Using Multiple Imputation to Adjust for Survey Non-Response

Missing or incomplete responses are a common feature with many censuses and sample surveys. The problem created by survey non-response is that data values intended by the survey design to be observed are in fact missing. These missing values not only mean less efficient estimates because of the reduced size of the database, but also that standard complete-data methods cannot be used to analyse the data. Originally proposed by Prof. D. Rubin, multiple imputation is the technique that replaces each missing or deficient value with two or more acceptable values representing a distribution of the possibilities. SOLAS for Missing Data Analysis will be used to illustrate new and innovative methods of multiple imputation as well as some current methods. Practical approaches to displaying missing data and creative graphical representations will also be highlighted.