Statistical methods for intensive longitudinal dyadic data
The use of intensive longitudinal approaches to investigate dynamic processes that shape and characterize dyadic relationships (e.g., romantic relationships, child–parent relationships) has become increasingly popular in recent years. Intensive longitudinal dyadic data are often collected by asking both partners of a dyad to report on their experiences multiple times a day for an extended period of time, or alternatively, by repeatedly assessing partners’ subjective experiences and/or physiological states during a lab experiment. Even though dyadic IL designs are already part of the toolkit of behavioral sciences researchers, key questions such as how many dyads should a study include, or for how long and when should participants be measured remain partially unanswered. In addition, the lack of methods to rigorously select the sample size can threaten the robustness and replicability of empirical findings. This PhD project will further develop methods for conducting power analysis for models typically applied to IL dyadic designs, such as the longitudinal actor-partner interdependence model, and will investigate how different design choices (e.g., the type of sampling scheme) influence the accuracy of statistical models applied to IL dyadic data. In addition, the PhD project will develop open-source software to make the methods easily available to researchers.
Prof. Dr. G. Lafit
Prof. Dr. E. Ceulemans
10 April 2023 – 10 March 2027