A brief measure of creativity among college students

College Student Journal, Dec, 2004 by Kathryn E. Kelly

This article presents three studies chronicling the developing and preliminary psychometric evaluation of a multidimensional self-report measure of creativity - the Scale of Creative Attributes and Behavior (SCAB). The results of these studies provide evidence that the SCAB is internally consistent, exhibits good initial construct validity through factor analysis, measures a stable trait, and is not related to social desirability. Areas for future research are discussed

**********

Previous research has generally assessed creativity using scales measuring divergent thinking (Torrance, 1974) and creative personality (Gough, 1979). Few easily administered, reliable, and valid self-report scales have attempted to measure creativity as a multidimensional phenomenon. The current research aimed to develop such a measure.

A multidimensional measure should assess the behavioral, cognitive, and personality factors which, when taken together, account for creativity. According to previous research, there exist at least five underlying components of creativity: creative engagement, creative cognitive style, spontaneity, tolerance, and fantasy (Eysenck, 1995; Glover, Ronning, & Reynolds, 1989; Gough, 1979; Sternberg, 1988; Torrance, 1974).

Creative engagement refers to enjoying creative activities and routinely spending time working on something creative. Creative cognitive style refers to the cognitive aspect of creativity which has often been linked with intelligence (divergent thinking and problem solving). Spontaneity is a style characterized by impulsivity and excitement seeking. Tolerance is the attitude of flexibility and openness to ideas and experience. And finally, fantasy is a mental activity of creativity, namely daydreaming and imagination. This article presents three studies which attempt to develop a scale measuring the aforementioned aspects of creativity.

Study 1

The purpose of Study 1 was to develop a brief measure of creativity. After obtaining informed consent, 89 females and 29 males enrolled in undergraduate psychology courses (Mean age = 24.0 years, SD = 6.6) were administered 127 items developed using a rational, brainstorming approach by the author and an assistant to measure the five proposed components of creativity mentioned previously.

Items were administered using a 7-point Likert scale ranging from 1 = "strongly disagree" to 7 = "strongly agree". Similar to the approach suggested by Jackson (1970), to select items for the final scale, items with lower item to total-scale correlations were systematically eliminated for each of the 5 creativity factors mentioned earlier. This was repeated until four items remained for each factor. The Coefficient Alphas were: .75 (total scale), .82 (Creative Engagement), .69 (Creative Cognitive Style), .75 (Spontaneity), .72 (Tolerance), and .70 (Fantasy). These items were subsequently named the Scale of Creative Attributes and Behavior (SCAB).

Study 2

Study 2 was designed to provide some evidence of construct validity by assessing the SCAB's factor structure. After obtaining informed consent, 226 female and 45 male undergraduates (Mean age = 25.8 years, SD = 7.6) were administered the SCAB using the 7-point scale described in Study 1. The Coefficient Alpha of the total SCAB was .82. A principal components factor analysis with a varimax rotation was calculated. This resulted in a five-factor solution accounting for 64% of the systematic variance. Items loaded on their expected factors as developed in Study 1. Factor 1 measured Spontaneity and accounted for 15.4% (eigenvalue = 3.1) of the variance, Factor 2 measured a Creative Cognitive Style (14.1%; eigenvalue = 2.8), Factor 3--Creative Engagement (13.2%; eigenvalue = 2.6), Factor 4--Tolerance (11.6%; eigenvalue = 2.3), and Factor 5--Fantasy (9.8%; eigenvalue = 2.0).

A second factor analysis was calculated using the total scores of the five factors. All five factors loaded on a single higher-order factor (interpreted as creativity), accounting for 38% (eigenvalue = 1.9) of the variance. These results provide some preliminary evidence for the construct validity of the SCAB.

Study 3

The purpose of Study 3 was to attain an estimate of the test-retest reliability of SCAB and explore the relationship between the SCAB and a potential confound of self-report measures, social desirability (Anastasi, 1988). After obtaining informed consent, 59 undergraduate students (Mean age = 25.0 years, SD = 6.7) were administered the SCAB as described in Study 2 and Reynold's (1982) 11-item version of the Marlowe-Crowne Social Desirability Scale. The Coefficient Alpha of the SCAB was .77. The total SCAB score did not significantly correlate with social desirability (r = -.08, p = .55). To attain test-retest data, the SCAB was re-administered to 36 of the same students one month later. The test-retest reliability coefficients were: .80 (total scale), .83 (Creative Engagement), .86 (Creative Cognitive Style), .90 (Spontaneity), .78 (Tolerance), and .70 (Fantasy). These results suggest that the SCAB measures a stable trait and is not evoking a social desirability response set.