems 22 (1) 45–83.Yang, H.L., Cheng, H.H., 2009. Creative self-efficacy and its factors: an empirical study of information system analysts and

programmers. Computers in Humber Behavior 25, 429–438.Zander, A., Forward, J., 1968. Position in group, achievement motivation, and group aspirations. Journal of Personality and Social

Psychology 8, 282–288.Zhou, J., George, J.M., 2001. When job dissatisfaction leads to creativity: encouraging the expression of voice. Academy of

Management Journal 44 (4) 682–696.