Enhancing creativity in group collaboration: How performance targets and feedback shape perceptions and idea generation performance

Computers in Human Behavior xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Computers in Human Behavior

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

Enhancing creativity in group collaboration: How performance targetsand feedback shape perceptions and idea generation performance

http://dx.doi.org/10.1016/j.chb.2014.02.0170747-5632/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +86 41184710336.E-mail addresses: [email protected] (X. Wang), christoph.schneider

@cityu.edu.hk (C. Schneider), [email protected] (J.S. Valacich).1 All authors contributed equally.

2 We admit that the creation of novel and creative solutions is only oneuses of computer-supported collaboration, and may not be the focusapplications, such as the use of computer-supported collaboration for educ

Please cite this article in press as: Wang, X., et al. Enhancing creativity in group collaboration: How performance targets and feedback shape percand idea generation performance. Computers in Human Behavior (2014), http://dx.doi.org/10.1016/j.chb.2014.02.017

Xuequn Wang a,⇑,1, Christoph Schneider b,1, Joseph S. Valacich c,1

a Global Institute of Management and Economics, Dongbei University of Finance & Economics, 217 Jian Shan Street, Sha He Kou District, Dalian 116025, Chinab Department of Information Systems, College of Business, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon Tong, Hong Kongc Department of Management Information Systems, Eller College of Management, University of Arizona, 430 McClelland Hall, Tucson, AZ 85721-0108, USA

a r t i c l e i n f o

Article history:Available online xxxx

Keywords:CreativityHuman–computer interactionSelf-determination theoryComputer-mediated idea generationPerformance targetsPerformance feedback

a b s t r a c t

As Information and Communication Technologies (ICTs) have become established tools for communica-tion, organizations increasingly use computer-mediated work groups to support various business pro-cesses and find creative solutions to organizational problems. In such a context, groups’ creativeperformance can greatly contribute to organizational success. Previous literature has examined the influ-ence of various factors on different outcomes of group collaboration. However, mechanisms throughwhich creativity can be improved, and how to design ICT’s interfaces to increase creativity have receivedlittle attention. In this study, we aim to understand the effects of two specific motivational affordances,namely, performance targets and performance feedback, on people’s perceived competence and creativitywithin the context of computer-mediated collaboration. Using computer-mediated idea generation as aninstantiation of collaboration systems, we test the effects of performance targets and different types offeedback on people’s perceived competence and creativity in a controlled laboratory experiment. Ourresults show that the difficulty of performance targets and the type of performance feedback interact,influencing people’s perceived competence, which in turn influences their creativity in group collabora-tion. We conclude our study with a discussion of implications for the design of human–computer inter-faces for computer-mediated idea generation.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Increasing globalization has created tremendous opportunities(such as the ability to reach new markets) and challenges (suchas increased competition) for organizations. In an attempt to gainand sustain competitive advantage, organizations need to continueinnovating by devising novel and creative products or services(Muller & Välikangas, 2002). In many contexts, generating creativeideas can be essential for survival and success of organizations. Inother words, ‘‘innovations that enable even modest increases in thequality of ideas available for consideration could be of immensepractical value’’ (Heslin, 2009, p. 129). To achieve such goal, orga-nizations are trying to bring together geographically distant indi-viduals from both inside and outside the organization to devisenovel and creative products or services; however, many organiza-tions have realized that geographical distance between group

members can create challenges for generating creative ideas. Toharness the creativity of diverse and often geographically sepa-rated individuals, a broad range of Information and Communica-tion Technologies (ICTs) are ever more utilized, withcollaboration becoming increasingly virtual, and novel and creativesolutions often being a crucial outcome of computer-supportedcollaboration (Goncalo & Staw, 2006).2

As prior group collaboration research has shown that the per-formance of each individual group member is an important con-tributor to success (e.g., Chidambaram & Tung, 2005), the effectsof various factors on group performance have been extensivelystudied. An area that has received little attention, however, is themechanisms through which individuals’ creativity can be en-hanced, and how to modify the design of ICTs to increase individ-uals’ creativity.

Therefore, our overarching question guiding this study is: Howcan the human–computer interfaces of collaboration environments

of manyof other

ation.

eptions

http://dx.doi.org/10.1016/j.chb.2014.02.017

mailto:[email protected]





http://www.sciencedirect.com/science/journal/07475632

http://www.elsevier.com/locate/comphumbeh


2 X. Wang et al. / Computers in Human Behavior xxx (2014) xxx–xxx

be designed to enhance individuals’ creativity in group collabora-tion? In this study, integrating research from human–computerinteraction, motivation, technology supported group work, andpsychology, we theoretically derive mechanisms for increasingindividuals’ motivation by increasing their perceived competence,which in turn serves to increase their creativity in a group collab-oration environment. We manipulate the interface of a computer-mediated idea generation system (a widely used collaborationsetting) to enhance the system’s motivational affordance—i.e., thesystem’s properties that fulfill users’ motivational needs (Zhang,2008a)—by providing different performance targets (moderate vs.challenging) in combination with different types of performancefeedback (informational vs. controlling) during an idea generationsession. We test the effectiveness of these mechanisms using a lab-oratory experiment. Using a simulated collaboration environment,we ask participants to generate creative ideas about how to helpsolve global warming. The analysis of the ideas’ creativity suggeststhat indeed, varying the difficulty of performance targets and typeof feedback has an interaction effect on people’s perceived compe-tence, which in turn influences their creativity.

In the next section, we provide an overview of the theoreticalbackground, followed by our hypotheses. We then describe our re-search methodology, and present our findings. Finally, we discussthe implications of the findings for the design of the human–com-puter interfaces of computer-mediated idea generation and collab-oration environments.

3 According to Zhang (2008a), a motivational approach can address questions suchas ‘‘Why do people use ICTs in different intensity?’’ ‘‘Why do people continue or stopusing ICTs?’’ Research in the area of adaptive learning and intelligent tutoring systems(e.g., Brusilovsky, 1996; Paramythis & Loidl-Reisinger, 2004) similarly attempts toassess how adaptive content can enhance learners’ motivation and learning perfor-mance; however, this study is primarily concerned with enhancing people’s creativityin the context of group collaboration, rather than on learning performance.

4 Note that by focusing on need for competence, we do not imply that other needsare not relevant for computer-supported idea generation. Indeed, we believe thatthese needs can also quite relevant and encourage future studies to further examinethese needs. Again, as Amabile (1983) suggests, examining these variables can ‘‘onlybe completed gradually’’ (p. 362). Further, we acknowledge that the desired outcomesof the users and the systems’ designers may differ; yet, in organizational settings, thedesired outcomes are likely to be aligned.

2. Theoretical background: motivational affordance and self-determination theory

Previous studies in Information Systems (IS) literature have as-sessed how ICTs can influence group collaboration and supportindividuals in generating high-quality and creative ideas in differ-ent contexts (e.g., e.g., Connolly, Jessup, & Valacich, 1990; Diehl &Stroebe, 1987; Jung, Schneider, & Valacich, 2010; Michinov & Pri-mois, 2005; Valacich, Jung, & Looney, 2006). These studies haveexamined various group-level (e.g., group size, anonymity) andenvironmental-level (e.g., the use of multiple dialogues) variableswhich may influence group performance in various tasks (see Junget al. (2010) for a brief review). However, relatively little is knownhow human–computer interfaces of ICTs can be designed to en-hance creativity (Jung et al., 2010).

According to Amabile’s (1983) componential theory of individ-ual creativity, three main components influence individual creativ-ity: domain-relevant skills, creativity-relevant skills, and taskmotivation. Woodman, Sawyer, and Griffin (1993) further proposeadditional group and organizational characteristics influencing cre-ativity in the context of organizations. Notwithstanding these find-ings, a comprehensive understanding of creativity can only bedeveloped ‘‘gradually, as progress is made in creativity research’’(Amabile, 1983, p. 362). As a step toward developing a more com-plete understanding, we focus on task motivation to examine howelements of the human–computer interface of collaboration sys-tems influence people’s motivation, which in turn influences theircreativity in group collaboration.

Owing to the positive effects of motivation on people’s levels ofenergy and persistence (Ryan & Deci, 2000), motivation is animportant topic in various disciplines (Amabile, 1983; Steers,Mowday, & Shapiro, 2004; Tubbs & Ekeberg, 1991; Venkatesh &Speier, 1999; Wang & Clay, 2012; Wasko & Faraj, 2005). For exam-ple, ‘‘managers see motivation as an integral part of the perfor-mance equation at all levels, while organizational researchers seeit as a fundamental building block in the development of usefultheories of effective management practice’’ (Steers et al., 2004, p.379). Similarly, researchers in IS are increasingly trying to

Please cite this article in press as: Wang, X., et al. Enhancing creativity in groupand idea generation performance. Computers in Human Behavior (2014), http:/

understand the antecedents and outcomes of motivation in variouscontexts, such as technology adoption and acceptance (Venkatesh& Speier, 1999) and content contribution (Wang & Clay, 2012;Wasko & Faraj, 2005), and recent studies have begun to adopt amotivational approach to understanding the design and use ofICTs.3 Specifically, Zhang (2008a) argued that in using ICTs peopleseek to fulfill psychological, cognitive, social, and emotional needs,and ICT should offer ‘‘motivational affordance’’ (p. 145) to supportand satisfy these needs. Following this premise, Jung et al. (2010)demonstrated that increasing a system’s motivational affordancecan effectively increase performance. Still, further studies are neededto understand how different designs can motivate people by satisfy-ing people’s needs (Jung et al., 2010; Zhang, 2008b) and improvetheir performance and, in particular, their creativity.

To understand how to satisfy people’s needs and enhance peo-ple’s creativity, we draw on Self-Determination Theory (SDT) (Deci& Ryan, 1985; Ryan & Deci, 2000). SDT emphasizes that humansare active and growth oriented organisms striving toward person-ality development and behavioral self-regulation (Ryan, Kuhl, &Deci, 1997), and explains individuals’ ‘‘inherent growth tendenciesand innate psychological needs that are the basis for their self-motivation as well as for the conditions that foster those positiveprocesses’’ (Ryan & Deci, 2000, p. 68). SDT proposes threepsychological needs to foster individuals’ propensities for growthand self-motivation: the need for competence (Harter, 1978;White, 1963), autonomy (DeCharms, 1968; Deci, 1975), and relat-edness (Baumeister & Leary, 1995; Reis, 1994). Specifically, theneed for autonomy is defined as the need to experience choiceand feel like the initiator of one’s own actions (DeCharms, 1968);the need for relatedness deals with people’s desire to establishand maintain a sense of mutual respect and care for each other(Baumeister & Leary, 1995; Harlow, 1958); lastly, the need forcompetence refers to the desire to succeed at optimally challengingtasks and being able to attain desired outcomes (Skinner, 1995;White, 1959). As the overarching purpose of our study is to under-stand how to enhance individuals’ creativity in group collabora-tion, we examine how to help people succeed in groupcollaboration (a challenging task) and generate creative ideas(i.e., the desired outcomes). Therefore, satisfying the desire to at-tain desired outcomes in challenging tasks appears to be most rel-evant in the context of the current study, so as to help improveindividuals’ creativity in a computer-supported idea generationsetting.4 Next, we develop our hypotheses to examine how differentdesigns can motivate people by satisfying their need for competence(i.e., increasing their perceived competence), which in turn helps en-hance creativity.

3. Hypotheses development

Prior studies have shown that the design of human–computerinterfaces can have strong effects on people’s motivation and per-formance; for example, in the context of group idea generation,Jung et al. (2010) demonstrated that providing participants

collaboration: How performance targets and feedback shape perceptions/dx.doi.org/10.1016/j.chb.2014.02.017


X. Wang et al. / Computers in Human Behavior xxx (2014) xxx–xxx 3

with performance targets and continuous, objective performancefeedback enhanced participants’ motivation, ultimately leading tohigher performance. Here, we build on Jung et al.’s study by theo-rizing about and testing different mechanisms designed to enhancepeople’s perceived competence, which in turn leads to increasedcreativity. Using computer-mediated idea generation as an instan-tiation of collaboration systems, we propose hypotheses related todiffering performance targets and types of feedback, which arelikely to influence people’s perceived competence, as well as re-lated to the relationship between perceived competence andcreativity.

3.1. Performance targets

Performance targets, representing challenges, are related to aperson’s need for competence (Jung et al., 2010). According toSDT, the need for competence is one of the fundamental needs ofpeople, representing people’s desire to expend effort to achieveand master optimal challenges (Zhang, 2008a). When presentedwith an optimally challenging target for a task, people are morelikely to get involved in the task, and performance targets canmotivate people to put forth effort and help them focus on the out-come (Locke & Latham, 1990). Relatedly, Jung et al. (2010) demon-strated that specific and achievable performance targets results inbetter performance. With a specific performance target, people candevote the effort necessary to achieve that target, and are morelikely to feel that they can achieve the task, which enhances theirperceived competence. People are likely to feel the strongest inter-est and greatest involvement if the difficulty of a performance tar-get precisely matches their skill level (Zhang, 2008b), so thatperceived competence should be highest. On the other hand, whenthe performance target is too challenging, people may feel thatthey are less likely to achieve that target, which is likely to havea detrimental effect on their perceived competence. In the contextof computer-mediated idea generation, where the desired outcomeis to produce creative ideas and solutions, people may feel that thetask is too challenging when the performance target is too difficult,leading to decreased perceived competence. Therefore, we arguethat:

H1. The level of performance targets will influence people’sperceived competence, such that people provided with overlychallenging targets will have lower levels of perceived competencethan people provided with moderate performance targets.

3.2. Performance feedback

Performance feedback allows individuals to evaluate their per-formance and compare it to standards of excellence. When thefeedback is positive and individuals can see that they are makinggood progress toward attaining a performance target, they willcontinue expending efforts to try to achieve the target, and theirperceived competence is likely to increase (Zhang, 2008a).

Performance feedback can be either informational or control-ling (Ryan, 1982). Informational feedback indicates people’s pro-gress toward attaining a performance target in a relativelyneutral fashion, so that people can monitor their progress andevaluate their performance (Zhang, 2008b). Thus, people arelikely to feel engaged and emotionally satisfied, which in turnpositively contributes to their perceived competence (Zhang,2008a). During a computer-mediated idea generation session,when people try to generate creative solutions, they (consciouslyor unconsciously) monitor how well they are progressing (e.g.,whether their creativity performance meets expectations). Asinformational feedback allows people to objectively understand


how they are performing, people perceiving to be performing atadequate levels are likely to have high levels of perceivedcompetence.

In contrast, controlling feedback typically includes statementsjudging people’s performance by attempting to steer people to-ward attaining the performance target. Thus, individuals are morelikely to interpret such feedback as pressure to attain a particularoutcome (Ryan, 1982), and are also more likely to feel anxiousand emotionally dissatisfied (Zhang, 2008b), leading to loweredlevels of perceived competence. In the context of computer-medi-ated idea generation, when receiving controlling feedback, peoplemay feel continuous pressure (e.g., a need to continuously producemore creative ideas), and are therefore less likely to meet a stan-dard of excellence (e.g., do not have many creative ideas), leadingto lower levels of perceived competence.

Taken together, we argue that:

H2. People provided with informational performance feedbackwill have higher levels of perceived competence than peopleprovided with controlling feedback.

Further, the effect of the performance feedback on people’sperceived competence is likely to be moderated by the level ofthe performance target. When people are provided with a rela-tively challenging performance target, they need to continuouslymonitor their progress to ensure that they can achieve such tar-gets (Zhang, 2008a). Informational feedback can serve to providethe information needed to monitor one’s progress (Zhang, 2008a).On the other hand, controlling feedback is likely to provide addi-tional pressure; given that a person is less likely to make good pro-gress under more challenging performance targets, controllingfeedback is more likely to be viewed as criticism about one’s perfor-mance. In such conditions, people are more likely to perceive thetargets to be too challenging, lowering their perceived competence.

When provided with moderate performance targets, however,people may feel that they do not need to put forth much effort toachieve that target. In such a context, informational feedback,which simply provides progress information, may not furtherencourage people to achieve more challenging targets and is thusless likely to increase their perceived competence. In other words,people may reduce their efforts after having reached a moderatetarget, and may not perceive the attainment of such moderate tar-get to be a big achievement. On the other hand, controlling feed-back pushes people to continuously attain more challengingtargets, and can encourage people to devote more efforts and tryto achieve higher levels of performance, leading to increased per-ceived competence.

To summarize, we argue that:

H3. Performance feedback and performance targets will interactsuch that when provided with controlling feedback, (a) a chal-lenging performance target will have a detrimental effect onperceived competence and (b) a moderate performance target willhave a positive effect on perceived competence.

3.3. Perceived competence and creativity

According to Amabile (1983), a ‘‘product or response will bejudged as creative to the extent that . . . it is both a novel andappropriate, useful, correct, or valuable response to the task athand’’ (p. 360). In idea generation (the domain of our study), cre-ativity typically refers to novel and useful ideas. Previousliterature on creativity has found that people who are indepen-dent of judgment, autonomous and self-confident (Barron &Harrington, 1981) are more likely to propose novel and creative



Fig. 1. Collaboration environment.


ideas. When people’s perceived competence is high, they feel thatthey can accomplish optimally challenging tasks and attain de-sired outcomes (Skinner, 1995; White, 1959), and are more likelyto feel confident toward their own abilities, leading to highercreativity.

Further, according to SDT, high perceived competence willlead to higher levels of motivation. When people have higherlevels of motivation, they tend to have higher levels of energyand persist longer on the task (Ryan & Deci, 2000), leading tovarious positive outcomes, such as general well-being (Ryan,Deci, & Grolnick, 1995), learning (Vallerand & Bissonnette,1992), dedication at work (Vansteenkiste et al., 2007), andknowledge sharing (Foss, Minbaeva, Pedersen, & Reinholt,2009). Further, based on Amabile’s (1983) componential theoryof individual creativity, motivation enhances individuals’ creativ-ity in two ways: first, when the task is presented, motivationwill influence the degree to which people engage in the task;second, during the generation of responses, motivation can en-hance the degree to which people attend to different aspectsof the environment which may be relevant to the solution ofthe task. In other words, motivation determines whether individ-uals begin and continue to search for creative solutions, andwhether they will search their environment to generate morecreative solutions. In the context of computer-mediated ideageneration, when people’s perceived competence is high, theyare more highly motivated, and probably initiate and continu-ously put more effort to generate creative ideas during thecollaboration. Thus, we argue that:

H4. Perceived competence is positively associated with creativity.

5 Whereas Ryan used self-administered vs. other-administered feedback in regularintervals, in this study we programmed the chat system to display the different typesof feedback to the participants.

4. Method

To test our hypotheses, we conducted a controlled laboratoryexperiment. Specifically, we created a group collaboration environ-ment in which each participant interacted with four simulatedgroup members. Using simulated group members allowed us tominimize within-group variance and helped to us to control for po-tential influences of group members upon others (see also Valacichet al., 2006). In the next section, we describe the collaborationenvironment, followed by a description of the experimentalprocedures.


4.1. Collaboration environment

We based the collaboration environment on an open sourceWeb-based chat system (AjaxChat; https://blueimp.net/ajax/). Thissystem allows participants to generate ideas, which are stored in aMySQL database. We incorporated a simulator into the chat sys-tem, which presented four streams of ideas (one stream for eachsimulated group member) to each participant. These streams weregenerated by participants of an earlier, interactive idea generationsession; we ensured that the sequence and timing of messagesresembled that of interactive sessions, so as to enhance the credi-bility of the system. We further modified the chat system to keeptrack of each participant’s performance (in terms of the quantityof contributions), which was displayed in the form of a bar chartto allow participants to monitor their progress towards the goalin real-time (see also Jung et al., 2010) (see Fig. 1).

During the session, the participants could see the agenda, atimer representing the remaining time in the session, as well asthe list of ideas generated. In addition, each participant could seehis/her own performance, the performance of the simulated groupmembers, as well as a line representing the performance target.

4.1.1. Performance target manipulationWe set the performance target following Jung et al. (2010). In a

pretest, participants generated on average 17 ideas (S.D. = 7). Thuswe set the moderate performance target as 17 contributions andthe challenging performance target as 31 contributions (i.e., aver-age plus two standard deviations) during a 15-min period.

4.1.2. Feedback manipulationIn addition, similar to Ryan’s (1982) study, we provided control-

ling or informational feedback (using a pop-up window, see Fig. 2)every four minutes, so that the feedback was presented three timesduring the session5; informational feedback compared a participant’sperformance to the average performance (based on a pretest, e.g.,‘‘You have contributed X ideas. On average, people with the same goalhave contributed Y ideas by this time.’’). Following Ryan (1982), thecontrolling feedback included the same information, plus one of five


https://blueimp.net/ajax/


Fig. 2. Pop-up providing controlling feedback.

Table 1Descriptive statistics.

Performance feedback Performance target

Moderate Challenging

InformationalGroup size 63 48Age 21.13 (SD 4.43) 20.08 (SD 1.15)% Female 39.7 (SD 0.49) 27.1 (SD 0.45)

ControlGroup size 56 52Age 19.98 (SD 1.27) 19.98 (SD 1.46)% Female 51.8 (SD 0.50) 23.1 (SD 0.43)


different evaluative statements based on the participant’s perfor-mance at the time of the feedback. In other words, we created threeinterim targets; the levels of the interim targets were based on the rateat which ideas are typically generated, with more ideas being gener-ated at the beginning of a session, and the rate of ideas generated slow-ing down as the session continues. Specifically, we set the first interimtarget (after four minutes into the session) at 35% of the performancetarget; we set the second interim target (after 8 min) at 60% of the per-formance target; we set the last interim target (after 12 min) at 80% ofthe performance target. Participants being more than four contribu-tions below an interim target received the following evaluative state-ment: ‘‘Very poor. You are not nearly making enough progress towardreaching the goal. You should try much harder.’’ Participants being be-tween two and four contributions below an interim target receivedthe following statement: ‘‘Poor. You are not making enough progresstoward reaching the goal. You should do better.’’ Participants beingbetween two contributions below and two ideas above an interim tar-get received the following statement: ‘‘Fair. You’re performing justadequately.’’ Participants being between two and four contributionsabove an interim target received the following statement: ‘‘Good.You are doing as you should.’’ Finally, participants being more thanfour contributions above an interim target received the followingstatement: ‘‘Excellent. You should keep up the good work.’’

4.2. Experiment

4.2.1. Participants and research designWe recruited 219 business students from a large state univer-

sity in the U.S. as participants for this study; the average age ofthe students was 20.34 years (S.D. 2.66), 35.5 percent were female.The participants received course credit as a token of appreciation.In addition, participating in the study provided them with theopportunity to experience an idea generation session similar tothose conducted in organizations. We examined students’ back-ground before the experiment. Overall, they had relatively littleexperience related to computer-supported collaboration and ideageneration, but had much experience with using computers forwork tasks and web surfing. We manipulated the level of the per-formance target (moderate vs. difficult target) and type of perfor-mance feedback (informational vs. controlling), as describedabove. Participants were randomly assigned to these four condi-tions. The descriptive statistics of each group is shown in Table 1.

4.2.2. ProcedureUpon arriving in the laboratory, each participant was assigned

to a workstation in a computer lab with fifty separate


workstations. Each participant was automatically assigned to aseparate channel within the chat system, where the participantwould generate ideas together with four simulated participants.

After all participants in a session arrived, the experimenter readaloud the rules for the experiment. Specifically, participants wereasked to generate ideas on how to solve the issue of global warm-ing; they were asked to generate as many ideas as possible, towithhold criticism, and to build upon the ideas of others (see Os-born, 1957). An idea generation session lasted 15 min, followingwhich the participants were redirected to a brief questionnaireand then released.

4.2.3. MeasuresThe final questionnaire contained items related to demographic

variables, and to the participant’s perceived competence; this wasmeasured using a 7 point Likert scale developed in prior SDT liter-ature (Baard, Deci, & Ryan, 2004; Deci et al., 2001). Specially, weasked participants about the degree to which they agreed withthe following statements according to their experiences duringthe session: (1) I felt good at what I did; (2) I think I did pretty wellin this session; (3) I felt very capable; (4) I felt a sense of accom-plishment from what I did; (5) I have been able to learn interestingnew knowledge; (6) This was an activity that I could not do verywell (reverse coded).

The dependent variables were creative-idea-count and idea-cre-ativity. Specifically, we used the number (quantity) and total score(quality) of creative ideas (after removing redundant, frivolous,and low-quality ideas) to operationalize creativity. A primary goalof idea generation is typically ‘‘the identification of a few good orinteresting ideas with a view to implementing one of them’’ (Barki& Pinsonneault, 2001); hence, consistent with many prior studies



Fig. 3. Mean of perceived competence.


(e.g., Connolly et al., 1990; Diehl & Stroebe, 1987; Jung et al., 2010),we used creative idea count and idea creativity for evaluating ide-ation performance (see also Reinig, Briggs, & Nunamaker, 2007).

We used crowdsourcing to assess the creativity of the ideasgenerated. Crowdsourcing (Howe, 2006) refers to using ordinarypeople to create an ad hoc distributed labor network (Zittrain,2008), based on the premise that people with a broad range of per-spectives can perform (at least) at the level of individual experts(Page, 2007; Surowiecki, 2004). Crowdsourcing is increasingly usedin organizations to perform ‘‘microtasks,’’ and research has shownthat high quality data can be obtained using micro-task marketssuch as Amazon’s ‘‘Mechanical Turk’’ (mturk) (e.g., Buhrmester,Kwang, & Gosling, 2011; Kittur & Kraut, 2008; Mason & Suri,2010; Mason & Watts, 2010). After we consolidated duplicateideas, we asked workers on Mechanical Turk to rate the creativityof each distinct idea on a 7-point scale (each idea was rated by fivedifferent raters). The workers received $.10 for each rating. Accord-ing to Amabile (1983), a ‘‘product or response is creative to theextent that appropriate observers independently agree it is crea-tive’’ (p. 359). Since there is a little agreement regarding to theappropriate subcategories to rate creativity, we did not provide ex-plicit details. Instead, we instructed raters to rate the creativityusing the general category of ‘‘creativity of solution’’, which is con-sistent with previous literature (e.g., Goncalo & Staw, 2006; Ocker,Hiltz, Turoff, & Fjermestad, 1995).

Given that different sets of raters rated different ideas, inter-rater reliability was assessed using Solomon’s (2004) rating reli-ability calculator. This program uses Ebel’s (1951) algorithm forassessing interrater reliability based on incomplete data, whichdoes not assume that all items are rated by the same set of raters.The reliability was .65, considered adequate by commonly ac-cepted standards (Landis & Koch, 1977).

4.2.4. Data analysisThe mean for creative-idea-count was 11.38 (SD = 5.99); the

mean for idea-creativity was 38.57 (SD = 20.72). The reliability ofthe items measuring perceived competence was .70. Followingprevious literature (Baard et al., 2004; Deci et al., 2001), we usedthe average score of these items as an indicator of perceived com-petence. Table 2 presents a summary of the means and standarddeviations; Fig. 3 graphically presents the means of perceived com-petence. A 2 � 2 ANOVA crossing performance feedback and per-formance target indicates a significant main effect ofperformance target on perceived competence (F (1, 215) = 4.606,p < 0.05; Cohen’s d = 0.29, a small effect), supporting Hypothesis1. On the other hand, the main effect of performance feedback onperceived competence was not significant (F (1, 215) = 0.744,p > 0.05; Cohen’s d = 0.11). Therefore, Hypothesis 2 was not sup-ported. The results further show a significant interaction effect be-tween performance target and performance feedback on perceivedcompetence (F (1, 215) = 7.767, p < 0.01; Cohen’s d = 0.38, a smalleffect). Thus, we proceeded with testing Hypotheses 3a and 3busing planned contrasts.

Hypothesis 3a, which stated that when provided with controllingfeedback, a challenging performance target will have a detrimental

Table 2Means and standard deviation of perceived competence.

Dependent variable Performance feedback Performance target

Moderate Challenging

Perceived competence InformationalM 4.73 4.80SD .81 .78ControllingM 4.93 4.40SD .70 .86


effect on perceived competence, was supported. Specifically, in thechallenging target condition, participants provided with controllingfeedback had significantly lower perceived competence (t(98) = 2.373, p < .05; Cohen’s d = 0.50, a medium effect). Hypothesis3b, which stated that when provided with controlling feedback, amoderate performance target will have a positive effect on per-ceived competence, was not supported; in the moderate target con-dition, the difference between informational and controllingfeedback was not significant (t (117) = 1.479, p > .05; Cohen’sd = 0.27, a small effect).

We then conducted multivariate linear regression analysis to testHypothesis 4. Hypothesis 4, stating that people’s perceived compe-tence is positively related to their creativity, was supported (F (2,216) = 4.180, p < 0.05). Specifically, perceived competence is signif-icantly related to both creative-idea-count (b = 1.321, t = 2.658,p < 0.01; Cohen’s d = 0.36, a small effect) and idea-creativity(b = 4.425, t = 2.658, p < 0.05; Cohen’s d = 0.35, a small effect).

5. Discussion

We conducted a controlled laboratory experiment to test the ef-fects of different levels of performance targets and different typesof feedback on people’s perceived competence, as well as the influ-ence of perceived competence on people’s creativity. Firstly, ourstudy demonstrated the importance of perceived competence byshowing a significant relationship between perceived competenceand creativity. Further, our results support Hypothesis 1, in thatparticipants provided with a too challenging target have a lower le-vel of perceived competence than those provided with a moderatetarget. On the other hand, the type of performance feedback didnot significantly impact peoples’ perceived competence, andHypotheses 2 was not supported.

Yet, the results should be interpreted in light of the significantinteraction between performance targets and feedback. Specifi-cally, providing participants with a challenging target and control-ling feedback had a detrimental effect on perceived competence.We attribute this finding to the fact that when provided with achallenging target, people are, on average, more likely to receiverelatively more critical feedback, leading to decreased perceivedcompetence. These results suggest that, especially when pairedwith challenging performance targets, controlling feedback mayhave detrimental effects on people’s motivation, and informational




feedback may better satisfy people’s perceived competence ingeneral.

5.1. Implications for research

Our study has several important implications for research. First,our study contributes to the current literature by suggesting mech-anisms that can help to enhance people’s creativity. By applyingSDT in the context of computer-mediated idea generation, ourstudy shows that perceived competence is positively associatedwith creativity. These results advance our understanding of howto increase people’s creativity in computer-supported ideageneration.

Second, our study shows that performance targets and type ofperformance feedback have an interaction effect on people’s per-ceived competence. People provided with challenging performancetargets have a higher level of perceived competence when pro-vided with informational feedback than those provided with con-trolling feedback. In fact, under challenging performance targetsand controlling feedback, people have the lowest level of perceivedcompetence, and the differences in perceived competence betweenthe challenging performance targets/controlling feedback condi-tion and all other conditions were significant.6 These results showthat the type of performance feedback can be an important factorinfluencing people’s perceived competence. In general, informationalperformance feedback can help support people’s perceived compe-tence, and controlling feedback may actually decrease perceivedcompetence, especially when paired with challenging targets.

5.2. Implications for practice

Our study also has important practical implications. Our studyshows that perceived competence has a significant effect on peo-ple’s creativity in computer-mediated idea generation, and practi-tioners should attempt to design systems so as to increase people’sperceived competence. For example, practitioners may manipulatehuman–computer interfaces of ICTs to enhance people’s perceivedcompetence and thus increase their creativity. Based on our re-sults, practitioners can enhance people’s perceived competenceby manipulating both performance feedback and targets whendesigning human–computer interfaces of ICTs. During computer-mediated idea generation, when people try to achieve more chal-lenging targets, practitioners may want ICTs to provide informa-tional feedback, so as to avoid potentially demotivating feedback.On the other hand, when the target is less challenging, practitio-ners may provide controlling feedback and thus attempt to createan encouraging atmosphere so that individuals keep trying toachieve more challenging performance targets. Our study demon-strates that depending on the level of the performance target, cer-tain types of performance feedback may have detrimental effectson people’s perceived competence, and providing controlling feed-back (including judgmental information) may have negative effectson people’s competence need satisfaction.

5.3. Limitations and opportunities for future research

Our study is not without limitations. First, we used a specificsubject population, as well as a specific task. We chose a task (gen-erate ideas about global warming) that is of concern to the

6 Independent sample t-tests show that people with moderate targets andcontrolling feedback have a significantly higher level of perceived competence thanthose with challenging targets and controlling feedback (t (106) = 3.504, p < .01), andpeople’s perceived competence is higher with moderate targets and informationalfeedback than those with challenging targets and controlling feedback (t(113) = 2.065, p < .05).


experimental participants and assessed a specific kind of creativity(creative ideas), and we believe that there is no a priori reason toassume an interaction between the setting and the variables ofinterest. Nevertheless, future research could test the relationshipsusing different subjects, different tasks, or different kinds of crea-tivity. In addition, we used a simulated idea generation environ-ment. While this helped us to minimize within-group variancedue to differences in group composition, we are cognizant thatthere is no real collaboration in such an environment, and thismay also have influenced participants’ behavior. However, the re-sults of the post-experiment questionnaire indicate that only veryfew participants seemed to have noticed the simulated nature ofthe interaction, and the vast majority indicated that there were be-tween 3 and 6 other members in the group (there were 4 simulatedparticipants). Yet, to completely rule out potential effects of thesimulator, future studies could conduct the study using interactinggroups. Another limitation is that we used crowdsourcing to ratethe creativity of ideas. While the results show reasonable reliabil-ity based on Ebel’s (1951) algorithm, future studies are needed toassess the reliability of ratings obtained by crowdsourcing. Further,we have focused on two specific interface characteristics, namely,performance targets and feedback.

Notwithstanding these limitations, our findings open up a num-ber of interesting opportunities for future research. One suchopportunity is related to the design of the interface. Future studiescould look at other interface designs and examine how differenttypes of interface designs can influence people’s perceived compe-tence and creativity. For example, following principles of positivepsychology (Avital & Boland, 2008), future studies could attemptto provide an optimally stimulating environment, e.g., by avoidingoverly critical feedback to participants falling short of an interimtarget, and providing further encouragement for participantsmeeting or exceeding the target.

By focusing on people’s needs for competence, our study showshow different interface designs can influence people’s perceivedcompetence, and how people’s perceived competence can increasetheir creativity. Future studies can focus on other needs proposedby SDT (e.g., need for autonomy), and examine if other needs canalso increase people’s creativity, and how different interface de-signs influence peoples’ other needs satisfaction.

Finally, our study did not consider individual differences. Re-search in psychology suggests that various individual differencesare also likely to influence individuals’ motivation, and we encour-age other researchers to theorize about and test mechanisms toinfluence people’s creativity, accounting for such differences. Forexample, the individual difference variables autonomy orientationand control orientation may influence the effects of type of feed-back on the satisfaction of the individuals’ need for competence(Deci & Ryan, 1985). Likewise, people with different prior skillsmay be optimally motivated by different difficulty levels of goals,such that, for example, more challenging goals may be more suit-able for people with higher levels of related skills or knowledge.Further, to continuously monitor the progress during group collab-oration, individuals may need relevant skills such as self-regulationskills and metacognitive abilities, which may not be present foreveryone.7 Future studies can examine how people with differentlevels of those skills process the provided feedback, and how thesedifferences will influence their creativity. Clearly, our findings havesome important implications for the design of the human computerinterface of group collaboration environments.

7 We thank one anonymous reviewer for pointing this out.




6. Conclusions

Integrating research from human–computer interaction, moti-vation, technology supported group work, and psychology, wehave demonstrated that the design of the human–computer inter-face of an information system can influence individuals’ perceivedcompetence, which in turn influences their creativity in a particu-lar task. Here, we used one specific task, system, and context in asimulated environment; future research could test these findingsin different settings, using different systems and tasks. Further,we have only focused on two specific components of the interface,namely, performance targets and performance feedback. We wouldencourage other researchers to theorize on and test how other de-signs as well as individual differences influence people’s variousneeds satisfaction and creativity. Clearly, many research opportu-nities remain.

Acknowledgements

The work described in this paper was substantially supportedby research grants from City University of Hong Kong (ProjectNo. 7002626 and 7004123) and by a grant from the ResearchGrants Council of the Hong Kong Special Administrative Region,China (Project No. CityU149512).

References

Amabile, T. M. (1983). The social psychology of creativity: A componentialconceptualization. Journal of Personality and Social Psychology, 45, 357–376.

Avital, M., & Boland, R. J. (2008). Managing as designing with a positive lens. InDesigning information and organizations with a positive lens. In M. Avital, R. J.Boland, & D. L. Cooperrider (Eds.). Advances in appreciative inquiry series (Vol. 2).Oxford: Elsevier Science.

Baard, P. P., Deci, E. L., & Ryan, R. M. (2004). Intrinsic need satisfaction: Amotivational basis of performance and well-being in two work settings. Journalof Applied Social Psychology, 34, 2045–2068.

Barki, H., & Pinsonneault, A. (2001). Small group brainstorming and idea quality: Iselectronic brainstorming the most effective approach? Small Group Research,32(2), 158–205.

Barron, F., & Harrington, D. (1981). Creativity, intelligence, and personality. In M.Rosenzweig & L. Porter (Eds.). Annual review of psychology (Vol. 32,pp. 439–476). Palo Alto, CA: Annual Reviews.

Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonalattachments as a fundamental human motivation. Psychological Bulletin, 117(3),497–529.

Brusilovsky, P. (1996). Methods and techniques of adaptive hypermedia. UserModeling and User-Adapted Interaction, 6(2–3), 87–129.

Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon’s mechanical Turk.Perspectives on Psychological Science, 6(1), 3–5.

Chidambaram, L., & Tung, L. L. (2005). Is out of sight, out of mind? An empiricalstudy of social loafing in technology-supported groups. Information SystemsResearch, 16(2), 149–168.

Connolly, T., Jessup, L. M., & Valacich, J. S. (1990). Effects of anonymity andevaluative tone on idea generation in computer-mediated groups. ManagementScience, 36(6), 689–703.

DeCharms, R. (1968). Personal causation. New York, NY: Academic Press.Deci, E. L. (1975). Intrinsic motivation. New York, NY: Plenum.Deci, E. L., & Ryan, R. M. (1985). The general causality orientations scale: Self-

determination in personality. Journal of Research in Personality, 19(2), 109–134.Deci, E. L., Ryan, R. M., Gagné, M., Leone, D. R., Usunov, J., & Kornazheva, B. P. (2001).

Need satisfaction, motivation, and well-being in the work organizations of aformer Eastern Bloc country. Personality and Social Psychology Bulletin, 27,930–942.

Diehl, M., & Stroebe, W. (1987). Productivity loss in brainstorming groups: Towardthe solution of riddle. Journal of Personality and Social Psychology, 53(3),497–509.

Ebel, R. L. (1951). Estimation of the reliability of ratings. Psychometrika, 16(4),407–424.

Foss, N., Minbaeva, D., Pedersen, T., & Reinholt, M. (2009). Encouraging knowledgesharing among employees: How job design matters. Human ResourceManagement, 48(6), 871–893.

Goncalo, J. A., & Staw, B. M. (2006). Individualism–collectivism and group creativity.Organizational Behavior and Human Decision Processes, 100, 96–109.

Harlow, H. F. (1958). Nature of love. American Psychologist, 13, 673–685.Harter, S. (1978). Effectance motivation reconsidered: Toward a developmental

model. Human Development, 1, 661–669.


Heslin, P. A. (2009). Better than brainstorming? Potential contextual boundaryconditions to brainwriting for idea generation in organizations. Journal ofOccupational and Organizational Psychology, 82(1), 129–145.

Howe, J. (2006). The rise of crowdsourcing. Wired.com, June 2006.Jung, J. H., Schneider, C., & Valacich, J. (2010). Enhancing the motivational

affordance of information systems: The effects of real-time performancefeedback and goal setting in group collaboration environments. ManagementScience, 56(4), 724–742.

Kittur, A., & Kraut, R. E. (2008). Harnessing the wisdom of crowds in Wikipedia:Quality through coordination. In Proceedings of the 2008 international ACMconference on supporting group work, San Diego, CA (pp. 37–46). New York, NY:Springer.

Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement forcategorical data. Biometrics, 33, 159–174.

Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and performance.Englewood Cliffs, NJ: Prentice Hall.

Mason, W., & Suri, S. (2011). Conducting behavioral research on Amazon’sMechanical Turk. SSRN eLibrary, 2010.

Mason, W., & Watts, D. J. (2010). Financial incentives and the ‘‘performance ofcrowds’’. ACM SIGKDD Explorations Newsletter, 11, 100–108.

Michinov, N., & Primois, C. (2005). Improving productivity and creativity in onlinegroups through social comparison process: New evidence for asynchronouselectronic brainstorming. Computers in Human Behavior, 21, 11–28.

Muller, A., & Välikangas, L. (2002). Extending the boundary of corporate innovation.Strategy and Leadership, 30(3), 4–9.

Ocker, R., Hiltz, S. R., Turoff, M., & Fjermestad, J. (1995). The effects of distributedgroup support and process structuring on software requirements developmentteams: Results on creativity and quality. Journal of Management InformationSystems, 12(3), 127–153.

Osborn, A. F. (1957). Applied imagination. New York: Scribner.Page, S. E. (2007). The difference: How the power of diversity creates better groups,

firms, schools, and societies. Princeton, NJ: Princeton University Press.Paramythis, A., & Loidl-Reisinger, S. (2004). Adaptive learning environments and e-

Learning standards. Electronic Journal on e-Learning, 2(1), 181–194.Reinig, B. A., Briggs, R. O., & Nunamaker, J. F. (2007). On the measurement of

ideation quality. Journal of Management Information Systems, 23(4), 143–161.Reis, H. T. (1994). Domains of experience: Investigating relationship processes from

three perspectives. In R. Erber & R. Gilmour (Eds.), Theoretical frameworks forpersonal relationships (pp. 87–110). Hillsdale, NJ: Erlbaum.

Ryan, R. M. (1982). Control and information in the intrapersonal sphere: Anextension of cognitive evaluation theory. Journal of Personality and SocialPsychology, 43(3), 450–461.

Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation ofintrinsic motivation, social development, and well-being. American Psychologist,55(1), 68–78.

Ryan, R. M., Deci, E. L., & Grolnick, W. S. (1995). Autonomy, relatedness, and the self:Their relation to development and psychopathology. In D. Cicchetti & D. J.Cohen (Eds.), Developmental psychopathology: Theory and methods(pp. 618–655). New York: Wiley.

Ryan, R. M., Kuhl, J., & Deci, E. L. (1997). Nature and autonomy: An organizationalview of social and neurobiological aspects of self-regulation in behavior anddevelopment. Development and Psychopathology, 9(4), 701–728.

Skinner, E. A. (1995). Perceived control, motivation, & coping. Thousand Oaks, CA:Sage.

Solomon, D. J. (2004). The rating reliability calculator. BMC Medical ResearchMethodology, 4(11).

Steers, R. M., Mowday, R. T., & Shapiro, D. L. (2004). The future of work motivationtheory. The Academy of Management Review, 29(3), 379–387.

Surowiecki, J. (2004). The wisdom of crowds. New York, NY: Doubleday.Tubbs, M. E., & Ekeberg, S. E. (1991). The role of intentions in work motivation:

Implications for goal-setting theory and research. The Academy of ManagementReview, 16(1), 180–199.

Valacich, J. S., Jung, J. H., & Looney, C. (2006). The effects of individual cognitiveability and idea stimulation on individual idea generation performance. GroupDynamics, 10(1), 1–15.

Vallerand, R. J., & Bissonnette, R. (1992). Intrinsic, extrinsic, and amotivational stylesas predictors of behavior: A prospective study. Journal of Personality, 60(3),599–620.

Vansteenkiste, M., Neyrinck, B., Niemiec, C. P., Soenens, B., DeWitte, H., & Van denBroeck, A. (2007). On the relations among work value orientations,psychological need satisfaction and job outcomes: A self-determinationtheory approach. Journal of Occupational and Organizational Psychology, 80(2),251–277.

Venkatesh, V., & Speier, C. (1999). Computer technology training in the workplace:A longitudinal investigation of the effect of mood. Organizational Behavior andHuman Decision Processes, 79(1), 1–28.

Wang, X., & Clay, P. F. (2012). Beyond adoption intention: Online communities andmember motivation to contribute longitudinally. Journal of OrganizationalComputing and Electronic Commerce, 22(3), 215–236.

Wasko, M., & Faraj, S. (2005). Why should I share? Examining social capital andknowledge contribution in electronic networks of practice. MIS Quarterly, 29(1),35–57.

White, R. W. (1959). Motivation reconsidered: The concept of competence.Psychological Review, 66(5), 297–333.

White, R. W. (1963). Ego and reality in psychoanalytic theory: A proposal regardingindependent ego energies. New York, NY: International Universities Press.


http://refhub.elsevier.com/S0747-5632(14)00077-6/h0005






































































































































Woodman, R. W., Sawyer, J. E., & Griffin, R. W. (1993). Toward a theory oforganizational creativity. Academy of Management Review, 18(2), 293–321.

Zhang, P. (2008a). Motivational affordances: Reasons for ICT design and use.Communications of the ACM, 51(11), 145–147.

Zhang, P. (2008b). Toward a positive design theory: Principles for designingmotivating information and communication technologies. In M. Avital, R. J.


Boland, & D. L. Cooperrider (Eds.), Designing information and organizations with apositive lens. Advances in appreciative inquiry (Vol. 2, pp. 45–74). Amsterdam: JAIPress.

Zittrain, J. (2008). Ubiquitous human computing. Philosophical Transactions of theRoyal Society A: Mathematical, Physical and Engineering Sciences, 366(1881),3813–3821.















Documents

Enhancing creativity in group collaboration: How performance targets and feedback shape perceptions and idea generation performance