Using invariance testing to study job crafting: a cross-cultural comparison of the well-documented relation between job crafting and work engagement

We test one of the classic job crafting scales in Russia (N = 285), Germany (N = 200), and Croatia (N = 192). Job crafting describes work behaviors employees engage in to create a better fit between their preferences and their jobs, typically by modifying existing norms, rules, and practices. We set out to highlight the importance of different invariance testing methods by using the example of job crafting and its well-documented relationship with work engagement. We use (a) strict invariance testing (MGCFA), assessing metric, configural, and scalar invariance, (b) more flexible invariance testing via the alignment methodology, and (c) Bayesian approximate measurement invariance testing (BAMI). The three investigated contexts differ systematically in features related to job crafting (e.g., hierarchy sensitivity), but we did not find that this moderated relations at the individual level. MGCFA indicated that only the job crafting subdimensions (increasing challenging job demands and increasing structural job resources) achieved metric invariance. The other two dimensions (increasing social job resources and decreasing hindering job demands) failed to reach even configural invariance. This questions the cross-cultural equivalence of the full four-factor structure. Alignment optimization further supported this conclusion. The results indicated that nearly half of the items within each non-invariant job crafting dimension displayed non-invariance. In contrast, the remaining two dimensions demonstrated stable factor loadings and an acceptable proportion of non-invariant parameters, including factor loadings and intercepts/thresholds. The different analytic approaches converged in showing that increasing challenging job demands and increasing structural job resources achieved scalar invariance in MGCFA, whereas decreasing hindering job demands and increasing social job resources failed to establish configural invariance. The consistency of this pattern across MGCFA, alignment optimization, and BAMI provide complementary evidence for the cross-cultural robustness of the former two dimensions only. On this basis, we discuss problematic items of job crafting as well as items that can be considered workable across cultural groups. We conclude that at its cross-cultural core, job crafting may be fundamentally about proactive, growth-oriented self-development (more than social negotiation or demand reduction), and highlight the benefits of multiple invariance testing methods to advance the theorizing on job crafting.