New Methods for New Methods

This collection provides work on the challenge of preparing doctoral students in education to develop competence in the advanced statistical methods they need to properly analyse the complex data that education generates.For example: Knowing whether a new curriculum program has impact over the status quo requires matching students in the innovation schools with similar students in a counterfactual situation (propensity matching), while taking into account that students are taught in groups which influence the nature of an innovation (nested hierarchical analysis) and that some students will be absent at data collection points (missing data analysis), and determining the effect of change (longitudinal value added analysis) may have to use non-equivalent measures of effectiveness (equating required) and take into account multiple confounding factors.Thus, this collection provides source documents and working documents around the idea that new ways of teaching and designing curriculum are probably needed to assist future post-doctoral researchers to cope with the complexity of educational data.