Estimating and correcting for measurement error using hidden Markov models
📍 Hybrid Event: MZES, Mannheim + Zoom
📆 September 20, 2023
Hidden Markov models (HMMs) are a group of latent class models that allow for the estimation and correction of measurement error in categorical, longitudinal data. The main advantage of these models is the fact that they do not rely on the availability of an error-free data source that is used as a benchmark to validate error-prone data. Instead, these models make use of the availability of multiple measures of the same indicator over time to extract information about the error from the data itself. In this workshop, I will provide an introduction to HMMs, discuss how they work and how they can be used in practice for measurement error correction. I will also show how standard HMMs can be implemented in R and how more complex specifications can be implemented in specialized software, specifically Latent Gold.
👤 Paulina Pankowska is an Assistant Professor at the Sociology Department of Utrecht University. Her research relates primarily to data and methods quality in the social sciences. In 2020 Paulina defended her PhD dissertation titled: 'Measurement error: estimation, correction, and analysis of implications', which investigated the feasibility of using hidden Markov models (a latent variable modelling technique) to account and correct for measurement error in survey and administrative data. The project was conducted in collaboration with Statistics Netherlands.