The purpose of this article is to discuss issues related to the analysis of missing data. Why do missing data occur in a data set? What percentage of the collected data constitutes missing data? What is the nature of missing data that emerges during data collection? The above questions are extremely important in assessing conducted surveys or in evaluating the quality of the collected data. A lack of reflection on the aspects mentioned above may lead to false conclusions and recommendations. This article presents not only an overview of the literature regarding missing data, but also shows how in a practical way an analysis of the randomness of missing data can be performed. The analysis presented in the article is based on data collected as part of the Polish General Social Survey carried out in 2008. The main recommendation of the author is to conduct an analysis of the randomness of missing data before analyzing the collected data.
Read full abstract