Logo Utrecht University

Dutch Mplus Users Group

Matthias Reitzle’s talk

Some considerations before you get seduced by the method

Enthusiasm for the seemingly unlimited possibilities of modern statistics have created a buoyant workshop activity. For the empirical researcher, there is an inevitable need to keep up with the latest developments of statistical procedures and statistical software packages. If senior researchers get lost in the flood of statistical innovations, the gifted doctoral students, at least, are sent out to bring the glitter of the “dernier cri” plus the skills for the latest machinery home. Unfortunately, enthusiasm sometimes turns into preoccupation, method sometimes overshadows content, and sophisticated analyses sometimes replace theoretical depth and sound measurement. The present talk is trying to present some kind of a counterweight from the perspective of a developmental researcher with regard to the conceptualization of development and change, and some measurement issues such as cross-cultural and longitudinal measurement invariance. This talk will definitely not offer statistical remedies for omitted or wooly thoughts. Instead, this talk will try to offer second thoughts which might be helpful before hopping into a statistical racing car without a driver license.