GESIS Training Courses
user_jsdisabled
Search

Scientific Coordination

Dr.
Sebastian E. Wenz
Tel: +49 221 47694-159

Administrative Coordination

Angelika Ruf
Tel: +49 221 47694-162

Course 5: Applied Multiple Imputation

Lecturer(s):
Dr. Ferdinand Geißler, Dr. Jan Paul Heisig

Date: 10.08 - 14.08.2020 ics-file

Location: Online via Zoom

About the lecturer - Dr. Ferdinand Geißler

About the lecturer - Dr. Jan Paul Heisig

Course description

[This is a 30 hour class.]
Missing data are a pervasive problem in the social sciences. Data for a given unit may be missing entirely, for example, because a sampled respondent refused to participate in a survey (survey nonresponse). Alternatively, information may be missing only for a subset of variables (item nonresponse), for example, because a respondent refused to answer some of the questions in a survey. The traditional way of dealing with item nonresponse, referred to as “complete case analysis” (CCA) or “listwise deletion”, excludes every observation with missing information from the analysis. While easy to implement, complete case analysis is wasteful and can lead to biased estimates. Multiple imputation (MI) seeks to address these issues and provides more efficient and unbiased estimates when certain conditions are met. Therefore, it is increasingly replacing CCA as the method of choice for dealing with item nonresponse in applied quantitative work in the social sciences.
 
The goals of the course are to introduce participants to the basic concepts and statistical foundations of missing data analysis and MI, and to enable them to use MI in their own work. The course puts heavy emphasis on the practical application of MI and on the complex decisions and challenges that researchers are facing in its course. The focus is on MI using iterated chained equations (aka “fully conditional specification”) and its implementation in the software package Stata. Participants should have a good working knowledge of Stata to follow the applied parts of the course and to successfully master the exercises. Participants who are not familiar with Stata may still benefit from the course, but will likely find the exercises quite challenging.
A detailed syllabus for this course is available for download here.


Keywords



Target group

Participants will find the course useful if:
  • use survey or other types of quantitative data and want to learn about MI as an alternative to CCA;
  • are already using MI but want to gain a better understanding of the underlying assumptions, of current best practice recommendations, and/or of how to solve specific problems that arise in its application (e.g., imputation diagnostics, convergence problems, imputation of transformed variables such as interactions, imputation of hierarchical data).


Learning objectives

By the end of the course participants will:
  • understand basic concepts of missing data analysis such as “missing at random”
  • be familiar with different approaches of how to handle item nonresponse and with their advantages and drawbacks;
  • have a solid understanding of the main assumptions and statistical theory underlying MI and of the main steps of an analysis involving MI (imputation, diagnostics, and analysis);
  • know how to implement MI using chained equations in Stata;
  • know how to deal with various (Stata-specific and general) practical complications that arise in the application of MI using chained equations.
 
Organizational Structure
This is a five-day course with a total amount of 30 hours of virtual class time.  Each day will begin with a three-hour lecture-like segment introducing the new material (9:30am-12:30pm). Exercises, most of them involving hands-on programming, will be distributed at the end of the lecture segment. Participants can start working on the exercises during the extended lunch break (12:30pm to 2:30pm). The first afternoon segment (2:30pm-4:30pm) will focus on the exercises. Participants will continue to work on the exercises, now with assistance from the lecturers, and eventually answers and solutions will be discussed with the full group. The final “flextime” segment of each day (4:30pm to 5:30pm) will serve to further discuss questions that have come up during the day and for lecturer-participant meetings that focus on individual questions and problems. Participants interested in individual consultations concerning their ongoing projects are encouraged to contact the lecturers before the course and provide a short description of the issues they would like to discuss. The individual-meetings can also be used for questions that arise during the course, however.


Prerequisites

  • Experience in the analysis of quantitative data
  • Good knowledge of regression analysis
  • Good working knowledge of Stata
  • Basic understanding of probability theory and sampling
  •  
    Software and Hardware Requirements  
    The practical examples and hands-on exercises will be done in Stata. Participants should have a recent version installed on their local computer. Version 15 or later would be ideal, although most examples should work in versions 12 and later. Participants who do not own a copy of Stata will be provided with access to a full Stata license by GESIS for the duration of the course. Stata will be installed and activated prior to the course by GESIS staff through remote access on the participants' machines.