Exam

For this course you will need to hand in a “written assignment”. In particular, you will need to use the ESS 2018 and several country-level data sources to replicate the following study:

Eicher, Véronique, Richard A. Settersten, Sandra Penic, Stephanie Glaeser, Aude Martenot, and Dario Spini. 2016. “Normative Climates of Parenthood across Europe: Judging Voluntary Childlessness and Working Parents.” European Sociological Review 32(1):135–50.

Note that the experiment reported in that study was repeated in ther ESS 2018. Your task is to investigate whether the central findings still hold by replicating their study as closely as possible. Note that you will not need to run any of the random slope / cross-level interaction models reported in that study. Moreover, also note that one of the central questions of the original analysis was actually not posed in the ESS 2018. Therefore please disregard the normative climate analysis and focus on the question how women and men judge women and men differently and at the country predictors (GDP, etc.).

You will need to write your exam using R Markdown and embed your code in the final paper. This is a prerequisite! Note that while the code shall be embedded in the final paper, it does not count as part of the exam's 10 pages; although embedded within the text, the code counts as Appendix. Tables and figures also do not count as part of the 10 pages. Also beware that you may not write more than 10 pages.

The main purpose of the exam is to show that you can now master R on your own. You need to show that you can engage with a research question using R, and prepare a dynamic R Markdown report of your analysis. You should embed your code, because it will be an important part of the evaluation. I will grade the written exam based on the following criteria, which I list in order of their importance:

1. Quality of the R code: Demonstrate that you have mastered the tidyverse, including, piping, joining, reshaping, recoding missing values, etc. Try to write short and legible code (i.e., avoid unnecessary long and repetitive code). Do not just paste your scrambled code from throughout the semester into the document.
2. Quality of sociological argument / deviation from template: Your essay should be an independent essay discuss the arguments proposed by Eicher et al. and also discuss own thoughts, critiques and ideas. The template I will upload is just an example of how you can write an RMarkdown document, but it is not a good essay.
3. Quality of the analyses: Present figures, tables and models that actually answer the questions you raise. Show that you can use R to produce statistics that answer your questions.
4. Quality of interpretation: Interpret your your results correctly and in a way that uninformed readers can follow why you present the results that you do present.
5. Quality of the output: Label your variables in tables, put figures into sizes and shapes that make them readable, label the axes, etc. You do not need to go beyond what I did in the slides, but you will find a lot in there.
6. Quality of the actual writing: Finally, good writing, logic argumentation, and a theoretical motivation for your paper by referring to at least some existing literature always matters, too. No thorough review is needed. But the text should have a good text flow and live up to the basic standards of a social science report.

Further information on exams

Please also refer to the Department of Sociology's detailed website on exams.