regTermTest & confint, survey package

[jose Bartolomei <surfprjab@hotmail.com>]

Dear R/S users,

 

I am running logistic regression on a complex sampling survey using the function “svyglm” of the survey package.

 

I have two questions.

1) I will like to confirm that regTermTest function is the adequate function to test for interaction?

Any recommendations for this or alternate test function on a complex survey?

 

2. Which is the formula/method use to calculate confident limit in the “confint” function?

 

I do not get results,

confint(brGlm1, method="Link", ddf=NULL)

 

Waiting for profiling to be done...

                  2.5 % 97.5 %

(Intercept)                   NA     NA

income2$10k-15k      NA     NA

income2$15k-20k      NA     NA

income2$20k-25k      NA     NA

income2$25k-35k      NA     NA

income2$35k-50k      NA     NA

income2$50k-75k      NA     NA

income2$75k+         NA     NA

income2Don't Know    NA     NA

income2Refused       NA     NA

 

Any guess in what could be happening?

 

 

Thanks,

Jose

---

Windows Live™ Groups: Create an online spot for your favorite groups to meet. Check it out.