Page 65 - Secondary school students’ university readiness and their transition to university Els van Rooij

Chapter 3
categorised all independent variables that were used in the 38 studies. Variables that did not perfectly t into a category were placed within the category they were most closely related to, e.g., ‘mathematics GPA in secondary school’ was placed in the category ability factors, as it can be seen as a sublevel of the ability factor secondary school GPA. Variables that did not t into an existing category were ICT skills (De Wit, Heerwegh, & Verhoeven, 2012), results of a mathematics test (Fonteyne et al., 2015), results of a mathematics and of a language test (Pinxten et al., 2015), and career guidance GPA and rst grade point average (Te Wierik, Beishuizen, & Van Os, 2015). ese variables were excluded from the analysis. e data synthesis resulted in an overview of all investigated variables in the Netherlands and Flanders per category. Second, for each variable in each study we noted if this variable was (positively or negatively) signi cantly related to student success, i.e., to GPA, EC, and/or persistence, while retaining the information of whether this correlate concerned a Dutch or Flemish, and a professional education or university sample. ird, a more comprehensive picture was obtained of which variables were most consistently related to academic outcomes, and whether or not this was outcome-speci c, country-speci c, or speci c for one of the educational levels. is was done by counting the number of positive, negative, and non- signi cant relationships and putting these together in one table per category. For reasons of expediency, but also because we wanted to compare results, variables that were only investigated by one study were excluded from these tables.
3.4 Results
3.4.1 Characteristics of the included studies
Table I in the Appendix presents an overview of the characteristics of the 38 included studies and their main ndings. Most studies were published quite recently, i.e., in the 2010s (69%). Ten per cent was published between 2000 and 2006 and 21 per cent from 2006 to 2010. More than three quarters of all studies were performed in the Netherlands (29 of 38). Most studies were based on samples of university students (29), eight focused on professional bachelor education, and one had a mixed sample. Almost half of the studies (44%) used a sample of students from several degree programmes. e most used outcome measure was EC, present in 27 studies. As students in Flanders can apply for a certain number of credits in the beginning of the year, the Flemish studies did not use EC as an absolute measure, but used the
64