Methodological Seminars

Semester

Semester 2, 2023-2024

Type of course

Methodological and Practical Courses

Date

May 13 and May 14, 2024

Location

Utrecht University

---

Duration

2 days

Maximum number of participants

30

ECTS

1 EC will be appointed for participation in the complete course

Staff

Marcello Gallucci (University of Milano-Bicocca)

Content, learning goals, preparation  

The goal of the workshop is to review from a theoretical and practical viewpoint the most useful applications of the linear model in psychology. This would allow students to understand and use the following statistical model: 

- ANOVA/regression, with linear and non linear effects (mediation and moderation) 

- Mixed models, applied to repeated measure designs and multilevel models 

- Generalized linear models, applied as logistic, Poisson, multinomial and similar models. 

- Generalized mixed models, which allows applying the Generalized linear model to repeated measure designs and multilevel models. 

In the examples, SPSS, R and jamovi software is used. In the practical sessions the students will use jamovi software. 

Seminars

1. A moderated view of the linear model: In this class we review the general linear model as applied to various research designs and problems. We discuss how the linear model accommodates continuous independent variables (regression), categorical independent variables (ANOVA) and a mix of them (ANCOVA). We discuss all fundamental techniques used in all types of linear models, such as moderation, mediation, simple effects, contrast analysis, post-hoc analysis.
2. A multilevel mix: mixed models, random effect models, and multilevel models: In this class we review theory and applications of the general mixed model, with fixed and random effects, as it applies to different research designs. The student will get familiar with statistical analyses of repeated measures designs, clustered data and multilevel samples.
3. Predicting categories: generalized linear and mixed models, for categorical dependent variables: In this class we review theory and applications of the generalized linear, such as logistic, multinomial and Poisson regression. The student will get familiar with statistical analyses aimed at modeling dichotomous, multi-categories, and count dependent variables. These models are discussed for a wide range of designs, including repeated measures and multilevel samples.

All seminars consists of a part in which the concepts are introduced and discussed and a practical part in which the students work together on examples and exercises aimed at clarifying the practical aspects of the analyses. 

Literature

None.