Survey data cleaning guidelines: (SPSS and Stata)

Survey data cleaning guidelines: (SPSS and Stata)

This paper focuses on the process of preparing data for analysis after data entry is completed, serving as a reference for those who are engaged in survey research.

The author clarifies that the computers to be used for data cleaning must have a current virus checker installed. On the other hand, the listing of the areas to be sampled must be available to verify that the coding for the identifying variables is correct.

The document figures that if data are entered into CSPro (Census and Survey Processing System), the corrections of any mismatches should be done in CSPro. Consequently, after importing all data from CSPro to the relevant software (SPSS or Stata), the syntax/do files should be used for checks. The checks should be run for each section, but not for the whole set. Nevertheless, cross file checks need to be done later to look at the variables that are related between different files.

The paper states that a track of the files used to clean data and files created from the cleaning should be kept. Still, the author emphasises that if there is no data entry error, and there are no notes to help determining where an error is, the data must be left as is.