Description

Author

Stefan König Department of Sports Science, University of Education, Weingarten, Germany

Abstract

Working in educational research, scientists often face the situation of nested data in both cross-sectional and longitudinal studies. With regard to repeated measurements, multilevel modeling has turned out to be an adequate approach to handle this issue of intact subgroups even in small-scale research, but it is stretched to its limits when predictors have to be explained beyond quantitative coefficients. This calls for the integration of qualitative data to enhance and to complement the validity and significance of statistical results. This article presents a multilevel mixed methods study from research on teaching in physical education (RT-PE). It deals with the problem of fitness education, an issue that has been discussed for years due to several alarming observations of children’s and juveniles’ personal and social development (e.g., obesity, poor motor performance, or an obvious decline in the ability to concentrate). Its focus is on a crossover quantitative-dominated mixed analysis that integrates qualitative data into a hierarchical multilevel regression model for change over time. This analysis yields some suggestions for future implications of mixed methods research designs.