31 2 Table 2.2 Estimates of the variance components of the GT model Parameter Estimate Standard Deviation Median σ j 2 1.000* - 1.000 σ t 2 0.368 0.056 0.362 σ i |j 2 0.275 0.029 0.273 σ jt 2 0.423 0.053 0.421 σ it |j 2 0.411 0.050 0.407 *The main effect of the teacher was set to 1 to identify the model. It was expected that most of the differences between scores from students would be explained by differences between teachers, because teachers differ in teaching quality (van de Grift, 2007, 2010). Furthermore, differences between students explained a lot of the total variance. Students may have a different perception of the quality of a particular teacher’s teaching, as the perceptions might be mediated by, for instance, their general impressions of the teacher, teacher and student gender, cultural backgrounds, or the student’s level of performance in the subject taught by the teacher (den Brok et al., 2005; Levy et al., 1992; Levy et al., 1997; Rowley, 2003). Figure 2.1 Reliability estimates for different numbers of students (N) and measurements. 0,5 0,6 0,7 0,8 0,9 1 2 3 4 5 6 7 Reliability Number of Measurements N=19 N=10 N=5
RkJQdWJsaXNoZXIy MjY0ODMw