- Analysis of grouped or longitudinal data using linear mixed models
- Basic principles of statistics
- Categorical data analysis using logistic regression
- Graphics in R (FLAMES workshop)
- Introduction to Internet Questionnaires with QUALTRICS
- Introduction to JMP Pro 14 software
- Introduction to Stata
- Method in data collection
- Method in research design
- Method in scale construction
- Multiple linear regression and ANOVA
- Multivariate statistics
- R Workshop

The use of scales as a means to quantify attributes is common in psychology, social science, and health studies. A typical way of constructing scales is by formulating a set of ordinal scaled items, assign scores to the corresponding ordinal categories, and add the scores as the scale value (so-called summated rating scale). Although elegant in its simplicity, summated rating scales have a number of disadvantages that invite for the serious consideration of alternative methods.

This course will discuss two alternatives, namely factor analytical models, and Item Response Theory. Students will receive theoretical background and will perform practical exercises on existing datasets using the statistical software package R.

At the end of this course, participants:

Know about basic aspects of measurement (validity, reliability, precision, discriminating power, measurement scale)

Understand the principles of Summated Rating Scales, perform an item analysis (incl. so-called reliability analysis) in R, and interpret the results

Understand the principles of factor analytical models, compute simple factor models (e.g., by principal components analysis) in R, and interpret the results

Understand the principles of Item Response Theory, compute a Rasch model in R, and interpret the results

The course will be taught in three days of learning-by-doing. The general set-up of this course involves lecturing and practice in computer lab.

This course is targeted at doctoral and post-doctoral researchers . Knowledge of general research methodology and basic statistics is recommended. The number of participants is limited to 20. No prior knowledge of R is required to participate in this course, but an understanding of basic statistics (t test, ANAOVA, regression) is an asset.