W.P. Zijlstra Outlier detection in test and questionnaire data for attribute measurement
Gedrukt boek
Outliers, often defined as observations, which are inconsistent with the remaining data, are influential to the degree in which they affect statistical results. Outlier research is often done with continuous variables, such as income and age. This project defines outliers for highly discrete data in test, questionnaire and survey data, and proposes several methods for detecting outliers. It is also investigated when an outlier is influential, and how to deal with influential cases. Simulated and numerous real data sets are used. The project will result in user - friendly software for handling data quality problems, including outlier analysis