This course is a refresher about regression. This means that I assume you know regression. You may just need more context, more in depth knowledge, maybe just a review before you take a more advanced class. This course is not for you if you do not know regressions. It is not the place to learn it. (We have a week-long regression course for that.) If you think it will be enough to take this quick course and, without much practice, move to a more advanced, class, please think again. You may not get much out of such a combo.
On the first day, we cover the conceptual foundations of statistical inference. Then we launch into the math underlying regression (in the most user-friendly way humanly possible – do not worry). Then we turn to regression’s multivariate form focused mainly on the functioning of ordinary least squares. We finish the day (or if we take too long, start the next day) with a review the assumptions of regression models both from the perspective of why these assumptions exist and what happens if you violate them. We cover practical tips on how to deal with assumption violations. We start by discussing model specification both from the perspective of omitted variable bias and the inclusion of unnecessary variables. We discuss measurement scales and measurement errors of the variables in the model. We cover homoscedasticity, autocorrelation, linearity of relationships and collinearity. Though not assumptions strictly speaking, we also go into good practices that prevent assumption violations like the treatment of outliers and diagnosis of all variable distributions. We cover these topics both in theory and (on Saturday) through an example (presented in R, though no knowledge of R is assumed).
You will have no use for a computer on the first day. You may be able to follow along on a computer if you have a background in R, on Saturday, but the emphasis will be less on learning how to run things and more on looking at the results of a real-world example. In this example, we place a strong emphasis on interpretation of model results and understanding the difference between bivariate and multivariate answers to the same questions. We finish the discussion of linear regressions by looking at interactions. Examples presented come from day-to-day life, not necessarily some niche of political science, and therefore generalise well to whatever application you want to use.
At the end of the course we briefly venture into the world of logistic regression and general linear models, both in theory and in practice (using R). We put a strong focus on intuitive interpretations offered by logistic regression models. (Or in other words, how can you translate your results into language even your grandmother will understand?)
We finish with a quick overview of where you can go from the point at which this class finishes. What else lies beyond in the world of statistical analysis?