If you participate in one of my courses you will be confronted by quantitative methods one way or another. My courses always contain some lectures on methods as well as lab sessions. However, time is scarce and usually there is not enough time to refresh and discuss in detail the simple steps of data analysis. Below I collected some info pages I use(d) as well.
There are excellent tutorials online which address how to load data in R, how to generate variables, how to create figures etc. Some of the sources I list below:
- UsefulStataCommands.pdf -> searching for a stata command?
- UCLA tutorials -> always check UCLA’s pages if you are running into trouble.
- polmeth.ch -> straightforward and plain R tutorial by Marco Steenbergen.
- cyclismo -> more encompassing R tutorial.
- tutorialspoint -> even more on R (loops as well).
- R bloggers -> tons of info on cool stuff you can do with R.
Thanks to Marco Steenbergen the IPZ also offers a “R helpdesk“. I encourage students to get in touch with the R helpdesk in case they face all kinds of R problems.
Students constantly run into troubles finding data sources. Usually google is your best friend. If you still struggle to find reliable information on a specific topic the following list might be of some help: data sources (Comparative Politics).
When You Crunch Those Numbers…
p-values (don’t be overly existed if p<0.05):
How to correctly interpret interaction terms:
Care about the design of your graphs:
If 95% confidence intervals (CIs) overlap, does this mean that two groups are not significantly different from each other? NO:
- Overlapping CIs and statistical significance
- A brief note on overlapping confidence intervals
- Overlapping confidence intervals or standard error intervals: What do they mean in terms of statistical significance?
- Use 83% CIs instead!
- Merging Graphics and Text to Better Convey Experimental Results: Designing an “Enhanced Bar Graph”
What is causality/a causal effect?
- Causal inference
- Statistics & causal inference
- Effects of causes and causes of effects
- 10 things to know about causal inference
Be careful with causal claims:
Some thoughts on modelling your data: