33 2 As far as the content of the items is concerned, the results are quite plausible. Item 2 (“The teacher explained the subject matter in such a way that I understood it well”), for example, contributed strongly to the scale (the item information values were 1.56, 1.30, 1.31, 1.56, and 1.15 respectively). It makes sense that this item reflects teaching quality well from a student perspective. This also applies to the aspect of teaching quality addressed by the item asking “If I did not understand the subject matter, the teacher made sure I understood it” (item 6). Item 6 had the highest contribution to the scale at all five reported points (the item information values were 1.56, 1.31, 1.32, 1.56, and 1.14, respectively). Formative assessment (checking for students’ understanding; item 4) contributed highly to the construct of teaching quality (the item information values were 0.88, 1.12, 1.11, 1.11, and 1.11, respectively); however, teachers in general hardly ever did this in their lesson, according to students (it was hard for teachers to receive a high score on this item). Either teachers indeed did not show this skill in the lesson or students did not notice teacher behaviour related to this skill. This also applies to item 5 (“The teacher asked questions about the subject matter, which made me think”) and item 12 (“The teacher ensured my active participation in the lesson”). Remarkably, item 14 contributed little to the teaching quality scale (the item information values were 0.22, 0.34, 0.49, 0.60, and 0.63, respectively). According to students, teachers rarely summarized what was learned at the end of the lesson (item 14), as it was very hard for teachers to receive a high score on this item (it had the highest Figure 2.1 Reliability estimates for different numbers of students (N) and measurements. has less effect on reliability. Local reliability Table 2.3 shows the discrimination ( -value) and location ( - (k (which are a function of the -value and the - latent scale (the distribution of teaching quality scores), namely at q item information value, the higher the contribution of the item to the teacher scale. 0,5 0,6 1 2 3 4 5 6 7 Number of Measurements N=10 N=5 k a k d k a k d -value, at 0.97). Teachers might routinely forget to do this, or not know or acknowledge the relevance of summarizing. As summarizing what was learned at the end of the lesson as found to be a characteristic of ffectiv teaching in our literature review, and because the item f it well in the model used for the analyses (see the average absolute differences in Appendix C, in tables C1, C2 and C3), in our opinion this item should remain in the questionnaire. Leaving this item in the Impact! questio naire might mak teachers more aware of the relevance of summarizing a lesson, and see it as point of improvement. The discrimination parameters of the items ranged from 1.11 to 2.40. This means that students could make a clear distinction between the different dimensions of teaching quality (which is also referred to as discriminant validity; Fauth et al., 2014) measured by means of the Impact! tool. This result is in line with the f indings from other studies on student perception questionnaires, for example, the studies by Kunter and Baumert (2006), Fauth et al. (2014) and van der Scheer et al. (2019).
RkJQdWJsaXNoZXIy MjY0ODMw