杏吧论坛

Prof Fiona Steele COL.7.12

This course is available on the BSc in Data Science and BSc in Mathematics, Statistics and Business. This course is available with permission as an outside option to students on other programmes where regulations permit and to General Course students.

A first course in statistics such as Elementary Statistical Theory (ST102), Elementary Statistical Theory I (ST109) or Quantitative Methods (Statistics) (ST107) and familiarity with multiple regression to the level of Applied Regression (ST211) or Statistical Models and Data Analysis (ST201).

Students who have no previous experience in R are required to take an online pre-sessional R course from the Digital Skills Lab before the start of the course. Please log into moodle.lse.ac.uk and self-enrol in the 'R for Statistics Pre-sessional Course’.

This course considers statistical methods for the analysis of data with a multilevel (clustered) structure with applications in social research.  Examples of multilevel structures include students nested within schools or universities, and individuals nested within households or geographical areas.  Multilevel structures also arise in longitudinal or panel studies where repeated measurements over time are taken on subjects (e.g. individuals or countries).  The course has an applied emphasis, and will consider practical issues such as the choice of an appropriate method of analysis for a given dataset and research question, descriptive analysis and modelling, model selection, and visualisation and interpretation of results. Students will gain hands-on experience of data analysis using R and datasets drawn from a range of social science disciplines.

The course will include the following topics:

• Introduction to multilevel data structures and research questions
• Simple model for a two-level structure; comparison of fixed effects and random effects approaches
• Random intercept models
• Handling level 2 endogeneity in random effects models
• Random slope models
• Higher-level explanatory variables: contextual, between-group and cross-level interaction effects
• Longitudinal data analysis
• Model selection
• Special topics, e.g. models for three-level structures and for non-hierarchical structures