Jerneja Pikelj (jerneja.pikelj@gov.si)
Statistical Office of the Republic of Slovenia
Abstract
The paper has two practical purposes. The first one is to analyze how successfully R can be used for data analysis on surveys carried out by the Statistical Office of the Republic of Slovenia. In order to achieve this goal, we analyzed the data of the Monthly Statistical Survey on Earnings Paid by Legal Persons. The second purpose is to analyze how the assumption on the nonresponse mechanism, which occurs in the sample, impacts the estimated values of the unknown statistics in the survey. Depending on these assumptions, different approaches to adjust the problem caused by unit nonresponse are presented. We conclude the paper with the results of the analysis of the data and the main issues connected with the usage of R in official statistics.
