0
$\begingroup$

Assume in a study the dependent variable is quantitative, while most independent variables are categorical, with some of them being quantitative. We aim to evaluate the relationship between the dependent variable and the main independent variable after eliminating the confounding effects of other variables. Which tests are recommended for this?

$\endgroup$
4
  • $\begingroup$ Are the categories ordinal? $\endgroup$ Mar 22 '19 at 17:13
  • $\begingroup$ @Acccumulation Dichotomous $\endgroup$
    – user241033
    Mar 22 '19 at 17:25
  • $\begingroup$ what is a "quantitative" variable? Be specific. $\endgroup$
    – AdamO
    Mar 22 '19 at 18:18
  • $\begingroup$ @AdamO continuous $\endgroup$
    – user241033
    Mar 22 '19 at 18:21
1
$\begingroup$

If I'm reading this correctly it sounds like your DV is continuous, your IV is dichotomous and categorical and you want to control for both categorical and continuous confounding variables.

You could either run a stepwise multiple regression with the first step including the variables you are hoping to control for and your second step adding in the IV. Any categorical variables would have to be dummy coded before adding that into the model.

Conversely, you could also run an ANCOVA. The categorical control variables make that a little trickier but still doable.

$\endgroup$
0
$\begingroup$

If you are working with SPSS (as it was mentioned in the original post), Univariate GLM procedure probably will do the trick: - set your dependent (and continuous) variable as dependent variable - set categorical independent variables as Fixed Factors and quantitative independent variables as Covariates.

$\endgroup$
1
  • $\begingroup$ thank you so much; i think i can use both of univariate GLM and regression--->linear procedures $\endgroup$
    – user241033
    Mar 22 '19 at 20:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy