Multiple complex problem-solving scenarios: The incremental validity of ability self-concept beyond reasoning in adults

Abstract

Complex Problem Solving (CPS) and reasoning are considered as correlated, but not equivalent (Stadler, Becker, Gödker, Leutner, & Greiff, 2015). However, empirical evidence for incremental validity of constructs beyond reasoning regarding CPS is scarce. This includes evidence for CPS self-concept (CPS SC; cf. Meißner, Greiff, Frischkorn, & Steinmayr, 2016). Strongest evidence would result from applying CPS scenarios as criteria that have been constructed to maximize correlations with conventional reasoning tests (e.g. MultiFlux, Kröner, 2001; Kröner, Plass, & Leutner, 2005; Zech, Bühner, Kröner, Heene, & Hilbert, 2017). Thus, we applied MultiFlux as a criterion to assess CPS performance of n = 240 university students and investigated incremental validity of CPS SC above and beyond reasoning according to a figural matrices test (DESIGMA, Becker et al., 2016; ω > .75 for all constructs). Results showed substantial unique effects of CPS SC on CPS above and beyond reasoning (β = .36; 16% incremental variance), both for a latent CPS factor and for all four CPS subdimensions. Despite a substantial amount of shared variance, reasoning and CPS proved to be distinct constructs and CPS SC showed a substantial unique effect beyond reasoning in explaining CPS. Avenues for exploring relevance of further constructs for explaining CPS variance are discussed.