Skip to main content Skip to navigation

QS105: Introduction to Social Analytics II

multipleregressionanalysis.gif

Timing and CATS

This module will run in the Spring Term and is worth 15 CATS.

Further information.

Module Convenor: Dr Florian Reiche

 

Module Description

This module builds on QS104 which is essential for taking this module. QS105 starts with cross-tabulations which help us to establish relationships between two categorical variables (such as gender and preference for ice-cream flavour, for example). Relationships between variables can equally be investigated by measures of correlation which the module will cover subsequently. A substantive part of QS105 is dedicated to regression analysis which is one of the standard tools of social science researchers. This method cannot only demonstrate whether relationships between variables exist, but also quantify the magnitude and direction of such a relationship. We will progress in small steps, starting with the intuition behind the method, then moving on to two-variable regression analysis. We will explore the assumptions of the Classical Model and how to assess model fit. The module will close with interval estimation and hypothesis testing. As in QS104, we will apply the methods you learn about in lectures (1h) in extended seminars (2h) where we will again be working with STATA on real, secondary data sets. Again, there will be weekly quizzes (non-assessed) to help you test your knowledge. Contrary to QS104, this module will have a 10% in-class test in week 7, and a 2h exam at the end (90%).