Strukturgleichungsmodelle mit Mplus

The impact of adolescents' self-efficacy and self-regulated goal attainment processes on school performance — Do gender and test anxiety matter?

“…The semi-restricted model yielded good fit indices as well (see Table 3). A χ 2 -difference test (Christ & Schlüter, 2012;Yuan & Bentler, 2004), conducted to compare the unrestricted model to the semi-restricted model, did not reach significance (χ 2 (26) = 18.75, p = .85). This result confirms measurement invariance and allows for group comparison, suggesting that the semi-restricted multi-group model fits the data not worse than the unrestricted model (Satorra & Bentler, 2001).…”

“…The indices indicated a good fit for the unrestricted and the semi-restricted model. Subsequently, both models were compared with a χ 2 -difference test (Christ & Schlüter, 2012;Yuan & Bentler, 2004). The test did not reach significance (χ 2 (26) = 30.08, p = .27).…”

“…The indices of the fully restricted model indicated a good fit as well. Again, the χ 2 -difference test (Christ & Schlüter, 2012;Yuan & Bentler, 2004) was used to compare the semi-restricted to the fully restricted model. This time, the test reached significance (χ 2 (9) = 31.42, p = .001), which indicates that the semi-restricted multi-group model fits the data better than the fully restricted model (Satorra & Bentler, 2001).…”

Promotion of Reading and Early Literacy Skills in Schools: a comparison of three European countries

“…Chi‐square testing is not used in this analysis for two reasons: The first reason is the sensitivity of the chi‐square test to the sample size. Secondly, being a very conservative model‐fit test, chi‐square results in a refusal of the H1 hypothesis even if the discrepancy between the model covariance matrix and the estimated population covariance matrix is negligible (Christ & Schlüter, ).…”

Support from teachers and trainers in Vocational education and training: The pathways to career aspirations and further career development

“…We used a rescaling-based robust estimator MLR in MPlus as recommended by Wang and Wang (2012) to deal with skewness and kurtosis in our data and obtain robust estimates. Measurement errors of the repeated measures at t1 and t2 (self-esteem, career aspirations) were allowed to correlate as recommended for autoregressive models (Christ and Schlüter, 2012). To assess the adequacy of the measurement model, we refer to the comparative fit index (CFI), Tucker-Lewis index (TLI) and the root-mean-square error of approximation (RMSEA).…”