Multilevel and Longitudinal Modelling
Volledige vakbeschrijvingMost datasets in European studies contain data whereby the traditional assumptions of ordinary least-squares regression are violated, primarily in a cross-sectional and longitudinal context and because the dependent variable is non-continuous. Individuals are grouped into countries, (Eurobarometer surveys), European Union decision-making is recorded on an annual basis (EUPOL dataset), or members of the European Parliament are asked in how far they agree (five answer categories) with the statement that the European Parliament should have more powers with regard to a particular policy item (EPRG MEP Survey). This course aims to provide an introduction into the most commonly used advanced statistical models to deal with clustered and (auto- or multi-)correlated data and categorical limited dependent variables. One learns statistics best by applying the techniques to a substantive topic of interest. Students are asked to choose a dataset (with one or more of the ‘violations’ mentioned above) on a research topic related to the seminar that runs parallel to this course and they will work on this dataset from the first week onwards. Students will acquaint themselves with a method suitable for their research purposes which may include clustered and panel corrected standard errors, random and fixed effects models, multilevel models and logit, ordered logit and multinomial logit models.
Doelstellingen van dit vakAfter this course students will be: • Introduced to the assumptions underlying advanced statistical methods (hence students are not expected to be able to ‘do the math’ themselves); • Able to identify the data structure and to recognize potential violations of traditional assumptions of ordinary least-squares regression; • Able to choose an appropriate statistical model in case of (1) times-series-cross-sectional data and/or (2) categorical and limited dependent variables; • Able to analyze commonly used datasets in European studies and properly apply statistical models in a relevant statistical package (Stata or SPSS).
Aanbevolen literatuurBeck, Nathaniel and Jonathan N. Katz. (1995). ‘What To Do (and Not To Do) with Time-Series Cross-Section Data’, American Political Science Review, 89: 634–47. Beck, Nathaniel, Jonathan N. Katz and Richard Tucker. (1998). ‘Taking Time Seriously: Time-Series–Cross-Section Analysis with a Binary Dependent Variable’, American Journal of Political Science 42: 1260–88. Hsiao, Cheng. (2003). Analysis of Panel Data. Second Edition. Cambridge: Cambridge University Press. [Specific chapters] Scott Long, J. (1997). Regression Models for Categorical and Limited Dependent Variables). London: Sage Publications. [Specific chapters] NB further suggested readings will be made available depending on methodological interests of students.
1 feb 2021
2 apr 2021
Taal van de opleiding:Engels
Trefwoorden:Quantitative skills, statistical models, time-series-cross-section datasets, multilevel models, limited and categorical dependent variables