I gave a presentation on the analysis results I have done on the survey to GLU1-7 students at the beginning of the spring semester of 2019, connected to the LATACME project. We have done an analysis of this survey before, with a cluster analysis approach. The aim was then to identify groups of respondents who largely reply in a similar fashion. The aim this time is to see if there is an association (i.e., non-parametric correlation) between how the respondents reply on the modelling claims, the argumentation claims, the ICT claims, and multilingual classroom claims. Examples of these claims are in the PowerPoint presentation. The presentation shows a significant association between all pairs of topics when selecting a handful of claims from each topic. This means that if a GLU1-7 student largely agrees on the selected claims in modelling, then there is a high probability that this student will also largely agree on a majority of the selected claims in argumentation, multilingual classrooms and ICT.
I find it interesting that the association between these four topics becomes so strong.
Fortsett å leseI hypothesised that several students might have responded as they believed expected by us teacher educators. The reason is that several of the claims do not have enough information to give a response as simple as the alternatives given in the survey. I also tried to connect this result with a theory on pre-service teachers’ beliefs. I tried to argue that this can be regarded as a beliefs system, with the association between the four topics. I still need to read more theory on beliefs, attitudes and so on, to fully justify these connections.
I received some helpful comments after the presentation. Among these was literature on the theory of beliefs and cognitions in general. One comment was related to my hypothesis that the pre-service teachers were influenced by teacher educators, to explain the significant association. The comment was that the survey was carried out at the beginning of the first semester of mathematics, to avoid this kind of speculation. I will discuss this in the article I am working on. I still think it is valid to pose the question. As mentioned above, it is difficult to justify some of the responses that have come in. Also, even though the students were at the beginning of their first semester of mathematics, they previously had a semester of pedagogy and choice subject. This could create a general culture that affects students’ attitudes. This last bit is also an example of how it can be fruitful to better understand the theory of beliefs and attitudes.
A follow-up comment on this discussion was that the whole point of setting up this survey was to better understand how we can better influence our students. If it shows up that the college environment actually influences them, this is a good thing.
A question was raised on whether it is necessary to speculate on the significant association between the four topics. The analysis shows that the association is significant. Is it necessary to ask why? My first thought was that we might not need to. But later on, I have started to think that it is necessary to at least raise the question of the reliability of the data. There is research that shows how pre-service teachers revert to their «old» beliefs when finishing their teacher education and start working as in-service teachers. So this is, in my opinion, highly relevant for us.
Another good point made by one of the listeners was that Philipp defines beliefs as «more cognitive than attitudes». Is it possible that a survey where students are asked to put X-es on which claim they agree to the most can reflect cognitive beliefs? This is something I also need to figure out.
One of the commenters also reminded me that there is still data from the November 2019 survey that should be analysed. This is true. There is a lack of respondents. Furthermore, we lack tools for directly comparing with the data from early 2019. But the correlation analysis should give out results that – if we are lucky – still show significance. It would also be interesting to see if the claims that show up as «selected claims» in the analysis are different from what we found here. This could give us a kind of indication of how the students are affected by their teaching.
– Nils Henry Williams Rasmussen