Development of a Method for Imputation of Missing Data Using ACS Exams as a Prototype

Abstract

Missing data is a regular issue that researchers and practitioners must consider for treatment. Commonly, cases for which data is missing are excluded from inclusion in larger data sets. However, this is not the only option and could artificially alter the sample. Other options are available for imputing missing data. Expanding on work previously reported, a method is presented here that not only preserves all observed data but also is shown to function for smaller data sets. As an example of the process, four ACS Exams are used as prototypes with a discussion on an expected noise level of any imputed sample.