A New Measure for the Assessment of Appreciation for Creative Personality

Abstract

This article introduces a new scale for the assessment of Appreciation for Creative Personality (ACP). The ACP scale is a brief 13-item forced-choice measure that assesses interindividual differences in the preference for interacting with creative people. ACP is considered an important factor of creative climate at the level of interpersonal interaction. Individuals who score high on ACP are thought to foster a creative climate in that they value creative traits in others. In two studies, the psychometric characteristics of the ACP scale were probed. The scale showed a clear unidimensional structure with evidence of good reliability and convergent, discriminant, and criterion validity. The ACP was substantially related to Big Five openness to experience, but predicted relevant criteria over and above openness, supporting the conceptual distinction between ACP and openness. In dyadic data analyses, participants’ openness to experience was significantly associated with their parents’ ACP, which shows that the ACP scale captured shared interpersonal variance. Moreover, parental ACP indirectly predicted participants’ everyday creative activities via the path of openness. These findings suggest that the ACP scale is a useful tool for the study of social-environmental climate for creativity from an interpersonal perspective.

Everybody values creativity, but does everybody appreciate creative people? They can be nonconforming, eccentric, and rebellious, and possibly not always easy to live with. Yet, creative people can only flourish in conducive environments. Different influential models of creativity emphasize the importance of environmental factors for the individual development and expression of creative potential (e.g., Eysenck, 1995; Guilford, 1950; Hennessey & Amabile, 2010; Rhodes, 1961). Calls have been made to emphasize social and cultural factors in the study of creative environments (e.g., Glăveanu, 2010), which can show considerable variation and affect creative outcomes (Chua, Roth, & Lemoine, 2015; Florida, 2002). Though elaborate models on conducive conditions for creativity have been developed for particular contexts such as organizational (Amabile, Conti, Coon, Lazenby, & Herron, 1996; Anderson & West, 1998; Ekvall, 1996), educational (e.g., Beghetto, 2010), or family settings (Kwaśniewska, Gralewski, Witkowska, Kostrzewska, & Lebuda, 2018), there is to date no context-independent account to the assessment of these conditions. Here, a scale is introduced that assesses appreciation for creative personality as an interpersonal, social-environmental factor for creativity.

All creative endeavors are embedded in a social context. The creative individual interacts with other people – be them colleagues, bosses, teachers, or family – who can vary in their appreciation for the creative personality. The immediate social-environmental conditions for creativity are usually referred to as climate for creativity (Hunter, Bedell, & Mumford, 2007). The acceptance of creative ideas, endeavors, and – ultimately – creative individuals is a key factor to understanding how beneficial or obstructive certain social surroundings are for the development or expression of creativity (MacKinnon, 1978). Though most people display positive attitudes towards creativity when being explicitly asked, creative ideas are not always viewed in a positive light (Mueller, Melwani, & Goncalo, 2012). The extent of appreciation by people in their closest surrounding can impact the creative individual dramatically, which is for instance well-document in the family context (Miller & Gerard, 1979; Simonton, 1984; see also Kwaśniewska et al., 2018). Thus, from an interpersonal perspective, appreciation for creative personality by important others can be considered a relevant social factor for people in various creative contexts, and make up a building block for the study of creative climate. Previous research on creative climate focused on the organizational, educational, and family contexts. In the following, theoretical models and research from these three contexts will be reviewed. It will be highlighted that interpersonal factors are being ascribed a central role for creative climate in different models and contexts, which motivated the construction of the Appreciation for Creative Personality scale. Finally, the construction and validation rationale of the scale will be presented.

In the workplace, a positive climate for creativity, as perceived by the creative person, is an important predictor of creative performance across different criteria, samples, and settings (Hunter et al., 2007). For instance, in Amabile’s KEYS model for creativity in organizations, the interpersonal factors encouragement of creativity (organizational/supervisory/work group), or autonomy/freedom were found to be important predictors of creative performance (Amabile et al., 1996). In a similar vein, the interpersonal factors participative safety and support for innovation were considered central determinants of creativity in West and colleagues research (e.g., Anderson & West, 1998). Similarly, the nine-factor model by Ekvall (e.g., Ekvall, 1996) emphasizes different interpersonal factors such as idea support (“ideas and suggestions are received in an attentive and supportive way by bosses and workmates”, p. 107), trust (“the emotional safety in relationships”, p. 107), or playfulness (“the spontaneity and ease that is displayed”, p. 108) as predictors of workplace creativity. Though all of these models also encompass context-specific factors that concern characteristics such as the organizational structure or specifics of the task, they have in common that they ascribe interpersonal aspects a central role for the flourishing of creative endeavors. In particular, all of these models highlight the crucial role of a safe surrounding which encourages individual spontaneity and creativity.

In the educational context, different scholars have repeatedly stressed the importance of a valuing climate for creativity (Beghetto, 2010; Cropley & Urban, 2002; Guilford, 1950). Ideally, such a climate would promote students’ intrinsic motivation, creative self-efficacy, and intellectual risk taking, while avoiding common fallacies, such as convergent teaching, or misconceptions of what creativity actually is (i.e., expecting too much; Beghetto, 2010). Unfortunately, this is not always the case. Empirical investigations showed that prospective teachers, for instance, prefer expected ideas over unexpected ideas (Beghetto, 2007). It has been observed early that there is a discrepancy between the desirable qualities of creative individuals and teachers’ views of the “ideal pupil”, which is conceived as compliant and conforming (Torrance, 1963). About 50 years later, in a similar vein, “creative” students were still viewed as less agreeable and conscientious than “good” students (Karwowski, 2010). Most strikingly, it was found that teachers prefer uncreative students, even though they explicitly endorse creativity as an educational goal (Westby & Dawson, 1995). This points to an important discrepancy between explicit cognitive attitudes towards creativity, and rather implicit affect-laden preferences for actually interacting with creative individuals. Or, as Runco (2007) put it: “No doubt they [teachers] do respect creativity, in the abstract, but not when faced with a classroom with 30 energetic children” (p. 172). The consequences of an anti-creative school climate can be drastic for the individual, as it was found that individual creativity can even increase the odds of school dropout (Kim & Hull, 2012).

Most recently, a measure of creative climate was also developed for the family context (Kwaśniewska et al., 2018). The scale comprises four theoretically derived factors, namely Encouragement to Experience Novelty and Variety (“I try to suggest to my child unconventional ways to solve problems”, p. 19), Encouragement of Nonconformism (“I am glad that my child has been taught not to break any rules”, p. 19; reverse-coded), Support of Perseverance in Creative Efforts (“I try to show my child different sides of the same situation”, p. 19), and Encouragement to Fantasize (“I encourage my child to fantasize”, p. 19). In a large sample of mothers, these factors were found to be moderately to strongly associated with the Big Five dimension of openness to experience, but less so with any other Big Five dimension. Notably, the four factors devised by the authors largely parallel those of the organizational models presented above, and also the factors that are thought to promote creativity in the classroom. They encompass an attitude that promotes autonomy and, at the same time, safety for creative endeavors. These are believed to establish a climate that is conducive to the child’s creative development. Though the scale shows evidence of good validity in terms of a match with existing theories on the factors promoting creativity, there is yet no direct evidence whether parental creative climate actually goes along with higher creativity in their children.

Taken together, the findings presented above suggest that research from different contexts (organizational, education, family) converges in that it assumes the existence of interpersonal factors, which promote the flourishing of creativity in the individual. Though expressed in different vocabularies, these factors describe the necessity of a safe and accepting interpersonal climate for the expression of creative ideas, and a tolerance for spontaneity, unconventionality, and nonconformity, which are inherent to the creative personality (Andreas, Zech, Coyle, & Rindermann, 2016; Batey & Furnahm, 2008; Eysenck, 1995; Feist, 1998). While most previous research focused on self-perceptions of climate for creativity, here, an interpersonal perspective to the study of creative climate was adopted: The Appreciation for Creative Personality scale was designed for assessing important others (such as bosses, teachers, or parents) in the surrounding of the creative individual. In this, the approach taken here is similar to that of Kwaśniewska et al. (2018), but not specific to parenting behavior. Rather, appreciation for creative personality is conceptualized as a context-independent individual difference variable.

Here, the development and analysis of a new scale for Appreciation for Creative Personality (ACP) is reported. The scale was originally developed in the course of a larger project on the selection of prospective teachers in Austria (see Koschmieder, Weissenbacher, Pretsch, & Neubauer, 2018), but is not limited to the educational context, and thus applicable to a wide range of contexts. The ACP aims to assess the personal preference for social contact with creative individuals. It is an affect-laden measure of personality, which is known to impact perceptions of climate for creativity (Karwowski, 2011) and is different from explicit attitudes, which can be biased by social desirability (particularly in the educational context; Runco, 2007; Westby & Dawson, 1995). Since the measure was constructed for the use in an admission test context, the latent constructs should not be readily transparent to test-takers in order to prevent faking. All of these goals motivated the development of the new ACP scale.

Items were devised through adaptation from similar inventories and theoretical considerations. These items reflect many of the factors reviewed above for the organizational, educational, and family contexts. In 13 forced-choice items, test-takers are requested to indicate their personal preference for interacting with individuals who display prototypically creative vs. less creative characteristics (see method for details). Evidence of reliability and validity of the ACP scale was tested in two independent samples.

In study 1, general psychometric characteristics (reliability and dimensionality) of the scale were assessed and its convergent, discriminant, and criterion validity were probed. Given the similarity of ACP with Big Five openness to experience (which includes the conceptually related facet openness to ideas), the new scale was expected to correlate substantially with openness in terms of convergent validity, but should show criterion validity over and above openness. In terms of discriminant validity, the ACP scale – conceptualized as an affect-laden measure of personality – should be relatively independent from the cognitive ability to recognize creative ideas (Benedek et al., 2016; see method for description). Finally, the ACP scale’s criterion validity was probed using criteria from the context of creative education (see method for details). It was tested whether the ACP scale can predict relevant creative education criteria above and beyond openness and further related variables, assuming conceptual specificity in terms of appreciating creative personality in others. Study 1 hence undertook a test of the psychometric characteristics as well as convergent, discriminant, and criterion validity of the ACP scale.

Study 2 probed whether the ACP scale – intended as an interpersonal measure of social-environmental climate for creativity – can actually capture interpersonal variance. For this, dyads composed of participants and one of their parents were sampled. Parents were chosen (rather than bosses, teachers, or other conceivable groups) as they represent significant others for the sample under study (see Study 2). It was investigated whether parental ACP actually impacts their children’s personality, creative potential, and real-life creative behavior. For this, the model of real-life creative behavior proposed by Jauk, Benedek, and Neubauer (2014) was adopted as a theoretical framework (see also Jauk, 2019). The model explains engagement in creative activities by variation in openness and cognitive creative potential (divergent thinking ability). The direct and indirect effects of parental ACP on their children’s openness, divergent thinking ability, and real-life creative activity were tested. It was assumed that parental ACP would impact participants’ personality (particularly openness, the arguably most distinctive personality characteristic of creativity individuals; Batey, Chamorro-Premuzic, & Furnham, 2010; Dollinger, Urban, & James, 2004) in the first place, which would then influence their engagement in creative activities (Jauk et al., 2014). Study 2 thus undertook an empirical test of the ACP’s capability to catch shared between-person variance as an interpersonal measure of social-environmental creative climate.

Study 1: scale Development

Method

Participants and procedure

Data were collected at the University of Graz between December 2015 and January 2016. Complete data were available from 224 individuals (146 women; mean age 22.26 [SD = 4.29] years) who took part in an online survey administered via Limesurvey (www.limesurvey.org). Participants had at least 12 years of schooling or a similar professional education. They gave written informed consent and took part on a voluntary basis. Students of psychology (86.2%) received course credits in exchange for their participation. The study was approved by the ethics committee of the University of Graz.

Measures

For initial item selection of the ACP scale, a pool of 37 items devised from existing self-report inventories (see next section) was used. All items were administered twice, using a regular instruction and a fake-good instruction (“now, please answer the questions in a way that leaves a good impression when applying for a job in the educational sector. Your answers do not need to reflect your actual personality.”). Moreover, the items were presented in forced-choice and rating scale response format. While both were found to show good psychometric characteristics, subsequent analyses focus on the forced-choice format, which seemed particularly suited for the admission context.

As convergent and discriminant validity measures, the German HEXACO-60 personality inventory (Ashton & Lee, 2009) and the German Big Five Inventory (BFI; John & Srivastava, 1999) as well as the previously established creativity evaluation test (CET; Benedek et al., 2016) were used. In contrast to the ACP scale probed here, the CET assesses the cognitive ability to accurately evaluate the creativity of ideas. Specifically, participants’ task in the CET is to classify answers from a divergent thinking task (for instance, alternate uses for a hat) as common (“using a hat for collecting donations”), inappropriate (“using a hat as cooking pot”), or creative (“using a hat as a Frisbee”). It was expected that the ACP, conceptualized as an affect-laden measure of personality, would be largely independent from the CET, as the cognitive ability for creativity evaluation was found to draw more upon cognitive than personality factors (Benedek et al., 2016). The CET was thus included as a discriminant validity measure.

For criterion validity assessment in the context of creative education (CE), two specifically constructed short scales were used. The first scale, in the following referred to as CE vignette, comprised vignettes of more and less creative children that were contrasted in six forced-choice format items.¹ In contrast to the ACP scale, the CE vignette scale specifically focused on the educational context: Participants were to imagine being a teacher and indicate which child they would prefer to teach. Item difficulties were between p = .23 and p = .92; SDs ranged from 0.27 to 0.50. The internal consistency of the six-item scale was α = .60 and did not change by exclusion of any single item (including the easiest one with p = .92). The corrected item-scale-correlations ranged from r = .20 to r = .44. The second measure, in the following referred to as CE attitude, consisted of seven five-point rating scale items that explicitly assessed preference for creative education (e.g., teaching a creative pupil, teaching a creative subject, discussing ideas and theories with pupils…). Item difficulties ranged from p = .49 to p = .81; SDs of these were between 0.84 and 1.39 (rating scale metric). The measure displayed an internal consistency of α = .77; exclusion of items would have lowered scale reliability. Corrected item-scale-correlations ranged from r = .29 to r = .67. The correlation between the CE vignette and CE attitude scales was r = .63, p < .001 (see Table 2), which provides evidence for their convergent validity. Taken together, the CE criterion scales differ from the ACP in that they are context-specific and rely on explicit and overt assessment of the latent construct (only CE attitude). In the validation study reported here, the CE scales were thus used to test whether the context-independent ACP scale displays evidence of validity in the educational context (assessment of prospective teachers) and across different response and measurement formats (explicit and overt).

TABLE 1 Item-level statistics of the Appreciation for Creative Personality (ACP) scale (study 1)

	Regular instruction				Fake-Good-Instruction
Item No.	p (SD)		ri-s		rCE_v		rCE_a	p (SD)
1	.64 (.48)	.38	.36	.27	.53 (.50)	.61	.12	.15
2	.43 (.50)	.49	.38	.31	.26 (.44)	.50	.19	.18
3	.61 (.49)	.41	.28	.42	.66 (.48)	.51	.15	.17
4	.74 (.44)	.27	.28	.32	.77 (.42)	.39	.13	.14
5	.49 (.50)	.38	.29	.32	.54 (.50)	.48	.16	.14
6	.71 (.46)	.36	.30	.33	.59 (.49)	.47	.18	.14
7	.57 (.50)	.59	.40	.40	.39 (.49)	.54	.19	.20
8	.67 (.47)	.54	.30	.38	.59 (.49)	.55	.19	.22
9	.60 (.49)	.61	.39	.33	.47 (.50)	.59	.17	.11
10	.63 (.48)	.21	.21	.17	.58 (.49)	.43	.17	.09
11	.39 (.49)	.32	.21	.22	.47 (.50)	.50	.15	.17
12	.74 (.44)	.23	.11	.16	.63 (.49)	.51	.14	.13
13	.79 (.41)	.33	.20	.14	.63 (.48)	.50	.18	.13

Note. Correlations above .17, .13, and .11 are significant at p < .01, p < .05, and p < .10, respectively. The item content can be found in Table A2.

TABLE 2 Descriptive statistics and intercorrelations of study 1 variables

	M (SD)	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
Appreciation for Creative Pers.
ACP (1)	0.62 (0.24)	.12	.56	.56	.11	.18	−.08	.17	.13	.03	.54	−.13	.14	.07	.02	.57
ACP fake good (2)	0.55 (0.29)		.27	.25	−.04	.05	−.05	.14	.04	−.01	.05	−.08	.10	.03	−.02	.05
Study validity measures
CE Vignette (3)	3.54 (0.38)			.63	.11	.23	−.12	.14	.06	−.02	.33	−.10	.13	.03	.04	.45
CE Attitude (4)	3.32 (0.65)				.02	.22	−.11	.16	.08	.05	.42	−.14	.15	.04	.08	.62
Creativity Evaluation Test (5)	75.17 (8.77)					.00	.15	−.08	−.07	.08	.03	.05	−.10	.05	.08	−.01
HEXACO Dimensions
Honesty-Humility (6)	3.51 (0.68)						.09	.03	.27	.10	.20	−.01	.04	.36	.09	.16
Emotionality (7)	3.33 (0.71)							−.33	.02	.01	−.04	.67	−.23	.24	−.07	−.15
Extraversion (8)	3.49 (0.70)								.06	.07	.04	−.58	.86	.17	.32	.18
Agreeableness (9)	3.20 (0.58)									−.02	.13	−.23	−.01	.69	−.03	.09
Conscientiousness (10)	3.59 (0.68)										.07	.04	.07	.10	.79	.19
Openness to Experience (11)	3.57 (0.64)											.01	.05	.02	.01	.73
Big Five Dimensions
Neuroticism (12)	2.88 (0.80)												−.43	−.15	−.16	−.12
Extraversion (13)	3.43 (0.82)													.14	.33	.16
Agreeableness (14)	3.65 (0.55)														.12	−.03
Conscientiousness (15)	3.54 (0.67)															.17
Openness to Experience (16)	3.61 (0.69)

Note. N = 224 for all variables. Correlations significant at p < .05 are printed in bold type. ACP = Appreciation for Creative Personality. CE = creative education.

Lastly, a creative attitude scale was administered. The scale has previously been constructed in the course of the project and is already implemented in the admission test for prospective teachers in Austria (why the characteristics of this scale cannot be disclosed here).

Item generation

The items of the ACP reflect short descriptions of creative persons, based on theoretical and empirical characterizations of the creative personality (e.g., Batey & Furnham, 2006, 2008; Eysenck, 1995; Feist, 1998). We generated or modified items based on existing self-report measures, which encompassed the openness subscale from the Big Five Aspects Scale (BFAS; DeYoung, Quilty, & Peterson, 2007), the Runco Ideational Behavior Scale (RIBS; Runco, Plucker, & Lim, 2001), the Creative Attitude Survey (Schaefer, 1991), and the Everyday Creativity Self-Concept Scale (Ivcevic, 2014; see also Patston, Cropley, Marrone, & Kaufman, 2017).

For the initial item selection pool, 37 items were derived in the following manner: The core content, for instance “my ideas are often considered ‘impractical’ or even ‘wild’” (from the RIBS; Runco et al., 2001, p. 399), was put in the third person “his/her ideas are often considered ‘impractical’ or even ‘wild’” (item no. 2). For the forced-choice format, then, a reverse-coded alternative was constructed in the same manner (“his/her suggestions are usually established ones, which have worked before.”). In cases where the core content appeared highly socially desirable (for instance “enjoys the beauty of nature”; adapted from the BFAS openness scale; DeYoung et al., 2007), a less desirable aspect was added (“often stops his/her car to photograph nature and is occasionally late because of it”; item no. 1) to achieve balanced item difficulty.²

The final set of 13 items (following item selection; see results section) including instructions to participants is provided in the Appendix (Tables A1 and A2). Participants’ task in the ACP is to indicate with whom of the two described persons they would prefer to have contact with. The ACP is available in a validated German version. For research purposes, an English translation (translated by two professionals), which is not yet validated, is also provided. For scoring the ACP, the forced-choice response alternatives A and B are assigned 0 and 1 points; three items need to be recoded first (see Tables A1 and A2). A total score is obtained by averaging across all 13 items.

Results

Item selection, internal consistency, and factor structure

Out of the initial pool of 37 items, 13 Items that best met a combination of several criteria were selected: (1) item difficulty between .20 < p_i < .80, (2) maximal correlation with both CE criterion variables (vignette and attitude) under (2a) regular instruction and (2b) fake-good instruction, and (3) maximal discriminatory power (item-scale-correlation). Table 1 displays item characteristics and criterion correlations. The 13-item-scale displayed an internal consistency of α = .77 which can be deemed acceptable. Exclusion of any single item would have lowered the scale’s internal consistency. The scale skewness (z = −1.68) did not deviate from normality at p < .01, the scale kurtosis (z = −2.70; p < .01) indicated a slightly platykurtic distribution. A principal components analysis yielded four components with eigenvalues > 1 which accounted for 53.14% of variance; the first component alone accounted for 27.87%. Velicer’s (1976) original and revised (Velicer, Eaton, & Fava, 2000) MAP test indicated a one-factor solution. This solution was confirmed by a confirmatory factor analysis for dichotomous indicators (WLSMV) which showed excellent fit to the data (χ²(65) = 75.48, p = .18; CFI = 0.99; WRMR = 0.77). Figure 1 depicts the confirmatory factor analysis model.

FIGURE 1 Confirmatory factor analysis model of the Appreciation for Creative Personality scale (study 1). See study 1 results for model fit information.

Validity evidence

Table 2 provides correlations between the ACP scale and convergent, discriminant, and criterion validity measures. Among the HEXACO personality dimensions, the ACP scale displayed a high correlation with openness (convergent validity) and low correlations with honesty-humility as well as extraversion. A similar pattern of results emerged for the BFI (except honesty-humility, which is not covered by the Big Five). Additional facet-level analyses using the HEXACO showed that the ACP’s factor-level correlation with openness generalized to all openness facets (aesthetic appreciation: r = .30, p < .001; inquisitiveness: r = .36, p < .001; creativity: r = .41, p < .001; unconventionality: r = .54, p < .001). The factor-level correlation of honesty-humility was driven by the subscales greed avoidance (r = .32, p < .001), the correlation of extraversion by social boldness (r = .18, p < .01). As expected, the ACP scale displayed a low and insignificant (p = .11) association with the CET (discriminant validity). Criterion correlations with the CE variables were generally high (note, however, that these were included as item selection criteria; see above).

Next, the incremental criterion validity of the ACP scale on the CE criteria over and above the HEXACO personality dimensions and the CET was investigated. For the CE vignette criterion, hierarchical multiple regression models showed that honesty-humility and openness were significant predictors, but extraversion was no longer significant when considered simultaneously (see Table 3; model 1). The CET could also explain additional variance as long as the ACP was not considered (model 2). In the final model, the ACP outperformed openness in predicting CE vignettes (model 3).

TABLE 3 Hierarchical multiple regression models for the prediction of creative education criteria (study 1)

	CE Vignette		CE Attitude
Model 1	β	p	β	p
Honesty-Humility	.20	< .001	−.02	.71
Emotionality	−.09	.20	−.01	.85
Extraversion	.10	.12	.02	.81
Agreeableness	−.04	.58	−.10	.11
Conscientiousness	−.07	.29	.19	< .001
Openness to Experience	.30	< .001	.40	< .001
	Δ R^2adj = .17, p < .001		Δ R^2adj = .21, p < .001
Model 2
Honesty-Humility	.20	< .001	−.02	.71
Emotionality	−.10	.12	−.04	.56
Extraversion	.11	.10	.02	.73
Agreeableness	−.03	.69	−.09	.16
Conscientiousness	−.08	.22	.17	.01
Openness to Experience	.29	< .001	.40	< .001
Creativity Evaluation Test	.13	.04	.18	< .001
	Δ R^2 adj = .02, p = .04		Δ R^2adj = .03, p = .04
Model 3
Honesty-Humility	.16	.01	−.04	.50
Emotionality	−.08	.16	−.03	.67
Extraversion	.04	.53	−.02	.77
Agreeableness	−.05	.40	−.10	.10
Conscientiousness	−.06	.29	.18	< .001
Openness to Experience	.04	.55	.26	< .001
Creativity Evaluation Test	.07	.22	.15	.01
ACP	.49	.00	.28	< .001
	Δ R^2adj = .16, p = .00		Δ R^2adj = .05, p = .00

Note. CE = Creative education. ACP = Appreciation for Creative Personality (ACP).

The CE attitude criterion was predicted by openness and conscientiousness in model 1. Again, the CET could explain incremental variance (model 2). Finally, all these variables and the ACP conjointly predicted CE attitude in model 3.

Faking

Under fake-good instruction, itemwise criterion correlations were reduced, but still mostly different from zero (see Table 1). The ACP displayed somewhat higher internal consistency (α = .85) under fake-good instruction than under regular instruction. There was no correlation between the ACP under fake-good and regular instruction conditions. Interestingly, the scale mean under fake-good instruction (see Table 2) was slightly lower (t₂₂₃ = 2.92, p < .01), not higher. Table 2 displays scale-level validity information. The ACP displayed some indication of criterion validity under fake-good instruction in terms of moderate correlations with the CE criteria. It was also related to HEXACO extraversion under fake-good instruction.

Discussion

Study 1 probed the psychometric properties of the new ACP scale. The short scale turned out to meet high psychometric standards in terms of internal consistency, unidimensionality, and validity. As expected for convergent and discriminant validity, the ACP was correlated mainly with the personality dimension of openness (and all of its facets) and was independent of the cognitive ability to recognize creative ideas (CET). This means that ACP was related to, but still distinct from the personality dimension of openness, and also different from the cognitive ability to evaluate creativity (discriminant validity evidence). Both of these results confirm the hypothesized nature of ACP as an affect-laden measure of personality that is related to yet separable from own creativity-related traits (openness) and the cognitive ability to recognize creativity in others. Of relevance to the use in the assessment context, faking analyses indicated that the latent construct was not readily transparent to test-takers as scores did not increase, and the scale retained some of its validity even under fake-good-instruction (which can be considered a worst-case scenario).

Most importantly, the ACP predicted relevant criteria over and above openness to experience, despite its substantial correlation with openness. This makes the ACP a useful tool for the study of creative climate at individual level: Individuals scoring higher on the ACP displayed a higher preference for teaching creative children (CE vignette) and endorsed creative education on an explicit level (CE attitude). The criterion validity of the ACP was higher for the CE vignette than for the CE attitude criterion. This might be attributed to common method variance among the ACP and the CE vignette criterion, as both used a forced-choice response format. However, this is only true for the multiple regression model (see Table 3), but not for the zero-order correlations (see Table 2). Thus, an alternative explanation might be that the CE attitude criterion, as compared to the CE vignette criterion, could be more readily explained by the personality variables included in the regression model.

As a potential limitation, study 1 demonstrated criterion validity only with respect to fictitious vignettes and attitude ratings. To further probe the validity of the ACP scale in an ecologically valid context, study 2 investigated the scale’s validity in parent-child-dyads.

Study 2: scale Validation

The main aim of study 2 was to validate the ACP using data not only from the participants themselves, but also from important others from their social environment. For this, participants’ parents’ ACP scores were used, as parents are arguably the most influential others during younger age (student sample). It was investigated whether participants’ personality, creative potential, and real-life creative behavior are correlated with their parents’ ACP scores. For this, the model of Jauk et al. (2014) was adopted as a conceptual framework, in which the engagement in creative activities is explained by openness and creative potential. In particular, this model suggests that openness, a key personality variable for creativity (Batey et al., 2010; Dollinger et al., 2004), lowers the behavioral threshold (Feist & Barron, 2003) for the exertion of creative activities, which makes openness a potential mediating factor between parental ACP and participants’ real-life creative activity. Thus, a mediation model corresponding to the one presented in Jauk et al. (2014) was tested.

Method

Participants and procedure

Study 2 used an independent sample of participants and their parents. Data were gathered between November and December 2016 in computer-based individual or small group test sessions. Participants were approached by students of the University of Graz. Participants were asked to nominate one of their parents (if possible, the one they had had more and closer contact with) to take part in a short survey. They were asked to convey parents’ email addresses; parents were then sent a link to an online survey (via Limesurvey; as in study 1).

The final sample consisted of 190 datasets including data from participants (114 women) and one of their parents (father or mother). The mean age was 21.95 years (SD = 4.40). Similar as in study 1, 96.80% of participants had at least 12 years of schooling or a similar professional education; 83.70% self-identified as students. Parent data stemmed mostly from participants’ mothers (148 datasets, or 77.90%). The mean age was 52.61 (SD = 5.40) years, 97.40% of parents had at least 12 years of schooling or a similar professional education; 88.90% of parents reported current employment. The study was approved by the ethics committee of the University of Graz.

Measures

Participants completed a German version of the Big Five Aspects Scales (BFAS; DeYoung et al., 2007). The BFAS differentiate two aspects within each broad personality dimension. With respect to the broad openness dimension, these are openness and intellect. While openness is more strongly related to cognitive processes underlying creativity, intellect is more tied to intelligence-related demands (Nusbaum & Silvia, 2011).

Creative potential was assessed using three divergent thinking tasks: alternate uses (find creative uses for a car tire), figure completion (imagine creative drawings starting from a square), and instances (what can be open). Time-on-task was three minutes for each task. Divergent thinking tasks were scored with respect to ideational fluency (number of ideas) and originality (rated originality of ideas; three trained raters). The average rating of all ideas generated during the three minutes on each task was used as an indicator of ideational originality.³ Interrater-reliability of the originality scores were .78, .70, and .72 for the alternate uses, figure completion, and instances tasks.

Intelligence was assessed using three scales (verbal, numerical, and figural) of the German Intelligence Structure Analysis (Intelligenzstrukturanalyse; Institut für Test- und Begabungsfoschung (ITB), & Gittler, 1998). The three scales were z-standardized and averaged to form a measure of general intelligence.

Finally, real-life creative activities and achievements were assessed using the Inventory of Creative Activities and Achievements (ICAA; Diedrich et al., 2018). The ICAA measures creative activities, defined in terms of everyday creative engagement, and creative achievement, defined as socially recognized creative accomplishments, across eight domains: literature, music, creative cooking, arts & crafts, sports, visual arts, performing arts, and science & engineering. The achievements scale of the ICAA is similar to the Creative Achievement Questionnaire (CAQ; Carson, Peterson, & Higgins, 2005) and is particularly suited to capture variation in accomplished samples. The activities scale captures variation in everyday creativity and can be used in nonprofessional samples (Diedrich et al., 2018). The following analyses, thus, focus on the prediction of creative activities. These may be studied either in a domain-general way (i.e., to investigate the factors determining any kind of creative activity, across domains; see also Jauk et al., 2014) or in a domain-specific way (Diedrich et al., 2018). Here, a domain-general model is presented first, and domain-specific results are reported in a complementary analysis (see results).

Participants’ parents completed the ACP scale, as described in study 1. The reliability of the 13-item scale was α = .74, similar to study 1. Figure 2 displays the factor structure, which is also similar to study 1.

FIGURE 2 Latent mediation model (study 2). See study 2 results for model fit information. Coefficients significant at p < .05 are displayed in bold type in the structural model part. All factor loadings are significant at p < .001 except ACP_10, which is significant at p < .05 (single factor loadings of openness and intellect are not displayed). ACP = Appreciation for Creative Personality. BFAS = Big Five Aspects Scale. res_open = residual term openness, res_intell = residual term intellect. ICAA = Inventory of Creative Activities and Achievements.

Results

Dyadic parent-participant correlations

Table 4 shows the associations between the study variables. Parents’ ACP significantly correlated with their childrens’ openness to experience, but not any other personality dimension. Closer inspection revealed that this correlation was due to the openness aspect (r = .22, p < .01), not the intellect aspect (r = .05, p = .46). Parents’ ACP was independent of their childrens’ divergent thinking ability (fluency and originality) or intelligence. Divergent thinking and intelligence correlated in an expected manner. There was no direct association between parents’ ACP and participants’ creative activities or achievements as measured by the ICAA. Associations between latent constructs are reported in the following.

TABLE 4 Descriptive statistics and intercorrelations of study 2 variables

	M (SD)	2	3	4	5	6	7	8	9	10	11
Appreciation for Creative Pers. (parents)
ACP (1)	0.56 (0.23)	−.04	−.01	.08	−.07	.17	.09	.13	.08	.12	.03
Big Five Dimensions (participants)
Neuroticism (2)	2.80 (0.65)		−.30	−.17	−.25	−.16	−.20	−.15	.03	.07	−.07
Extraversion (3)	3.60 (0.49)			.21	.24	.45	.06	−.02	.03	.14	.22
Agreeableness (4)	4.01 (0.42)				.08	.24	−.01	.12	.01	.15	.09
Conscientiousness (5)	3.13 (0.50)					.04	.03	−.15	−.03	−.06	−.03
Openness/Intellect (6)	3.77 (0.52)						.30	.23	.25	.31	.42
Cognitive Abilities (participants)
General Intelligence (7)	0.00 (0.73)							.31	.19	.10	.22
Divergent Thinking Originality (8)	0.82 (0.20)								.05	−.09	.20
Divergent Thinking Fluency (9)	8.54 (3.19)									.13	.12
Real-Life Creative Behavior (participants)
Creative Activities (10)	65.49 (27.01)										.53
Creative Achievements (11)	51.53 (35.50)

Note. N = 190 for all variables. ACP = Appreciation for Creative Personality. The first row presents dyadic correlations between participants’ personality and their parents’ ACP scores. See method study 2 for more detailed information. Bold indicates significance at p < .05.

Mediation model

Based on the findings presented above and previous models of creativity (e.g., Jauk et al., 2014), a latent variable mediation model was set up to test whether parents’ ACP might predict creative activities via the openness or intellect aspects from the BFAS. Specifically, it was assumed that openness would act as a personality mediator between social-environmental climate (here: parental ACP) and the exertion of creative activities, as openness is thought to lower the behavioral threshold for engaging in creative activities (Feist & Barron, 2003; see also Jauk et al., 2014).

The ACP scale was modeled itemwise using WLSMV for dichotomous indicators, as in study 1. Openness and intellect were also modeled itemwise; the ICAA activities scale was modeled using two item parcels in exact accordance with Jauk et al. (2014). The model showed good fit to the data (χ²(555) = 740.49, p = .00; CFI = 0.81; WRMR = 1.08); though the χ² test was significant, the ratio of χ²/df was well below 2 (Byrne, 1989). All factor loadings were significant at p < .001; loadings of the ACP were generally of similar magnitude to those obtained in study 1 (see Figure 2). The residual correlation between openness and intellect captures the shared variance among both variables, which are regarded as aspects of a higher order openness factor (DeYoung et al., 2007), controlling for ACP.

Figure 2 shows the latent mediation model parameter estimates. Parents’ ACP had a significant effect on participants’ openness (β = .28; p < .01), but not intellect (β = .07; p = .43). Participants’ openness, in turn, was significantly associated with their amount of creative activities (β = .41; p < .001). Intellect was not associated with creative activities (β = .00; p = .99). The direct effect of ACP on creative activities was not significant (β = .01; p = .95), but there was a significant indirect effect via openness (β = .11; p = .01).

As an additional exploratory analysis, a domain-specific model in which parents’ ACP was related to participants’ openness and intellect, as well as the eight ICAA subscales (see Table 5), was investigated. Modeling of the ACP and openness/intellect personality dimensions corresponds exactly to the model reported above. The ICAA subscales were entered as manifest criterion variables in order to avoid overparameterization of the model. The model converged to an admissible solution and showed similar fit to the data as the model reported above (χ²(732) = 938.02, p = .00; CFI = 0.81; WRMR = 1.03). Table 5 displays the coefficient estimates for the effects of openness and intellect on the ICAA subscales, as well as the direct and indirect effect of ACP on the subscales. All ICAA subscales except sports and science & engineering were significantly related to openness; literature and science & engineering were significantly related to intellect. As in the domain-general model presented above, the ACP did not display direct effects on any of the ICAA subscales, but did show significant indirect effects via openness on creative activities in the domains literature, music, arts & crafts, visual arts, and performing arts.

TABLE 5 Domain-specific model for the prediction of participants’ creative activity by means of openness and intellect, their parents’ ACP, and the indirect effects of parents’ ACP via participants’ openness and intellect

	Literature	Music	Arts & Crafts	Creative Cooking	Sports	Visual Arts	Performing Arts	Science & Engineering
Openness	.36	.43	.43	.21	.05	.37	.38	−.10
Intellect	.21	−.05	−.12	−.06	.01	−.07	−.10	.23
ACP direct	−.02	−.02	.02	.07	−.04	−.04	.03	.02
ACP via Openness	.10	.12	.12	.06	.02	.10	.10	−.03
ACP via Intellect	.02	−.04	−.01	.00	.00	−.01	−.01	.02

Note. Standardized regression weights are displayed. Coefficients marked in bold are significant at p < .05. ACP = Appreciation for Creative Personality.

Discussion

Study 2 showed that there is a significant association between parents’ ACP and participants’ openness. This result was specific to openness rather than any other personality aspect, including intellect. This shows that the ACP scale could capture shared variance between individuals, making it a candidate tool for the study of the social-environmental climate for creativity from an interpersonal perspective. Moreover, the latent mediation model showed that parents’ ACP affected participants’ exertion of creative activities, indirectly, via the path of openness. This means that parental ACP, in terms of a social-environmental factor for creative climate, selectively impacted the personality dimension of openness, which, in turn, had an effect on the engagement in creative activities. The mediation effect is of moderate size but quite notable given that it represents covariation across parent-child dyads. Complemental domain-specific analyses showed that parental ACP selectively supported the exertion of creative activities in arts-related domains, but not in creative cooking, sports, or science & engineering. While this yields evidence for the criterion validity of ACP in the arts, the missing correlation with science & engineering could be seen as a limiting factor (see limitations section for discussion).

While parental ACP was associated with participants’ personality and, indirectly, behavioral tendencies (creative activities), the ACP did not affect any cognitive measures such as creative potential or general intelligence. This indicates that ACP is specifically related to personality traits rather than cognitive abilities, which might be fostered by other factors.

General Discussion

Many contemporary models of creativity emphasize the role of a valuing social environment in terms of a positive climate for the development and expression of human creativity. Yet, to date, there is no context-independent measure that allows for the quantification of the social-environmental climate for creativity on an individual level. Here, the construction and validation of a new measure for the assessment of Appreciation for Creative Personality, the ACP scale, were reported. The scale allows for the quantification of social-environmental conditions for creative individuals from an interpersonal perspective.

The ACP scale is a concise, one-dimensional measure of appreciation for creative personality that meets high psychometric standards. Validity analyses indicated that the scale, as expected, correlated highly with the Big Five trait openness to experience (convergent validity). Yet, it cannot be reduced to openness, as it displayed incremental criterion validity in the prediction of relevant criteria (preference for creative education). This means that, while open people are generally also open to social interactions with creative people, ACP can be useful to predict creativity-related outcomes (CE criteria in study 1) over and above openness. Besides openness, the scale also displayed small correlations with honesty-humility and extraversion. Closer inspection revealed that the correlation with extraversion could be explained by the social boldness facet. Individuals scoring higher on social boldness might be more at ease facing the nonconforming aspects of creative individuals. The correlation with honesty-humility was due to the facet greed avoidance, which circumscribes a low interest in acquiring wealth, luxury, and social status (Lee & Ashton, 2004). It seems plausible that a low interest in these materialistic goals also implies valuing intellectual and creative endeavors. However, further research is needed to clarify this link.

The ACP scale was largely independent (discriminant validity) of the cognitive ability to recognize creative ideas as measured by the previously published CET (Benedek et al., 2016). This supports the conceptualization of ACP as an affect-laden measure of preference for interacting with creative people, which is different from the cognitive ability to recognize creative products or people. Yet, both variables simultaneously predicted creative education attitude (see Table 3). This indicates that both measures capture distinct and relevant portions of variance in creative education attitude that are more related to personality and ability aspects, respectively. This pattern of results suggests that ACP and CET scales are best used together in contexts where both aspects are relevant, such as in the educational or also in the organizational context. In both of these contexts, teachers or bosses are not only required to appreciate creativity on the personality level, but also to correctly recognize creative ideas or individuals as such.

Under fake-good instruction, the ACP largely lost its validity, although there was still some evidence for criterion validity (see Table 1). These correlations, however, should not be overrated, as they were part of the item selection process (see study 1 methods) and would possibly not replicate in an independent sample. More importantly, however, itemwise faking analyses indicated that the latent construct is not readily apparent to test-takers, as some item means increased, while others decreased under fake-good instruction (see Table 1). This notion is further supported by qualitative information from test-takers during the pilot phase, who were unable to identify the latent construct. This makes the ACP a useful tool that is not limited to research purposes but is also suited for the diagnostic context in terms of low susceptibility to socially desirable responding.

Taken together, the ACP differs from existing measures of creative climate in two ways: First, it takes an interpersonal approach to the study of creative climate by directly assessing individuals from the social surrounding of the creative individual. This is what differentiates the ACP from organizational measures of creative climate, such as Amabile and colleagues’ KEYS (Amabile et al., 1996) or conceptually similar inventories (Anderson & West, 1998; Ekvall, 1996). Second, the ACP was designed as a context-independent measure that targets the level of personality traits rather than attitudes in a specific context. As such, the ACP is different from Kwaśniewska et al. (2018) recently published Climate for Creativity in Parent-Child Relationship Questionnaire. The scale presented here aims to assess ACP as a personality trait in a behaviorally anchored manner (“indicate with whom of them you would prefer to have contact with”; see Table A2). This makes the ACP not only a versatile instrument that can be used in different contexts, but also goes along with a low susceptibility to socially desirable responding (as it is of relevance for instance in the educational context; Runco, 2007; Westby & Dawson, 1995)

Study 2 found that – in parent-child dyads – parents’ ACP predicted their childrens’ openness, and, in turn, everyday creative activities. While openness can be conceived an important personality prerequisite for creative activity (Jauk et al., 2014), the findings reported here add to previous literature in that they provide evidence for an indirect effect from parents’ ACP to their childrens’ creative activities via the path of openness. This yields an interesting insight into the putative developmental conditions for creative behavior, which might be indirectly promoted by parents who are accepting and supportive towards their creative children (Kwaśniewska et al., 2018). It was also found that, while parents’ ACP impacted their childrens’ openness, there was no association with intellect, which demonstrates the specificity of the effect and conforms to prior research (Nusbaum & Silvia, 2011). However, personality demands may vary between different fields of creative endeavor (Kaufman et al., 2016), and complementary domain-specific analysis indicated that the results obtained here might only be valid for artistic creativity (see limitations).

Taken together, the correlation between parents’ ACP and their childrens’ personality shows that the ACP scale can capture covariance between a target person and individuals from their respective social environment. This opens many possibilities to the use of the ACP scale as a measure of creative climate, such as for instance the quantification of ACP in organizations, schools, or families. The ACP scores of single individuals or groups of individuals might then be related to creative outcome variables in other groups of individuals (for instance employees, pupils, or children).

Limitations

Though the ACP scale displayed good psychometric properties in the studies reported here, there are some limitations to the measure. First, the scale is rather short. While this allows for an economic assessment of the main construct, it does not allow to disentangle more fine-grained facets, which might emerge when the construct was measured with a longer and more nuanced scale (e.g., Kwaśniewska et al., 2018). However, as long as a single score is needed, as it was the case in the admission test context, the short scale seems well suited. Moreover, while appreciation of creative personality arguably represents a fundamental aspect of the social climate for creativity at the interpersonal level, a more comprehensive assessment may consider further factors such as trust, or the readiness to support and cooperate with creative people.

Although the parent-participant correlation between ACP and openness obtained in study 2 is encouraging, the data are of cross-sectional nature and thus cannot speak to causality. It cannot be inferred that parents’ ACP causally influences their offsprings’ openness. It could also be the case that parents who have open and creative children develop higher ACP (i.e., adjust their preference for creative personality in accordance to their childrens’ traits). Though the former causal path may presumably be stronger than the latter, the issue must remain subject to future studies. Also, common genetic factors could account for part of the shared variance. Future studies should thus validate the ACP also in the organizational or educational context, where common genetic variance is not an issue. Importantly, irrespective of the causal nature of the findings reported here, the data demonstrate that the ACP can successfully capture interindividual covariance, which renders it a useful tool for the study of social environments for creativity.

Due to time restrictions and practicability, no personality traits apart from ACP were assessed in participants’ parents in study 2. Above all, it would have been interesting to include Big Five openness to investigate incremental validity of the ACP. However, incremental validity of the ACP beyond openness could be established in study 1. Future research could directly examine the incremental validity of the ACP above and beyond Big Five traits in interpersonal datasets. Such studies could be carried out in organizations, schools, or families, and might help to unveil important aspects of the social-environmental climate for the development and expression of creativity.

Lastly, further research is needed on the validity of the ACP in different domains; particularly in the arts and sciences, which go along with different demands for creative personality (e.g., Kaufman et al., 2016). The subscale analyses of the ICAA showed that parental ACP is associated with their children’s creative activities in the arts, but not in science & engineering. This can be seen as a potential limitation of the ACP scale to capture appreciation for the personality of people interested in scientific creativity. It has to be noted, though, that common measures of everyday creativity also tend to show a broader coverage of various artistic domains, maybe due to their higher prevalence in non-professional samples (Diedrich et al., 2018). This renders the comparison between arts and sciences unbalanced in this study. Future studies could use in the ACP in dedicated artistic and scientific professionals.

Acknowledgments

We express our special thanks to Hannah Wolf and Magda Gerhold for translating the scale into English. This work was supported by the HRSM Fund from the Austrian Federal Ministry of Science, Research, and Economy as part of the Project PädagogInnenbildung Neu – Development and Implementation of a common selection procedure for teacher students.

Additional information

Funding

This work was supported by the HRSM Fund from the Austrian Federal Ministry of Science, Research, and Economy as part of the Project PädagogInnenbildung Neu–Development and Implementation of a common selection procedure for teacher students. The authors acknowledge the financial support by the University of Graz.

Notes

¹ The forced-choice format was chosen instead of a rating scale format to eliminate the general preference for teaching children. An example item is: “An ambitious and diligent child, who always follows the teacher’s instructions, but does not usually come up with own ideas” vs. “An active and rebellious child, who does not always follow the teacher’s instructions, but has many own ideas”.

² While these items include two latent traits (in this case, openness to aesthetics, and tardiness), which could be considered problematic in standard psychometric testing, the item content of the ACP does not refer to the test-taker, but to a prototypically creative person. As creative personality is a multidimensional construct including also less desirable traits, this seems adequate here. Importantly, test-takers’ ACP can still be unidimensional (see results section).

³ As an alternative score, max3 scoring, an adaptation of the top3 scoring method (Benedek, Mühlmann, Jauk, & Neubauer, 2013; Silvia, 2011) in which the three ideas with the highest ratings are averaged to an indicator of divergent thinking ability, was also probed. Average scoring and max3 scoring did not differ with respect to the effects of interest in this study.

Table A1. Appreciation for creative personality scale (German)

Item No.	Response Alternative A	Response Alternative B
1*	Bleibt bei weiteren Autofahrten öfter stehen, um die Natur zu fotografieren, und kommt daher manchmal zu spät.	Wenn sie mit dem Auto unterwegs ist, versucht sie möglichst schnell an ihr Ziel zu kommen und hat keine Zeit die Umgebung zu bewundern, auch wenn die Natur noch so schön ist.
2*	Ihre Ideen werden oft als „unpraktisch“ oder sogar „gewagt“ bezeichnet.	Ihre Vorschläge sind meist bewährte, die schon früher gut funktioniert haben.
3	Sie mochte in der Schule gerne Sportunterricht, weil es da immer klare Spielregeln gab.	Kunst war eines ihrer Lieblingsfächer in der Schule.
4*	Bewundert KünstlerInnen und SchriftstellerInnen gleich wie ÄrztInnen und JuristInnen.	ÄrzteInnen und JuristInnen leisten ihrer Meinung nach wichtigere Arbeit als KünstlerInnen und SchriftstellerInnen.
5	Zeichnet nur Bilder von echten Personen oder Objekten gerne.	Zeichnet nur abstrakte Bilder gerne.
6	Ist nett und ruhig, hat aber nicht so viele Ideen.	Ist lebhaft und voll von Ideen und tut daher oft nicht das, was andere erwarten.
7	Normalerweise tut sie Dinge auf akzeptierte Weise.	Normalerweise versucht sie, sich neue Wege einfallen zu lassen, etwas zu tun.
8	Beschäftigt sich lieber mit Fakten.	Beschäftigt sich lieber mit Ideen und Theorien.
9	Ist die Art von Person, die es bevorzugt, alltägliche Probleme auf bewährtem Weg zu lösen.	Ist die Art von Person, die es bevorzugt, alltägliche Probleme auf außergewöhnlichem Weg zu lösen.
10	Die erste Idee, die man für die Lösung eines Problems hat, ist ihrer Meinung nach fast immer die richtige.	Braucht in der Regel lange für die Lösung von Problemen, weil sie sich eine Vielzahl möglicher Lösungen ausdenkt.
11	Fragt vor Geburtstagen gerne nach, was der/die zu Beschenkende als Geschenk gebrauchen könnte.	Verschenkt zu Geburtstagen gerne selbst hergestellte Dinge, auch wenn der/die Beschenkte diese gar nicht brauchen kann.
12	Kauft nützliche Einrichtungsgegenstände, deren Aussehen ist ihr nicht so wichtig.	Hat eine Vorliebe für ausgefallene Kombinationen von Einrichtungsgegenständen, die andere manchmal zum Schmunzeln bringen.
13	Lässt sich spontan Witze einfallen, die andere vielleicht nicht lustig finden.	Erzählt lieber bewährte Witze.

Note. Item numbers correspond to those in Table 1. Response alternative A is assigned zero points, response alternative B is assigned one point. Items marked with an asterisk are reverse-coded. Scale score is calculated by averaging across all items. The German scale instruction is as follows: “Nachfolgend werden Ihnen Beschreibungen von jeweils zwei Personen dargeboten. Bitte stellen Sie sich für jede Aussage eine Person vor und geben Sie an, mit welcher der beiden Personen Ihnen der Umgang angenehmer wäre.”

Table A2. Appreciation for creative personality scale (english)

Item No.	Response Alternative A	Response Alternative B
1*	Stops often to photograph nature during long car rides, and is therefore sometimes late.	When driving by car, he/she tries to reach the destination as quickly as possible and has no time to admire the surrounding nature, no matter how beautiful it is.
2*	His/her ideas are often considered “impractical” or even “wild”.	His/her suggestions are usually established ones, which have worked before.
3	In school, the person enjoyed physical education very much, because the rules were always explicit.	Art was one of her favorite subjects at school.
4*	Admires artists and writers the same amount as medical doctors and lawyers.	Is of the opinion that the work of medical doctors and lawyers is more important than that of artists and writers.
5	Likes to draw pictures only of real people and real objects.	Likes to draw abstract pictures only.
6	Is nice and calm, but does not have so many ideas.	Is lively and full of ideas and therefore does often not do what others might expect.
7	Does usually do things in a widely accepted manner.	Normally tries to find new ways to do things.
8	Likes to concern oneself with facts.	Likes to concern oneself with ideas and theories.
9	Is the kind of person who prefers solving everyday problems in well proven ways.	Is the kind of person who prefers to solve everyday problems in unconventional ways.
10	Believes that the first idea that comes to mind when trying to solve a problem is almost always the right one.	Typically takes a long time to solve problems because he/she thinks about numerous different possible solutions.
11	Before making a gift, asks what the person actually needs for his/her birthday.	Likes to give self-made birthday presents, even if the person does not need them.
12	Buys practical furniture, the appearance of which is not so important.	Has a preference for unusual combinations of furniture, which sometimes makes other people chuckle.
13	Comes up with jokes spontaneously which others might not find funny.	Likes to tell established jokes.

Note. Item numbers correspond to those in Table 1. Response alternative A is assigned zero points, response alternative B is assigned one point. Items marked with an asterisk are reverse-coded. Scale score is calculated by averaging across all items. The English scale instruction is as follows: “In the following, you will see descriptions of two persons. Please imagine a person for each of these descriptions and indicate with whom of them you would prefer to have contact with.”