My PhD project is part of an ERC Consolidator Grant under supervision of Ellen Hamaker. The ERC project is called: ‘Coming-of-Age of Process Research: Connecting Theory with Measurement and Modelling’. Process research refers to research that focuses on psychological dynamics: the interplay between mental, behavioral, and environmental factors that take place over multiple timescales.
Studying psychological dynamics requires many repeated measurements over time, i.e., intensive longitudinal data. Examples of intensive longitudinal data are daily diaries, experience sampling method, and ambulatory assessment. Although gathering these data can be time-consuming, advances in technology have eased measurements (Connor & Barrett, 2012). Moreover, developments in statistical modelling help analyze dynamics in intensive longitudinal data (e.g., Chow, 2019).
The aim of my PhD is to contribute to connecting theory, measurement, and modelling in intensive longitudinal methodology by means of four projects. The first project focuses on understanding a specific psychological process: interpersonal interactions. A literature review reveals popular techniques that analyze dyadic data (e.g., Butler, 2011). These techniques are described in-depth so they become more accessible to other researchers, including empirical examples to illustrate how one can use these methods. Furthermore, data simulations will demonstrate what dynamical patterns these techniques can discern.
The second project examines data features using data generation and adjusting or developing new methods to match intensive longitudinal measurements with statistical models. Specifically, statistical models are assessed that incorporate measurements with different timescales. For example, how to handle sleep data, usually one or two measurements at the beginning and/or end of the day, and ESM data, usually measured multiple times during the day (see also de Wild-Hartmann et al., 2013; Nebauer et al., 2020).
The third project looks at ways to better connect models and theories through comprehensive explanations of the statistical models and their diverse model components. For example, simulation studies of various models will attempt to reproduce hysteresis, a phenomenon of a system where it switches between two distinct states, which has been suggested to play a role in the onset of psychopathology (Borsboom, 2017). Moreover, the amount of data required to detect hysteresis in practice is assessed.
The fourth project evaluates linking theory and measurements by looking at specific challenges. Measurements for intensive longitudinal data are distinguished as self-report, expert-ratings, and objective (e.g., physiological) measures, which differ not only in how they are measured but also in what timescales are possible. An overview of diverse measurements that have been used in the context of social interactions will be provided together with a thorough assessment of their strengths and weaknesses.
Prof.dr. E. Hamaker
Dr. N. Schuurman
ERC Consolidator Grant