Measurement Models in Quantitative Research (UvA)

Course Objective

When scholars operationalize their core concepts, they implicitly or explicitly construct measurement models. This involves the measurement of concepts that are not directly observable (i.e. latent) such as loneliness and religiosity. Constructing such models is not straightforward, as one needs to apply sometimes complex statistical models using input from theory.
This course introduces students to the basic principles and implications of latent variable modelling, including factor analysis and latent class analysis.
By the end of this course, students should understand the basic principles of measurement models and distinguish between the main types of this models as well as reflect critically on the benefits and limitations of them. Additionally, students should be able to apply confirmatory factor analysis and latent class analysis in the programming/statistical software R, and to communicate the results of these models to others. Thereby, this course contributes to students’ mastery of relevant research methods and techniques, their ability to translate research problems into relevant research designs, to adopt a clear position in academic discussions on methodological designs regarding measurement models, to use relevant computer software (R in particular), and to present their social science research findings in writing – as formulated in the exit qualifications of the RMSS.

Course Content

This course is about measurement models that are based on latent variable modelling. Measurement models are constructed to optimally link theoretical concepts to empirical observations. This can refer either to estimating a continuous latent variable that measures a concept from a number of observed variables (i.e. a variable-centered approach) or estimating categories of a latent variable that define population subgroups of a number of observed variables (i.e. person-centered approach). Measuring theoretical concepts with statistical models requires advanced methodological knowledge, conceptual clarity, correct use of theoretical input as well as scholarly creativity. This course provides the necessary methodological knowledge but also trains students to balance all these aspects in the application of measurement models. This course begins by provides a broad overview of different types of measurement models and by refreshing/training students in basic statistical procedures that are needed to apply these models (i.e. linear and logistic regression). Moreover, the course elaborates on two important measurement models: confirmatory factor analysis and latent class analysis. Both methods are extensively discussed both theoretically and empirically (in R). For latent class analysis, we also discuss some advanced extensions of it in longitudinal data (i.e. hidden Markov models

Teaching Methods

Teaching involves 2 weekly meetings of 3 hours each on Mondays and Wednesdays. Typically, the first part of each meeting is devoted to the presentation of the statistical model or a particular measurement model. In the second part, students are trained in the application of the model in R using real data. In the first week, we will deviate from this structure, as students need to be introduced in R.
During the course, we will exclusively use the software R. As many students are not familiar with the program, we offer a series of online tutorials that can help students to familiarize themselves with the basic aspects of the program.

Method of Assessment

The assessment of the course includes 3 parts:
1. A participation grade (10% of the final grade). Students are expected to be present and actively participate in all meeting. Failing to do so in a meeting (without a reason) will result in the deduction of 1/8 of this part of the final grade.
2. Completing an individual assignment on confirmatory factor analysis (45% of the final grade). This assignment will be given in week 2 and should be submitted via Canvas until the end of week 3. A pass grade for this assignment is required to get a pass grade for the course.
3. Completing an individual assignment on latent class analysis (45% of the final grade). This assignment will be given in week 3 and should be submitted via Canvas until the end of week 4. A pass grade for this assignment is required to get a pass grade for the course.
For the resit students will be given the opportunity to improve one or both their assignments

Literature

Agresti, A. (2018) An Introduction to Categorical Data Analysis, 3rd Edition, John Wiley & Sons, Inc, Hoboken, New Jersey.
Brown, T. A. (2015) Confirmatory Factor Analysis for Applied Research. Second Edition. The Guilford Press.
Collins, L. and Lanza S.T. (2010) Latent Class and Latent Transition Analysis: With Applications in the Social, Behavioral, and Health Sciences, John Wiley & Sons, Inc, Hoboken, New Jersey.
Darlington, R. B. and Hayes, A. F (2017) Regression Analysis and Linear Models: Concepts, Applications, and Implementation. London: The Guilford Press
Sijtsma, K. (2015). Classical test theory. In S. J. Henly (Ed.), Routledge international handbook of advanced quantitative methods in nursing research. Abingdon, UK: Routledge.

Target Audience

Research Master