Direct Ascription of Missing Categorical Values in Survey Research Data

Abstract

The complete datasets are a prerequisite for sustainable analyses, robust an- alytics and unbiased interpretation of results. Missing values in a survey occur when no data value is stored for the variable in an observation. Missing data can have a significant effect on the conclusions that can be drawn from the data. Direct ascription is the process of replacing missing data with predicted values. The aim of this work is to describe an approach to direct ascription of missing categorical values in survey research data based both on the assumption that values in a data set are missing at random and on the implementation of the correspondence analysis.