Page 78 - Secondary school students’ university readiness and their transition to university Els van Rooij

master theses.
e fact that many studies included multiple independent variables did
cause another limitation though: By investigating the impact of several variables on the outcome of interest simultaneously, e.g., in stepwise regression analysis, e ects of individual predictors can be concealed. Fortunately, a large majority of papers that used regression analysis also included correlation matrices, so the potential distortion in this regard is limited.
Last, we did not take into account di erences between elds of study. Some 3 studies that used a sample of students from di erent degree programmes also
performed separate analyses per programme and found small di erences between
these programmes. However, due to reasons of e ciency we only looked at their
general ndings.
3.5.6 Recommendations for future research
Some in uential variables in international research were hardly present in the Dutch and Flemish studies, such as need for cognition and mastery and performance goals. It would be interesting if Dutch and Flemish researchers would take speci c notice of these variables in future research. We also found that some categories were more o en being investigated in the Netherlands and others in Flanders. Few respectively none of the Flemish studies included engagement or learning environment variables and relatively little attention in Dutch studies was paid to demographic factors, the subject uptake in secondary school, and motivation. Flemish and Dutch researchers could include these relatively understudied topics, since they showed relationships with academic success.
Furthermore, future research should consider the conceptual and methodological issues that came to light in this review, namely that too o en constructs are not clearly de ned and/or measured by newly developed instruments. To obtain reliable and comparable data, and consequently work towards a comprehensive picture of what matters for student success and the extent to which certain factors have a higher impact in certain countries or education systems, it would be extremely helpful if researchers would use uniform de nitions of factors and the same instruments when they investigate the same factors. Nevertheless, the problem with translations remains and adaptations to the context are sometimes necessary, because some items may not be appropriate in a certain context.
Since we were not able to draw a rm comparison between the Netherlands
Systematic review of rst-year success
77