Longitudinal Measurement Invariance: From between to within comparisons (PhD project)
It should not matter whether the height of an egg, a chick, or a chicken is measured, as long as the same ruler is used in a standardized fashion. Photo: Colourbox
About the project
Based on common sense, measurements can be naturally compared if the same instrument has been used to make these measurements. The implicit assumption made here is that the measurement instrument remains invariant under different conditions. So it should not matter whether the height of an egg, a chick, or a chicken is measured, as long as the same ruler is used in a standardized fashion. Yet, even in this case measurement invariance is not that obvious to establish, as it is not exactly clear how ‘height’ is conceptualized for each entity in this triplet.
The current procedures for looking at measurement invariance are based on situations in which comparisons are made between persons that belong to different groups in the population. In a longitudinal context the focus is no longer on making comparisons between persons, but instead the focus is on making comparisons within persons by using repeated measures collected at different time points. The first part of the project aims to explore and study possible modifications to adjust the between-group measurement invariance procedures to the longitudinal within-group design.
The project runs from November 2017 until October 2021.
- Parent project (Latent Variable Mixture models to track Longitudinal Differentiation Patterns)
The project is funded by the Research Council of Norway