Problems with the Big Five assessment in the World Values Survey

Most psychological studies rely on student samples. Students are usually considered as more homogenous than representative samples both within and across countries. However, little is known about the nature of the differences between student and representative samples. This is an important gap, also because knowledge about the degree of difference between student and representative samples may allow to infer from the former to the latter group. Across 59 countries and 12 personality (Big-5) and attitudinal variables we found that differences between students and general public were partly substantial, incoherent, and contradicted previous findings. Two often used cultural variables, embeddedness and intellectual autonomy, failed to explain the differences between both groups across countries. We further found that students vary as much as the general population both between and within countries. In summary, our results indicate that generalizing from students to the general public can be problematic when personal and attitudinal variables are used, as students vary mostly randomly from the general public. Findings are also discussed in terms of the replication crisis within psychology.

I first provide an overview of general issues in personality measurement across cultures (e.g., measurement bias and equivalence, levels of test adaptation or indigenization). Then, in keeping with the organization of this special issue, I discuss personality measurement from alternative theoretical perspectives that have addressed, in varying degrees, personality and its measurement in cross-cultural perspective (i.e., trait perspectives, projective techniques, cultural psychology and constructivist perspectives, evolutionary perspectives). An integrated measurement approach is advocated, incorporating diverse aspects or levels of personality, while drawing on the complementary strengths of alternative approaches.