We attempt to investigate the causal effects of instructional time and amount of professional development of teachers on reading comprehension. Cross-sectional studies are frequently inflicted by issues as selection bias and omitted variables, which make causal inferences impossible. However, selection bias can be dealt with by aggregating variables to a higher level, whereas relating change in certain characteristics to change in student outcomes diminishes to a great extent the issue of omitted variables as most confounding variables remain constant over time (Gustafsson, 2007). This implies that country-level longitudinal studies (such as PIRLS-waves) are promising databases that allow for causal inferences (see e.g. Nilsen, Gustafsson, & Yang-Hansen, 2017). In this study we use two waves (2006 and 2016) of PIRLS (Progress in International Reading Literacy Study). The dependent variable is the score for reading comprehension of Grade 4 students in 2006 and 2016. Twenty-nine countries participated in these two cycles and have comparable and reliable trend data. We use a difference-in-difference approach with both correlation matrices and regression models with fixed effects. The regression relationships are shown in the form of scatterplots. We use plausible values and weighted country aggregates of the variables. As indicators of instructional time we consider the percentage of instructional time used for language instruction in general and for reading instruction more specifically. The other variable is the number of hours spent for professional development on reading instruction during the last two years. We also consider changes in age of the students. In studying the effects of age and schooling, we come to a slightly different ranking of the countries in PIRLS 2016. Some of the results are that changes in achievement are linked to changes in professional development but not to changes in the instructional time. At ECER we will discuss the results.

Gustafsson, J.E. (2007). Understanding causal influences on educational achievement through analysis over time within countries. In T. Loveless (Ed.), Lessons learned : What international assessments tell us about math achievement (pp. 37-63). Washington, DC : The Brookings Institution. Nilsen, T., Gustafsson, J. E., & Yang-Hansen, K. (2017, August). Changes in teacher quality and school climate influencing changes in achievement from 2011 to 2015. Paper presented at the ECER conference, Copenhagen.