Comparing datasets: checking double entry
In brief
If a questionnaire was entered twice, "Edit data" → "Compare datasets" shows every cell in which the two entries differ. File 1 is the open data, file 2 is chosen (SPSS, Statable project, Stata, CSV or Excel). The data do not change.
Rules
- Key column: a column in both files that names each case uniquely (for example a number). If a key occurs twice, Statable does not compare and names the keys.
- Without a key the files are compared by row number, only if the row counts are equal, with a note.
- Both cells empty = equal. Numbers are equal if the value read is equal (1 and 1.0). Text must match exactly; the option "Ignore spaces at the edges and upper/lower case" is off at first.
- Result: differences with key, column, value in file 1 and file 2; cases only in one file; columns only in one file (not compared); number of differences per column. The list can be saved as CSV.