Sociology is the study of how human attitudes and behavior are shaped by the social environment and how, vice versa, the social environment emerges from human (inter-)action. The investigation of context effects, where an environmental feature (e.g., a characteristic of a neighborhood or country) affects processes at a lower level (e.g., that of the individual), is therefore central to the discipline. This is also true for the topic of racist and xenophobic attitudes against immigrants, new citizens, and Black and People of Color (BPOC). For example, are native-born white Europeans who are exposed to immigrants or BPOC in their neighborhood or at work less racist and xenophobic, as Contact Theory (Pettigrew 1998) has it, or rather more, as Group Threat Theory (cf., Wimmer 1997) implies?
In recent decades, scholars have increasingly examined such context effects using quantitative statistical methods. However such analyses face serious challenges. That is, single observations are not independent from each other but are (auto-)correlated as siblings, neighbors, classmates, contemporaries, and so on. These complex dependencies among observations make contemporary data sources more sociological and thus exciting than ever before. But they also pose statistical challenges that need to be addressed via multilevel modeling techniques.
In this course, we will discuss sociological theory and research on the contextual sources of racist and xenophobic attitudes, and you will learn how to conduct statistical multilevel analyses testing these theories yourself. In other words, this is an introduction to multilevel analyses, applied to the topic of racism and xenophobia.
The default for this class is that we meet physically. However, I will try to set up a streaming of our course via Zoom. You will find the link on Absalon.
The goals for this course are twofold. First, I hope you will gain a solid understanding of contemporary quantitative research on racist and xenophobic attitudes. Second, I hope you will learn how statistical multilevel modeling works and how to apply these techniques yourself to answer research questions on context effects.
Overall, you have five tasks in this seminar: Thorough preparation of class readings, active participation in the seminar’s discussion, and three assigned portfolios.
This class is based on Snijders and Bosker (2012) as textbook on multilevel modeling. Copies should be available for purchase at the university bookstore. I suggest to buy a copy of the book, so you can work through it carefully.
Apart from this book, we read contemporary empirical studies of racist and xenophobic attitudes. You will find these readings on Absalon. I have selected the readings such that for each session they are an application of those multilevel modeling techniques that we have learned in the previous session. It is thus important that you read these empirical applications closely. On the one hand they illustrate how the abstract techniques can be put to use. On the other hand, they can be considered as a test of whether you have mastered the methodological teachings well enough to understand and parse a typical empirical application.
| Session | Topic | Required readings | Skim Reading |
|---|---|---|---|
| 1 | Intro: Racism & Xenophobia | Clair and Denis (2015), and Hervik (2015) | Bonilla-Silva (2015), and ESS ’14 questionnaire |
| 2 | Recap: (OLS) Regression | Gelman and Hill (2007, Ch. 3 & 4) | |
| 3 | Explaining Racist & Xenophobic Prejudice | Dinesen and Hjorth (2020), and Pettigrew (1998) | |
| 4 | OLS Replication | Helbling and Kriesi (2014), and Larsen and Schaeffer (2020) | |
| 20 Feb | Submit Portfolio 1 | Racism/Xenophobia and threat in Denmark. Analyse individual predictors of the Danish ESS 2014 using OLS. | Submit your Portfolio 1 to Peergrade on Absalon until Sunday 23:59 o’clock following Session 4. Afterwards, give your peers feedback on Peergrade until the following Wednesday 12 o’clock. |
| 5 | Clustered data | Snijders and Bosker (2012, Ch. 2 & 3.1.-3.4) | Gelman and Hill (2007, Ch. 11) |
| 6 | Random Intercepts | Snijders and Bosker (2012, Ch. 4) | Gelman and Hill (2007, Ch. 11) |
| 7 | RI Replication 1 | Schlueter, Masso, and Davidov (2019) | |
| 8 | Visit by Thorkil Klint from Epinion | ||
| 9 | RI Replication 2 | Hiers, Soehl, and Wimmer (2017) | |
| 10 | Work on Portfolio 2 | ||
| 13 Mar | Submit Portfolio 2 | Contextual and individual predictors of Racism/Xenophobia across Europe 1. Analyse the ESS 2014 using a random intercept model. | Submit your Portfolio 2 to Peergrade on Absalon until Sunday 23:59 o’clock following Session 9. Afterwards, give your peers feedback on Peergrade until the following Wednesday 12 o’clock. |
| 11 | Random Slopes | Snijders and Bosker (2012, Ch. 5), and Heisig and Schaeffer (2019) | |
| 12 | RS Replication | Green et al. (2020) | |
| 13 | Statistical inference | Elff et al. (2020) | |
| 14 | Model Assumptions | Snijders and Bosker (2012, Ch. 10) | |
| 24 Apr | Submit Portfolio 3 | Contextual and individual predictors of Racism/Xenophobia across Europe 2. Analyse the ESS 2014 using a random intercept and slope model. | Submit your Portfolio 3 to Peergrade on Absalon until 22 November 23:59 o’clock. Afterwards, give your peers feedback on Peergrade until 29 November 12 o’clock. |
| 7 Jun | Final paper submission | Before 12 o’clock! |