Kurssin kuvaus

One of the important tasks of social science research is to understand change in behaviour and attitudes of individuals over time. Longitudinal data measures repeated observations on the same individuals (or other subjects). This course describes how to manage, describe, and model longitudinal (and multilevel) data as it pertains to the types of research carried out in the social sciences. It offers practical experience with real world examples and an overview on the estimation of models for continuous and categorical dependent variables, focusing either on the level of these variables at different points in time or on their change over time including fixed and random effects models and event history models.

The course is intended for students of MDPINVEST and DPINVEST. The course is also open to all graduate students and researchers in INVEST and Department of Social Research.

Learning outcomes

After the course, the student will understand the basic properties of three methods suitable for longitudinal and hierarchical data: panel regression, multilevel models, and (discrete) event history analysis. The student is able to read relevant methods literature and applied empirical research articles. The student is able to conduct their own research using at least one of the methods and communicate their results orally and in a written report.

-        Understanding and managing longitudinal/multilevel data.

-        Describing longitudinal/multilevel data.

-        Formulating and answering an appropriate research question using longitudinal/multilevel data and appropriate methods.

-        Using Stata for managing and editing longitudinal/multilevel data and estimating basic models for such data.

Study methods

The course is realized as a pre-course assignment, lectures (including group work and other in-class assignments), guided Stata labs, and weekly homework during period 3 (Master students: lectures 30 h, labs 16 h, independent work 80 h; PhD students: lectures 30 h, labs 16 h, independent work 45 h). These are followed by a short report, a seminar presentation, and a final report during period 4 (90 h / 44 h). The student is expected to participate in all lab sessions and the seminars, and assessment is a mixture of attendance, lab exercises, homework, and a final report. The learning platform is TIM: https://tim.education/view/utu/kurssit/ykt/sosiaalitieteet/longitudinal/2022/home

Pre-requisites

Multivariate methods course or equivalent. Understanding of multivariate linear and logistic regression models. Applied data analysis using a statistical programme package; some knowledge of STATA is helpful.