An examination of the relationship between conscientiousness and group performance on a creative task

Journal of Research in Personality 44 (2010) 222–231

Contents lists available at ScienceDirect

Journal of Research in Personality journal homepage: www.elsevier.com/locate/jrp

An examination of the relationship between conscientiousness and group performance on a creative task Christopher Robert a,*, Yu Ha Cheung b a

Dept. of Management, Robert J. Trulaske, Jr. College of Business, University of Missouri-Columbia, 329 Cornell Hall, Columbia, MO 65211, United States Dept. of Management, School of Business, Hong Kong Baptist University, The Wing Lung Bank Building for Business Studies, 34 Renfrew Road, Kowloon Tong, Kowloon, Hong Kong, China b

a r t i c l e

i n f o

Article history: Available online 28 January 2010 Keywords: Creativity Conscientiousness Groups Performance Personality

a b s t r a c t The positive relationship between conscientiousness and job performance appears to be robust for individuals and groups across most contexts. This research examined the possibility that this positive relationship might reverse in groups engaged in a creative task. In Study 1 (N = 55 four-person groups), a significant negative relationship was observed between group conscientiousness and group performance. To test our hypothesis that this effect is driven by the activation of maladaptive or adaptive behaviors for high vs. low conscientiousness groups (respectively), in Study 2 (N = 59 four-person groups), we manipulated task instructions to make the approach to the task appear more ‘‘flexible” or ‘‘structured.” Task instructions interacted with group conscientiousness, and provide support for this theoretical explanation. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction Over the last two decades, research has shown conscientiousness to be a consistent predictor of individual performance across jobs (e.g., Barrick & Mount, 1991; Hurtz & Donovan, 2000; Salgado, 1997). Conscientious people are high on achievement striving, dependability, volition, competence, and self-discipline (Barrick & Mount, 1991; Waung & Brice, 1998), and are organized, taskfocused, and persistent (Costa & McCrae, 1992; Goldberg, 1992; Mount & Barrick, 1995). Explanations for the positive conscientiousness–performance relationship tend to highlight the intuitive relationships between these specific traits and adaptive workplace behaviors. Indeed, Barrick, Mount, and Judge (2001) note ‘‘it is hard to conceive of a job where it is beneficial to be careless, irresponsible, lazy, impulsive and low in achievement striving” (p. 11). The extension of this robust relationship to group performance contexts appears straightforward. It has long been assumed that the performance of a group is a function of the potential productivity of individual group members, defined as the sum of initial resources individuals bring to the group task (e.g., Steiner, 1972). Because conscientious individuals bring resources such as effort and persistence to the group’s task, it is perhaps not surprising that many studies in the literature on groups (e.g., Barrick, Stewart, Neubert, & Mount, 1998; Neuman, Wagner, & Christiansen, 1999; Neuman & Wright, 1999), as well as a recent meta-analysis (Bell, * Corresponding author. Fax: +1 573 884 6857. E-mail addresses: [email protected] (C. Robert), [email protected] (Y.H. Cheung). 0092-6566/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jrp.2010.01.005

2007) have concluded that group conscientiousness is also positively related to group performance. Indeed, on the basis of her meta-analytic results, Bell (2007) recommended that practitioners consider composing groups with highly conscientious individuals to obtain high performance. In this research, however, we test the boundaries of this recommendation. In particular, we examine the hypothesis suggested by Driskell, Hogan, and Salas (1987) that group conscientiousness will be negatively associated with performance on creative tasks. To accomplish this, we designed Study 1 to confirm that such a negative relationship could be observed, and to explore the hypothesis that high conscientiousness groups perform worse on creative tasks because they are less likely to engage in idea and information sharing, and because they utilize an excessively structured approach to the task. In Study 2, we designed an experiment to examine an alternative hypothesis that highly conscientious groups might do worse on creative tasks because such task contexts fail to activate conscientiousness-relevant traits. 1.1. The potential for negative conscientiousness–performance relationships Despite the many studies and meta-analyses that support the existence of a robust positive conscientiousness–performance relationship, negative relationships between conscientiousness (or its facets) and performance have been observed at the individual and group levels. Tett and Burnett (2003) cite 11 published or unpublished studies in which significant and meaningful negative relationships between conscientiousness and individual perfor-

C. Robert, Y.H. Cheung / Journal of Research in Personality 44 (2010) 222–231

mance are reported. Included in that list is a study by Robertson, Gibbons, Baron, MacIver, and Nyfield (1999), who found that conscientiousness was negatively related to a performance factor that included creativity and innovation, personal motivation, and an action orientation. At the group level, LePine (2003) and LePine, Colquitt, and Erez (2000) found that groups that were high on the dependability facet of conscientiousness performed worse than low dependability groups after a mid-task change in group members’ ability to communicate with each other. In addition, Waung and Brice (1998) reported that groups composed of highly conscientious individuals who caucused before performing a novel task performed worse than low conscientiousness groups and high conscientiousness groups with no opportunity to caucus. These results are suggestive of possible negative effects of conscientiousness on group performance, although the literature offers little in the way of theoretical guidance. However, one common theme is that high conscientiousness individuals and groups will employ procedures that are inappropriate for the task at hand. For example, LePine (2003) suggested that high dependability groups performed poorly in his study because they were ‘‘less willing to abandon old patterns of behavior, even after it is apparent that the behavior is inappropriate” (p. 31). Similarly, in explaining the poor performance of high conscientiousness groups in their study, Waung and Brice (1998) suggested that highly conscientious groups with an opportunity to caucus adopted performance norms that included systematic evaluation of group members’ contributions to the group, even though a less structured approach to idea generation would have been more appropriate. Such arguments suggest that individuals and groups that are high in conscientiousness prefer to adopt systematic procedures for task completion, and become locked into their use even though such procedures are not always appropriate for the task. An important elaboration on this basic idea is that task contexts in which creativity is a key feature might be problematic for high conscientiousness groups and individuals. Driskell et al. (1987) were perhaps the first to hypothesize that because conscientious people ‘‘perform well on tasks requiring routine, systematic, or rule-guided performance” (p. 105) high conscientiousness groups will perform poorly on tasks requiring creativity. Similarly, George and Zhou (2001) and Tett (1998) suggest that highly conscientious people have tendencies toward conformity and control that are incompatible with performance on creative tasks, because creativity requires that people embrace flexibility, ignore rules, and reject the status quo. These assertions are consistent with Barry and Stewart’s (1997) suggestion that group conscientiousness might be more clearly linked to performance on ‘‘planning” tasks (McGrath, 1984), for which it is easy to identify clear procedures for completion. The literature on creativity highlights the incompatibility between systematic and structured approaches to creative tasks and creative output. For example, Taggar (2002) maintains that creative tasks require a more nimble thought process, the ability to take new perspectives, and flexibility in the approach to problem solving ‘‘carried out via discovery rather than via a predetermined step-by-step procedure” (p. 315). Similarly, Simonton (2003) notes that creative outcomes are more likely to occur when a certain amount of chance, randomness, or unpredictability are infused in the process. Low conscientiousness individuals and groups might adopt less planful and systematic approaches to creative tasks, which, ironically, might result in better task performance. In combination, the tendency for the highly conscientious to apply systematic task performance processes across tasks, and the incompatibility of these task processes with the requirements of creative tasks, would suggest that high conscientiousness will result in low performance on creative tasks because of an excessive focus on applying systematic procedures to task completion.

223

1.2. Conscientiousness, groups, and creative task performance Although the incompatibility between conscientiousness and creative tasks is likely relevant at both the individual and group levels, we believe that it will be particularly pronounced when groups are asked to complete creative tasks. Information sharing and communication are thought to be key elements in group performance (Bunderson & Sutcliffe, 2002). Group performance on creative tasks requires that group members adopt processes that facilitate cross-fertilization, whereby group members’ contributions are combined and elaborated upon in unique ways (McGrath, 1984). Idea sharing has been shown to be an important factor in enhancing creativity and innovation in groups (Paulus & Yang, 2000). It tends to stimulate group members’ generation of more ideas (Dugosh, Paulus, Roland, & Yang, 2000) and results in the synergistic effects for which groups are touted. In other words, groups that fail to engage in idea and information sharing on tasks requiring creative solutions are likely to exhibit poor performance relative to groups that share ideas and information. When a group is comprised of high conscientiousness individuals, that group might engage in less idea and information sharing for two reasons. First, idea expression in creative group contexts involves a degree of risk (e.g., criticism from others) that might be inconsistent with the ‘‘cautiousness” and ‘‘conformity” characteristics of conscientiousness (Tegano & Moran, 1989). Waung and Brice (1998) found that members of high conscientiousness groups who had an opportunity to develop performance norms were more likely to inhibit the expression of ideas during a creative task requiring groups to generate ideas for novel uses of objects. The authors suggested that individuals feared their ideas were either too impractical, or too similar to ideas that were already expressed, and therefore feared being criticized by other group members. Second, although people who are high on conscientiousness are believed to be able to stay on task, and to march steadily toward task completion, the unstructured nature of idea and information sharing might be perceived as counterproductive. Rather, high conscientiousness group members might have a preference for directing their effort toward activities that are clearly ‘‘on-task” (e.g., writing down or elaborating on ideas that have already been proposed) and regulating themselves from engaging in ‘‘off-task” activities (Kanfer, Ackerman, Murtha, Dugdale, & Nelson, 1994). Thus, we predict that high conscientiousness groups might not actively participate in idea and information sharing processes, which are thought to lead to good performance in creative settings. In summary, we predict a negative relationship between group conscientiousness and performance on a creative task, and that this relationship will be mediated by a tendency to adopt a more systematic and planful approach to creative tasks, coupled with a tendency to engage in less idea and information sharing within the group. Formally: Hypothesis 1. Group conscientiousness will be negatively associated with group performance on a creative task. Hypothesis 2a. Idea and information sharing will mediate the relationship between group conscientiousness and group performance on a creative task, such that group conscientiousness will be negatively associated with idea and information sharing, which, in turn, will be positively associated with performance. Hypothesis 2b. Use of systematic task processes will mediate the relationship between group conscientiousness and group performance on a creative task, such that conscientiousness will be positively associated with systematic task processes, which, in turn, will be negatively associated with group performance.

224

C. Robert, Y.H. Cheung / Journal of Research in Personality 44 (2010) 222–231

2. Study 1 2.1. Method 2.1.1. Participants Participants were 229 undergraduate students at a large midwestern university enrolled in an introductory psychology course and participating for course credit. Participants (138 females, 89 males, 2 unreported) were randomly assigned to 1 of 3 three-person groups, or 1 of 55 four-person groups. The data from the 55 four-participant groups were used for hypothesis testing, and data from the three three-person groups were used to help train raters to rate task performance. The average age of the participants was 18.5 years (SD = 1.23).

2.1.2. Procedures Participants completed a pre-task survey assessing conscientiousness, openness to experience, extraversion, and GPA before they were randomly assigned to groups. Each group was asked to develop a ‘‘marketing plan” for five products that were relatively new and/or unique (e.g., a colored pen with a spongy grip, a sleeve of titanium women’s golf balls with a breast cancer research logo). Students were primarily freshman psychology majors without training in marketing, so participants’ responses were unlikely to be significantly influenced by expertise (Baumann & Bonner, 2004). However, compared to typical team brainstorming tasks, this type of task is a more realistic representation of creative tasks in real-life settings (Amabile, 1982). The creative task required participants to generate original ideas and develop marketing plans as a group for each of the five products. A booklet was given to each group to record their responses to questions such as ‘‘What kind of slogans or catch phrases would you use to help market the product?”, ‘‘How would your advertising look?”, and ‘‘How would the packaging of your product appear?” In addition, participants were asked to indicate who they would market to, using what media (e.g., TV ads, newspapers, etc.) and how their marketing would make the product distinctive. They were asked to respond in writing and/or with drawings and finish the task in 50 min. Participants were told that their group’s performance would be evaluated by a professor and graduate students in the business school. After the experiment, participants were asked to complete a post-task survey that measured their perceptions of the degree to which they employed a systematic approach to task completion, and the extent to which their group engaged in idea and information sharing.

2.1.3. Group performance The marketing plans were rated using a coding scheme developed specifically for this task. Three raters rated each marketing plan along three dimensions, which captured the novelty, originality, and creativity of ideas, how well the ideas matched the target consumers to whom the plans would be directed, and how exciting, convincing, and persuasive the marketing plans would be to the target consumers. Although the most fundamental definition of creativity highlights the novelty of ideas, we adopted a broader definition of creativity (e.g., Mumford & Gustafson, 1988) that includes the notion of innovation, which acknowledges the importance of novel ideas that are also valuable and useful (e.g., Scott & Bruce, 1994). Data from the 3 three-participant groups were used to train raters to rate group performance. A measure of overall task performance was calculated by summing across the evaluation criteria for each product, and then averaging across products and raters.

2.1.4. Conscientiousness Conscientiousness was measured with a 10-item scale from the IPIP-BF (‘‘Big Five Markers”) and the International Personality Item Pool (IPIP), developed by Goldberg (1999), and available to the public (http://www.ipip.ori.org). Participants rated how accurately each item described them on a five-point scale (1 = ‘‘Very inaccurate” to 5 ‘‘Very accurate”). Sample items include ‘‘Follow a schedule,” and ‘‘Am exacting in my work”. We operationalized group conscientiousness in two ways: as the mean of individual scores, and as the minimum individual score in the group. The mean was used based on Bell (2007), who found that the mean operationalization of a group characteristic resulted in the strongest relationships with performance. However, Barrick et al. (1998) advised that alternative operationalizations of group characteristics might be more theoretically appropriate, depending on the characteristics of the task. The task in the current study is disjunctive (i.e., key creative contributions of just one member might result in high performance by the group), and we hypothesized that low conscientious members will contribute most to creative performance because they are not excessively rigid and structured. Therefore, consistent with Barrick et al.’s (1998) recommendations, it follows that groups that include an individual with a low conscientiousness score will perform well, suggesting that the group’s minimum score is a theoretically meaningful operationalization of group-level conscientiousness for this task. 2.1.5. Idea and information sharing The idea and information sharing scale included eight items adopted from Campion, Medsker, and Higgs’s (1993) and Katz’s (1999) task interdependence scales, and Riordan and Weatherly’s (1999) work group communication scale. Participants reported the extent to which they agreed with each item on a five-point scale (1 = ‘‘Strongly disagree” to 5 ‘‘Strongly agree”). A sample item is ‘‘When ideas and information were expressed, people listened patiently.” 2.1.6. Systematic strategy approach Four items were developed to measure a systematic task approach. Participants reported their perception of the degree to which the group adopted a systematic strategy to complete the task using a five-point scale (1 = ‘‘Strongly disagree” to 5 ‘‘Strongly agree”). A sample item is ‘‘Our group stuck to the task at hand rather than fooling around.” 2.1.7. Control variables Extraversion, openness to experience, and grade point average (GPA) were treated as control variables. Extraversion and openness were included as controls because the literatures on group performance and creativity suggest they are relevant to group performance on creative tasks. For example, extraversion has been linked with group performance (e.g., Barrick et al., 1998; Barry & Stewart, 1997; Mohammed, Mathieu, & Bartlett, 2002), and Barrick et al. (2001) hypothesized that extraversion might be particularly relevant in group contexts. Researchers have also hypothesized and demonstrated that openness to experience is relevant for performance on creative tasks (e.g., Feist, 1998; George & Zhou, 2001). In addition, researchers have suggested that creativity in teams might be stimulated by certain combinations or patterns of characteristics. Baer, Oldham, Jacobsohn, and Hollingshead (2008) found that when groups were high on what they called ‘‘creative confidence,” those with more members who were high on extraversion and openness to experience, and low on conscientiousness were more creative. Buchanan (1998) also took a pattern approach, but found a different ‘‘optimal” pattern, whereby groups that were moderate on extraversion and high on both openness and consci-

225

C. Robert, Y.H. Cheung / Journal of Research in Personality 44 (2010) 222–231

entiousness performed best. These findings, although not consistent with each other, suggest that in order to isolate the relationship between conscientiousness and creativity in groups, it is important to control for extraversion and openness. Self-reported GPA was obtained prior to group assignment. College students’ self-reports of GPA are a good approximation of actual GPA (Kuncel, Credé, & Thomas, 2005). GPA is correlated with general mental ability (Roth & Bobko, 2000), which has been found to be consistently related to group task performance (e.g., Barrick et al., 1998; LePine et al., 2000; Mohammed et al., 2002; Neuman & Wright, 1999). In this context, we believe that controlling for GPA makes the selection implications of the results more relevant to real-world contexts. Specifically, although employers might indeed select employees based on personality criteria such as conscientiousness, rigorous selection procedures typically also involve non-personality measures of general mental ability such as GPA.

Table 2a Hierarchical regression analysis for group mean conscientiousness predicting group performance (Study 1). Variables

B

SE B

GPA Extraversion Openness to experience

1.59 .01 ..05

.61 .04 .06

.35* .05 .11

GPA Extraversion Openness to experience Mean conscientiousness

2.08 .01 .07 .12

.61 .04 .06 .05

.45*** .04 .17 .34*

b

Step 1

Step 2

Note: Results are based on N = 55 groups. R2 = .14 for Step 1; DR2 = .10 for Step 2 (p < .05). * p < .05. ** p < .01. *** p 6 .001.

2.2. Results Table 1 presents descriptive statistics and correlations for the variables in this study. We note that variables in this table represent values aggregated from individual scores to the level of the group, and that such aggregation tends to inflate correlations (Ostroff, 1993; Robinson, 1950). Reliabilities for the three personality measures were acceptable (.85, .89, and .74 for conscientiousness, extraversion, and openness to experience, respectively). Reliabilities for the idea and information sharing, and systematic task strategy variables were acceptable (a = .84 and a = .70, respectively, and mean rwg(1) statistics for the aggregated idea and information sharing, and systematic task strategy variables were .87 and .85, respectively, indicating sufficient within group agreement. For the group performance variable, interrater reliability was acceptable (ICC1 = .82, ICC2 = .84: Bliese, 1998; James, Demaree, & Wolf, 1984). Correlations between scores on the three rated dimensions ranged from r = .62 to r = .80, indicating substantial overlap in the rated dimensions. Hypothesis 1 predicted that group conscientiousness would be negatively related to performance. Regression analyses revealed that group conscientiousness was negatively associated with group performance, as operationalized by the mean conscientiousness scores (b = .34, p < .05), or as operationalized by the minimum conscientiousness score in the group (b = .39, p < .05), controlling for GPA, extraversion, and openness to experience (see Tables 2a and 2b, respectively). Post-hoc analyses revealed that removing openness and extraversion from the analyses did not change these relationships, nor did removing GPA as a control for the regression using minimum conscientiousness. However, removing GPA from the regression equation using the mean conscientiousness score resulted in a non-significant parameter estimate for conscientiousness. Hypotheses 2a and 2b predicted that two process variables, idea and information sharing and a systematic strategy approach, would

Table 2b Hierarchical regression analysis for minimum group conscientiousness predicting group performance (Study 1). Variables

B

SE B

GPA Extraversion Openness to experience

1.59 .01 ..05

.61 .04 .06

.35* .05 .11

GPA Extraversion Openness to experience Minimum conscientiousness

2.07 .01 .06 .10

.58 .04 .06 .03

.45*** .04 .14 .39**

b

Step 1

Step 2

Note: Results are based on N = 55 groups. R2 = .14 for Step 1; DR2 = .14 for Step 2 (p < .01). * p < .05. ** p < .01. *** p 6 .001.

mediate the relationship between group conscientiousness and group performance. However, although idea and information sharing and use of a systematic task strategy were both related to group performance (r = .31, p < .05, and r = .33, p < .05, respectively), neither idea and information sharing nor systematic task processes were significantly related to group mean conscientiousness or minimum conscientiousness. Therefore, additional tests for mediation were not conducted, and Hypotheses 2a and 2b were rejected.

2.3. Discussion As hypothesized, our results suggested that low conscientiousness groups, defined either in terms of the group mean or the minimum score within a group, tended to outperform groups with high

Table 1 Means, standard deviations, and correlations of Study 1. Variables

Mean

SD

1. 2. 3. 4. 5. 6. 7. 8.

3.59 36.84 29.56 35.45 35.39 4.11 3.92 8.33

.26 3.34 4.76 4.41 2.71 .40 .48 1.20

GPA Conscientiousness (mean) Conscientiousness (minimum) Extraversion Openness to experience Idea and information sharing Systematic strategy approach Group performance

1

2 .34* .28* .03 .14 .15 .08 .36**

Note: Results are based on N = 55 groups. Two-tailed significance levels are provided. * p < .05. ** p < .001.

3

.76** .07 .22 .04 .07 .15

4

.03 .12 .03 .10 .24

5

.32* .43** .02 .02

.01 .14 .14

6

7

.34* .31*

.33*

226

C. Robert, Y.H. Cheung / Journal of Research in Personality 44 (2010) 222–231

average levels of conscientiousness on a task requiring creativity. This result represents clear support for Driskell et al.’s (1987) original hypothesis of a significant and negative relationship between group performance on a creative task and group conscientiousness. However, our hypothesis that the negative relationship between conscientiousness and performance would be mediated by a tendency to take a systematic approach to the task, and a failure to share ideas and information, was not supported (using either operationalization of group conscientiousness). In fact, although the relationship between idea and information sharing and creative performance was positive as expected, the relationship between use of a systematic approach to the task and creative performance was also positive. This finding suggests that one of our fundamental assumptions about performance on creative tasks— that its better to take a free-wheeling and unstructured approach—is perhaps faulty. Moreover, we also assumed that high conscientiousness (i.e., high mean or high minimum) groups would be more likely to engage in a more systematic approach to the task, and this too was not supported by the data. To reconcile the predicted finding of a negative relationship between conscientiousness and creative task performance with the unexpected finding that using a systematic approach to the task was positively associated with task performance, we turned to a different theoretical approach; Trait-activation theory (Tett & Burnett, 2003; Tett & Guterman, 2000). Trait-activation theory takes an ‘interactionist’ middle ground between trait theories (i.e., personality traits are associated with particular behaviors), and situationist theories (i.e., situations dictate the appropriate behaviors to be enacted), by suggesting that trait-relevant behaviors are cued by particular types of situational characteristics (Tett & Guterman, 2000). Tett and colleagues argue that differences between people in the expression of trait-relevant behaviors (e.g., orderliness) will only be manifest when the situation cues orderly behavior, without either demanding it (e.g., boot camp) or prohibiting it (e.g., a water balloon fight). For example, when provided with office space, some faculty members will choose to keep their belongings organized in a very orderly fashion, whereas others do not. Because there are not strong prohibitions against orderliness of one’s office, nor external rewards for maintaining order, the personality trait of orderliness is allowed to become manifest (Tett & Burnett, 2003). The hypotheses developed for Study 1 followed most closely to a trait theory. We hypothesized that the trait of conscientiousness would cause participants to behave in a manner that is inconsistent with adaptive approaches to completing creative tasks. An alternative explanation for the results of Study 1 that is consistent with trait-activation theory is that the creative task context did not provide highly conscientious individuals with sufficient cues for activating trait consistent behaviors (e.g., rule-following, effort, persistence) that might have resulted in high levels of performance for high conscientiousness groups. In contrast, for low conscientiousness groups, the creative task context might have activated more disorderly, flexible, unstructured, and non-linear behaviors that are actually more appropriate for creative contexts. As such, trait-activation theory suggests that high group conscientiousness does not result in the expression of behaviors that are incompatible with performance on creative tasks, but rather, the creative task context inhibits (or fails to activate) the expression of behaviors that are compatible with performance on creative tasks in high conscientiousness groups.

3. Study 2 To examine the Study 1 finding of a conscientiousness–performance reversal in groups completing creative tasks, Study 2 in-

cluded two key design features. First, we created groups that included individuals who scored either low or high on conscientiousness. Part of the rationale for using this approach was to examine groups composed of highly conscientious individuals, similar to those that would be created in organizational contexts if managers were to follow the seemingly logical advice that groups be composed of highly conscientious individuals (e.g., Bell, 2007). Second, we created two experimental conditions by manipulating instructions for the creative task: one designed to help activate conscientiousness to promote performance on the creative task (‘‘systematic” condition), and a second set of instructions, essentially replicating task instructions from Study 1, which we believed would not activate conscientiousness (‘‘flexible” condition). Manipulations of the instructions were developed with guidance from Tett and Burnett’s (2003) detailed table (p. 508), which describes task, social, and organizational job demands, distracters, constraints and releasers that should be associated with activation (or suppression) of performance relevant behaviors for each of the Big Five personality dimensions. In the ‘‘systematic” condition, task instructions placed constraints on the order in which each aspect of the task must be completed by the group (i.e., the order in which questions about characteristics of the marketing plan must be answered), as well as prohibiting them from returning to parts of the task that were already completed. With regard to Tett and Burnett’s (2003) specific guidelines, the systematic task instructions manipulation was designed to activate performance-enhancing behaviors in high conscientiousness groups associated with ‘rule-following,’ and ‘precision,’ simultaneously reducing attention to the ‘demand for creativity,’ and increasing the emphasis on ‘short term goals’ (i.e., writing down something for each product before moving onto the next product). In the ‘‘flexible” condition, participants were explicitly told to be flexible in their approach to the task, and no constraints were placed on the manner in which the task was completed. We predicted that when task instructions cue the need for a systematic approach to task completion, the trait-relevant performance-enhancing behaviors associated with low conscientiousness (i.e., demand for creativity, attention to rules, need for precise and explicit communication) will be suppressed (or not activated), resulting in lower performance than in low conscientiousness groups provided with more flexible instructions for task completion. Stated formally: Hypothesis 3. Under explicitly flexible task instructions, low conscientiousness groups will outperform high conscientiousness groups. Hypothesis 4. Group conscientiousness (high vs. low) will interact with task instructions (systematic vs. flexible); high conscientiousness groups in the systematic condition will outperform high conscientiousness groups in the flexible condition, and that pattern will reverse for low conscientiousness groups. 3.1. Method 3.1.1. Participants Pre-test questionnaires assessing conscientiousness were completed by undergraduate students enrolled in an introductory psychology course at a large mid-western university. The same 10-item conscientiousness scale used in Study 1 was used to measure respondents’ conscientiousness levels. Respondents who scored one standard deviation or more above (i.e., high in conscientiousness) or below the mean (low in conscientiousness) were contacted via e-mail, approximately 1 month after completing the conscientiousness measure, to solicit their participation for course credit. A total of 358 low and 400 high conscientiousness

227

C. Robert, Y.H. Cheung / Journal of Research in Personality 44 (2010) 222–231

individuals were invited to sign up for experimental sessions. We recruited 236 participants (82 male, 154 female), and assigned them to 1 of 59 four-person groups (i.e., all high or all low pre-test conscientiousness). As the number of female participants was higher than male participants, we attempted to balance all groups such that there were no all-male or all-female groups, to control for team gender composition and avoid possible confounding based on group gender composition. The average age of the participants was 18.7 years (SD = 2.5). 3.1.2. Procedures The experimental design was a two (high vs. low conscientiousness), by two (systematic vs. flexible task instructions) betweensubjects design. Upon arriving at the laboratory, participants were asked to complete a pre-task questionnaire assessing their GPA and demographic data before being randomly assigned to groups with the constraint on gender assignment noted above. Using the same procedures as in Study 1, each group was asked to generate original ideas and develop marketing plans for three relatively new or unique products (e.g., a jar of glow-in-the-dark hair gel). Participants were asked to respond in writing and/or with drawings, and to record their responses in a booklet. Task instructions were manipulated to cue or suppress trait-relevant behavior associated with high or low conscientiousness, as described above. In the systematic task instruction condition, groups were instructed to develop marketing plans for three products in a predetermined order, and to answer the associated questions in the order presented in the response booklet. They were also explicitly told not to revisit previous products or questions once they had answered them. In the flexible condition, groups were told that they could work on the items in any sequence, could respond to the questions about their marketing plans in any order, and were welcome to go backward to revisit items/issues they had dealt with previously. Instructions were provided in writing, and were read aloud prior to beginning the task to reinforce the manipulation. 3.1.3. Group performance The marketing plans were rated using the same coding scheme and criteria used in Study 1, and involved two raters, blind to experimental condition, rating each marketing plan on three dimensions. The overall task performance was calculated by summing across the three evaluation criteria, and then averaging across criteria and raters (i.e., the metric is the same as Study 1). Interrater reliability was sufficient (ICC1 = .71, ICC2 = .76). Correlations between scores on each dimensions ranged from r = .47 to r = .77.

Table 3 Means, standard deviations, and correlations of Study 2. Variables

Mean

SD

1

1. GPA 2. Conscientiousnessa 3. Group performance

3.56 1.56 8.98

.35 .50 1.44

.51** .07

.10

Note: Results are based on N = 59 groups. Two-tailed significance levels are provided. * p < .05. ** p < .001. a Conscientiousness is dichotomized into high conscientiousness = 2 and low conscientiousness = 1.

Table 4 Means and standard deviations of group performance in Study 2. Task instructions conditions

Group conscientiousness High

Systematic Flexible

Low

Mean

SD

Mean

SD

9.36 8.88 (d = .32)

1.57 1.39

8.34 9.39 (d = .78)

1.27 1.41

test for the interaction was significant, F(1, 54) = 4.33, p < .05, and the hypothesized cross-over form of the interaction was obtained (see Fig. 1). For the low conscientiousness groups, mean scores for the systematic and flexible instructions conditions were 8.34 and 9.39, respectively (d = .78). For the high conscientiousness groups, mean scores for the systematic and flexible instructions conditions were 9.36 and 8.88 (d = .32), respectively. This result suggests that while low conscientiousness groups performed better under the flexible task instructions condition than under the systematic task instructions condition, the direction of the means was reversed for high conscientiousness groups. However, results of simple comparisons indicated that the mean for the high conscientiousness groups in the systematic instructions condition was not significantly greater than the mean for the high conscientiousness groups in the flexible instructions condition, F(1, 24) = .68, n.s. The low conscientiousness groups assigned to the flexible task instructions condition performed significantly better than low conscientiousness groups assigned to the systematic instructions con-

Performance on Creative Task

3.2. Results

Systematic condition

Flexible condition

12

Team Performance Mean Score

Table 3 presents descriptive statistics for the variables examined in this study. Hypothesis 3 predicted that under the explicitly flexible task instructions condition, low conscientiousness groups would outperform high conscientiousness groups (a conceptual replication of Study 1). Although the trend of the data was in the predicted direction (i.e., M = 9.39 for low conscientiousness groups vs. M = 8.88 for high conscientiousness groups, d = .32), results of one-way ANOVA using GPA as a covariate indicated that the difference was not statistically significant. Hypothesis 4 predicted that group conscientiousness and task strategy would interact, whereby high conscientiousness groups would perform best in the systematic task instructions condition, and low conscientiousness groups would perform best in the flexible instructions condition. The performance of groups in each condition is presented in Table 4. Two-way ANOVA was used to test the hypothesis, with group mean GPA entered as a covariate. The

2

10

8

6 Low

High Conscientiousness

Fig. 1. Interaction between group conscientiousness and task instructions (Study 2).

228

C. Robert, Y.H. Cheung / Journal of Research in Personality 44 (2010) 222–231

dition, F(1, 31) = 5.0, p < .05. A post-hoc contrast comparing means in the activated trait conditions (i.e., the high conscientiousness/ systematic instructions condition and the low conscientiousness/ flexible instructions conditions) against the means in the non-activated conditions (i.e., low conscientiousness/systematic instructions and high conscientiousness/flexible instructions) indicated that the mean across the activated trait conditions (M = 9.38) was significantly higher than the mean across the non-activated trait conditions (M = 8.56, d = .59; F(1, 57) = 4.96, p < .05). Additional post-hoc analyses revealed that removing GPA as a covariate had no effect on the statistical significance of the interaction or comparisons of individual cell means. 3.3. Discussion Study 2 was designed to test the prediction that changes in task characteristics could activate or suppress trait-relevant behaviors that could be applied to the creative task, and was developed using Tett and Guterman’s (2000) trait-activation theory. We predicted that high conscientiousness groups would perform particularly well when the performance-enhancing behaviors consistent with high conscientiousness (e.g., those characterized by persistence, task focus, high effort) were activated by task instructions that cued a more rigid and systematic approach, whereas groups low in conscientiousness would perform particularly well when task instructions were flexible and cued the performance-enhancing behaviors associated with creativity (e.g., trial-and-error, flexible/ non-systematic approaches to task completion). Although the simple effects tests showed that the difference between the high and low conscientiousness groups in the flexible task instructions condition (replicating the conditions in Study 1) were not significantly different from each other, the significant cross-over interaction pattern that was obtained was consistent with our predictions. Specifically, the pattern observed in the interaction showed that high conscientiousness groups given systematic task instructions, and the low conscientiousness groups given flexible task instructions tended to perform best at the task, and groups given the opposite instructions (i.e., those not activating the traits) tended to perform worse. Consistent with the predictions, groups that were directed to complete the task in a manner that cued trait-relevant behaviors that could be applied to the creative task outperformed groups that were cued to suppress those trait-relevant behaviors. 4. General discussion In the period since Barrick and Mount (1991) concluded that there was a robust and positive relationship between conscientiousness and performance across work contexts, considerable attention has been paid to the potential importance of conscientiousness in organizational contexts: and with good reason. However, theoretical and empirical ‘‘hints” about contexts in which conscientiousness might actually interfere with performance have surfaced sporadically in the literature, and have provided some reference points for defining the boundaries of the positive conscientiousness–performance relationship. Building on the proposition that groups with highly conscientious members might actually perform worse on creative tasks, first put forth by Driskell et al. (1987), we conducted two studies designed to (a) document that such a result could be obtained in a controlled setting with groups engaged in a creative task and (b) understand why such a result might be obtained. The results of Study 1 provided support for Driskell et al.’s (1987) proposition that high conscientiousness groups will perform worse than low conscientiousness groups on creative tasks. Notably, this was done by defining the group in terms of both

the average conscientiousness score, and by using the minimum score of all group members. However, our results did not support our prediction that high conscientiousness groups would perform poorly because of their tendency to apply an excessively systematic approach to task completion, or because of a failure to share ideas and information among group members. Study 2 was guided by an alternative hypothesis for the negative conscientiousness– performance relationship derived from Tett and Guterman’s (2000) trait-activation theory. Specifically, in a ‘‘systematic” task instructions condition, we manipulated the task instructions in such a way that the creative task, which might normally trigger an assumption that the task requires a flexible approach unbounded by rules, could appear to necessitate a more fixed process or more rigid procedures for task completion. This manipulation was designed to activate traits associated with high conscientiousness, and thus to result in higher performance for more conscientious groups. In contrast, the other task instructions condition emphasized the potential for participants to approach the task more flexibly, without constraints on how the task was carried out. This manipulation was designed to be a conceptual replication of the task conditions from Study 1, and to result in a pattern whereby the low conscientiousness groups would outperform the high conscientiousness groups. The results of Study 2 provided some support for our underlying theoretical approach, which suggests that groups would perform best when the characteristics of the task (manipulated through instructions in this study) activated traits that could be applied effectively to the task. This was manifest in both the significant interaction between conscientiousness and task instructions, and in the post-hoc analysis which demonstrated that when the task instructions matched the behavioral proclivities of the group (i.e., flexible instructions for low conscientiousness groups and systematic instructions for high conscientiousness groups) those groups performed significantly better than groups receiving instructions that did not match (and thus did not activate) the group’s trait. Stated differently, our findings suggest that it is not necessarily the case that high conscientiousness groups will do the wrong or inappropriate things while engaged in creative tasks as hypothesized in Study 1. Instead, it seems that creative task contexts might typically activate behaviors associated with task performance in low conscientiousness groups in particular (e.g., to approach the task in a more flexible manner, unconstrained by rules or rigid procedures). Our results in Study 2 were not supportive of the opposing perspective: creative contexts do not adequately activate behaviors associated with high performance in high conscientiousness groups unless the task is structured specifically to do so. Although the means were in that direction, they were not significantly different from each other. However, it is also important to interpret the overall pattern of results in light of the fact that the systematic task instructions were not just descriptively different from the flexible task instructions, they also imposed constraints on the behavior of participants by forcing them to address specific items in a preset order and to not go backwards and reconsider responses that had already been recorded. This is a distinct handicap, and represents a reasonable alternative explanation for why the level of performance observed in high conscientiousness groups in the systematic instructions condition was not statistically greater than the level of high conscientiousness groups in the flexible condition. Given the fairly well-accepted notion that creativity might require non-linear approaches to tasks (Simonton, 2003; Taggar, 2002), as well as evidence that low conscientiousness groups performed significantly better under flexible than systematic task instructions, the fact that high conscientiousness groups under the systematic task instruction condition performed similar to groups in other conditions, despite the handicap, is notable.

C. Robert, Y.H. Cheung / Journal of Research in Personality 44 (2010) 222–231

4.1. Practical implications Given the consistent relationship observed between conscientiousness and performance, researchers and practitioners have been encouraging the use of conscientiousness-related personality measures in selection contexts (Behling, 1998; Morgeson, Reider, & Campion, 2005). However, successful completion of creative tasks is becoming more important for organizational success and innovation (Brown & Eisenhardt, 1995; Taggar, 2002), and those creative tasks are more often being assigned to groups (Edmondson, 2002; Paulus & Yang, 2000). Therefore, it is important that managers and researchers understand how to best leverage the strengths of groups asked to accomplish creative tasks. Our results imply somewhat counterintuitive advice for managers of work groups engaged in high creativity tasks. Specifically, managers might have an instinct to impose structure and increase monitoring on groups engaged in creative tasks that have been unreliable, and are assumed to be composed of individuals who are low in conscientiousness. However, for such groups, perhaps characterized by less rule-oriented and more artistic sensibilities (Feist, 1998) imposing structure and order (e.g., rules, deadlines, progress checks) might serve to reduce creativity. Indeed, offering freedom, flexibility, and autonomy to groups who are lower in mean conscientiousness, or have some members who are, might be the most promising strategy. In contrast, the theory underlying Study 2, which was based on the notion of trait activation, suggests that groups composed of high conscientiousness individuals might actually benefit from providing highly conscientious groups with more structure. This could be done through task features such as time deadlines, checkpoints, or intermediate progress reports, which activate conscientious group members’ natural inclinations to march steadily toward completion of tasks. While our Study 2 results did not fully support this conclusion, it is interesting to note that when we imposed formal task constraints on high conscientiousness groups, the performance of those groups did not suffer, and indeed, was somewhat better (though not significantly) than high conscientiousness groups that did not have such constraints imposed. 4.2. Limitations and future directions Like other research conducted in laboratory settings using somewhat artificial tasks, the two studies have limited realism and generalizability. In particular, the durations of the experiments were very short relative to the time-horizons of most groups engaged in creative tasks for organizations. With longer time periods, the willingness of high conscientiousness individuals to exert sustained effort might have proven important. Undoubtedly, many creative tasks (e.g., performing arts, writing) require persistence and practice to achieve high performance, qualities more likely to be found in high conscientiousness individuals. Also, although we chose to measure creativity in a manner that captures the practical usefulness and value of ideas (Mumford & Gustafson, 1988), and highlighted idea quality, our task context did not allow us to measure the quantity of creative ideas as is done in much of the research on creativity (e.g., Buchanan, 1998). Again, this might have biased the results against high conscientious participants, whose diligence and industriousness might have resulted in higher quantities of creative output on different tasks. We also note some methodological limitations. First, the flexible task instruction condition in Study 2 did not preclude groups from choosing a very systematic and rigid approach to the task. However, the fact that our results were obtained despite this possible bias might actually increase confidence in our results. Second, in Study 2 we used an extreme groups approach, whereby only individuals who scored low or high on a pre-test measure of con-

229

scientiousness were assigned to groups. Scholars have noted that the use of such an approach is associated with a number of potential problems (Preacher, Rucker, MacCallum, & Nicewander, 2005), including the fact that regression to the mean is more likely to occur when extreme scores are used. In addition, this approach precludes an examination of potentially interesting effects of variability in conscientiousness scores on group performance. However, our hypotheses in Study 2 were based on the assumption that low and high conscientiousness individuals would respond differently to task instructions, and our extreme groups approach allowed us to examine this research question in a straightforward way. In addition, we believe that the creation of extreme groups on the ‘‘highly conscientious” side are a potential reality in organizational contexts if managers follow advice suggesting that only highly conscientious individuals be assigned to groups (e.g., Bell, 2007). Our sample also had some limitations. First, the use of undergraduate participants is often seen as a weakness with regard to the generalizability. However, an important advantage to using undergraduates is that sources of variance in task performance such as differences in education levels, experience, and age are essentially controlled in a population that is relatively homogeneous on those dimensions. Second, our sample included more women than men. We attempted to control for this by balancing the gender composition of our groups, but this also might have consequences. Researchers have observed that mixed gender composition in groups tends to increase the salience of gender identity, which subsequently influences behavior within the group (Cota & Dion, 1986; Randel, 2002). Importantly, this would suggest that personality effects might be attenuated in our study, so the fact that personality effects were still observed is notable. Given the intriguing nature of the current results, additional research is needed that uses more realistic tasks, perhaps with intact groups. Another possible direction for future research might be the use of hierarchical linear modeling to explore the potential impact of both individual and group-level conscientiousness on group performance. Such an analysis might provide a more sophisticated approach to understanding the contributions of personality to group-level performance. A particularly important issue for future research is the possibility that an examination of facet-level effects might provide meaningful insight into the relationship between conscientiousness and creative task performance. A recent meta-analysis conducted by Dudley, Orvis, Lebiecki, and Cortina (2006) indicated that facetlevel measures of conscientiousness had very different associations with a variety of performance measures across a range of performance contexts, although creative contexts were not included in their analysis. A number of authors suggest that the distinction between achievement and dependability facets might be particularly important (Hough, 1992; LePine, 2003). Indeed, Eysenck (1993) differentiated between internal motivation and conformity/nonconformity as unique predictors of creative behavior: two traits that seem to parallel the achievement and dependability facets. Importantly, our hypotheses about the negative relationship between conscientiousness and performance were most in-line with the negative impact of characteristics associated with dependability rather than achievement, and our Study 2 manipulation was derived from an interpretation of conscientiousness that focused primarily on the dependability or orderliness aspects rather than achievement. Notably, the measure we employed in both studies appears to be weighted strongly toward the dependability facet (i.e., most items also show up on facet measures of orderliness and purposefulness in other IPIP scales) and it is possible that our results would not be replicable if we employed a measure weighted more strongly toward achievement. Indeed, including GPA as a control variable in Study 1 appears to have resulted in a

230

C. Robert, Y.H. Cheung / Journal of Research in Personality 44 (2010) 222–231

suppressor effect. Because GPA appears to be a good proxy for achievement and achievement motivation (Robbins et al., 2004), it is possible that the weak zero-order negative relationship between our measures of conscientiousness and creative group performance was strengthened by the inclusion of GPA as a control (which was positively correlated with the performance measure). This pattern is consistent with the notion that although achievement and dependability are likely related to each other in a higher order conscientiousness construct, they might very well have unique relationships with creative behavior. Concern about our Study 1 findings relying on a suppressor effect were mitigated somewhat by the finding that the minimum conscientiousness operationalization of group conscientiousness was also negatively related to group creative performance without GPA as a control. In addition, we also believe that including GPA as a control variable has realworld parallels to organizational contexts. Selection processes in organizations often utilize non-personality measures of cognitive abilities (including GPA), resulting in some degree of restriction of range on the achievement elements associated with conscientiousness, and thus relatively more emphasis on non-achievement relevant facets such as dependability and orderliness. Nonetheless, our results suggest that future research examining creativity– conscientiousness relationships should pay particular attention to the potential for facet-level effects, most notably differences in relationships between achievement and dependability facets. 4.3. Conclusion The conclusion that conscientiousness is positively related to task performance seems both logical and inescapable. However, the current research provides some initial evidence that helps define the boundaries of that relationship. Although prior research suggested that a failure to adapt might be the cause of a breakdown in the conscientiousness–performance relationship in some task contexts (LePine, 2003; Waung & Brice, 1998), our results support a somewhat different view. Specifically, our results suggest that it is not just the underlying task itself that is either conducive or problematic for performance, but rather, the interaction between personality characteristics and the characteristics of the task. In addition, our results suggest that even modest changes to the task itself (i.e., instructions on procedures but not underlying content) might serve to activate the positive characteristics associated with traits such as conscientiousness. This finding has important implications for managers, particularly if they employ selection strategies by which conscientiousness is used as a selection criterion. References Amabile, T. M. (1982). Social psychology of creativity: A consensual assessment technique. Journal of Personality and Social Psychology, 43, 997–1013. Baer, M., Oldham, G. R., Jacobsohn, G. C., & Hollingshead, A. B. (2008). The personality composition of teams and creativity: The moderating role of team creative confidence. Journal of Creative Behavior, 42, 255–282. Barrick, M. R., & Mount, M. K. (1991). The Big Five personality dimensions and job performance: A meta-analysis. Personnel Psychology, 44, 1–26. Barrick, M. R., Mount, M. K., & Judge, T. A. (2001). Personality and performance at the beginning of the new millennium: What do we know and where do we go next? International Journal of Selection and Assessment, 9, 9–29. Barrick, M. R., Stewart, G. L., Neubert, M. J., & Mount, M. K. (1998). Relating member ability and personality to work-team processes and team effectiveness. Journal of Applied Psychology, 83, 377–391. Barry, B., & Stewart, G. L. (1997). Composition, process, and performance in selfmanaged groups: The role of personality. Journal of Applied Psychology, 82, 62–78. Baumann, M. R., & Bonner, B. L. (2004). The effects of variability and expectations on utilization of member expertise and group performance. Organizational Behavior and Human Decision Processes, 93, 89–101. Behling, O. (1998). Employee selection: Will intelligence and conscientiousness do the job? Academy of Management Executive, 12, 77–86.

Bell, S. T. (2007). Deep-level composition variables as predictors of team performance: A meta-analysis. Journal of Applied Psychology, 92, 595–615. Bliese, P. D. (1998). Group size, ICC values, and group level correlations: A simulation. Organizational Research Methods, 1, 355–373. Brown, S. L., & Eisenhardt, K. M. (1995). Product development: Past research, present findings, and future directions. Academy of Management Review, 20, 343–378. Buchanan, L. B. (1998). The impact of Big Five personality characteristics on group cohesion and creative performance. Unpublished Doctoral Dissertation. Virginia Polytechnic Institute and State University. Bunderson, J. S., & Sutcliffe, K. M. (2002). Comparing alternative conceptualizations of functional diversity in management teams: Process and performance effects. Academy of Management Journal, 45, 875–893. Campion, M. A., Medsker, G. J., & Higgs, A. C. (1993). Relations between work group characteristics and effectiveness: Implications for designing effective work groups. Personnel Psychology, 46, 823–850. Costa, P. T., & McCrae, R. R. (1992). Revised NEO personality inventory (NEO-PI-R) and NEO five-factor inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources, Inc. Cota, A. A., & Dion, K. L. (1986). Salience of gender and sex composition of ad hoc groups: An experimental test of distinctiveness theory. Journal of Personality and Social Psychology, 50, 770–776. Driskell, J. E., Hogan, R., & Salas, E. (1987). Personality and group performance. In C. Hendrick (Ed.). Group processes and intergroup relations: Review of personality and social psychology (Vol. 9, pp. 91–112). Newbury Park, CA: Sage. Dudley, N. M., Orvis, K. A., Lebiecki, J. E., & Cortina, J. M. (2006). A meta-analytic investigation of conscientiousness in the prediction of job performance: Examining the intercorrelations and the incremental validity of narrow traits. Journal of Applied Psychology, 91, 40–57. Dugosh, K. L., Paulus, P. B., Roland, E. J., & Yang, H. (2000). Cognitive stimulation in brainstorming. Journal of Personality and Social Psychology, 79, 722–735. Edmondson, A. C. (2002). The local and variegated nature of learning in organization: A group-level perspective. Organization Science, 13, 128– 146. Eysenck, H. J. (1993). Creativity and personality: Suggestions for a theory. Psychological Inquiry, 4, 147–178. Feist, G. J. (1998). A meta-analysis of personality in scientific and artistic creativity. Personality and Social Psychology Review, 2, 290–309. George, J. M., & Zhou, J. (2001). When openness to experience and conscientiousness are related to creative behavior: An interactional approach. Journal of Applied Psychology, 86, 513–524. Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure. Psychological Assessment, 4, 26–42. Goldberg, L. R. (1999). A broad-bandwidth, public domain, personality inventory measuring the lower-level facets of several five-factor models. In I. Mervielde, I. Deary, F. De Fruyt, & F. Ostendorf (Eds.). Personality psychology in Europe (Vol. 7, pp. 7–28). Tilburg, The Netherlands: Tilburg University Press. Hough, L. M. (1992). The ‘‘Big Five” personality variables-construct confusions: Descriptions versus predictions. Human Performance, 5, 139–155. Hurtz, G. M., & Donovan, J. J. (2000). Personality and job performance: The Big Five revisited. Journal of Applied Psychology, 85, 869–879. James, L. R., Demaree, R. G., & Wolf, G. (1984). Estimating within-group interrater reliability with and without response bias. Journal of Applied Psychology, 69, 85–98. Kanfer, R., Ackerman, P. L., Murtha, T. C., Dugdale, B., & Nelson, L. (1994). Goal setting, conditions of practice, and task performance: A resource allocation perspective. Journal of Applied Psychology, 79, 826–835. Katz, T. Y. (1999). Self-construal as a moderator of the effects of task and reward interdependence on group performance (Doctoral dissertation, Columbia University, 1999). Dissertation Abstracts International, 60, 6-B. Kuncel, N. R., Credé, M., & Thomas, L. L. (2005). The validity of self-reported grade point averages, class ranks, and test scores: A meta-analysis and review of the literature. Review of Educational Research, 75, 63–82. LePine, J. A. (2003). Team adaptation and postchange performance: Effects of team composition in terms of members’ cognitive ability and personality. Journal of Applied Psychology, 88, 27–39. LePine, J. A., Colquitt, J. A., & Erez, A. (2000). Adaptability to changing task contexts: Effects of general cognitive ability, conscientiousness, and openness to experience. Personnel Psychology, 53, 563–593. McGrath, J. E. (1984). Groups: Interaction and performance. Englewood Cliffs, New Jersey: Prentice Hall. Mohammed, S., Mathieu, J. E., & Bartlett, A. L. (2002). Technical–administrative task performance, leadership task performance, and contextual performance: Considering the influence of team- and task-related composition variables. Journal of Organizational Behavior, 23, 795–814. Morgeson, F. P., Reider, M. H., & Campion, M. A. (2005). Selecting individuals in team settings: The importance of social skills, personality characteristics, and teamwork knowledge. Personnel Psychology, 58, 583–611. Mount, M. K., & Barrick, M. R. (1995). The Big Five personality dimensions: Implications for research and practice in human resources management. Research in Personnel and Human Resources Management, 13, 153–200. Mumford, M. D., & Gustafson, S. G. (1988). Creativity syndrome: Integration, application, and innovation. Psychological Bulletin, 103, 27–43. Neuman, G. A., Wagner, S. H., & Christiansen, N. D. (1999). The relationship between work-team personality composition and the job performance of teams. Group and Organization Management, 24, 28–45.

C. Robert, Y.H. Cheung / Journal of Research in Personality 44 (2010) 222–231 Neuman, G. A., & Wright, J. (1999). Team effectiveness: Beyond skills and cognitive ability. Journal of Applied Psychology, 84, 376–389. Ostroff, C. (1993). Comparing correlations based on individual-level and aggregated data. Journal of Applied Psychology, 78, 569–582. Paulus, P. B., & Yang, H. (2000). Idea generation in groups: A basis for creativity in organizations. Organizational Behavior and Human Decision Processes, 82, 76–87. Preacher, K. J., Rucker, D. D., MacCallum, R. C., & Nicewander, W. A. (2005). Use of the extreme groups approach: A critical reexamination and new recommendations. Psychological Methods, 10, 178–192. Randel, A. E. (2002). Identity salience: A moderator of the relationship between group gender composition and work group conflict. Journal of Organizational Behavior, 23, 749–766. Riordan, C. M., & Weatherly, E. W. (1999). Defining and measuring employees’ identification with their work groups. Educational and Psychological Measurement, 59, 310–324. Robbins, S. B., Lauver, K., Le, H., Davis, D., Langley, R., & Carlstrom, A. (2004). Do psychosocial and study skill factors predict college outcomes? A meta-analysis. Psychological Bulletin, 130, 261–288. Robertson, I. T., Gibbons, P., Baron, H., MacIver, R., & Nyfield, G. (1999). Understanding management performance. British Journal of Management, 10, 5–12. Robinson, W. S. (1950). Ecological correlations and the behavior of individuals. American Sociological Review, 15, 351–357. Roth, P. L., & Bobko, P. (2000). College grade point average as a personnel selection device: Ethnic group differences and potential adverse impact. Journal of Applied Psychology, 85, 399–406.

231

Salgado, J. F. (1997). The five factor model of personality and job performance in the European community. Journal of Applied Psychology, 82, 30–43. Scott, S. G., & Bruce, R. A. (1994). Determinants of innovative behavior: A path model of individual innovation in the workplace. Academy of Management Journal, 37, 580–607. Simonton, D. K. (2003). Scientific creativity as constrained stochastic behavior: The integration of product, person, and process perspectives. Psychological Bulletin, 129, 475–494. Steiner, I. D. (1972). Group process and productivity. New York: Academic Press. Taggar, S. (2002). Individual creativity and group ability to utilize individual creative resources: A multilevel model. Academy of Management Journal, 45, 315–330. Tegano, D. W., & Moran, J. D. (1989). Sex differences in the original thinking of preschool and elementary school children. Creativity Research Journal, 2, 101–110. Tett, R. P. (1998). Is conscientiousness always positively related to job performance? The Industrial–Organizational Psychologist, 36, 24–29. Tett, R. P., & Burnett, D. D. (2003). A personality trait-based interactionist model of job performance. Journal of Applied Psychology, 88, 500–517. Tett, R. P., & Guterman, H. A. (2000). Situation trait relevance, trait expression, and cross-situational consistency: Testing a principle of trait activation. Journal of Research in Personality, 34, 397–423. Waung, M., & Brice, T. S. (1998). The effects of conscientiousness and opportunity to caucus on group performance. Small Group Research, 29, 624–634.