Introduction to Stata

This course is suited to students who have no experience with Stata, but who bring in basic statistical skills, e.g. some knowledge of descriptive, bivariate and multivariate statistics.

Its aim is to show how Stata can be best used to conduct statistical analyses. Stata relies strongly on the command line interface. Thus, the course gives an insight to general construction of Stata code (syntax). The instructor and teaching assistant will help you to write clear and parsimonious Stata commands, and improve do-files and do-file documentation.

You will learn to use do-files as a simple lab-book, where annotations and ideas can be recorded in addition to the raw code, and to facilitate replicable analyses.

First, we introduce the software package:

• What is it?
• What does the Stata interface look like?
• How does it work?

To demonstrate, we give an applied overview of the various screens and windows.

We then move on to data management (importing and exporting data from various sources, recoding variables and labels, generating new variables), uni- and bivariate statistics.

On Saturday morning, we will cover the basics of regression models.

Depending on the specific interests of participants, the Saturday afternoon session will focus on more advanced topics, such as variations of regression models (e.g. logistic regression), graphs and tables of results, and loops and macros.

• No previous knowledge of Stata required
• Basic statistical knowledge is assumed (e.g. descriptive statistics, basics of regression analysis)
• Knowledge of syntax-based software (e.g. SPSS, R) helpful but not required

This course introduces standard statistical procedures in Stata. It does not give detailed insight into the statistical procedures and underlying mathematics.