A Latent Profile Analysis of the Criterion-related Validity of a Divergent Thinking Test

Abstract

The criterion-related validity of divergent thinking (DT) tests has been questioned because of relatively low correlations between DT test scores and creative performance. A possible inhibiting factor in previous correlations—heterogeneity of the sample—has been ignored. If groups with different DT profiles are involved in this line of research, low correlations would not necessarily be indicative of actual relationships, and criterion-related validity would appear to be lower than it is. Thus, this study identified groups with their own unique DT profiles and examined their creative performance. To this end, elementary children’s (N = 325) responses in 3 DT tasks were analyzed using latent profile analysis. Results suggested 3 groups with unique DT profiles: One group scored moderate on fluency and low on originality; another group scored moderate on fluency and high on originality; the third group scored uniformly low across fluency, originality, and flexibility. Surprisingly, only the low-originality group excelled in creative achievement, but the hypothesis of this investigation was supported by the fact that only the high-originality group significantly excelled in ideation. These results confirm that low correlations between DT indices and creative performance do not necessarily indicate a lack of the criterion-related validity of DT tests. Differences between the creative achievement and the ideation criterion measures are detailed, and idiosyncratic properties in DT profiles and their corresponding creative strengths are discussed as educational implications.