Page 50 - Secondary school students’ university readiness and their transition to university Els van Rooij

the entropy statistic. Lower AIC, BIC, and ABIC values indicate a better tting
model (Flaherty & Ki , 2012). If the VLMRT is signi cant, this means that the
current number of groups is a better t to the data than the model with one group
fewer (To ghi & Enders, 2008). Last, higher entropy statistics are better, as they
signify less classi cation error (Collins & Lanza, 2010). Needless to say, we also 2 checked whether the groups that were identi ed by the analysis were meaningful,
i.e., whether they made sense and could have theoretical and practical value. A er identifying the optimal number of groups, we assigned all students to the group they most likely belonged to and performed ANCOVAs to investigate group di erences with regard to academic adjustment and achievement one year later in university. ANCOVAs were used so that we could control for the background variables age, gender, and coursework.
2.3.3 Framework analysis
Chapter 6 was a qualitative study by means of which we investigated secondary school teachers’ beliefs about university readiness and their classroom practices regarding university preparation. We used framework analysis to analyse the transcripts of the 50 teacher interviews. is is a form of thematic analysis that uses ve clear-cut steps to bring back a whole lot of data to meaningful answers to the research questions posed. ese ve steps, as developed by Ritchie and Spencer (1994), are 1) familiarisation; 2) identifying a thematic framework; 3) indexing; 4) charting; and 5) mapping and interpretation. We used the four-part model of college readiness by Conley (2008) as a starting framework to categorise the data on secondary school teachers’ beliefs and practices regarding university preparation, but also let the data speak for themselves by giving room to themes that arose from the transcripts but were not part of Conley’s model.
Method
49