Analysing Comparative Longitudinal Survey Data Using Multilevel Models
Description:
Many surveys spanning multiple countries--or many regions within a single country--are now being fielded multiple times over the course of years or even decades. Examples include the European Social Survey, International Social Survey Programme, and (across states) the U.S. General Social Survey. The range of topics that can be studied using data from these surveys is extremely broad: from health to religiosity to political attitudes and behaviours. This course will show students how to analyse these comparative longitudinal survey data (CLSD) using multilevel models that exploit any or all of three different kinds of variation: differences between countries, change within countries over time, and variation across individuals.
We will begin by considering the structure of CLSD, and then what fixed effects and random effects (multilevel) models each reveal about the variation between and within groups in data characterised by clustering. We will see how CLSD can be understood as doubly hierarchical (or clustered), and therefore how we can analyse them with models partitioning between and within effects. We will also consider the capabilities of societal growth curves, and the insights that can be gained from models with random (country-specific) slopes. The course will emphasise the use of graphical analysis throughout, and note some risks that analysts of CLSD need to avoid. In the lab sessions will use the open-source R software and environment for statistical computing, including some easily installable add-on packages.
Prerequisites:
Basic understanding of multilevel models, and some experience of designing and fitting them. Students unfamiliar with multilevel/random effects models, but who are strongly familiar with other techniques for panel data analysis, are also welcome to join. Students are not required to have any prior familiarity with R.
Useful background readings and illustrative applications:
Bell, Andrew, and Kelvyn Jones. 2015. "Explaining Fixed Effects: Random Effects Modeling of Time-Series Cross-Sectional and Panel Data." Political Science Research and Methods 3[1]: 133-153. DOI: 10.1017/psrm.2014.7.
Fairbrother, Malcolm. 2014. "Two Multilevel Modeling Techniques for Analyzing Comparative Longitudinal Survey Datasets." Political Science Research and Methods 2[1]: 119–40. DOI: 10.1017/psrm.2013.24.
Fairbrother, Malcolm, and Isaac W. Martin. 2013. "Does Inequality Erode Social Trust? Results from Multilevel Models of US States and Counties." Social Science Research 42: 347–360. DOI: 10.1016/j.ssresearch.2012.09.008.
Schmidt-Catran, Alexander W. 2014. "Economic Inequality and Public Demand for Redistribution: Combining Cross-Sectional and Longitudinal Evidence." Socio-Economic Review: 1–27. DOI: 10.1093/ser/mwu030.
Detailed schedule:
(1) 6 July 2016 (15.00 to 18.00)
- understanding and characterising the structure of CLSD
- fixed effects and random effects (multilevel) models
- between and within relationships
- Lab:
- merging survey and macro data
- mean-centering
- graphing CLSD
- fitting fixed effects and random effects models to two-level data
(2) 7 July 2016 (15.00 to 18.00)
- cross-sectional and longitudinal relationships in CLSD
- societal growth curves
- random slopes
- Lab:
- fitting multilevel models to CLSD
Instructor's bio:
Malcolm Fairbrother (seis.bris.ac.uk/~ggmhf/) is Senior Lecturer in Global Policy and Politics and an associate member of the Centre for Multilevel Modelling at the University of Bristol. He holds a PhD in sociology from the University of California, Berkeley, and has been a visiting researcher at institutions in Canada, Mexico, Sweden, Spain, and Italy. His methodological research has concentrated on the use of multilevel models in analysing comparative survey data. In his substantive research, he focuses on understanding public attitudes towards environmental degradation and protection, and the correlates of social trust. He has also used qualitative comparative-historical methods to investigate the political origins of economic globalisation. His work has appeared in journals such as the European Sociological Review, American Journal of Sociology, and Political Science Research and Methods. He teaches environmental policy and politics, political economy, and research methods in Bristol's School of Geographical Sciences.