10.3389/fpsyg.2012.00132

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Interest is a key driver of performances, decisions and choices, for example starting an educational program. The German O*NET Interest Profiler Short Form is a recent RIASEC-based instrument. However, its evidence is limited in German speaking subpopulation. We analyze its psychometric properties in <i>N</i> = 173 students from rehabilitation education programs. Internal consistency is adequate in all six scales (<i>ω</i> = .80-.90) and structural validity shows partial support for the RIASEC circumplex. Intercorrelations follow the expected pattern only partly and multidimensional scaling indicate a good overall fit (RSQ = .94; stress ≈.11), although deviations from a perfect hexagonal is observed. Discriminant validity is supported by generally small-to-medium associations with Big Five traits below the discriminant threshold (<i>r</i> < .85). Criterion-related validity is weak. Using Euclidean distance for congruence analysis, associations with study satisfaction (<i>r</i> = -.08, -.09) and dropout intention (<i>r</i> = .06) are near zero. Using angular agreement for congruence analysis, small but statistically significant correlations exist for study satisfaction (<i>r</i> = .17-.18, <i>p</i> < .05). In summary, results show partial evidence for structural validity, clear support for discriminant validity, and limited criterion-related validity. Findings further suggest that conclusions about validity depend on sample characteristics and the chosen operationalization of vocational interest congruence.

Abstract: PubMed · Datensatz