Select or construct an appropriate dataset that has the appropriate variables:

• For an independent t-test or ANOVA, you need two or more groups/ categories as the IV and a continuous variable as the DV.
• For a dependent t-test, you need the same observations with two continuous variables. (Check the slides or request advice if you are unsure.)
• For a regression, you need a continuous DV. Ideally, you would also have a continuous IV, but you may consider using dummy variables to capture categories. (Yes, that would really just an ANOVA in disguise.)
• Again, using the European Social Survey is completely fine, but students are asked to create their own dataset by subsetting/ filtering their data
  • Selecting different variables than in the ESS data subset
  • Selecting different countries than in the ESS data subset (keeping all countries may be perfectly fine, depending on the conceptual model.)
  • Removing missing values

1. Briefly describe the dataset, e.g. who created it, why and how did they create it?
2. Briefly explain what concepts are represented by each variable and how each variable is measured.
3. Explain what relationship you expect to find between the variables and why you expect to find that relationship. State this expectation as a set of hypotheses.
4. Generate one organized table of descriptive statistics that includes relevant information from all variables used in the model. Include any additional tables or visuals, as appropriate, e.g. frequency tables, histograms, etc.
5. Analyze the relationship between the variables using the appropriate statistical test. Present all relevant information from these results in a tidy table, visualize your results, as appropriate.
6. Interpret your results as supporting or not supporting your research hypothesis.
7. Interpret your results as supporting or not supporting your research hypothesis.