Multicollinearity, VIF and interaction

[Wah Myint]

Hello
I have seen many posts (questions and responses) related to multicollinearity, VIF and interaction. After reading all these threads, I still do not get my answer.
Background: I'm currently using a large survey data set in which I tried to test multicollinearity between two independent variables (both are dummies). My outcome/dependent variable is a dummy variable.

My questions are
(Q1) Can I use Vif for this situation? I saw many posts here and in publications that Vif is used, but most of them are used for the continuous outcome/dependent variables.
I am also aware that the interaction term can also be used to check for collinearity. But, mostly I've seen using "regress" but again, their outcome variables are continuous variables.
(Q2) Currently, I used interaction terms to see if there is any interaction between them.
So, here is my code for interaction. Is the code below correct for checking interaction (outcome var= dummy variable; two independent var checking for collinearity are also dummy variables?

. svy, subpop (if COUTYP4==3): logistic illyr i.new_amdeyr##i. anysuideation2, or

Below is the output.
-----------------------------------------------------------------------------------------

illyr | Odds ratio std. err. t P>|t| [95% conf. interval]
--------------------------+----------------------------------------------------------------
new_amdeyr |
Yes | 2.368648 .1665209 12.27 0.000 2.056722 2.727881
1.anysuideation2 | 4.202448 .3737046 16.14 0.000 3.515055 5.024265

new_amdeyr#anysuideation2 |
Yes#1 | .5716726 .0724573 -4.41 0.000 .4431859 .7374096
|
_cons | .155493 .0033261 -87.01 0.000 .1489538 .1623192
-------------------------------------------------------------------------------------------



(Q2.1) If it is not correct, can anyone suggest the correct code?

(Q3) if it is correct, can the code be used for or checking interaction (outcome var= categorical variable; independent variables checking for collinearity are also dummy variables, categorical and independent var?


Thank you so much.

[Carlo Lazzaro]

Wah:
foreword: I never ever challenged myself with survey. Therefore, please consider what follows as a tentative reply.
That said, I would rely on -estat vce, corr- to grasp the correlation patterns:

. use https://www.stata-press.com/data/r17/nhanes2b.dta

. svy: logistic highbp height weight age female
(running logistic on estimation sample)

Survey: Logistic regression

Number of strata = 31                            Number of obs   =      10,351
Number of PSUs   = 62                            Population size = 117,157,513
                                                 Design df       =          31
                                                 F(4, 28)        =      368.33
                                                 Prob > F        =      0.0000

------------------------------------------------------------------------------
             |             Linearized
      highbp | Odds ratio   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
      height |   .9657022   .0051511    -6.54   0.000     .9552534    .9762654
      weight |   1.053023   .0026902    20.22   0.000     1.047551    1.058524
         age |   1.050059   .0019761    25.96   0.000     1.046037    1.054097
      female |   .6272129   .0368195    -7.95   0.000     .5564402     .706987
       _cons |    .716868   .6106878    -0.39   0.699     .1261491    4.073749
------------------------------------------------------------------------------
Note: _cons estimates baseline odds.

. estat vce, corr

Correlation matrix of coefficients of logistic model

             | highbp                                           
        e(V) |   height    weight       age    female     _cons 
-------------+--------------------------------------------------
highbp       |                                                  
      height |   1.0000                                         
      weight |  -0.6406    1.0000                               
         age |   0.4553   -0.0988    1.0000                     
      female |   0.5294   -0.1320    0.4688    1.0000           
       _cons |  -0.9741    0.4880   -0.5754   -0.6242    1.0000 

.

[Wah Myint]

Carlo: Thank you so much. I appreciate it. I will try this way.