The second part of this course is about the entanglement of psychometrics and data science. Most psychometric models are designed to estimate ability levels based on dichotomous or polytomous response data. With other data sources available, like for example textual data, videos or logfiles, the question arises how these sources of information can be used to get insight in the ability levels of the students. Therefore, new models from the field of data science can be applied.
Text mining and machine learning models will be studied as new ways to measure individual traits. State of the art tools will be applied to handle various data sources.  Finally, the integration of psychometrics and data science is studied. One approach to combine them is to develop a kind of architecture where both kinds of models are treated as different modalities that provide information about the same latent variables. A different approach is to combine these new types of measurement with item response theory within a Bayesian framework.