Not everyone responds to a survey request, and those that do may not answer every question. To correct for this missing data, it may be necessary to use imputation and weighting strategies.
Unit non-response is best handled with weighting, to ensure underrepresented populations are reflected in the final data set. This course considers how these weights work along with other survey weights (e.g. design and post-stratification).
Item non-response is best handled with imputation, filling in missing responses with usable data. The course starts by looking at the pros and cons of older simplistic methods of imputation, then more sophisticated methods.
For both of these topics, the course presents the debates in the literature, and includes real life examples. There is no focus on specific software but rather on the decisions to be made in choosing and implementing and appropriate method. Practical workshop sessions will help to put learning into practice.
By the end of the course, participants will:
- Have knowledge of a variety of data imputation methods and their pros, cons and trade-offs,
- Have knowledge of simple, sophisticated and complex weighting schemes and their pros, cons and trade-offs.
- Be able to make informed choices about which particular imputation methods and which particular weighting methods are best for their data
- Missing data in surveys
- Are your data missing at random or not, and the implication of types of missing data
- Review of basic cell weighting
- Imputation strategies, from more simple to more complicated methods, and the pros and cons of each
- 3 stages of weighting – design weights; non-response weights; post-stratification to known population total weights
- Examples of complex weighting schemes
- Further details on weighting: stage 2 unknown eligibility and non-response (deterministic [cell weighting] versus stochastic [response propensity score weighting]; propensity score versus classification trees), stage 3 calibration methods (GREG estimator, ratio estimator (single auxiliary variable), post-stratified estimator, raking ratio estimator).
- Developments, issues, and debates in weighting
Who will benefit
Participants who want to use imputation and weighting strategies on their data, and also participants who want some knowledge of these topics for critiquing other’s work.
Participants need knowledge and experience of survey statistics and sampling. The SRA course on “Sampling and an Introduction to Weighting” would also be a good prerequisite.
- A better knowledge of post data collection adjustment strategies,
- Knowledge of the pitfalls of each of the methods discussed and
- An understanding of how this information can lead to better quality survey results
Dr. Tarek Al Baghal is a Research Fellow at the Institute for Social and Economic Research, University of Essex, and has a PhD in Survey Research and Methodology from the University of Nebraska. He is the lead questionnaire designer for the Understanding Society: the UK Household Longitudinal Study. He has taught a number of methods courses, including Survey Sampling and Questionnaire Design.
This course contributes 6 hours to the MRS CPD programme