FAQ/selectif - CBU statistics Wiki
location: FAQ / selectif

Filtering outlying observations from analyses in SPSS

To select out outliers (either low or high) of a variable such as age e.g. highlighted by a stem and lreaf display (see the EDA Graduate Statistics talk) you can go to DATA>SELECT CASES clicking the IF BUTTON and putting AGE >= 3 AND AGE <= 10 in the top box in the resultant SELECT CASES: IF window and clicking CONTINUE and OK. Both ends should then be filtered out from then on until you cancel the filter. The syntax equivalent is below which you can get by clicking on the Paste button. In this example ages under 3 or greater than 10 are removed from all subsequent calculations.

FILTER OFF.
DATASET ACTIVATE DataSet0.
USE ALL.
COMPUTE filter_$=(AGE >= 3 AND AGE <= 10).
VARIABLE LABELS filter_$ 'AGE >= 0 AND AGE <= 10 (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMATS filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

None: FAQ/selectif (last edited 2015-01-23 10:13:46 by PeterWatson)