group creativity: the effects of extrinsic, intrinsic, and...

Group Creativity: The Effects of Extrinsic, Intrinsic,and Obligation Motivations

Randolph B. CooperUniversity of Houston

Bandula JayatilakaState University of New York at Binghamton

ABSTRACT: Creativity-reducing extrinsic motivationgenerally has been associated with rewards tied to taskperformance. However, there is also evidence that cre-ativity-affecting motivation may also result from ex-trinsic rewards that are not tied to task performance.This type of motivation may be due to feelings of obli-gation. A research model is developed that examinedhow such obligation motivation differs from extrinsicand intrinsic motivations in terms of influencing cre-ativity. Given the importance of groups in organiza-tions, the focus is on motivation and creativity withininteracting groups. A laboratory experiment is per-formed that involved groups determining requirementsfor an academic information system. The results sup-port a conceptual differentiation between the threetypes of motivation.

The creativity of interacting groups is becoming in-creasingly important in organizations (West, 2002;Williams & Yang, 1999) and has gained an increasingamount of research attention. This research has exam-ined the effects of group characteristics (size, diversity,cohesiveness, potency, autonomy, cooperation, type ofleader support, and degree of structure: Abbey & Dick-son, 1983; Amabile, Conti, Coon, Lazenby, & Herron,1996; Craig & Kelly, 1999; Nemeth & Kwan, 1985;Redmond, Mumford, & Teach, 1993; Sosik, Avolio, &Kahai, 1997, 1998; Thornburg, 1991) and task charac-teristics (time pressure and group support systems:Karau & Kelly, 1992; Kelly & Karau, 1993; Kelly &McGrath, 1985; Ocker, Hiltz, Turoff, & Fjermestad,1995–1996; Satzinger, Garfield, & Nagasundaram,1999). However, the influence of motivation on group

creativity has received little attention to date; this is in-teresting since motivation has been a major focus ofindividual creativity research (Eisenberger &Cameron, 1996; Hennessey & Amabile, 1998). Groupprocess theorists have argued that there is value in in-vestigating whether individual-level phenomena alsooccur at the group level (Hinsz, Tindale, & Vollrath,1997), and so it is reasonable to examine the effects ofmotivation on group-level creativity.

Two types of motivation are typically addressed inthe creativity literature. Extrinsic motivation resultsfrom individuals perceiving an instrumental connec-tion between their behavior and their receipt of extrin-sic rewards, such as money and praise (Deci & Ryan,1985). This motivation typically reduces individuals’creativity (Amabile, 1988; Collins & Amabile, 1999;Hennessey, 2000) because they tend to focus on the re-wards rather than on the task itself (Amabile, 1996). Asa result, individuals may be distracted from creativ-ity-relevant aspects of the task and/or may exhibitfunctional fixedness or other forms of cognitive rigid-ity, with their efforts narrowly focused on the task asoriginally defined and on common algorithms thathave worked well in the past (Hogarth, 1987; Staw,1990; Woodman, Sawyer, & Griffin, 1993). In con-

Creativity Research Journal2006, Vol. 18, No. 2, 153–172

Copyright © 2006 byLawrence Erlbaum Associates, Inc.

Creativity Research Journal 153

We thank David Taylor for his valuable contributions early on in ourwork. In addition, we thank Wynne Chin, Rudy Hirschheim,Surinder Kahai, Richard Scamell, and Andrew Schwarz for theirvaluable contributions. Finally, we thank Mark Runco and the anon-ymous reviewers for their helpful comments.

Correspondence and requests for reprints should be sent toRandolph B. Cooper, C. T. Bauer College of Business, University ofHouston, Houston, TX 77204–6282. E-mail: [email protected]

trast, intrinsic motivation arises from positive reactionsto qualities of the task itself; intrinsically motivated in-dividuals engage in a task primarily out of their own in-terest in it (Lepper, Greene, & Nisbett, 1973). Becausethey enjoy the task itself and the process of searchingfor new solutions, intrinsically motivated individualsare more likely to expend energy exploring the prob-lem and find creative solutions (Hennessey, 2000;Staw, 1990; Woodman et al., 1993).

There may be a third type of motivation importantfor creativity that comes from individuals’ feelings ofobligation. This motivation has characteristics incommon with extrinsic motivation in that it is linkedto extrinsic rewards. However, in contrast to extrinsicmotivation, rewards that may result in obligation mo-tivation are not contingent on task performance. Forexample, rewards that can result in obligation motiva-tion are given for merely being present during a task.This article examines the existence of obligation mo-tivation and how it compares to extrinsic and intrinsicmotivations in its influence on group creativity.

Theory and Research Model

This section provides a deeper level of understand-ing of creativity and motivation, including issues sur-rounding the conceptualization of motivation at thegroup level. The research model and hypotheses arethen described.

Creativity

Organizational creativity results from individualsworking together on a task in a complex social systemand taking a more heuristic than algorithmic approachwith an outcome consisting of a useful and novel prod-uct, service, procedure, or process (Amabile, 1996;Paulus, 2000; Woodman et al., 1993). Useful outcomesare those appropriate for the task and task goals. Noveloutcomes can be illustrated in terms of paradigm modi-fication (Nagasundaram & Bostrom, 1995), which canbe viewed as changes to organizational elements andchanges to relationships among those elements. Thehighest degree of novelty occurs when substantialchanges are made to organizational elements and theirrelationships. For example, Cooper (2000) describedcreativity associated with new imaging informationtechnology in terms of radical changes in departments

by facilitating the changing or eliminating of depart-mental boundaries, job descriptions, and work flows.In contrast, he described a lack of creativity as beingreflected in an automation of manual processes with-out significant changes to current departmental bound-aries, job descriptions, and work flows.

The capacity for cross-fertilization of ideas alongwith thorough problem exploration makes interactinggroups potentially more creative than individuals ornominal groups (Paulus, 2000; Thompson, 2003; West,2002). Group members’ task-relevant knowledge andtechnical skills as well as creativity-relevant cognitiveskills and personality traits are important for suchcross-fertilization; however, task motivation is key togroup creativity since it reflects group members’ driveand determines what they will do (Amabile, 1990;Barron & Harrington, 1981; Hennessey, 2000; Wood-man et al., 1993). For example, sufficient motivationcan reduce the degree of social loafing and free riding(Baron, Kerr, & Miller, 1992; Kerr & Bruun, 1983;Latané et al., 1979; Sweeney, 1973), which would oth-erwise reduce group member participation and therebyinhibit the degree of cross-fertilization. Motivationalso affects creativity by influencing the use of mem-bers’ knowledge and cognitive and technical skills,thereby affecting the degree of exploration and thelikelihood that alternative (and potentially more novel)response possibilities will be examined (Amabile,1988).

Individual Motivation

Prior to examining group motivation, motivation atthe individual level is explored. The most popularnonacademic conceptualization of motivation is interms of the verb to want (Higgins & Kruglanski,2000a). Excluding habits and automatic responses (e.g.,Cooper&Bhattacherjee,2001;Mischel&Shoda,1995;Zajonc, 1980), it is motivation that converts cognitioninto action (Siegel & Ramanuaskas-Marconi, 1989;Sorrentino & Higgins, 1986). Theories that have exam-ined this conversion process commonly focused on an-ticipatory thought in the form of expectancy models(Bandura, 1986; Bowditch & Buono, 1990; Higgins &Kruglanski, 2000b; Naylor et al., 1980).

For example, Atkinson (1957) described an individ-ual’s motivation to perform a task as being a functionof:

154 Creativity Research Journal

R. B. Cooper and B. Jayatilaka

Motivation to perform a task =∑ni = 1 f(Motivei × Expectancyi × Incentivei)

where i refers to a potential outcome associated withthe task, Motivei is a relatively stable energizing drive(e.g., maximizing satisfaction associated withachievement or minimizing pain associated with hu-miliation) associated with outcome i, Expectancyi isthe subjective probability that the task will be fol-lowed by outcome i, and Incentivei is the relative at-tractiveness or unattractiveness (i.e., valence) of out-come i. In addition to task choice, motivation iscrucial to the “vigor of the (task activities) … and forits tendency to persist over time” (Atkinson, 1957).Therefore, faced with a choice between task A and B,an individual will tend to choose A if the motivationassociated with A is larger than that associated withB. In addition, the time and energy applied to task Aactivities would then be a positive function of task Amotivation.

Individuals’ motivations have been described ascoming from both extrinsic and intrinsic sources

(Bandura, 1986; Deci & Ryan, 1985). Some theoristshave been unclear whether extrinsic and intrinsic mo-tivations are conceptually distinct; for example, theyhave treated intrinsic and extrinsic motivation as op-posite ends of a single dimension (e.g., Harter, 1981;Newman, 1990). However, many theorists have dis-tinguished between intrinsic and extrinsic motivation(e.g., Lepper & Henderlong, 2000; Naylor et al.,1980; Staw, 1977) and have proposed that this dis-tinction is important for creativity (e.g., Amabile,Hill, Hennessey, & Tighe, 1994; Woodman, et al.,1993). Atkinson’s model described above focuses onan individual’s general evaluation of motive and in-centive associated with task outcomes. In contrast,Staw’s (1977) expectancy model differentiates be-tween extrinsic and intrinsic outcomes and adds thepotential value associated with task behavior for itsown sake. From Staw’s perspective (see Figure 1):

Motivation to perform a task =f[∑ni = 1 P1P2a(EVi) + ∑ni = 1 P3a(EVi) + IVbeh +

P1(IVacc)]


Group Creativity

Figure 1. Motivation model.

where EVi = extrinsic valence associated with extrinsicoutcomei; IVbeh = intrinsic valence associated with taskbehavior; IVacc = intrinsic valence associated with taskaccomplishment; P1 = probability that task behaviorwill lead to task accomplishment; P2a = probabilitythat task accomplishment will lead to extrinsicoutcomei; P3a = probability that task behavior will leaddirectly to extrinsic outcomei.

Here, extrinsic outcomei includes monetary re-wards, desirable working conditions, and supervisorrecognition. These outcomes can result from task ac-complishment. For example, monetary bonuses andrecognition may result from an individual successfullydeveloping a new information system. In this case,P1P2a(EVi) represents the associated expected valence.Extrinsic outcomei can also result from task behaviorsalone. For example, an individual might earn a salaryor supervisor recognition for merely attempting to de-velop a new information system, and P3a(EVi) repre-sents the associated expected valence. These valencesinvolve Atkinson’s notions of motive and incentive,and are related to how well the outcomes are perceivedas satisfying the individual’s needs. For example,Maslow (1954) recognized an individual’s need forself-esteem, which can be satisfied by recognitionfrom an individual’s supervisor.

Need satisfaction can also be derived from out-comes intrinsic to task behavior. For example, enjoy-ment of task behaviors has been found to be importantfor creativity (e.g., Amabile, 1996; Elam & Mead,1990). These outcomes are independent of any rewardsor recognition by others and help to satisfy individuals’needs for pleasure. Because the probability that a be-havior will result in its associated intrinsic outcome ap-proximates one, IVbeh represents the expected intrinsicvalence. In addition, need satisfaction can come fromoutcomes intrinsic to task accomplishment. For exam-ple, accomplishment of the task independent of any re-wards or recognition by others can satisfy an individ-ual’s need for self-efficacy. In this case, an individualmay enjoy the challenge associated with successfullycreating information systems solutions that satisfyconflicting user requirements; P1(IVacc) represents theassociated expected intrinsic valence.

Cognitive evaluation theory suggests that extrinsicmotivation occurs when behaviors are perceived as in-strumental in getting rewards (Deci & Ryan, 1985).From this perspective, rewards given that are not contin-genton taskperformance(e.g., rewardsgivenformerely

being present during a task) should not be associatedwith instrumental task-accomplishment behavior andthereforeshouldnot result inextrinsicmotivation.How-ever, this view appears to be contradicted by recent find-ings that creativity-reducing motivation can be fosteredeven when rewards are received prior to task engage-ment (Amabile, Hennessey, & Grossman, 1986). In ad-dition, individuals’ self-evaluations have been found todecrease creativity (Szymanski & Harkins, 1992).Taken together, this suggests thatobligationmayplayanimportant role in motivation. Such obligation motiva-tion differs from extrinsic motivation in that it is linkedto rewards that are not contingent on task performance;it differs from intrinsic motivation in that it is linked toextrinsic rewards.

There appears to be a powerful rule of reciprocationthat is pervasive in all human societies, and which re-sults in our feeling obligated to the future repayment ofhelp, favors, and gifts (Cialdini, 1993; Eisenberger etal., 2001; Gouldner, 1960). This sense of obligationhas been credited with the notion of being human(Leakey & Lewin, 1978), with enabling the division oflabor thereby facilitating the human adaptive ability(Tiger & Fox, 1971), and with facilitating systems ofaid, gift giving, defense, and trade (Cialdini, 1993).Therefore, human societies train their members tocomply with and believe in it (Cialdini, 1993) andmoral tales of reciprocity proliferate in literature andlore, such as that by Aesop and Dr. Seuss (Whatley,Webster, Smith, & Rhodes, 1999). Such obligation canbe felt toward organizations as well as individuals(Eisenberger et al., 2001; Rhoades, Eisenberger, &Armeli, 2001).

Obligation results in part from internalized pres-sures to be consistent with one’s value system (Whatlyet al., 1999) and to help maintain one’s positiveself-image as one who repays debts (Eisenberger et al.,2001; Harrison, 1995). Failing to fulfill such obliga-tions can result in self-reproach and guilt (Whately etal., 1999). Though internal pressure may not be instru-mental in receiving extrinsic rewards, it can result inbehavior much like extrinsic motivation since an indi-vidual’s freedom to act is constrained (Goei et al.,2003) because he or she feels pressure to behave inways he or she believes others (the reward givers)would like him or her to behave (Gouldner, 1960). Inaddition, meeting obligations can be instrumental inthe sense that it helps individuals avoid the socialstigma associated with the reciprocity norm’s violation



(Eisenberger et al., 2001; Whatley et al., 1999). Thesetwo pressures on individuals to form their behaviors inaccord with others’needs and expectations suggest thatthe effect of obligation can be conceptualized in a man-ner similar to extrinsic motivation.

From Staw’s perspective (see Figure 1), obligationcan be added to extrinsic and intrinsic motivation in thefollowing way:

Motivation to perform a task =f(•) + g[∑ni = 1 P1P2b(OVi) + ∑ni = 1 P3b(OVi)]

where f(•) = the motivation function that includes in-trinsic and extrinsic motivations; OVi = obligation va-lence associated with obligation outcomei; P1 = proba-bility that task behavior will lead to taskaccomplishment; P2b = probability that task accom-plishment will lead to obligation outcomei; P3b = prob-ability that task behavior will lead directly to obliga-tion outcomei.

Here, obligation outcomei can be the repayment ofrewards that are not contingent on task activities by ac-complishing a task in a high quality manner. Thismight occur, for example, when an individual success-fully develops a new information system without re-ceiving any associated bonuses or recognition. In thiscase P1P2b(OVi) represents the expected valence. Obli-gation outcomei can also result from task behaviorsalone. For example, an individual might believe thatthe reward can be repayed by merely being conscien-tious in attempting to develop a new information sys-tem, and P3b(OVi) represents the expected valence.These valences are related to perceptions of the indi-vidual as to how well the outcomes satisfy importantneeds of the reward giver and thereby fulfill reciprocitynorm requirements.

Aggregating Individuals’ Motivations

Although the previous model focuses on individu-als’ tasks, it can be augmented to reflect individuals’motivations to perform group tasks (Shamir, 1990).First, the model can be expanded to include the expec-tation that individual effort will result in the group’stask performance. This expectation would be affectedby an individual’s evaluation of other group members’abilities as well as group leadership, group cohesive-ness, and group effectiveness. Second, when consider-ing extrinsic and obligation outcomes directly affected

by an individual’s task behaviors, social acceptance ofan individual by the group can be included because itoften depends on social judgments about the individ-ual’s contributions and adherence to norms, which typ-ically relate to behavior independent of task perfor-mance.

A question remains as to whether and how individu-als’ motivations can be aggregated to enablegroup-level analyses. Hackman and Morris (1975) andWeingart (1992) refer to a general notion of group mo-tivation, which reflects the amount of time and energygroup members are willing to apply to a group task.However the translation of individual members’ moti-vations to group motivation is not necessarily straight-forward. Aggregation of individual-level measures toform or reflect group-level constructs can be appropri-ate when the individual-level measures demonstrate ashared understanding among group members aboutsome group or contextual characteristic. This approachhas been used to reflect group aspirations (Zander &Medow, 1963), collective self-esteem (Crocker &Luhtanen, 1990), and group potency (Guzzo, Yost,Campbell, & Shea, 1993). Such approaches typicallybase their validity on the assumption that there is lowvariance among the individual measures within eachgroup (e.g., Sosik et al., 1997).

However, a shared understanding of some group orcontextual characteristic is not of interest here. Rather,of interest is a measure that reflects the time and energy(i.e., motivation: Naylor et al., 1980) that group mem-bers are willing to devote to a group task. If one wereinterested in developing a group attribute called moti-vation, a sophisticated model of motivation-orientedgroup dynamics that includes notions of cohesion,norms, and leadership (Hackman & Morris, 1975;Shamir, 1990; Woodman et al., 1993) might be usedalong with the measures of individual group members’motivations (LePine, Hollenbeck, Ilgen, & Hedlund,1997). Nevertheless, without such a model, one canstill aggregate individual members’ motivations withina group and employ them as an estimate in group-levelanalyses with the understanding that these motivationsare resources that can be employed by the group andare related to the level of time and energy that the groupwill employ on a task (LePine et al., 1997).

An aggregation approach that has been used insuch situations is that proposed by Steiner (1972),which is based on the assumption that significantvariation does exist among individuals within a


Group Creativity

group. This approach has been employed to charac-terize groups in terms of individual members’ creativ-ity, intelligence, physical strength, and conscientious-ness (e.g., Barrick, Stewart, Neubert, & Mount, 1998;Colquitt, Hollenbeck, Ilgen, LePine, & Sheppard,2002; LePine et al., 1997). Steiner indicates that dif-ferent tasks may require different uses of member ca-pabilities and therefore different aggregationschemes. With additive tasks, each member’s contri-bution is important; and the sum (or average withgroups of equal sizes) of these contributions is an ap-propriate group-level indicator. Additive aggregationis appropriate here because, as described earlier, thecapacity for cross-fertilization of ideas along withthorough problem exploration is what makes interact-ing groups potentially more creative than individualsor nominal groups. Thus, the extent to which each in-dividual provides effort is important and can increasegroup creativity. Therefore, in accord with Barrick etal. (1998), an additive aggregation employing groupmeans is adopted.

In summary, motivation reflects an individual’s will-ingness to apply time and energy to a task. An individ-ual’s motivation is determined by his or her anticipatoryexpectations of outcomes that can accrue, and the resul-tant satisfaction of needs. A reasonable estimate of theamount of time and energy a group is willing to apply toa task can be measured by an additive aggregation of the

individuals’ motivations within the group. The follow-ing discussions propose the research constructs andtheir relationships as illustrated in Figure 2.

Extrinsic Motivation and Creativity

As described earlier, extrinsic motivation isother-directed in that it arises from sources outside ofthe task itself, such as monetary reward and supervisorrecognition (Lepper & Henderlong, 2000). In addition,it reduces individuals’ freedom in that their behaviorstend to be confined to actions instrumental in gainingthe rewards. This is likely to decrease individuals’ cre-ativity because their attentions are narrowly focused onthe task as originally defined and they typically rely oncommon, well-worked algorithms that they havelearned for doing the particular task. However, thereare important exceptions to this negative relationshipbetween extrinsic motivation and creativity. Extrinsicmotivation can provide individuals the focus and en-ergy necessary for completing a creative task whenthere are important “tedious hurdles” that are not them-selves intrinsically interesting (Amabile, 1988, 1996;Brophy, 1998). For example, when the task is develop-ing information technology requirements, careful vali-dation of ideas can be very important and can be lessintrinsically interesting than interviewing and analyz-ing. This need to effectively perform the less interest-



Figure 2. Research model. *Must be significant for the associated a and b hypotheses to be supported.

ing activities may counter the potentially negative ef-fect of extrinsic motivation on creativity during suchtasks. In addition, positive effects of extrinsic motiva-tion have been found when participants have been ex-plicitly told to be creative (Eisenberger & Shanock,2003). However, in the experiment reported here,groups were not provided the opportunity to carefullyvalidate their ideas and they were not told to be cre-ative. This leads to the following hypothesis.

H1. Extrinsic motivation will decrease group cre-ativity.

(This positive relationship is depicted in Figure 2 aspositive.)

Obligation Motivation and Creativity

Obligation motivation is based on the reciprocitynorm, and results from an aversive psychological ten-sion that reduces individuals’ freedom of action (Goeiet al., 2003). As described earlier, this tension resultsfrom individuals’ needs to be consistent with theirvalue systems and to maintain positive self-images.Reciprocating the favor releases this aversive tension,thereby regaining the individual’s freedom of action(Goei et al., 2003). In addition, meeting obligationshelps individuals avoid the social stigma associatedwith violation of the reciprocity norm. These needs tobe consistent with value systems and to avoid social os-tracism result in a motivation to act in ways that can re-pay favors. For example, when rewards are given formerely being present in a group activity, repaymentcan take the form of acting in ways that individuals be-lieve the reward giver would like them to act. Thisother-directed behavior and constraint on actions issimilar to the effects of extrinsic motivation. There-fore, for reasons described earlier associated with theextrinsic motivation hypothesis, the following parallelhypothesis is proposed:

H2. Obligation motivation will decrease group cre-ativity.

Intrinsic Motivation and Creativity

Intrinsic motivation arises from positive reactionsto qualities of a task itself; intrinsically motivated in-dividuals engage in a task primarily out of their own

interest in it (Lepper et al., 1973). There is agreementamong theorists that intrinsic motivation is a psycho-logical state characterized by deep involvement andplayfulness (Amabile, 1996). Enjoyment of task be-haviors is an intrinsic outcome that provides individu-als with the sense of engaging in play rather thanwork and it helps satisfy their needs for pleasure(Amabile, 1996; Higgins & Kruglanski, 2000a; Staw,1977). Because they enjoy the task itself and the pro-cess of searching for new solutions, intrinsically mo-tivated individuals (a) devote more attention to a taskfor its own sake rather than as a means to an end, (b)have a willingness and ability to explore alternativecognitive pathways (to step away from a problem tosee the nonobvious sides of issues) in problem solv-ing, and (c) attend to apparently incidental aspects ofthe tasks and environment (Amabile, 1996; Higgins& Kruglanski, 2000a; Staw, 1977). Therefore intrinsi-cally motivated individuals will be more likely ex-pend energy exploring the problem and find creativesolutions (Crutchfield, 1962; Hennessey, 2000; New-ell, Shaw, & Simon, 1962; Staw, 1990; Woodman etal., 1993). This leads to the following hypothesis:

H3. Intrinsic motivation will increase group cre-ativity.

Rewards and Motivations

Cognitive evaluation theory suggests that extrinsicmotivation occurs when behaviors are perceived as in-strumental in getting rewards (Deci & Ryan, 1985).For example, rewards tied to task performance will re-sult in greater extrinsic motivation than rewards formerely being present when task activities occur. Thisimpact of instrumentality on motivation has substantialempirical support (Lepper & Henderlong, 2000). Re-wards can also be perceived as informational, wherebyreward receipt conveys information about the recipi-ent’s competence and self-determination (Deci &Ryan, 1985). This informational aspect can mitigatethe negative extrinsic effects by increasing intrinsicmotivation (Deci & Ryan, 1985). However, rewardsare received after task performance in the experimentreported here, eliminating the potential information ef-fect. This leads to the following hypothesis:

H4. Extrinsic motivation increases as rewards aretied to task performance.


Group Creativity

In contrast, the literature on reciprocity suggeststhat rewards dependent on individuals’ behavior (e.g.,individuals feel that rewards are tied to task perfor-mance) reduce the level of obligation motivation(Cialdini, 1993; Eisenberger et al., 2001; Gouldner,1960). This leads to the following hypotheses:

H5. Obligation motivation decreases as rewards aretied to task performance.

(This negative relationship is depicted in Figure 2 asH5a being negative and H5 being positive.)

Decreased intrinsic motivation resulting from ex-pectations of extrinsic rewards has been proposed byseveral theories (e.g., cognitive evaluation theory:Deci & Ryan, 1985; overjustification hypothesis andself-perception theory: Lepper, 1981, Bem, 1972; so-cial cognitive theory: Bandura, 1986) and is empiri-cally well documented (e.g., Boone & Cummings,1981; Deci, Koestner, & Ryan, 1999; Jordan, 1986;Staw, 1974). These theories suggest that when behav-ior is perceived to be instrumental in receiving re-wards, individuals can feel less personal interest, lesscompetent, and more behaviorally controlled. This re-sults in a negative emotional tone, the experience ofpressure and tension, a decrease in enjoyment, andthereby a decrease in intrinsic motivation (Bandura,1986; Deci & Ryan, 1985; Lepper, 1981; Szymanski& Harkins, 1992). In addition, research examiningextrinsic rewards that are not tied to task performancehas found no significant influence on intrinsic moti-vation (Amabile et al., 1986). This leads to the fol-lowing hypothesis:

H6. Intrinsic motivation decreases as rewards aretied to task performance.

(This negative relationship is depicted in Figure 2 asH6a being negative and H6 being positive.)

Further, rewards have been found to decrease moti-vation if they are not perceived as being reasonablecompensation for the task (Abbey & Dickson, 1983;Amabile, 1996). Therefore three reasonableness hy-potheses are included:

H4b. Extrinsic motivation increases as rewards areperceived as being more reasonable.

H5b. Obligation motivation increases as rewards areperceived as being more reasonable.

H6b. Intrinsic motivation increases as rewards areperceived as being more reasonable.

Method

Experimental Participants

One hundred thirty-eight senior-level undergradu-ate students enrolled in multiple sections of a Informa-tion Technology Systems Analysis and Design class ata large urban university were participants in the labora-tory experiment. Participants were recruited from threesemesters to participate in single sessions. Forty-twostudents participated the first semester, 45 in the sec-ond semester, and 51 in the third semester. Participantswere randomly assigned to 46 teams of mixed gender,each team consisting of three students.

Experimental Procedure

The experimental procedure was as follows. Allparticipants attending an experimental session (typi-cally 9 participants) initially met together in one room.At this time, they were told that the objective of thestudy was to identify factors that can affect informationtechnology (IT) requirements determination. Each par-ticipant was then provided with a copy of the generalinstructions (see Appendix A.1). The experimenterread the instructions out loud along with the partici-pants and asked if there were any questions. Partici-pants were not told that the experiment concerned cre-ativity. This is in accord with Amabile (1996) and isprobably more reflective of IT requirements tasks ingeneral (Mumford, 2003). Participants were told thatthey would have 60 min to complete their task, afterwhich they would return to the meeting room and fillout a questionnaire regarding their group experiences.Participants were then randomly assigned to groupsand to stakeholder roles. Groups were then randomlyassigned to treatments and rooms. Each participantwas provided with an individual instruction sheet (seeAppendix A.2 for an example), which described his orher reward and stakeholder role. Students were in-formed that they could quit the experiment at any time;and if they quit, they would receive only one extracredit point. Each group met for 60 min in its assignedroom, which contained chairs, one table, and a chalk-



board. The experimenter notified each group whenthere were 15 min remaining.

Task

The task for each group involved the determinationof changes aimed at improving the University’s admin-istrative information systems. Each participant in agroup was randomly assigned to represent one of threestakeholders: working students, nonworking students,and faculty. Working and nonworking student repre-sentatives were given the following set of stakeholderimprovement issues, with “working” or “nonworking”appropriately included:

1. Classes desired by working/nonworking stu-dents are not available because they are alreadyclosed.

2. Classes desired by working/nonworking stu-dents are not available because they are not of-fered.

3. Classes desired by working/nonworking stu-dents are in conflict with each other becausethey are offered at the same day and time.

4. After registering for a class, working/nonworking students are dropped from theclass because it is cancelled by administration,for example, due to too few students taking theclass.

5. Classes desired by working/nonworking stu-dents are offered at inconvenient days and/ortimes.

Faculty representatives were given the following setof stakeholder improvement issues:

1. Assigned classes are scheduled for inconve-nient days and/or times for faculty members.

2. Assigned classes will require too much workon the part of faculty members.

3. Assigned classes do not interest faculty mem-bers.

4. Assigned classes are not appropriate becausethe faculty member does not understand thematerial.

5. Assigned classes result in an overload of workto the faculty member due to large class sizes.

6. Required equipment and material for the as-signed classes are not available.

At the end of the task, each group handed in a reportthat described suggestions for administrative informa-tion systems improvements in terms of what the systemshould do and how the proposed changes help resolvethe student and faculty issues.

This task is in accord with Torrance’s (1965) sug-gestion that product (in this case administrative infor-mation systems) improvement is an appropriate test forcreative behavior. In addition, Amabile (1996) sug-gested that the task should (a) be heuristic–open endedenough to permit considerable flexibility and noveltyin response, (b) result in an observable product or re-sponse that can be recorded and later judged on cre-ativity, and (c) be feasible in the sense that virtually allparticipants in the study can produce something thatcan be assessed by judges. In accord with Amabile, thetask (a) is open ended, with multiple issues and fewconstraints on solutions, (b) results in a written reportto be judged, and (c) is feasible due to students’ famil-iarity with the issues and the administrative informa-tion system. (Note that these issues were identifiedfrom an earlier survey of similar students.)

Operationalization of Constructs

Reward. Hypotheses 4, 5, and 6 require a differ-entiation between rewards perceived as being tied totask performance and rewards perceived as being inde-pendent of task performance. Negotiations with the in-structors teaching classes from which student partici-pants were drawn resulted in the following rewardconditions. In addition to receiving one (guaranteed)extra credit point for showing up at the experiment, stu-dents were given:

1. Task performance-independent condition—guaranteed receipt of 2 extra credit points inde-pendent of the quality of work.

2. Task performance condition—the potential ofreceiving up to 2 extra credit points dependingon the quality of work.

Based on cognitive evaluation theory (Deci & Ryan,1985), extrinsic motivation should be manifest in thetask performance condition due to the instrumentalityof behavior. Key to this theory is that the more an indi-vidual perceives his or her behavior as being tied totask performance, the greater the degree of extrinsicand the lesser the degree of intrinsic motivations. Per


Group Creativity

the aforementioned theoretical discussions, the morean individual perceives his or her behavior as being in-dependent of task performance, the more his or her ob-ligation motivation is expected to be. Therefore, the re-ward construct includes one formative dummy itemthat contrasts the two conditions. In addition, Hypothe-ses 4b, 5b, and 6b require an indication of whether re-wards are perceived as being reasonable for the experi-mental task. Therefore the reward construct alsoincludes a formative item from the posttask question-naire (see Table 1) that asks the degree to which the re-ward is reasonable. Because the loadings of these itemsare significantly different for the motivation con-structs, one reward construct is employed for each mo-tivation construct.

Motivation. The motivation constructs are mea-sured using posttask questionnaire items (see Table 1).Extrinsic motivation items are in accord with Amabileet al. (1994). Items involving obligation motivation arein accord with Harrison’s (1995) direct approach as

well as Eisenberger et al. ’s (2001) focus on job perfor-mance. Items involving intrinsic motivation are inaccord with Amabile (1996), Ryan et al. (1983), andSansone, Sachau, and Weir (1989).

Creativity. A holistic approach, asking expertsto judge creativity, was taken to measure the writtenproducts (Amabile, 1996). As described later, expertsalso judged novelty and usefulness to help evaluate thevalidity of creativity. The experts were administrativepersonnel who understood student and faculty issuesand were very familiar with the administrative infor-mation technology currently in use.

In accord with Massetti (1996) and Amabile (1996),the following judging procedure was employed. (a) Toreduce distractions, experts were provided with com-monly formatted word-processed copies of team re-ports with spelling and grammatical errors corrected.(b) Reports were in a different random order for eachexpert to eliminate order effects. (c) Experts were allprofessionals within the University administration, and



Table 1. Posttask Questionnaire Items by Construct

Reference Extrinsic Motivation

Amabile et al., 1994 1. The opportunity to receive a reward for this activity played a significant role in directingmy efforts.

Amabile et al., 1994 2. The opportunity to receive extra credit for this activity played a significant role indirecting my efforts.

Amabile et al., 1994 3. During the group activity, I was motivated by the opportunity to receive a reward.

Obligation Motivation

Eisenberger et al., 2001 1. I felt that it was important to do a good job because of the extra credit I was guaranteed toreceive.

Harrison, 1995 2. The extra credit I was guaranteed to receive made me feel obligated to perform well.Eisenberger et al., 2001 3. I felt compelled to perform well because of the extra credit I was guaranteed to receive.

Intrinsic Motivation

Keys Instrument–Amabile et al., 1996 1. During the group activity, I felt considerable pressure to meet someone else’sspecifications on how to do our work. (reverse coded to represent autonomy).

Ryan et al., 1983; Sansone et al., 1989 2. My own interest motivated my work during the group activity.Sansone et al., 1989 3. The group activity seemed like fun.

Reward Reasonableness

The reward offered is reasonable for this activity.

Guaranteed Reward

I am guaranteed to receive the following amount of extra credit: 0 points, 1 point, 2 points, 3points, 4 points, 5 points.

Note. All items except guaranteed reward were responded to on a 7-point Likert scale, from 1 (completely disagree) to 7 (completely agree).

were chosen due to their familiarity with the problemsand potential solutions facing the students. (d) Expertswere asked to work independently from each other andto evaluate each report in terms of creativity, novelty,and usefulness on 7-point scales. In accord with thedefinition of creativity, experts were asked to use theirown subjective notions of creativity. They were askedto evaluate novelty in terms of the degree to which theybelieve that the team’s solutions are new and unusual.They were asked to evaluate usefulness in terms of thedegree to which they believe that the team’s solutionswill help solve the associated problems. (e) Expertswere asked to make their evaluations relative to all theother teams, rather than on some absolute standard. Tofacilitate this relative evaluation, they were asked toread all solutions for at least five teams (approximately20%) before judging any single team.

Control. The model subjected to partial leastsquares (PLS) analysis has one construct to control fordifferences across the three semesters that the experi-ment occurred. This construct has paths to all otherconstructs and has two dummy items that represent thethree semesters.

Results

Descriptive statistics are provided in Tables 2a and2b. In addition, three analyses of covariance (with rea-sonableness as a covariate) were run, examining the in-

fluence of reward on motivations. Reward was found tohave a marginally significant effect (p = .07) on extrin-sic motivation and nonsignificant effects on obligationand intrinsic motivations. These results are further ex-plored later.

The researchmodel inFigure2 isanalyzedusingPar-tial Least Squares (PLS—version 3.00 build 1017), amultivariate analysis technique for testing structuralmodels that allows the simultaneous analysis of multi-ple criteria and predictor constructs (Barclay et al.,1995; Wold, 1985). PLS is especially good when thereare not many data points; it can be used when there are asfew as five for each path leading to the construct that hasthe most incoming paths or the formative construct withthe most indicators, whichever is greater (Falk & Miller,1992). In PLS, construct indicators may be modeled asreflective or formative (Fornell & Bookstein, 1982),with reflective indicators determined by the constructthey represent and formative indicators determining theconstruct they represent (Chin & Gopal, 1995). Indica-tors for the extrinsic and obligation motivation con-structs in Figure 2 are reflective; those for the creativity,intrinsic motivation, and reward constructs are forma-tive. Many of the reliability and validity discussions be-low are supported by PLS analyses.

Construct Reliability and Validity

Reward. To determine if participants were ableto distinguish between the reward conditions, a corre-lation was run relating the reward condition dummy


Group Creativity

Table 2a. Descriptive Statistics for Performance-Independent Reward Condition

N Minimum Maximum M SD

Extrinsic motivation 24 7 19 12.8 3Obligation motivation 24 10.7 48.9 25.5 10.2Intrinsic motivation 24 11 17.7 14.3 1.7Reasonableness 24 4.3 7 5.6 0.73Creativity 24 –1.4 1.0 0.1 0.6

Table 2b. Descriptive Statistics for Performance Reward Condition

N Minimum Maximum M SD

Extrinsic motivation 22 6.7 18.3 14.3 2.9Obligation motivation 22 10.5 48.9 29 11.2Intrinsic motivation 22 10.7 17.7 14.7 2.0Reasonableness 22 3.3 7 5.5 0.85Creativity 22 –1.1 1.4 0.1 0.6

item to a posttask question asking how many extracredit points participants were guaranteed to receive;the result is significant (r = .70, p = .01). This sug-gests that participants appropriately perceived signifi-cant differences between the reward conditions,thereby supporting validity of the reward construct.

Motivation. PLS enables the assessment of mea-surement components by providing principal compo-nents factor loadings of items. The factor loadings pro-vided by the PLS analysis and presented in Table 3indicate adequate reliability for the motivation con-structs. The loadings are all greater than 0.6, demon-strating that less than half of any indicator’s variancewas due to error (Bagozzi & Youjae, 1988). In addition,composite scale reliabilities (an internal consistency es-timate similar to Cronbach’s α) for the motivation con-structsexceed therecommendedcutoffof0.7 (Fornell&Larcker, 1981). Finally, average variances extracted bymotivation constructs from their items exceed the rec-ommended cutoff of 0.5 (Fornell & Larcker, 1981).

The items of motivation constructs also demon-strate convergent and discriminant validity accordingto a criterion similar to a multitrait/multimethod anal-ysis (Carmines & Zeller, 1979). Specifically, Table 4illustrates that the constructs share more variancewith their items than with the other constructs.

Creativity. Recent research has demonstratedthat acceptably reliable assessments can be obtained

with as few as three experts for verbal and problem solv-ing tasks (Amabile, 1996). Eight experts were availableto us each semester. Because some experts were differ-ent in different semesters, their responses are standard-ized (Chin, Marcolin, & Newsted, 1996) and combinedsuch that each team had one measure for creativity, onefor novelty, and one for usefulness. Based on IntraclassCorrelation Coefficients (McGraw & Wong, 1996), theinterexpert reliability for the first semester is .86 for cre-ativity, .77 for usefulness, and .76 for novelty; that forthe second semester is .81 for creativity, .80 for useful-ness, and .81 for novelty; and that for the third semesteris .77 for creativity, .81 for usefulness, and .82 for nov-elty.Thesedatademonstrateverygoodreliability for thesubjective judgment measures.

According to Amabile (1996), interexpert reliabil-ity (demonstrated earlier) is also equivalent to con-struct validity because creativity is defined as thatwhich is agreed to among experts to be creative. In ad-dition, novelty and usefulness (the two components ofcreativity: Amabile, 1996) are significantly related tocreativity, accounting for 87% of its variance. This fur-ther supports creativity validity.

Analysis

Figure 3 depicts PLS analysis of the researchmodel. PLS generates estimates of standardized re-gression coefficients for the paths in a model’s struc-tural component. To determine the significance of



Table 3. PLS Construct Indicators, Factor Loadings, Average Variances Extracted, and Composite Scale Reliabilities forReflective Constructs

Construct IndicatorsRange of

Factor LoadingsAverage Variance

ExtractedComposite Scale

Reliability

Extrinsic motivation Questionnaire itemsa 0.92 to 0.96 0.88 0.96Obligation motivation Questionnaire itemsa 0.91 to 0.93 0.84 0.94Intrinsic motivation Questionnaire itemsa Not applicable for formative indicators

aSee Table 1 for questionnaire items.

Table 4. Average Variance Extracted by PLS Constructs (Diagonal Elements) and Shared Variance Between Constructs(Off-Diagonal Elements)

Extrinsic Motivation Obligation Motivation Intrinsic Motivation

Extrinsic motivation 0.88Obligation motivation 0.77 0.84Intrinsic motivation 0.25 0.31 Not applicable

these paths, jackknifed standard error estimates for thepaths are obtained using the blindfolding procedurewith an omission distance of 11 (Sambamurthy &Chin, 1994). Figure 3 also shows R-square values inparentheses for motivations and creativity, which is theproportion of variance of a construct explained by con-structs having paths leading to it. This analysis leads tothe following hypothesis evaluations.

Effects on creativity were significant, with the ef-fects of extrinsic and intrinsic motivations supportinghypotheses (H1 and H3). However, obligation motiva-tion increased creativity in contrast with expectations(H2). The effect of reward on extrinsic motivation wassignificant (H4a), whereas rewards did not significantlyinfluence obligation or intrinsic motivations (H5a andH6a). In contrast, there were consistently strong effectson all motivations due to reward reasonableness (H4b,H5b, and H6b).

Discussion

The results provide substantial support for the re-search model. However, three findings were contraryto expectations: Rewards did not significantly affectintrinsic and obligation motivations and obligationmotivation affected creativity in a manner more similar

to intrinsic rather than extrinsic motivation. These is-sues are addressed next.

Rewards and Motivation

Due to constraints set by instructors of the studentparticipants, both reward conditions included 1 extracredit point. The performance condition offered theopportunity to receive up to 2 more extra credit pointsdepending on report quality. The performance-inde-pendent condition gave 2 more extra credit pointsmerely for being present during the task. Comparedto the performance-independent condition, the perfor-mance condition significantly increased extrinsic mo-tivation. However, the contrast between these twoconditions did not significantly affect obligation orintrinsic motivations. Because both conditions offeredextrinsic rewards, the lack of effect on intrinsic moti-vation may not be that surprising. The lack of effecton obligation motivation may be due to perceptionsthat, in terms of engendering obligation, the differ-ence between being guaranteed 1 or 3 extra creditpoints was not significant.

Because both reward conditions included somekind of reward, it appears that variations in perceptionsof reward reasonableness, rather than the reward con-dition, had more influence on obligation motivation. In


Group Creativity

Figure 3. PLS results. All paths are significant (two-tailed) at p = .01 except those noted as “ns” and that with a single asterisk. The singleasterisked path is significant at p = .05; “ns” indicates not significant. § Item weights. ** Direction contrary to hypothesis.

general, Abbey and Dickson (1983) and Amabile(1996) have noted a direct impact of reward reason-ableness on motivation. The potential for such effectsin this study can be seen in a survey (administered tosimilar students who did not participate in the experi-ment) that was used to determine the levels of extracredit rewards. Over 20% of the respondents indicatedno difference between the level of extra credit thatwould entice them to participate and that which wouldmake them work hard. If the rewards were enough tohave them sign up for the experiment, additional com-pensation would presumably not affect their degree ofobligation motivation. This would suggest that formany participants there may be a threshold, which,when met, would get them involved, and greater com-pensation (within the limits of this study’s extra creditrewards) may not have increased effects on obligationmotivation. For such individuals, perceptions of re-ward reasonableness—indicating whether the rewardmet the threshold—could have more of an impact onobligation than the reward treatment itself.

Obligation Motivation and Creativity

Contrary to expectations, obligation motivation hadeffects on creativity opposite that of extrinsic motiva-tion and in accord with intrinsic motivation. The nega-tive effects of extrinsic motivation on creativity comelargely from individuals’ focus on behaviors that aremore likely to gain the extrinsic rewards; these behav-iors typically do not include extended explorations ofthe problem space, but rather focus on the technical orrule-bound aspects of task performance (Amabile,1988; Hogarth, 1987; Woodman et al., 1993). In con-trast, the positive effects of intrinsic motivation on cre-ativity come largely from individuals’ behaviors thatincrease their understanding of the problem itself. Thiscan occur when individuals have an interest in theproblem for its own sake (e.g., due to the potential forprofessional growth or because it is pleasurable) andhave the necessary autonomy to enable extensive prob-lem space exploration (Amabile, 1988; Hogarth, 1987;Woodman et al., 1993).

Implications for behavior due to obligation motiva-tion can be more ambiguous than that associated withextrinsic motivation. In the case of this experiment, ob-ligation came from the receipt of extra credit that wasguaranteed, independent of task performance. The ap-propriate reciprocation of this favor was not described.

If participants believed that participation in the taskwould be reasonable reciprocation, then they wouldfeel freer to explore the problem space than would ex-trinsically motivated participants. In addition, becausethe act of participating is an act of repayment, partici-pation could reduce the aversive psychological tensioncaused by obligation, thereby increasing pleasure asso-ciated with the task itself. Therefore, the type of obli-gation motivation provided here could result in a de-gree of pleasure and autonomy more associated withintrinsic rather than extrinsic motivation. Though thisrationale is clearly speculative, it is supported by theanalysis illustrated in Figure 4: Intrinsic motivation isdecreased by extrinsic and increased by obligation mo-tivation.

Conclusion

The purpose of this article was to examine the de-gree to which obligation motivation differs from ex-trinsic and intrinsic motivations in terms of their ef-fects on group creativity. Results of the laboratoryexperiment support the existence of obligation moti-vation and indicate that it can have effects on creativ-ity that are more in line with intrinsic rather than ex-trinsic motivation. In addition, the experiment



Figure 4. Effects on intrinsic motivation. Paths are significant(two-tailed) at p = .01.

demonstrated that reward reasonableness can be avery important factor influencing the degree of allthree motivations.

Several potentially fruitful areas of research aresuggested by the experiment. (a) Though the creativityof interacting groups was examined, this examinationwas at a rudimentary level because simple aggrega-tions of individual-level responses were employed torepresent group motivation. Though this is a reason-able first step, future research should attempt to modelmotivation-oriented group dynamics, including no-tions of cohesion, norms, and leadership (Hackman &Morris, 1975; Shamir, 1990; Woodman et al., 1993).(b) The potential for motivations to affect group cre-ativity suggests the potential for examining relation-ships among the three motivations and group charac-teristics and processes that lead to creativity. (c) Thesurprisingly strong effects of reward reasonableness onthe motivations suggest research questions involvingwhat affects perceptions of reward reasonableness andwhy such perceptions affect motivation.

References

Abbey, A., & Dickson, J. W. (1983). R&D work climate and innova-tion in semiconductors. Academy of Management Journal, 26,362–368.

Amabile, T. M. (1988). A model of creativity and innovation in orga-nizations. In B. Staw & L. L. Cummings (Eds.), Research in or-ganizational behavior (Vol. 10, pp. 123–167). JAI.

Amabile, T. M. (1990). Within you, without you: The social psychol-ogy of creativity and beyond. In M. A. Runco & R. S. Albert(Eds.), Theories of creativity (pp. 61–91). Newbury Park, CA:Sage.

Amabile, T. M. (1996). Creativity in context: Update to the socialpsychology of creativity. Boulder, CO: Westview.

Amabile, T. M., Conti, R., Coon, H., Lazenby, J., & Herron, M.(1996). Assessing the work environment for creativity. Acad-emy of Management Journal, 39, 1154–1184.

Amabile, T. M., Hennessey, B. A., & Grossman, B. S. (1986). Socialinfluences on creativity: The effects of contracted-for reward.Journal of Personality and Social Psychology, 50, 14–23.

Amabile, T. M., Hill, K. G., Hennessey, B. A., & Tighe, E. M.(1994). The work preference inventory: Assessing intrinsic andextrinsic motivational orientations. Journal of Personality andSocial Psychology, 66, 950–967.

Atkinson, J. W. (1957). Motivational determinants of risk-taking be-havior. Psychological Review, 64, 369–372.

Bagozzi, R. P., & Youjae, Y. (1988). On the evaluation of structuralequation models. Academy of Marketing Science, 16, 74–94.

Bandura, A. (1986). Social foundations of thought and action: A so-cial cognitive theory. Englewood Cliffs, NJ: Prentice Hall.

Barclay, D., Higgins, C., & Thompson, R. (1995). The partial leastsquares (PLS) approach to causal modeling: Personal computeradoption and use as an illustration. Technology Studies, 2,285–309.

Baron, R. S., Kerr, N. L., & Miller, N. (1992). Group process, groupdecision, group action. Pacific Grove, CA: Brooks/Cole.

Barrick, M. R., Stewart, G. L., Neubert, M. J., & Mount, M. K.(1998). Relating member ability and personality to work-teamprocesses and team effectiveness. Journal of Applied Psychol-ogy, 83, 377–391.

Barron, F. B., & Harrington, D. M. (1981). Creativity, intelligence,and personality, Annual Review of Psychology, 32, 439–476.

Bem, D. (1972). Self-perception theory. In L. Berkowitz (Ed.), Ad-vances in experimental social psychology (Vol. 6, pp. 1–62).New York: Academic Press.

Boone, K. B., & Cummings, L. L. (1981). Cognitive evaluation the-ory: An experimental test of processes and outcomes. Organi-zational Behavior and Human Performance, 20, 289–310.

Bowditch, J. L., & Buono, A. F. (1990). A primer on organizationalbehavior (2nd ed.). New York: Wiley.

Brophy, D. R. (1998). Understanding, measuring, and enhancingcollective creative problem-solving efforts. Creativity ResearchJournal, 11, 199–229.

Carmines, E., & Zeller, R. (1979). Reliability and validity assess-ment. Sage University paper series on quantitive applicationsin the social sciences (Vol. 07-017). Beverly Hills, CA: Sage.

Chin, W., & Gopal, A. (1995). Adoption Intention in GSS: RelativeImportance of Beliefs. Database, 26(2&3), 42–63.

Chin, W., Marcolin, B. L., & Newsted, P. R. (1996). A partial leastsquares latent variable modeling approach for measuring inter-action effects: Results from a monte carlo simulation study andvoice mail emotion/adoption study. In J. DeGross, S. Jarvenpaa,& A. Srinivasan (Eds.), Proceedings of the Seventeenth Interna-tional Conference on Information Systems (pp. 21–41). Cleve-land, OH.

Cialdini, R. B. (1993). Influence: Science and practice (3rd ed.).Boston: Scott.

Collins, M. A., & Amabile, T. M. (1999). Motivation and Creativity.In R. J. Sternberg, Handbook of creativity (pp. 297–312). NewYork: Cambridge University Press.

Colquitt, J. A., Hollenbeck, J. R., Ilgen, D. R., LePine, J. A., &Sheppard, L. (2002). Computer-assisted communication andteam decision-making performance: The moderating effect ofopenness to experience. Journal of Applied Psychology, 87,402–410.

Cooper, R. B. (2000). Information technology development creativ-ity: A case study of attempted radical change. MIS Quarterly,24, 245–276.

Cooper, R. B., & Bhattacherjee, A. (2001). The effect of automaticresponses to authority on information technology use. DataBase, 32, 36–50.

Craig, T. Y., & Kelly, J. R. (1999). Group Cohesiveness and CreativePerformance. Group dynamics: Theory, research, and practice,3, 243–256.

Crocker, J., & Luhtanen, R. (1990). Collective self-esteem andingroup bias. Journal of Personality and Social Psychology, 58,60–67.


Group Creativity

Crutchfield, R. (1962). Conformity and creative thinking. In H.Gruber, G. Terrell, & M. Wertheimer (Eds.), Contemporary ap-proaches to creative thinking (pp. 120–140). New York: Ather-ton.

Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta-analytic re-view of experiments examining the effects of extrinsic rewardson intrinsic motivation. Psychological Bulletin, 125, 627–668.

Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-de-termination in human behavior. New York: Plenum.

Eisenberger, R., Armeli, S., Rexwinkel, B., Lynch, P. D., & Rhoades,L. (2001). Reciprocation of perceived organizational support.Journal of Applied Psychology, 86, 42–51.

Eisenberger, R., & Cameron, J. (1996). Detrimental effects of re-ward: Reality or myth? American Psychologist, 51, 1153–1166.

Eisenberger, R., & Shanock, L. (2003). Rewards, intrinsic motiva-tion, and creativity: A case study of conceptual and method-ological isolation. Creativity Research Journal, 15, 121–130.

Elam, J., & Mead, M. (1990). Can software influence creativity? In-formation Systems Research, 1, 1–22.

Falk, R. J. & Miller, N. B. (1992). A primer for soft modeling. Akron,OH: University of Akron Press.

Foote, D. (2001). IT professionals rake in bucks despite downturn.Computerworld, 35, 23.

Fornell, C., & Bookstein, F. (1982). Two structural equation models:LISREL and PLS applied to consumer exit-voice theory. Jour-nal of Marketing Research, 19, 440–452.

Fornell, C., & Larcker, D. (1981). Evaluating structural equationmodels with unobservable variables and measurement error.Journal of Marketing Research, 18, 39–50.

Goei, R., Massi, L., Lisa, L., Boster, F. J., Skalski, P. D., & Bowman,J. M. (2003). The mediating roles of liking and obligation on therelationship between favors and compliance. CommunicationResearch, 30(2), 178–197.

Gouldner, A. W. (1960). The norm of reciprocity: A preliminarystatement. American Sociological Review, 25, 161–178.

Guzzo, R. A., Yost, P. R., Campbell, R. J., & Shea, G. P. (1993). Po-tency in groups: Articulating a construct. British Journal of So-cial Psychology, 32, 87–106.

Hackman, R. J., & Morris, C. G. (1975). Group tasks, group interac-tion process, and group performance effectiveness: A reviewand proposed integration. In L. Berkowitz (Ed.), Advances inexperimental social psychology (Vol. 8 pp. 45–99). New York:Academic.

Harrison, D. A. (1995). Volunteer motivation and attendance deci-sions: Competitive theory testing in multiple samples from ahomeless shelter. Journal of Applied Psychology, 80, 371–385.

Harter, S. (1981). A new self-report scale of intrinsic versus extrinsicorientation in the classroom: Motivational and informationalcomponents. Developmental Psychology, 17, 300–312.

Hennessey, B. A. (2000). Rewards and creativity. In C. Sansone & J.M. Harackiewicz (Eds.), Intrinsic and extrinsic motivation:The search for optimal motivation and performance (pp.55–78). New York: Academic.

Hennessey, B. A., & Amabile, T. M. (1998). Reward, intrinsic moti-vation, and creativity. American Psychologist, 53, 674–675.

Higgins, E. T., & Kruglanski, A. W. (2000a). Motivational science:The nature and functions of wanting. In E. T. Higgins & A. W.

Kruglanski (Eds.), Motivational science: Social and personal-ity perspectives (pp. 1–20). Philadelphia: Psychology Press.

Higgins, E. T., & Kruglanski, A. W. (2000b). Bridging the gap be-tween knowing and doing. In E. T. Higgins & A. W. Kruglanski(Eds.), Motivational science: Social and personality perspec-tives (pp. 147–149). Philadelphia: Psychology.

Hinsz, B. B., Tindale, R. S., & Vollrath, D. A. (1997). The emergingconceptualization of groups as information processors. Psycho-logical Bulletin, 121, 43–64.

Hogarth, R. (1987). Judgement and choice (2nd ed.). New York:Wiley.

Jordan, P. C. (1986). Effects of an extrinsic reward on intrinsic moti-vation. Academy of Management Journal, 29, 405–412.

Karau, S. J., & Kelly, J. R. (1992). The effects of time scarcity andtime abundance on group performance and interaction. Journalof Experimental Social Psychology, 28, 542–571.

Kelly, J. R., & Karau, S. J. (1993). Entrainment of creativity in smallgroups. Small Group Research, 24, 179–198.

Kelly, J. R., & McGrath, J. E. (1985). Effects of time limits and tasktypes on task performance and interaction of four-persongroups. Journal of Personality and Social Psychology, 49,395–407.

Kerr, N., & Bruun, S. (1983). The dispensability of member effortand group motivation losses: Free-rider effects. Journal of Per-sonality and Social Psychology, 44, 78–94.

Latané, B., Williams, K., & Harkins, S. (1979). Many hands makelight the work: The causes and consequences of social loafing.Journal of Personality and Social Psychology, 37, 822–832.

Leakey,R.,&Lewin,R. (1978).Peopleof the lake.NewYork:Anchor.LePine, J., Hollenbeck, J. R., Ilgen, D. R., & Hedlund, J. (1997). Ef-

fects of individual differences on the performance of hierarchi-cal decision-making teams: Much more than g. Journal of Ap-plied Psychology, 82, 803–811.

Lepper, M. R. (1981). Intrinsic and extrinsic motivation in children:detrimimental effects of superfluous social controls. In W. A.Collins (Eds.), Aspects of the development of competence: TheMinnesota Symposia on Child Psychology (Vol 14, pp.155–214). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Lepper, M. R., Greene, D., & Nisbett, R. E. (1973). Underminingchildren’s intrinsic interest with extrinsic reward: A test of theoverjustification hypothesis. Journal of Personality and SocialPsychology, 28, 129–137.

Lepper, M. R., & Henderlong, J. (2000). Turning ‘play’ into ‘work’and ‘work’ into ‘play’: 25 years of research on intrinsic versusextrinsic motivation. In C. Sansone& J. M. Harackiewicz(Eds.), Intrinsic and extrinsic motivation: The search for opti-mal motivation and performance (pp. 257–307). New York:Academic.

Maslow, A. (1954). Motivation and personality. New York: Harper& Row.

Massetti, B. (1996). An empirical examination of the value of cre-ativity support systems on idea generation. MIS Quarterly, 20,83–97.

McGraw, K., & Wong, S. P. (1996). Forming inferences about someintraclass correlation coefficients. Psychological Methods,1(1), 30–46.

Mischel, W., & Shoda, Y. (1995). A cognitive-affective system the-ory of personality: Reconceptualizing situations, dispositions,



dynamics, and invariance in personality structure. Psychologi-cal Review, 102, 246–268.

Mumford, M. D. (2003). Taking stock in taking stock. Creativity Re-search Journal, 15, 147–151.

Nagasundaram, M., & Bostrom, R. P. (1995). Structuring of creativeprocesses using GSS. Journal of Management Information Sys-tem, 11, 87–114.

Naylor, J. C., Pritchard, R. D., & Ilgen, D. R. (1980). A theory of be-havior in organizations. New York: Academic.

Nemeth, C. J., & Kwan, J. L. (1985). Originality of word associa-tions as a function of majority versus minority influence. SocialPsychology Quarterly, 48, 277–282.

Newell, A., Shaw, J., & Simon, H. (1962). The processes of creativethinking. In H. Gruber, G. Terrell, & M. Wertheimer (Eds.),Contemporary approaches to creative thinking (pp. 63–119).New York: Atherton.

Newman, R. S. (1990). Children’s help-seeking in the classroom:The role of motivational factors and attitudes. Journal of Edu-cational Psychology, 82, 71–80.

Ocker, R., Hiltz, S. R., Turoff, M., & Fjermestad, J. (1995). The ef-fects of distributed group support and process structuring onsoftware requirements development teams: Results on creativ-ity and quality. Journal of Management Information Systems,12(3), 127–153.

Paulus, P. B. (2000). Groups, teams, and creativity: The creative po-tential of idea-generating groups. Applied Psychology, 49(2),237–263.

Ravichandran, T., & Rai, A. (2000). Quality management in systemsdevelopment: An organizational system perspective. MISQuarterly, 24(3), 381–415.

Redmond, M. R., Mumford, M. D., & Teach, R. (1993). Putting cre-ativity to work: Effects of leader behavior on subordinates cre-ativity. Organizational Behavior and Human Decision Pro-cesses, 55, 120–151.

Rhoades, L., Eisenberger, R., & Armeli, S. (2001). Affective commit-ment to the organization: The contribution of perceived organi-zationalsupport.JournalofAppliedPsychology,86,825–836.

Ryan, R. M., Mims, V., & Koestner, R. (1983). Relation of rewardcontingency and interpersonal context to intrinsic motivation:A review and test using cognitive evaluation theory. Journal ofPersonality and Social Psychology, 45, 736–750.

Sambamurthy, V., & Chin, W. (1994). The effects of group attitudestoward alternative GDSS designs on the decision-making per-formance of computer-supported groups. Decision Sciences,25, 215–241.

Sansone, C. A. (1986). Question of competence: The effects of com-petence and task feedback on intrinsic interest. Journal of Per-sonality and Social Psychology, 51, 918–931.

Sansone, C., Sachau, D. A., & Weir, C. (1989). Effects of instructionin intrinsic interest: The importance of context. Journal of Per-sonality and Social Psychology, 57, 819–829.

Satzinger, J. W., Garfield, M. J., & Nagasundaram, M. (1999). Thecreative process: The effects of group memory on individualidea generation. Journal of Management Information Systems,15(4), 143–160.

Shamir, B. (1990). Calculations, values and identities: The sourcesof collectivistic work motivation. Human Relations, 43,313–332.

Shein, E. (1998). Team spirit. PC Week, 15(19), 69.Siegel, G., & Ramanauskas-Marconi, H. (1989). Behavioral

acounting. Cincinnati, OH: South-Western.Sorrentino, R. M., & Higgins, E. T. (1986). Motivation and cogni-

tion: Warming up to synergism. In R. M. Sorrentino & E. T.Higgins (Eds.), Handbook of motivation and cognition: Foun-dations of social behavior (Vol. 1, pp. 3–19). New York:Guilford.

Sosik, J. J., Avolio, B. J., & Kahai, S. S. (1997). Effects of leadershipstyle and anonymity on group potency and effectiveness in agroup decision support system environment. Journal of AppliedPsychology, 82, 89–103.

Sosik, J. J., Kahai, S. S., & Avolio, B. J. (1998). Transformationalleadership and dimensions of creativity: Motivating idea gener-ation in computer-mediated groups. Creativity Research Jour-nal, 11, 111–122.

Staw, B. M. (1974). The attitudinal and behavior consequences ofchanging a major organization reward. Journal of Personalityand Social Psychology, 29,743–751.

Staw, B. M. (1977). Motivation in organizations: Toward synthesisand redirection. In B. M. Staw & G. R. Salancik (Eds.), New di-rections in organizational behavior (pp. 55–96). Chicago: St.Clair.

Staw, B. M. (1990). An evolutionary approach to creativity and inno-vation. In M. West & J. Farr (Eds.), Innovation and creativity atwork: Psychological and organizational strategies (pp.287–308). New York: Wiley.

Steiner, I. D. (1972). Group process and productivity. New York: Ac-ademic.

Sweeny, J. (1973). An experimental investigation of the free-riderproblem. Social Science Research, 2, 277–292.

Szymanski, K., & Harkins, S. (1992). Self-evaluation and creativity.Personality and Social Psychology Bulletin, 18, 259–265.

Thompson, L., & Brajkovich, L. R. (2003). Improving the creativityof organizational work groups. Academy of Management Exec-utive, 17(1), 96–112.

Thornburg, T. (1991). Group size and member diversity influence oncreative performance. Journal of Creative Behavior, 25,324–333.

Tiger, L., & Fox, R. (1971). The imperial animal. New York: Holt.Torrance, E. P. (1965). Rewarding creative behavior: Experiments in

classroom creativity. New Jersey: Prentice-Hall.Weingart, L. R. (1992). Impact of group goals, task component com-

plexity, effort, and planning on group performance. Journal ofApplied Psychology, 77, 682–693.

West, M. (2002). Sparkling fountains or stagnant ponds: An integra-tive model of creativity and innovation implementation in workgroups. Applied Psychology: An International Review, 51,355–424.

Whatley, M. A., Webster, J. M., Smith, R. H., & Rhodes, A. (1999).The effect of a favor on public and private compliance: How in-ternalized is the norm of reciprocity? Basic and Applied SocialPsychology, 21, 251–259.

Williams, W. M., & Yang, L. T. (1999). Organizational creativity. InR. J. Sternberg (Ed.), Handbook of creativity (pp. 373–391).New York: Cambridge University Press.

Wilson, T. Quoted in (1999) How Rosy is IT’s Future? InfoWorld,21(41), 99.


Group Creativity

Wold, H. (1985). Systems analysis by partial least squares. In P.Nijkamp, H. Leitner, & N. Wrigley (Eds.), Measuring the un-measurable. Dordrecht, The Netherlands: Martinus Nijhoff.

Woodman, R. W., Sawyer, J. E., & Griffin, R. W. (1993). Toward atheory of organizational creativity. The Academy of Manage-ment Review, 18, 293–321.

Zajonc, R. B. (1980). Feeling and thinking: Preferences need no in-ferences. American Psychologist, 35, 151–175.

Zander, A. W., & Medow, H. (1963). Individual and group levels ofaspiration. Human Relations, 16, 89–105.

Zingheim, P. Quoted in (1999) How Rosy is IT’s Future? InfoWorld,21(41), 99.

Appendix A.1: General Instructions

Students are faced with problems when attemptingto register and take classes at the University of Hous-ton. These problems may be different depending onwhether a student is also working at a job. For example,nonworking students may have trouble getting classesbecause the ones they want are not being offered.Working students may have trouble getting classes be-cause the ones they want are being offered at the timesthey have to go to work. In addition, faculty membershave problems with classes they are assigned. For ex-ample, they may be assigned classes for which they donot understand the material.

You will be assigned to a group; each group willconsist of 3 students. One of you will represent facultyand will be provided with a list of faculty problems thatneed to be solved. One will represent working students,and will be provided with a list of working-studentproblems that need to be solved. One will representnonworking students, and will be provided with a listof nonworking-student problems that need to besolved. The task for your group is to develop Informa-tion System Specifications to help solve these prob-lems. Please do not attempt to solve problems otherthan the ones you are given.

Due to budget constraints, additional faculty, TAs,classrooms, etc. cannot be provided. However, the Uni-versity administration has suggested that an Informa-tion System might help that can perform activities suchas scheduling classes, assigning classrooms, allocatingequipment and material, etc. Your group’s task is toprovide the administration with a list of InformationSystem functions (what the Information System shoulddo) to help solve the student and faculty problems. Re-member, your solutions MUST BE information systemsolutions. Therefore, for example, having the Univer-sity provide more classes is NOT an appropriate solu-

tion because it means that additional faculty must behired. HIRING MORE FACULTY IS NOT ANINFORMATION SYSTEMS SOLUTION.

Deliverables

Please hand in the following in the ’Group Report’forms provided to you.

1. Information Systems Functions: Describe whatthe Information System should do. Provideenough detail to resolve potential conflictsamong nonworking students, working stu-dents, and faculty.

2. Problems: Specifically relate how the Informa-tion System functions help solve the problemsof nonworking students, working students, andfaculty members that were provided to you inthe individual instruction sheets.

3. Prioritization: Due to budget constraints, theUniversity administration may not be able toimplement all of your described functions.Therefore, prioritize the Information Systemfunctions in order of their importance.

You will have 1 hour to complete your answers.Please be sure that your answers are appropriate, havethe right level of detail, and are complete.

Please do not discuss any part of this exercise withany other students until we announce in class that theexercise is over. (Thanks!!)

Note: The Third Deliverable (Prioritization) wasnot examined in this article.

Appendix A.2: Example of the TASKINDEPENDENT Instruction Sheet

WORKING STUDENTS

Important: Though you may discuss these instruc-tions, please do not show these instructions to othermembers of your group

In class, it was announced that you could earnfrom 1 to 3 extra credit points. HOWEVER,Each member of your group WILL receive 3 ex-tra credit points. Your receipt of these 3 points isguaranteed (you will get the 3 points no matter



what), and does NOT depend on the quality ofwork that you do.

Students who are attending the university and areworking at a full-time job are faced with problemswhen attempting to register and take classes at the cam-pus. You represent these working students in the groupactivity you are about to perform. Other members ofYOUR group represent faculty members and part-timestudents.

The following lists the problems faced by the work-ing students you represent:

1. Classes desired by working students are notavailable because they are already closed

2. Classes desired by working students are notavailable because they are not offered

3. Classes desired by working students are in con-flict with each other since they are offered atthe same day and time

4. After registering for a class, working studentsare dropped from the class because it is cancel-led by administration, e.g. due to too few stu-dents taking the class

5. Classes desired by working students are offeredat inconvenient days and/or times.

Appendix BJudges’ Instruction Sheet

Instructions for Group Evaluations

BACKGROUND:

Undergraduate Seniors in the College of BusinessAdministration were put into groups of three studentseach, and asked to provide information systems solu-tions to help solve one or more of the following 11problems.

1. Classes desired by working/nonworking stu-dents are not available because they are alreadyclosed.

2. Classes desired by working/nonworking stu-dents are not available because they are not of-fered.

3. Classes desired by working/nonworking stu-dents are in conflict with each other since theyare offered at the same day and time.

4. After registering for a class, work-ing/nonworking students are dropped from theclass because it is cancelled by administration,e.g. due to too few students taking the class.

5. Classes desired by working/nonworking stu-dents are offered at inconvenient days and/ortimes.

6. Assigned classes are scheduled for inconve-nient days and/or times for faculty members.

7. Assigned classes will require too much workon the part of faculty members.

8. Assigned classes do not interest faculty mem-bers.

9. Assigned classes are not appropriate becausethe faculty member does not understand thematerial.

10. Assigned classes result in an overload of workto the faculty member due to large class sizes.

11. Required equipment and material for theclasses assigned to faculty are not available.

Each group has proposed solutions to one or moreof the above problems. Groups have indicated whichproblem or problems each solution is proposed to helpsolve. We would like you to evaluate their solutions interms of Creativity, Usefulness, and Novelty. Pleaseuse your own subjective definition of Creativity. Use-fulness should be evaluated in terms of the degree towhich you believe that their solutions will help solvethe associated problems. Novelty should be evaluatedin terms of the degree to which you believe that theirsolutions are new and unusual.

Each solution has the following format:

This IS function should:

This reduces the following problem

for

by doing, performing, etc. the following:

Most groups have included a number of solutions inthe above format. For each group, please evaluate thegroup’s overall Creativity, Usefulness, and Novelty,considering all of their solutions together. When doing


Group Creativity

this evaluation, please judge each group’s Creativity,Usefulness, and Novelty relative to all the othergroups.

PROCEDURE

1. To get an initial feel for relative Creativity,Usefulness, and Novelty, please read all of thesolutions for at least the first 5 groups before

doing any evaluations. Then, go back to Group1, and evaluate all groups in terms of Creativity,Usefulness, and Novelty

2. Please do your evaluations in Group NumberOrder (evaluate Group 1, then Group 2, etc.).

3. Please do NOT talk to your colleagues aboutthis exercise until after all of the evaluationshave been returned.



group creativity: the effects of extrinsic, intrinsic, and...

Documents