Abstract
Students entering higher education need to integrate themselves into a new study community. Those coming from minority ethnic backgrounds may experience this phase more intensely since they may feel themselves to be different or consider that they are being treated differently. This article approaches ethnic identity as a dynamic concept. The differences characterizing students with varying ethnic self‐definitions are explored as subjective factors that are important for good study progress. Finally, study progress after one year is analyzed.
Notes
Indonesia is a former Dutch colony, in which many Dutch citizens lived, having been born there or having an Indonesian and a Dutch parent.
It should be noted, however, that those figures are based either on “crude” data from faculties, or, if no information from this source was available, on the survey data collected by the authors.
If a “list‐wise” deletion were to be performed, only 466 respondents could be used.
Model A (all parameters free) compared to model B (regression coefficients set equal for both groups). Checking the significance of chi‐square B—chi‐square A, with df B—df A, if that chi‐square has a significance of p < 0.05, then use model A, if p>0.05, use model B.
The chi‐square measure of fit is very sensitive to sample size. It is therefore necessary to report three other fit measures that are less sensitive to sample size (see Hox and Bechger, Citation1998, Arbuckle and Wothke, Citation1995). NFI is the normed fit index: it has values from 0 to 1, with 1 indicating a perfect fit. A rule of thumb is that an NFI below 0.90 requires improvement of the model. The Tucker Lewis Index (TLI) is quite similar to the NFI, but it compensates for the complexity of the model. A TLI above 0.90 indicates a moderately good fit, and above 0.95 indicates a good fit. Finally, the root mean square error of approximation (RMSEA) is designed to assess the approximate fit of a model in relation to the degrees of freedom. The further below 0.05 RMSEA is, the more a close fit is suggested.