Page 76 - Secondary school students’ university readiness and their transition to university Els van Rooij

instances (six of 19). International research from this century quite consistently
showed that female students outperform male students in higher education
(e.g., Conger & Long, 2010; Hillman & Robinson, 2016; Richardson et al., 2012),
although the gender gap in higher education is not as large as it is in primary
and secondary education (Voyer & Voyer, 2014). Our results indicate that, at
least in the Netherlands and Flanders, the gender gap is larger in professional
education than in university education. Other di erences were that the level of
prior education and factors within the categories personality factors and learning 3 strategies were more o en related to success outcomes in university than in
professional education.
3.5.4 Limitations of Dutch and Flemish rst-year student success research
Many articles did not clearly de ne the measured constructs and/or did not describe thoroughly how the constructs were measured. Moreover, sometimes di erent names were given to constructs with a comparable de nition. For example, Meeuwisse, Severiens, and Born (2010) de ned informal peer interaction as interaction among students regarding personal matters, whereas Severiens and Wol (2008) labeled this exact same de nition as informal social integration. ese di erences in naming and de ning constructs, as well as di erences in the operationalisation of constructs, makes it di cult to evaluate and compare previous research ndings. Furthermore, rather than using (inter) nationally validated instruments, many studies used instruments developed by the researchers themselves, which also makes it more di cult to compare results from di erent studies.
Another issue concerns the outcome variables used. We found that the presence and strength of a relationship with academic success depends on how academic success is measured. Motivational factors, for example, o en related to GPA and credit obtainment, but not to persistence. Also, most studies used EC as the only outcome measure, which is also re ected in the general results: e clearest evidence concerns the relationships with EC, whereas only for a few variables it is clear that they are related to persistence. It would be worthwhile if more studies would use multiple outcome variables, to investigate di erential e ects.
3.5.5 Limitations of this review
One limitation of this review study is that the number of Flemish studies that t the inclusion criteria was too low to compare the Dutch and Flemish studies in
Systematic review of rst-year success
75