PUBLISHED PROJECTS
COMPARING ESTIMATION METHODS FOR PSYCHOMETRIC NETWORKS WITH ORDINAL DATA
Co-Author: Dr. Mijke Rhemtulla
Little research has investigated how psychometric network models (which are used to explore the associations between variables) perform when estimated on ordinal data, which are common in psychological research. We compared the performance of three network estimation techniques when applied to ordinal data, and evaluate how performance changes under different characteristics of the ordinal data.
​
COLLECTING LONGITUDINAL DATA: PRESENT ISSUES AND FUTURE CHALLENGES
Chapter in APA Handbook of Research Methods in Psychology (2nd Edition)
Co-Authors: Rohit Batra & Dr. Emilio Ferrer
In this chapter, we provided an overview of the various decisions that researchers might have to make when collecting longitudinal data (such as making decisions related to measurement instruments, selection of persons, or how many assessments to collect) and the associated challenges. We also provide some recommendations for researchers on how to deal with such challenges.
​
Find the book (and our chapter) HERE
CONSEQUENCES OF SAMPLING FREQUENCY ON THE ESTIMATED DYNAMICS OF AR PROCESSES USING CONTINUOUS-TIME MODELS
Co-Authors: Rohit Batra, Dr. Meng Chen, & Dr. Emilio Ferrer
Continuous-time (CT) models provide a flexible approach to modeling as they do not rely on a single time interval for interpretation, but assume and model a continuous underlying process. The parameters of CT models can theoretically be rescaled to any common time interval, but little work has investigated how well CT models can recover the true dynamics of a process when there is a mismatch between the sampling interval and the time scale of the true generating process. Our work evaluated the ability of a CT autoregressive model to recover the dynamics across different generating intervals, sampling frequencies, AR strengths, and number of observations.
​
Find the paper HERE!