Discussion Session: Missing Values and Multiple Imputation
9th of February, 2017
11:00-12:30 at Badia, Emeroteca
In quantitative analysis, there are often missing values in the variables of interest. While missing data do not bias estimation as long as they are completely random, we often face situations where ‘missings’ appear nonrandom.
A certain type of respondents may not answer a certain type of questions in surveys. GDP data in a few countries may be missing because of political turmoil. What consequences might this missing data have on estimation? And how can we tackle this problem? In this session, we discuss these questions in the context of the following readings:
Lall, Ranjit. 2016. “How Multiple Imputation Makes a Difference.” Political Analysis 24 (4): 414-33.
Honaker, James, and Gary King. 2010. “What to Do about Missing Values in Time-Series Cross-Section Data.” American Journal of Political Science 54 (2): 561–81.
Cranmer, Skyler J., and Jeff Gill. 2012. “We Have to Be Discrete About This: A Non-Parametric Imputation Technique for Missing Categorical Data.” British Journal of Political Science 43 (2): 425–49.