Introduction to Inferential Statistics

With the ubiquitous increase in the availability of quantitative data in political science, there is a growing need for literacy in analyzing and interpreting it. The goal of this course is to teach students the basic statistical techniques used in the field, which serve as a basis for moving on to more advanced statistical techniques. We will first go over various descriptive statistics which help organize and summarize data and then move on to inferential statistics which help answer general questions about unknown populations on the basis of limited information gathered from samples, the case for most of social science research.

The first part of the course will introduce the distinction between population and sample and discuss sampling error. We will then touch upon issues of measurement and scale, including the relationship between concepts and indicators, as well as validity and reliability. Next, we review frequency distributions and different measures of central tendency. Finally, we cover three measures of variability: the range, standard deviation and variance and go into the foundations of inferential statistics by discussing probability distributions, the normal distribution, the central limit theorem and standard error.

The second part of the course introduces some basic inferential techniques used in social science research. We will cover hypothesis testing, the difference between type I and type II errors and focus on two types of tests for comparing groups which use sample means and mean differences to draw inferences about the corresponding population parameters: the t-test for two independent and two related samples. We finally look at the association between categorical variables with the chi-square test for goodness of fit and the chi-square test for independence.

The final part of the course covers associations between continuous variables, namely bivariate correlation and simple linear regression. We will discuss the principles behind these two statistical techniques, the interpretation of results, the difference between correlation and causation and finish with model assumptions and the implications of violating them.

At the end of the course, participants should have a good understanding of the descriptive statistics used for organizing and summarizing results and of the principles of inferential statistics used for establishing relationships between samples and populations. Although no previous statistical knowledge is required, given the breadth of material covered, participants are expected to prepare for the course by completing the assigned readings. The lab sessions will be held in SPSS and will follow closely the structure of the material presented in class, giving participants the possibility to apply in practice the techniques learned in class.

Although this is a refresher course and it is expected that most participants have been exposed to statistics courses in the past, the course is designed such that no statistical knowledge is pre-required.