When building a regression, I used SPSS Weighted Least Squares Regression to adjust for heteroscedaticity.
When building the WEIGHT variable, does SPSS allow me to include multiple layers and multiple independent variables in producing one WEIGHT?
See below.
*COMMENT To generate the unstandardized residuals.
REGRESSION
/MISSING LISTWISE
/STATISTICS COEFF OUTS R ANOVA
/CRITERIA=PIN(.05) POUT(.10)
/NOORIGIN
/DEPENDENT Holistic
/METHOD=ENTER Outside
/METHOD=ENTER HoursWeekReading2 HoursWeekListening2 HoursWeekConversing2
/SAVE RESID.
*COMMENT Set Absolute value for AbsRES.
COMPUTE AbsRES=abs(RES_1).
EXECUTE.
*Comment This generates the predicted residuals.
REGRESSION
/MISSING LISTWISE
/STATISTICS COEFF OUTS R ANOVA
/CRITERIA=PIN(.05) POUT(.10)
/NOORIGIN
/DEPENDENT AbsRES
/METHOD=ENTER Outside
/METHOD=ENTER HoursWeekReading2 HoursWeekListening2 HoursWeekConversing2
/SAVE PRED.
*COMMENT Create Weight variable.
COMPUTE WEIGHT=1/(PRED_1**2).
EXECUTE.
*COMMENT Run the Weighted Least Squares Model.
REGRESSION
/MISSING LISTWISE
/REGWGT=WEIGHT
/STATISTICS COEFF OUTS R ANOVA
/CRITERIA=PIN(.05) POUT(.10)
/NOORIGIN
/DEPENDENT Holistic
/METHOD=ENTER Outside
/METHOD=STEPWISE HoursWeekReading2 HoursWeekListening2 HoursWeekConversing2.
Thanks and stay safe,
Stafford
Stafford Cox
Academic Technology Services
Location: AS-120
Email: scox@csulb.edu
------------------------------
Stafford Cox
------------------------------
#SPSSStatistics