Abstract
The innovations of creative individuals are regarded as vital for business functioning and survival. To this end, efforts have been made to design measures of creative personality in hopes of predicting creative performance. Current measures of creative personality all reside at the explicit level, yet theory and research both suggest that a large proportion of personality can also be conceptualized at the implicit level. We address this issue by presenting a theoretical basis for creative personality that operates on an implicit level. Using conditional reasoning methodology, we describe five cognitive biases that serve as justification mechanisms for creative personality. Next, we link implicit creative personality to creative abilities through a developmental process. We then test this model and our new measure of creative personality in five different studies. Our results provide evidence in support of an implicit component of creative personality and suggest that it is a substantial predictor of creative performance. Finally, we describe the management and human resources implications of the conceptualization of creative personality as an implicit construct.
In response to changing market conditions and a cutthroat business environment, organizations face increasing pressure to use their creative and innovative assets, including creative employees, to maintain a competitive advantage and ensure their survival (Crossan & Apaydin, 2010). Answering the recent call for development of new theories to describe creativity (Anderson, Potočnik, & Zhou, 2014), we present a theoretically based model of creative personality using the conditional reasoning (CR) framework (e.g., James, 1998; James & LeBreton, 2012) and a new implicit measure, the Conditional Reasoning Test for Creative Personality (CRT-CP). Our theory posits that creative personality includes a substantial implicit component, which differs from current measures emphasizing the explicit components of personality. Our work suggests that individuals vary with respect to creative personality. Although many factors affect the creativity of one’s work, we contend that individuals with a high level of implicit creative personality are more likely to engage in processes leading to creative outcomes.
We begin by providing a rationale for developing an implicit measure of creative personality. We then discuss the CR framework and describe five rationalizations, or justification mechanisms (JMs), that creative individuals may use in order to cognitively defend their creative efforts. Next, we present hypotheses for the nomological network relating implicit creative personality to creative performance, explicit assessment of creative personality, and creative ability. We then present empirical evidence from five studies designed to test our hypotheses and explore measurement-related matters regarding the CRT-CP. We conclude by discussing the theoretical and practical implications of our model.
The primary contributions of this article are (a) to draw attention to the implicit component of creative personality as to provide a holistic view of personality, (b) to address the insufficient predictive validity of existing explicit measures by developing a new theory of creative personality using a CR framework, (c) to validate the CRT-CP, and (d) to advance the research agenda on creativity and the CR literatures. Taken together, these contributions further our understanding of creative personality and enhance the prediction of workplace creativity.
Rationale and Theoretical Development
Explicit and Implicit Personality
Efforts have been made to develop measures of creative personality and creative self-perceptions, perhaps most notably openness from the five-factor model (FFM; McCrae, 1987) and the Creative Personality Scale (CPS; Gough, 1979). Creative personality is similar, both theoretically and in item content, to creative self-efficacy (Tierney & Farmer, 2002), creative role identity (Farmer, Tierney, & Kung-McIntyre, 2003), and cognitive style (Jabri, 1991; Kirton, 1976), all of which rely on individuals to self-report their own perceptions.
Self-report techniques are the dominant paradigm for assessing personality in the organizational sciences (James & LeBreton, 2012), and in some contexts, this strategy may be fruitful in predicting creative performance. However, self-reported responses may not accurately represent a person’s entire personality or the way he or she views and responds to external stimuli in situ (Hogan, 2005; James & LeBreton, 2012), in part because people can only report on those aspects of their personality about which they are aware. Explicit personality is the part of personality individuals can access and thus represents the way individuals view themselves. In contrast, implicit personality constitutes the aspect of personality that lies outside of individual awareness and is not accessible via introspection (James, 1998; James & LeBreton, 2012).
Implicit and explicit aspects of personality represent complementary attributes of the personality system but, interestingly, are often not highly correlated with one another (James & LeBreton, 2012; McClelland, 1987). Although personality psychologists have argued that both elements should be considered, and not discounting the contributions that have been made, to date organizational scientists have almost exclusively focused on the explicit assessment of personality, leaving a deficit in our conceptual view and empirical assessment of personality (cf. Bing, LeBreton, Davison, Migetz, & James, 2007; Bing, Stewart, et al., 2007; Frost, Ko, & James, 2007; James, 1998; James & LeBreton, 2012; McClelland, 1987; Winter, John, Stewart, Klohnen, & Duncan, 1998). To this end, a more holistic perspective of creative personality is needed. Inclusion of the implicit components of creative personality allows for a deeper understanding of human functioning and, ultimately, better prediction of creativity.
Although implicit and explicit aspects of personality are both important, the implicit perspective suggests that some facets of personality reside more deeply in the unconscious and are defended more heavily via rationalizations and psychological defenses (McClelland, 1987). Thus far researchers have examined implicit aspects of aggression, power/dominance, achievement motivation, self-defeat, fear of failure, and addiction proneness (e.g., Bowler, Bowler, & Cope, 2013: James, 1998; James & LeBreton, 2012; James et al., 2005). All of these constructs, including the potentially compulsive nature of achievement-motivated individuals to place hours of hard work ahead of enjoyment with friends and family, require individuals to rationalize and justify their actions. This defensive process buffers self-esteem and allows individuals to maintain positive self-views (James & LeBreton, 2012). People wish to believe that the decisions they make, based upon the ambiguous or conflicting social information that surrounds them, are nevertheless sound. Indeed, even individuals who engage in activities running counter to cultural norms or that conflict with their personal self-views prefer to view their behaviors as rational (James, 1998). Because creative action is often not sanctioned by the environment, creative personality may be heavily guarded by such cognitive biases and is worthy of consideration from an implicit perspective.
The Implicit Nature of Creative Personality
Creativity is “any act, idea, or product that changes an existing domain, or that transforms an existing domain into a new one” (Csikszentmihalyi, 1996: 28). Creative efforts must also be regarded as appropriate in the given context, because creative ideas must be implementable in order to be valuable; however, creative endeavors require effort, time, and continued engagement to persevere, typically beyond the mastery of achievement motivation, even in the face of failure (Bandura, 1997; Ford, 1996). The confidence that creative individuals exhibit regarding their novel ideas and the magnitude of effort they invest are often regarded by others as irrational, at least until they have attained success (Vance, 2015). To illustrate, when Werner Forssmann proposed the idea to use a catheter to reach inside the heart of a living person, he was initially ignored. Forssmann persisted and was allowed to test his ideas on animals (Dunn, 2015). He was successful, asked to try his technique on humans, but was denied. He then successfully tried the technique on himself and captured an X-ray image as proof. While ignored in his home country and unbeknownst to him, Forssmann’s work laid the foundation for the minimally invasive techniques that revolutionized surgery, and he was awarded a Nobel Prize nearly 30 years later.
Further, individuals who are classified as highly creative frequently sidestep social norms and ignore rules (Csikszentmihalyi, 1996; Ford, 1996) and in so doing may be regarded as dissatisfied malcontents, social outcasts, deviants, and oddballs (Dunn, 2015; Isaacson, 2011; Vance, 2015). Societal and peer pressures attempt to compel individuals to conform to normative standards, yet creative individuals view things differently, recognize a need for change, and regard existing solutions as suboptimal when others are satisfied, placing creative individuals at odds with existing norms (Meyerson & Scully, 1995).
Despite innumerable books and articles extolling the virtues of creativity in both the popular press and academia, unconventional ideas are often labeled as risky, foolish, dangerous, or radical (Gilson, 2008; Kerr, 1975; Meyerson & Scully, 1995; Staw, 1995). Research suggests that individuals, on average, are averse to novelty (Mueller, Melwani, & Goncalo, 2012; Mueller, Wakslak, & Krishnan, 2014) and instead prefer practical ideas proven to work. Risk-taking and risk perception research indicates that managers are, on average, risk averse and that concepts associated with risk are linked to negative affective reactions like fear and thoughts of failure (March & Shapira, 1987).
Indeed, creativity appears to be an espoused but infrequently enacted organizational value. We recognize that some organizations do applaud creative efforts; however, our argument is that these efforts are typically incremental in nature and are encouraged so long as they do not disturb the organizational status quo (Campbell, 2000; Grant & Ashford, 2008), as creative efforts resulting in organizational changes are often complex, costly, and time-consuming. Thus, organizations often harbor a bias against risk and creativity (Kerr, 1975; Lewis, 2003; Mueller et al., 2012; Mueller et al., 2014; Staw, 1995). Despite these obstacles, some individuals press forward with their ideas (e.g., Bandura, 1997; Dunn, 2015; Meyerson & Scully, 1995).
Individuals who gravitate toward novel endeavors, especially those that conflict with social norms, also wish to view their choices as logical and sensible (James, 1998; James & LeBreton, 2012). These individuals may continue to persevere in the face of resistance by developing justifications that allow unorthodox choices to appear reasonable (James, 1998). We argue that the dispositional characteristics that support creative action develop as part of a defensive strategy to enhance decisions to engage in behaviors that run counter to organizational or societal expectations. Societal ambivalence toward creativity (Staw, 1995) forces the development of creative personality to exist, in part, at the implicit level. Individuals who make creative advancements often rationalize their behaviors not as creative but as the “right” thing to do (Csikszentmihalyi, 1996; Dunn, 2015).
Toward a New Theory of Creative Personality
Creativity undoubtedly includes a cognitive ability component given that creative individuals entertain insights and ways of thinking that others do not (Bink & Marsh, 2000; Guilford, 1967). Creative endeavors possess a motivational component as well, as individuals must also make efforts to implement their ideas if they hope to make a contribution. The mere conception of a novel idea does not lead individuals to be viewed by others as creative unless they also act in some way (even such limited action as voicing their view) to pursue the idea. Thus, the motivation to engage in creative activity is also an aspect of creative personality. 1 As novel ideas are meaningless in the absence of action, the underlying driving force of the individual must work in conjunction with the ability to develop creative ideas before an individual may be viewed as creative. We therefore propose that creative individuals possess a distinctive personality structure that engenders sustained creative action.
An indirect measurement system is needed to capture the implicit creative personality we have described. The CR framework, which provides both a measurement technique and a theoretical paradigm, is most conducive to our construct of interest. CR is used to describe how individuals automatically frame information around them, quickly and implicitly rationalize their actions, and defend the decisions they make, which may include the decision to engage in creative action. Given that we conceptualize implicit creative personality as partly motivational, CR is especially advantageous in identifying how individuals rationalize creative behaviors (James & LeBreton, 2012; Uhlmann et al., 2012).
The cornerstone of the CR theory of personality is the use of JMs. Individuals develop JMs in order to “enhance the logical appeal” (James, 1998: 131) of their decisions to engage in certain behaviors over others. JMs align with each individual’s latent desires in such a way that individuals with particular underlying desires likely develop similar methods for rationalizing their preferred behaviors, while individuals with different underlying desires develop different rationalizations or JMs (James, 1998). Social information is inherently ambiguous and can be interpreted in a variety of ways. Thus, individuals who use different JMs differentially interpret the information provided by social situations and, based upon these interpretations, may act differently from one another. JMs are applied automatically, with little apparent effort, and outside of conscious awareness (James, 1998; James & LeBreton, 2012).
JMs for Creative Personality
To identify JMs, James and his colleagues (James, 1998; James & LeBreton, 2012) recommend consulting numerous sources. We reviewed biographies and biographical summaries about creative individuals (e.g., Csikszentmihalyi, 1996; Dunn, 2015; Isaacson, 2011; Vance, 2015), fictional works about creative characters (e.g., Rand, 1943), academic descriptions of creativity, creators, and organizational radicals (e.g., Amabile, 1996; Bandura, 1997; Guilford, 1950; Meyerson & Scully, 1995), theoretical descriptions of the need for cognition (Cacioppo & Petty, 1982) and learning orientations (e.g., Dweck, 1975), and reality television shows depicting individuals engaged in creative activities (e.g., Weinstein et al., 2011). In sum, five JMs were identified (see Figure 1). Unlike aggression and achievement motivation, previously examined using implicit techniques (McClelland, 1987; Murray, 1938), we relied on inductive top-down theorizing (Shepherd & Sutcliffe, 2011) to develop these JMs. The five JMs described herein likely do not represent an exhaustive list, and we continue our efforts to extend our understanding of creative personality and the identification of JMs that support creative efforts. The five JMs identified to date are the impact bias, the exclusivity bias, the novelty appreciation bias, the efficacy of tenacity bias, and the malleability of social norms bias. Brief descriptions are presented in Figure 1, and detailed descriptions are provided in the online supplement. These JMs allow creative individuals to rationalize working diligently for extended periods on enigmatic problems while defending their efforts against burnout and perceptions of foolishness. Reliance upon these JMs additionally enables creative individuals to discount the status quo while also taking pride in working on projects that others may deem as unique, obscure, or unnecessary.
JMs for Creativity
JMs are not factors in the sense of the traditional factor analytic model often used in the organizational sciences (e.g., the FFM). Individuals may use multiple JMs to rationalize their behaviors, and different individuals may use different sets of JMs. CR items are often written to evoke the use of multiple JMs, and situations may evoke different JMs in different people. Thus, JMs may be regarded as theoretical tools for conceptualizing the way in which individuals rationalize their behavior as well as for item development (we provide three sample items with explanations in the online supplemental). The manner in which JMs combine into potentially separable factors is an area that is in the exploratory stages in CR research (see James & LeBreton, 2012). We provide a preliminary exploratory factor analysis (EFA) of the CRT-CP in Study 5 in the online supplement.
Hypotheses
Individuals do not have access to their implicit reasoning and judgment processes (Greenwald & Banaji, 1995; Haidt, 2001; James, 1998; McClelland, 1987; Nisbett & Wilson, 1977); thus, implicit and explicit personality are distinct entities that are unlikely to correlate highly (Frost et al., 2007; James & LeBreton, 2012; Uhlmann et al., 2012). 2 Although in some cases self-perceptions may concur with implicit biases, more frequently they are incongruent. Even when implicit and explicit aspects of personality are congruent, individuals rarely possess true insight into their own reasoning processes, and there is little theoretical basis for these aspects of personality to be interrelated. Thus, we propose that implicit and explicit measures of creative personality will not correlate highly (Frost et al., 2007; Greenwald & Banaji, 1995; Haidt, 2001; James, 1998; James & LeBreton, 2012; McClelland, 1987; Murray, 1938).
Hypothesis 1: Scores on measures of implicit and explicit creative personality will not be highly correlated.
It is well established that personality motivates behavior via exhibition of motivated behaviors such as effort, direction, and persistence and that motivation results in a greater likelihood of high performance (Kanfer, Chen, & Pritchard, 2008). Along the same lines, it follows that an individual’s implicit creative personality, supported by the previously discussed JMs, is related to greater effort in creative endeavors and thus greater creative performance.
Hypothesis 2: Scores on the CRT-CP will be positively correlated with assessments of creative performance.
The self-report nature of current measures of creative personality renders them susceptible to a number of potential issues (e.g., Podsakoff & Organ, 1986). The influence of social desirability is of particular concern, and resulting response biases along with the limited theoretical basis for current explicit measures of creative personality (cf. Block, 1995) are likely culprits of the modest correlations between explicit measures of creative personality and creative performance (Schoen, 2015). Implicit measures circumvent issues such as social desirability and faking (LeBreton, Barksdale, Robin, & James, 2007) and typically exhibit higher correlations with the criteria of interest than do explicit measures (James & LeBreton, 2012).
In addition to methodological issues, self-report assessments of creative personality may not be highly related to creative performance for theoretical reasons. Self-reported personality represents identity or self-presentations (Hogan, 2005; James & LeBreton, 2012). Crutchfield (1962) argues for differences between task-involved and ego-involved motivation for creativity. Ego-involved individuals (i.e., those who perceive themselves to be creative) are “too self-consciously bent on being a ‘creator’ [and are] not apt to succeed” (Crutchfield, 1962: 122). Thus, people who self-report as being highly creative may view themselves in this way and attempt to align their behaviors with these self-views. They may develop solutions that are novel but not particularly useful (i.e., bizarre) and thus may be unlikely to be viewed as creative. In contrast, people who work in a more task-involved way, driven by their biased views that existing solutions are idiosyncratic, change is necessary, and persistence to achieve an ideal state is required (i.e., those with an implicit creative personality), may exhibit high levels of creativity. Research suggests that explicit aspects of creative personality are positively related to creative performance; however, implicit aspects of creative personality, because they lead individuals to be more task-involved, are likely to be stronger predictors of creative performance. Indeed, in a study by Schoen (2015), an explicit assessment of achievement motivation did not predict creative performance, whereas an implicit measure of achievement motivation did.
Although we have primarily argued that creative performance is a time-consuming process, we still expect differences in implicit creative personality to be associated with variation in creative performance of individuals as assessed over both shorter and longer periods of time. Individuals with a creative personality will likely have had some (and possibly considerable) experience with creativity; thus, when given tasks that provide latitude for creative performance, even short-duration projects, they are more likely to be creative. Implicit processes are assumed to be automatic and fast-acting, affecting both proximal and distal performance, whereas creativity self-views often operate better over longer time spans as individuals work to align their self-views with their behaviors (e.g., Helson, Roberts, & Agronick, 1995). In sum, we propose the following:
Hypothesis 3: Scores on the CRT-CP will better predict creative performance than explicit creative personality.
Development of Creative Personality
The cultivation of a creative personality is complex and likely follows a developmental cycle. Variability in creative performance, creative ability, and creative personality are likely the result of a self-reinforcing cycle in which developments and advancements in one area support innovations in other areas (Schoen, DeSimone, & James, 2011). Thus, creative performance, creative ability, and creative personality should be interrelated. To illustrate this developmental and self-reinforcing process, consider the example of a child who is likely to engage in a number of artistic exercises as part of his or her early education. If this child demonstrates a special talent and is praised for these novel efforts, he or she may find such approval quite rewarding and may decide to continue to work in the arts, frame it as enjoyable, and pursue this path to the exclusion of other activities (e.g., sports, academics, etc.). Investing time and effort in artistic activities is likely to be rationalized by this individual as enjoyable, rewarding, interesting, and fun. Engagement in creative activities establishes the foundation for further creative skill refinement and development of cognitive structures that support creativity in other areas.
Whether it be art, music, technical ability, or another domain, individuals who pursue novel and creative efforts are likely to learn fairly early that peers, family, and other authority figures scrutinize such unconventional activities. Even with a supportive family environment, social pressures may drive individuals toward mainstream behaviors. Engaging in artistic activities or music is often regarded as activities for children or viable only as hobbies. Career choices in those fields are viewed as limited, unstable, and are associated with a low likelihood of success. Furthermore, scientific endeavors allowing for creativity require arduous advanced training, and science, technology, engineering, and mathematics educations are often stigmatized as nerdy or geeky. When individuals experience success in a domain in which creativity is expected because of their creative abilities, they then likely develop JMs to support their continuous pursuits. Thus, the cognitive abilities that support creativity and reliance upon relevant JMs are mutually reinforcing (Schoen et al., 2011) and should thus be correlated.
If this developmental model is correct and similar to that of constructs like intelligence and achievement motivation, which are related (James, 1998; Schoen et al., 2011), then cognitive abilities related to creativity and creative personality should be related to one another. Two cognitive aspects are prevalent in the study of creativity: divergent thinking, which is the generation of novel ideas, and convergent thinking, which is the consolidation of ideas (McCrae, 1987; Mednick, 1962). It is thus expected that the implicit assessment of creative personality will be significantly related to both of these creativity-relevant cognitive abilities.
Hypothesis 4: Scores on the CRT-CP will be positively correlated with divergent thinking ability.
Hypothesis 5: Scores on the CRT-CP will be positively correlated with convergent thinking ability.
Creative Personality and Entrepreneurship
The concept of an entrepreneurial personality has been of interest in the organizational sciences for decades (Kets De Vries, 1977). An FFM-based entrepreneurial profile composed of high levels of extraversion, conscientiousness, and openness and low levels of agreeableness and neuroticism has met with some success in predicting entrepreneurs’ status, career satisfaction, and gross income (Schmitt-Rodermund, 2004; Zhao & Seibert, 2006). Implicit assessments of achievement motivation have also been linked to entrepreneurial success (Zhao & Seibert, 2006). Creative behavior is frequently regarded as a critical component of entrepreneurship (Zhao & Seibert, 2006), and a high level of openness to experience, also associated with creative personality (McCrae, 1987), is part of the entrepreneurial profile. Our conceptualization of implicit creative personality, which includes (a) a strong association between persistence and success, (b) enjoyment of working on unconventional tasks and projects, and (c) a desire to make an impactful contribution, suggests that this construct may predict entrepreneurial success. Thus, we expect that individuals with higher levels of implicit creative personality will be more likely than others to seek out entrepreneurial activities and to garner greater entrepreneurial success.
Hypothesis 6: Scores on the CRT-CP will be higher for individuals engaged in entrepreneurial activities than for individuals not engaged in entrepreneurial activities.
Hypothesis 7: Scores on the CRT-CP will be correlated with assessments of entrepreneurial success.
Present Studies
Thirty items were initially developed for the CRT-CP. Participants in Study 1 were undergraduate students in a regional public college in the southeastern United States. This study assessed item and scale properties and served as an initial test of Hypotheses 1 through 5. A few of the initial items did not function as intended; thus, new items were developed for inclusion in a revised version of the CRT-CP. Study 2 participants were drawn from the same population as Study 1, and this study assessed item and scale properties of the revised CRT-CP and served as a confirmatory test of Hypotheses 1 through 5. Study 3 consisted of individuals competing in one of three different entrepreneurship competitions, and data were used to test Hypotheses 6 and 7. Study 4 participants were recruited from Amazon Mechanical Turk (MTURK), and data served to test the relationship between the CRT-CP and an implicit measure of achievement motivation. Finally, Study 5 included participants from Studies 2 and 3 and an additional sample from MTURK, and relevant data were used to conduct an EFA of the revised CRT-CP.
Placing implicit measures within a nomological network is not a straightforward process as implicit scales are not expected to correlate with explicit scales of similar constructs. The major focus shifts to criterion-related validity (James, 1998); thus, individual item properties such as item-scale correlations, proportion of individuals who select various responses to the items, and scale reliability take precedence. As implicit and explicit measures are not expected to be highly correlated, demonstration of discriminant and incremental validity are also important. To clarify the presentation of samples and results, an overview of all five studies can be found in Table 1. Details regarding Studies 4 and 5 are included solely in the online supplement.
Summary of Study Results
Note: CRT-CP = Conditional Reasoning Test for Creative Personality; CPS = Creative Personality Scale; DVs = dependent variables; MTURK = Amazon Mechanical Turk; CRT-RMS = Conditional Reasoning Test for Relative Motive Strength.
Study 1
Study 1 Method
Participants
Individuals were recruited from undergraduate courses in algebra, statistics, principles of management (PM), and organizational behavior (OB). Participants received either course credit as part of a research requirement (PM and OB courses) or course extra credit (algebra and statistics). Of approximately 220 students asked to participate, 159 (72.3%) participated in some aspect of the study, and 54% of the sample was female. Other demographic information was not collected; however, the regional college in which these data were collected is diverse and serves a large nontraditional student population. Most of the students are employed on a part- or full-time basis. Information on gender was collected in this sample, and correlations are reported in Table 2. Gender was tested as a control, but it did not substantively alter the findings and is not discussed further.
Study 1 below the diagonal and Study 2 above the diagonal. Study 1: N = 159 (146 for implicit creative personality, 140 for GPA, 53 for divergent thinking, and 82 for creative performance; small variations exist within these sample sizes based upon missing data. Mean replacement was used where possible). Study 2: N = 122 (107 for implicit creative personality, 79 for divergent thinking, 60 for riddles, 92 for Duncker, and 55 for creative performance; small variations exist within these sample sizes based on missing data; mean replacement was used where possible).
Reliabilities are in parentheses on the diagonal when applicable (Study 1, Study 2).
Gender: 0 = female, 1 = male.
Individuals who completed the divergent thinking task did not complete the creative performance task for Study 1.
Ratings for Study 1 only.
p ≤ .10.
p ≤ .05.
p ≤ .01.
Procedure
Participants were first asked to complete several test instruments online and later asked to participate in a laboratory session. There was at least a 2-week gap between the online component and participation in the laboratory session. Participants were asked to take part in one of two different lab studies depending upon the course from which they were recruited.
Measures
The online instruments consisted of the CRT-CP, several self-report scales, and a convergent thinking task. The CRT-CP contains 30 items, with the aim being to assess the primary content using six items for each JM (secondary and tertiary content coverage was not controlled). A 30-minute time limit was enforced for the CRT-CP, and the questionnaire proceeded automatically if the time limit was reached. This manipulation serves to enhance the plausibility of CRTs as assessments of inductive reasoning ability (James, 1998). Self-report personality instruments included (a) openness from the FFM, (b) intellectance from the Hogan Personality Inventory, and (c) the CPS. Included outcome measures were (a) the Creative Behavior Inventory (CBI) and (b) the Remote Associates Test (RAT). Openness (α = .85) and intellectance (α = .69) were assessed using 20-item and 10-item scales, respectively, from the International Personality Item Pool (IPIP). Items were rated using a 7-point scale (1 = strongly disagree to 7 = strongly agree). A sample openness item is “I have a vivid imagination,” and a sample item from the intellectance scale is “I like to solve complex problems.” The CPS is a 30-item checklist consisting of 18 positively and 12 negatively scored adjectives in which a higher score indicates a higher level of creative personality (Gough, 1979). The CBI (α = .90) is a self-report scale that consists of 28 objective behaviors indicative of past creative activities (Hocevar, 1979). Items are rated on a 4-point scale (1 = never did this, 2 = did this once or twice, 3 = 3-5 times, and 4 = more than 5 times), and a sample behavior is “published literature.” Finally, participants completed 10 items from the RAT to assess convergent thinking ability (Mednick, 1962). RAT items consist of three words, and participants are asked to supply a fourth word that links the other three together. An example is “cottage, blue, and mouse.” The answer is “cheese.” RAT items were scored “1” if correct and “0” if incorrect or left blank, and a total score was computed by summing the correct responses. The RAT was restricted to 10 minutes, and the questionnaire proceeded automatically if this time limit was reached.
Participants from algebra and statistics courses were asked to take part in a laboratory study with a divergent thinking task, and those from PM and OB courses were asked to complete an open-ended in-basket exercise, each lasting 30 minutes. Of the 159 individuals who took part in the online part of the study, 135 (84.9%) completed a laboratory session (divergent thinking, n = 53; in-basket exercise, n = 82). In the divergent thinking exercise, participants were asked to develop ideas for improving a refrigerator, expanding campus parking (an ongoing problem), and rewarding employees. Participants were encouraged to work on the problems in any order they wished and to switch between tasks as necessary in order to maintain a sense of autonomy and self-determination. The number of ideas generated across the three problems was averaged to yield the dependent measure for divergent thinking.
Participants from the PM and OB courses, representing more advanced students, were asked to complete three memos from an open-ended in-basket work simulation. This exercise, used in a number of creativity studies (Schoen, 2015; Shalley, 1991, 1995; Shalley & Perry-Smith, 2001; Zhou, 1998), allows for complex problem solving and was designed for the assessment of creative performance (Shalley, 1991, 1995). Individuals are asked to assume the role of a human resources management director and respond to a series of issues. The issues may be addressed in various ways and do not necessarily require a strict human resources interpretation (for more detail on this task, see Schoen, 2015; Shalley, 1991, 1995), and performance was assessed using the Consensual Assessment Technique (Amabile, 1982). Three raters with a PhD in either business or Industrial Organizational (I/O) psychology or with business experience rated creativity using a scale ranging from 1 (not at all creative) to 7 (exceptionally creative). 3 Agreement was assessed using rwg (James, Demaree, & Wolf, 1984), and acceptable agreement was attained for the ratings across all memos and all participants (rwg = .85).
Two supplemental measures were also collected. First, as anecdotal evidence suggests that creative students may outperform noncreative students in the secondary educational system (Bain, 2012), college GPA was obtained via academic records. Second, ideas generated from divergent thinking tasks were assessed for creativity (Friedman & Förster, 2005). Correlations, means, and standard deviations for Study 1 are displayed below the diagonal in Table 2.
Study 1 Results
CRT-CP
Respondents who select too many distractors may not sufficiently comprehend the items or may not be taking the test in earnest. Distractor items are intended to be easily excluded and are rarely selected (see LeBreton et al., 2007). CR testing guidelines suggest excluding individuals who select five or more distractors (James & McIntyre, 2000); however, the original CRTs (the Conditional Reasoning Test for Relative Motive Strength [CRT-RMS] and Conditional Reasoning Test for Aggression [CRT-A]) contained fewer items. Five items from the original CRTs are about one quarter of these tests and represent a reasonable rule of thumb (LeBreton, personal communication, August 5, 2015); thus, a cutoff of seven was deemed a reasonable tradeoff between power and excluding random or incomplete responses and represented approximately one quarter of the test. As a result, 14 participants were excluded. 4
Of the 30 items developed, 23 displayed acceptable scale properties based on the same forms of analysis set out by James (1998) for the original CRTs. Because items are dichotomously scored, item-scale correlations were assessed via a biserial correlation. In total, seven items were rejected. Five of these items exhibited negative or small (rbis = .07) item-scale correlations. Two additional items were rejected that exhibited negative correlations with divergent or convergent thinking or with the dependent measures (i.e., creativity assessments). These negative correlations are especially problematic considering that predictive validity is a primary consideration when using the CR methodology (see James, 1998, p. 157 point 2.). Finally, James (1998) suggested excluding items whose keyed responses were selected by fewer than 5% or more than 95% of respondents. Of the seven excluded items, one was eliminated for this reason, and it also exhibited negative correlations with other variables. Two items selected by 96% of participants were retained because of other positive item qualities. Thus, the final scale contained 23 items, and with the exception of two items, all item-scale correlations were .27 or greater with an average of rbis = .40. The scale reliability for the 23 retained items, using the KR-20 formula suggested by James and LeBreton (2012), attained an acceptable level of .75.
Hypothesis tests
Hypothesis 1, which suggested that measures of implicit and explicit creative personality would not correlate highly, was supported. The correlation between the CRT-CP and intellectance (r = .22) was significant but did not meet Cohen’s (1992) definition of a medium correlation. This was also true for the CRT-CP and openness (r = .18) and the CRT-CP and the CPS (r = .18). These results support James and LeBreton’s (2012) claim that implicit and explicit assessments of the same construct often exhibit positive but small correlations.
Hypothesis 2 suggested that the CRT-CP would be positively correlated with creative performance. Based on the correlations presented in Table 2, this hypothesis was fully supported. Correlations between the CRT-CP and creative performance on the in-basket exercise (r = .27), between the CRT-CP and the CBI (r = .33), and between the CRT-CP and the supplemental creativity ratings of the idea generation task (r = .29) all provided support for Hypothesis 2.
Hypothesis 3 proposed that the CRT-CP would better predict creative performance than explicit assessments of creativity. The results of the hierarchical regression, shown in Table 3, confirmed this hypothesis. Addition of the CRT-CP to the regression was significant over openness, intellectance, and the CPS for both the in-basket exercise ratings and the CBI as the dependent measures. Supplemental analysis of creativity ratings in the divergent thinking exercise (not shown) also provided support for Hypothesis 3, as the CRT-CP added significantly to the model over openness and was marginally significant over intellectance and the CPS.
Hierarchical Multiple Regression Results for Hypothesis 3 in Study 1
p ≤ .10.
p ≤ .05.
p ≤ .01.
Hypotheses 4 and 5 suggested that scores on the CRT-CP would be positively correlated with divergent and convergent thinking abilities, respectively. Hypothesis 4 was partially supported, as the correlation (shown in Table 2) was in the correct direction but was marginally significant (r = .26), likely because of the small sample size. The correlation between the CRT-CP and the RAT was positive and significant (r = .25). Thus, Hypothesis 5 was supported.
Finally, the supplemental analysis exploring the correlation between implicit creative personality and school performance was supported, as there was a modest but significant correlation between the CRT-CP and GPA (r = .18).
Study 1 Discussion
Study 1 provided an initial test of the CRT-CP, and all of the tested hypotheses were either fully or partially supported. While some of the correlations were less robust than one might hope based upon the results of other CRTs (i.e., James et al., 2005), a consistent pattern of results nevertheless emerged. These preliminary results suggest that our theory represents an advancement in the ability to predict creative performance. A revision of the CRT-CP and second data collection were undertaken in order to both confirm and improve upon the findings from Study 1.
Study 2
Study 2 Method
Participants
Participants in this study were recruited from undergraduate PM and OB courses and received course credit as part of a research requirement. Of the 210 students solicited, 134 (63.8%) took part in the first part of the study. The sample was 55% male, and the average age was 27.4 years. Among the participants, 47% identified as White or Caucasian, 30% as Black or African American, 6% as Asian, 6% as Hispanic or Latino, and 2% as mixed race (the remaining 9% did not disclose this information). Correlations with age and gender are reported above the diagonal in Table 2. Age and gender were tested as controls but did not substantively alter the findings and are not discussed further.
Procedure
Participants were first asked to complete several test instruments online. The course instructors incorporated a number of exercises into their normal course activities for this study. PM students completed a smaller subset of exercises than the OB students, and participation levels varied based upon class attendance.
Measures
The online instruments consisted of the revised CRT-CP, several self-report scales, and a convergent thinking task. A time limit of 30 minutes was again imposed for the CRT-CP. Openness to experience (α = .86), intellectance (α = .77), the CPS, and the CBI (α = .92) were all assessed similar to Study 1. The RAT was also assessed; however, 20 items were featured instead of 10; thus, the time limit was extended to 20 minutes from 10 minutes.
Students in the PM classes were given an idea generation task to assess divergent thinking, along with Duncker’s (1945) candle problem, a well-known visual insight problem believed to assess convergent thinking. The divergent thinking exercise was similar to Study 1 but featured different primes (e.g., “list as many uses for a brick” and “What would happen if people didn’t need to sleep?”). Students in the OB classes received the same set of exercises, as well as the in-basket exercise used in Study 1 and a set of 10 riddles, verbal insight problems, intended to assess convergent thinking (DeYoung, Flanders, & Peterson, 2008; Duncker, 1945). A sample riddle is: “A man has married 20 women in a small town. All of the women are still alive and none of them are divorced. The man has broken no laws. Who is the man?” (The man is a preacher/priest.) Students worked individually on the exercises and were given 10 minutes to work on the idea generation task, 5 minutes to complete Duncker’s candle problem, 10 minutes to solve the riddles, and 20 minutes to finish the in-basket exercise. Duncker’s candle problem has a fixed solution; hence, students scored a 1 if they solved it correctly and 0 if they did not. The riddles also have fixed solutions; thus, each correctly solved riddle was scored as a 1, and all others received a score of 0. Scores on the correct answers were then summed to form a scale. Similar to Study 1, performance on the in-basket exercise was assessed using the Consensual Assessment Technique (Amabile, 1982). Two raters either with a PhD in business or with business experience rated the creativity of the responses to the memos using a scale ranging from 1 (not at all creative) to 7 (exceptionally creative). 3 Agreement was assessed using rwg (James et al., 1984), and acceptable agreement was attained across all memos and all participants (rwg = .84). GPA was collected for supplemental analysis. The correlations, means, and standard deviations for Study 2 are presented above the diagonal in Table 2.
Study 2 Results
CRT-CP
In keeping with CR testing protocol, 15 participants were excluded because they selected too many distractors or omitted too many questions on the CRT-CP. An additional 12 participants were excluded because they had participated in a CRT study the previous semester and had been informed about the nature of CRTs during the debriefing. CRTs are suggested to no longer act as implicit tests of personality if individuals are aware of their nature (LeBreton et al., 2007). Although a sensitivity analysis showed that no appreciable differences resulted by including or excluding the 12 individuals who had participated previously, these data were excluded, which provided for more statistically conservative tests.
Of the 30 items in the revised version of the CRT-CP, 29 had acceptable scale properties. Of these, 23 of the items were those retained for analysis in Study 1, and seven new items were also included. The additional items were developed in an effort to ensure coverage of the five JMs. As with Study 1, item-scale correlations and other theoretically expected relationships were explored. One newly written item for the revised version was rejected, as it exhibited a negative item-scale correlation and negatively correlated with other variables in the analysis.
As noted in Study 1, James (1998) suggested dropping items if fewer than 5% or more than 95% of participants selected the keyed response. We retained two items from the original CRT-CP that were selected by 96% of participants. These two items were selected by a large proportion of individuals in Study 2 but displayed the expected regression to the mean and thus met the criteria set out by James, as did the other 27 items. The scale reliability for the 29 retained items, using the suggested KR-20 formula from James and LeBreton (2012), was .82.
Hypothesis tests
Hypothesis 1 suggested that the measures of implicit and explicit creative personality would not correlate highly. Based upon the correlations presented in Table 2, this hypothesis was supported. The CRT-CP’s correlations with intellectance (r = .21), openness (r = .23), and the CPS (r = .05) were similar in magnitude to those in Study 1.
Hypothesis 2 suggested that the CRT-CP would be positively correlated with creative performance. Again, this hypothesis was supported based on the correlations presented in Table 2. Correlations between the CRT-CP and rated creative performance on the in-basket exercise (r = .31) and between the CRT-CP and the CBI (r = .33) were both positive and significant.
Hypothesis 3 suggested that the CRT-CP would be a better predictor of creative performance than explicit assessments of creative personality. This hypothesis was tested using hierarchical regression, shown in Table 4, and was also supported. Addition of the CRT-CP into the regression equation was significant over openness, intellectance, and the CPS when both the in-basket exercise ratings and the CBI were used as the dependent measures.
Hierarchical Multiple Regression Results for Hypothesis 3 in Study 2
p ≤ .05.
p ≤ .01.
Hypotheses 4 and 5 suggested that CRT-CP scores would be positively correlated with divergent and convergent thinking abilities, respectively. As shown in Table 2, results showed that the CRT-CP was positively related to the divergent thinking exercise (r = .23), offering support for Hypothesis 4. Similarly, results confirmed a positive relationship between the CRT-CP and the RAT (r = .25); thus, Hypothesis 5 was also supported. The correlation between the CRT-CP and the number of riddles solved was positive but only marginally significant (r = .23), likely because of the small sample size. Additionally, the correlation between the CRT-CP and success in solving Duncker’s candle problem was positive but not significant (r = .16). Finally, the supplemental analysis exploring the correlation between implicit creative personality (i.e., the CRT-CP) and school performance (i.e., GPA) was supported (r = .27).
Study 2 Discussion
Study 2 provided a partial replication of the evaluation of the CRT-CP with a revised scale. Hypotheses 1 through 5 were fully supported. The revised scale displayed a higher level of reliability (KR-20 = .82) and comparable or slightly better predictive capability than the original CRT-CP. As with the first study, although some sample sizes were small, a consistent pattern of results emerged across multiple measures consistent with our theory, thus suggesting that an important implicit component of creative personality has gone unexplored until now. The following study further explores the predictive and discriminant validity of the CRT-CP.
Study 3
Study 3 was conducted primarily to test Hypotheses 6 and 7. Data were collected from individuals engaging in entrepreneurship competitions outside of the college setting. Although some individuals were students, these data primarily represent a nonstudent sample.
Importantly, the entrepreneurship competitions were targeted toward actual entrepreneurship rather than small business or lifestyle business ownership. One of the authors attended sessions for two of the three entrepreneurship competitions, without being identified as a researcher, to verify that the business pitches in these competitions consisted of novel products, processes, or business formats with high growth potential and high investor interest. One of the competitions specifically included possible investors in the judging process, and these potential investors evaluated and provided feedback on all of the proposals. Information on age and gender was collected in this sample, and correlations are reported in Table 5. Inclusion of age did substantively alter the results. The effects of age in this model are described further in the Discussion section, and additional analyses are available from the first author.
Means, Standard Deviations, and Correlations Among Study Variables for Study 3 a
Note: N = 43 (39 for implicit creative personality and 40 for the FMM; small variations exist within these sample sizes based upon missing data; mean replacement was used where possible). aReliabilities are in parentheses on the diagonal when applicable. bGender: 0 = female, 1 = male. cOwnership Status: 0 = have not started a company, 1 = started a company.
p ≤ .10.
p ≤ .05.
p ≤ .01.
Participants
Approximately 250 individuals were recruited from three entrepreneurship competitions, and 43 (17.2%) participated in the study. The participants were predominantly male (55%), and the average age was 35.6 years. The racial makeup of the sample was 51.2% White or Caucasian, 18.6% Black or African American, 4.7% Asian, 7.0% Hispanic or Latino, 1.5% mixed race, and the remaining 13.8% declined to disclose their race.
Procedure
Individuals were asked to complete an online questionnaire by the directors of the different entrepreneurship competitions. Participation was voluntary and respondents were automatically entered into a drawing for one of two cash prizes of $200. Participants could also select whether they wished to learn their scores on the online questionnaire, and 91% elected to receive this information. Individuals were given 2 months to complete the study; two of the three competitions sent reminders about the questionnaire every 2 weeks, and one competition only sent the initial request. In the online questionnaire, individuals first completed the CRT-CP (limited to 30 minutes), a self-report measure of personality, objective and subjective questions about their entrepreneurial efforts, and lastly, questions assessing demographic information.
Measures
Personality instruments included the revised version of the CRT-CP and the FFM from the IPIP that featured a 5-point scale (1 = strongly disagree to 5 = strongly agree). The FFM consists of openness (α = .82), extraversion (α = .89), conscientiousness (α = .80), agreeableness (α = .82), and neuroticism (α = .87), each assessed with 10 items. Sample items include “feel comfortable around people” (extraversion), “have a good word for everyone” (agreeableness), “am always prepared” (conscientiousness), “I have a vivid imagination” (openness), and “insult people” (neuroticism). The questions about entrepreneurial activities were based on previous entrepreneurship research (Schmitt-Rodermund, 2004; Zhao & Seibert, 2006). A sample objective question is “Do you currently own a business?” A sample subjective question is “How satisfied are you with the economic development of your business?” assessed using a 7-point scale ranging from 1 (totally and completely dissatisfied) to 7 (totally and completely satisfied). Correlations, means, and standard deviations are displayed in Table 5.
Study 3 Results
CRT-CP
The standard exclusion criteria were again applied to the CRT-CP, and the same set of 29 items were scored as in Study 2. As a result, four individuals were excluded from the analyses based on excessive selection of distractors. The KR-20 reliability was .85.
Entrepreneurship profile
An entrepreneurial profile suggested to relate to entrepreneurial activities was developed from the FFM (Brandstätter, 2011; Schmitt-Rodermund, 2004). This profile is composed of high levels of openness, extraversion, and conscientiousness and low levels of agreeableness and neuroticism. The entrepreneurial profile was computed by reverse-scoring agreeableness and then averaging across all items (Schmitt-Rodermund, 2004).
Hypothesis tests
Hypothesis 6 suggested that individuals engaged in entrepreneurial activities would display higher CRT-CP scores than those not engaged in such activities. To test this hypothesis, mean CRT-CP scores of participants from Studies 2 and 3 were compared, both of whom completed the revised version of the CRT-CP. The mean difference (means reported in Tables 2 and 5, respectively) was significant, with the entrepreneurship sample exhibiting higher scores than the student sample, t(144) = 2.80, p < .01. Thus, Hypothesis 6 was supported.
Hypothesis 7 suggested that the CRT-CP would be associated with measures of entrepreneurial activities. Inspection of the correlations in Table 5 indicates that Hypothesis 7 was fully supported. Although the sample size was small, CRT-CP scores were significantly correlated with whether or not the individuals reported already owning a business (r = .44) and whether they had started more than one business (r = .37). The correlation between the CRT-CP and the number of businesses started was marginally significant (r = .30). The CRT-CP was not significantly correlated with the level of satisfaction individuals reported with regard to their entrepreneurial efforts. This study also provided additional support for Hypotheses 1 and 3 (relevant results and analyses are provided in the online supplement).
Study 3 Discussion
Creative performance and entrepreneurial success are difficult to predict (Amabile, 1996; Schmitt-Rodermund, 2004); however, interest in personality as a predictor of entrepreneurship has come back into favor after being dismissed at one time (Brandstätter, 2011). Research on the influence of personality on entrepreneurial activity and success has, as in many other areas, relied heavily upon self-reports, and effects of the FFM and the entrepreneurial profile appear to be rather small (Brandstätter, 2011). In light of the limited sample size, these results should be interpreted with some caution; however, our effect size was substantial, and evidence in support of Hypotheses 6 and 7 was found. Implicit measures do not frequently correlate highly with self-report measures of performance (James & LeBreton, 2012). That the CRT-CP did not correlate with self-reported satisfaction, although disappointing, was not unexpected.
General Discussion
We have proposed and tested a novel theory of creative personality with a new measure designed to assess creative personality at the implicit level. Our results show a consistent pattern across multiple, diverse samples and measures, thus indicating that the CRT-CP is a valid and reliable predictor of various forms of creative performance. The CRT-CP is distinct from both self-report measures of creative personality and also from an implicit measure of achievement motivation (see online supplement). Additionally, we proposed that creative personality likely stems from a developmental process including cognitive abilities and life experiences. While we could not fully test it, our results supported our developmental model as the CRT-CP was related to cognitive abilities that are critical to creativity as predicted by our theory.
James et al. (2005) reported medium to large relationships between implicit measures of personality and the behaviors they were designed to predict. Although the relationships we report between the CRT-CP and creative performance are also sizable compared to self-report, most of our effects are in the medium range by Cohen’s (1992) definition and are, on average, not quite as large as those reported by James and his colleagues (James & LeBreton, 2012; James et al., 2005). One possibility is that the ephemeral and serendipitous nature of creativity means that creative performance is driven by a myriad of factors, including luck, such that any single factor may not have a large effect. A second possibility, and one that needs to be explored further, is that we have not yet captured all of the JMs that make up the implicit creative personality, and thus, to date, the CRT-CP constitutes a partial assessment.
Theoretical Implications
First, many current models of creativity were developed to address the poor predictive capabilities of existing trait models of creative performance (e.g., Amabile, 1996; Ford, 1996; Woodman, Sawyer, & Griffin, 1993). Although these models have advanced the field, researchers have called for a renewed focus on individual differences (e.g., Shalley, Gilson, & Blum, 2009). The most prominent of these models suggests that intrinsic motivation is the most important factor in predicting creative performance (Amabile, 1996), but it has been criticized for a lack of predictive capability (e.g., George, 2007; Shalley, Zhou, & Oldham, 2004). Our theory and measure facilitate the prediction of creativity and answer both the call for new theory development in creativity and for greater empirical focus on individual differences (Anderson et al., 2014; Shalley et al., 2009).
Second, group composition (cf., Miron-Spektor, Erez, & Naveh, 2011), coworkers and networks (cf., Perry-Smith & Shalley, 2003), leadership (cf., Mumford, Scott, Gaddis, & Strange, 2002), and other factors are vital for creativity. Although environmental factors clearly play a role in stimulating creativity (Amabile, 1996; Shalley et al., 2004), our model additionally suggests that intrinsic aspects of creative individuals have not yet been thoroughly studied. Consequently, organizational scientists may be downplaying the role of personality as a predictor of creativity in groups, teams, and organizations when, with better theory and measurement, we may discover that dispositional characteristics are more pivotal than currently believed.
Third, this research highlights the importance of implicit personality as the counterpart to self-views in predicting creative potential, and ultimately actual creative performance. Thus, we argue that in order for creativity to occur and to be recognized, one must not only possess the efficacy to generate novel ideas but also be motivated to actively pursue these creative ideas.
Individuals who possess the cognitive capabilities that fuel creativity are typically inventive in a few select areas. People are unlikely to enjoy everything they try or to succeed in numerous domains; thus, they tend to focus on activities that are most rewarding to them. Time constraints preclude people from developing the domain knowledge and talent required for novel performance in a multitude of widely varying fields (Amabile, 1996; Csikszentmihalyi, 1996). Although cognitive abilities play a role in determining creative performance, these abilities are not substitutes for creative personality. As with cognitive abilities, the justifications that underlie creative personality are likely to be similar across the diverse domains in which creative endeavors take place. Thus, the JMs for creative personality may be relevant to artistic originality, scientific discovery, and creativity in other fields.
Practical Implications
Organizations frequently wish to improve their creative output and innovation (Amabile, 1996), and one of the most effective ways to accomplish this is by enlisting employees with creative talent. Many organizations would likely be reluctant to base hiring decisions on current measures of creative personality (cf., Morgeson et al., 2007), and even when creativity is used as a selection criterion, organizations frequently rely on indicators of past creative achievement (e.g., number of publications in top academic journals, number of patents, etc.). This emphasis on past achievement delays detection of creative potential until tangible manifestations have been displayed. However, once individuals become recognized for creative achievement, they may attain status, fame, or substantial incentives in their current organization, making it more difficult for competing organizations to recruit them. Thus, the capability to predict creative achievement before it has been publicly displayed offers a unique benefit by allowing organizations to identify creative individuals before they establish a track record of innovative organizational contributions. Advocating the use of CRTs for the purposes of employee selection is consistent with past research (LeBreton et al., 2007). 5 Moreover, CR measures have demonstrated resistance to faking (Bowler et al., 2013; LeBreton et al., 2007), and the increased criterion-related validity exhibited by implicit measures (James et al., 2005; James & LeBreton, 2012) is tremendously beneficial.
A second application in which the CRT-CP may prove beneficial is professional development via job placement and training. Creative individuals often find themselves in rigidly structured jobs requiring strict rule following and little latitude for deviation, thus hindering creativity. Organizations could benefit by placing these individuals in positions that provide outlets through which their creative potential may be exercised. In addition, providing creative individuals with focused training would benefit organizations, as domain-relevant knowledge appears to be an important component of creativity (e.g., Amabile, 1996; Schoen, 2015). Employees who experience low person-organization and person-job fit are at greater risk of attrition (e.g., Kristof, 1996). Thus, organizations could customize the placement and training of their implicitly creative employees in ways that optimize creativity, thereby reducing the voluntary turnover of these individuals and aiding company performance.
Future Research
Our theory suggests that implicit creative personality stems from a developmental process that likely begins at a young age and continues into early adulthood (Feist & Barron, 2003). The outcome of this developmental process is that forms of creative ability should be related to implicit creative personality. Although divergent and convergent thinking were not correlated in our sample (Study 2), both abilities were related to creative performance. The fact that both abilities were correlated with our implicit measure of creative personality and creative performance as hypothesized provides cogent support for our model. Longitudinal data collection is required to fully test our model and thus represents a future research opportunity.
Our model suggests that individuals with high levels of implicit creative personality are likely to be mismatched in some work organizations. We believe it is likely that many of these individuals, after wishing to express their creativity and experiencing frustration in their work environments, are then more likely to strike out on their own. This type of pattern may explain the relationship between age and CRT-CP scores evidenced in Study 3. In addition, older individuals have several advantages over their younger counterparts in that they have had the benefit of a greater amount of time and opportunity to start at least one business, and possibly multiple businesses. Thus, although age is related to the variables of interest in Study 3, it is likely not the causal mechanism. This represents a research opportunity as our work could be used to better understand employee fit and attrition within organizations as well as factors affecting how and why individuals become entrepreneurs.
Lastly, creative individuals may also present a double-edged sword as they are less likely to conform to social norms (Crutchfield, 1955, 1962). Based on our theory, creative individuals may view social norms as malleable and thus prove problematic in some organizations (anecdotal data abounds; see Dunn, 2015; Isaacson, 2011; Vance, 2015). In positions with little latitude for creativity, creative individuals may exhibit seemingly destructive behaviors as they attempt to make changes, improve processes and increase efficiency, and otherwise make an impact. When organizational agents attempt to constrain or reprimand creative individuals for their actions, they may not be receptive and, framing the situation through several of the biases described here, may label these individuals as unnecessarily restrictive or shortsighted. This could explain forms of counterproductive workplace behaviors not captured by other models. Our theory now helps to explain the long recognized link between creativity and nonconformity.
Conclusion
The identification of creative individuals has become increasingly vital, as creative human resources have emerged as a strategic factor in organizational success. However, current measures of creative personality have only been marginally successful in predicting creative performance. We contend that one possible explanation for this may be that existing measures have solely focused on explicit aspects of personality. The present research addresses this gap in the literature by proposing that examining implicit processes may enhance our understanding of creative personality. Moreover, we addressed another shortcoming of this research stream by establishing a solid theoretical basis for creative personality based on CR methodology. To this end, this manuscript presents a model of implicit creative personality based on five cognitive biases that support engaging in creative efforts, as well as empirical data from five samples exploring implicit creative personality and demonstrating its efficacy in predicting creativity.
Footnotes
Acknowledgements
This article was accepted under the editorship of Patrick M. Wright. Portions of this work were presented at the Southern Management Association Annual Meeting in New Orleans, LA in 2013 and the Society of Industrial Organizational Psychology Annual Meeting in Philadelphia, PA in 2015. We would like to thank several people for their assistance in helping us to bring this research to fruition, including James LeBreton, Rustin Meyer, Jennifer George, our Associate Editor J. Craig Wallace, and two anonymous reviewers. Above all, however, we believe that advancing conditional reasoning research serves to honor the memory of our good friend and mentor Lawrence R. James, whose impact in this and so many other domains will not be forgotten.
Supplemental material for this article is available with the manuscript on the JOM website.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
