Schult, J., Fischer, F. T., & Hell, B. (2016). Tests of scholastic aptitude cover reasoning facets sufficiently. European Journal of Psychological Assessment, 32(3), 215-219. http://dx.doi.org/10.1027/1015-5759/a000247
Johannes says: "Structural equation modelling, mediation analysis..."
Data analysis finished: 2015-09-07
Article published: 2015-05-21
Replication files published: 2015-06-19
Software used: Mplus 7.11, Stata 13.1
Below you find the syntax files used for the analysis. NB: The raw data is not available online due to data privacy laws. You can run a similar (but not identical) analysis by inputting the correlation matrices reported in Table 1 in Mplus (e.g., DATA: FILE IS ncorr.dat; TYPE IS CORRELATION; NOBSERVATIONS ARE 218;) and using the maximum likelihood estimator (ANALYSIS: estimator = ML;). The correlations are included in the files ncorr.dat (science sample) and wcorr.dat (economics sample).
crfacetsV1.do reads in raw data files and creates an interim data set for each of two samples.
The analysis is the same for both samples. The models with one intelligence factor (N1.inp, W1.inp) were compared to models with three content factors (N3.inp, W3.inp). Then, fully mediated models were estimated (N3nodirect.inp, W3nodirect.inp) Models without mediation effects were also run (N3nomediation.inp, W3nomediation.inp; see Footnote 1 in the paper). The models mentioned in Footnote 2 were N1nodirect.inp and W1nodirect.inp.
Please visit http://www.jutze.com/research for more materials on publications and possibly newer versions.
Written by Johannes Schult (email@example.com)
Last updated: 2015-06-19