Cross-cultural measurement invariance of divergent thinking measures

An emerging trend in current creativity research is the cross-cultural comparison of creative potential. However, no studies have addressed the cross-cultural measurement invariance of creativity instruments, which refers to the instrument's ability to assess the same construct across different cultures. This study investigated the latent structure and the degree of measurement invariance of two divergent thinking (DT) tests (the line meaning test and the real-world problem test) between American (n = 302) and Chinese (n = 316) college students. A two-factor model of the two DT tests was supported in both samples based on either the fluency or originality scores, suggesting variations between different types of DT tests. Multi-group confirmatory factor analyses revealed that although the assumptions of configural and weak invariance were met for both the fluency and originality indices, the partial strong measurement invariance was marginally satisfied only for the real-world problem fluency across the two cultural groups. Further, latent mean comparison showed similar fluency between American and Chinese college students on the real-world problem test. In contrast, results from the observed mean comparison indicated higher performance of American individuals on the real-world problem fluency and originality; the opposite trend was revealed for both indices on the line meaning test. Overall, the present study demonstrates the importance of examining the equivalence of the measurement instrument before interpreting cross-cultural differences in creativity. Suggestions for future cross-cultural creativity research are provided.