Propensity Score Matching

Propensity Score Matching and Structural Equation Modelling using R

Structural Equation Modeling (SEM) and Propensity Score Matching (PSM) are widely used by researchers in the social, behavioral, educational, and business sciences. SEM is a combination of psychometrics and econometrics. On the psychometric side, SEM allows for latent variables with multiple indicators. On the econometric side, SEM allows for multiple equations. Structural equation models (SEM) allow both confirmatory and exploratory modeling, meaning they are well suited to both theory testing and theory development. 

In the statistical analysis of observational data, Propensity Score Analysis is a statistical technique that attempts to estimate the effect of a treatment, policy, or other intervention by accounting for the covariates that predict receiving the treatment. It attempts to reduce the bias due to confounding variables that could be found in an estimate of the treatment effect obtained from simply comparing outcomes among participants that received the treatment versus to those that did not. 

The seminar uses R to demonstrate the implementation of propensity score analysis and structural equation modelling. The R software can be downloaded for free.

The course is run by Dr Daniel Boduszek who has used propensity score matching and structural modelling in his numerous psychological, social science, and medical research publications. The Quantitative Research Methods Training Unit (QRM-TU) will also invite Associate Members to lead the training.

Course outline

  1. Why and when Structural Equation Modelling and Propensity Score Matching are used
  2. Theoretical Introduction to Structural Equation Modelling Propensity Score Matching
  3. Introduction to R environment 
  4. Confirmatory Factor Analysis in R 
  5. Model specification, model fit, and model modification
  6. Structural Equation Modelling in R
  7. Propensity Score Matching in R (optimal and greedy matching)
  8. Post-matching Structural Equation Modelling
  9. Interpretation and reporting of results for publication purpose
  10. Practical session

Who should attend?

The course is designed for researchers and postgraduate students who are engaged in research with large data sets. The prerequisite for taking this seminar is basic knowledge of regression analysis. Researchers from economics, public health, epidemiology, psychology, sociology, social work, medical research, education, and similar disciplines are welcome.

Why Quantitative Research Methods Training Unit?

  • We know that learning statistical analysis can be a daunting and unpleasant experience at times. This is why we present complicated procedures in a simple way, avoiding jargon and confusing mathematical formulas at all costs! We teach researchers everything they need to know about structural modelling in order to finish their projects in a relaxed and friendly atmosphere where they are helped and encouraged every step of the way.
  • The uniqueness of Quantitative Research Methods Training Unit is the focus on the practical application of SEM and PSM, not mathematical procedures.
  • Dr Daniel Boduszek and Associate Members of QRM-TU have an extensive experience in application of SEM and PSM techniques which is demonstrated by their research outputs.
  • The course is delivered by means of lectures and practical sessions. All analyses are conducted under supervision of Dr Daniel Boduszek and three tutors (Dr Susie Kola, Dr Katie Dhingra, and Ms Kathryn Sharratt) to maximise learning outcomes (we offer a 1:5 instructor student ratio). Each course concludes with a question and answer session, and there is always the opportunity for participants to discuss their own studies with the tutors.

Location

The Quantitative Research Methods Training Unit (QRM-TU), Ramsden Building, University of Huddersfield.

Book now

To find out when the next training session will be held, please visit our online store where you can also book your place.