Balancetest with logit....

[Kasper Barfoed]

Hi Statalist,

Writing a small assignment based on a survey i made.

It was set up as experimental treatment survey, where one respondents where sorted in to two groups based upon their birth date.
Then they received one simple treatment regarding salary- or career options, and answered some background questions. A very smal survey only intended for education purposes.

Balance test:
First i have made several descriptive statistics showing that the two groups are of equal variance and composition on the background variables.
- But how do i perform the balance test with the use of "logit"?

Hope it makes some sence, and someone can help a novice out:-)

Br, Kasper

[Phil Bromiley]

You'll increase your chances of a helpful answer if you follow the FAQ on asking questions -- provide Stata code in code delimiters, Stata output, and sample data using dataex.

I guess you'd have a dummy variable for experimental group and then run a logit with that dummy as the dv and your other variables as independent variables. This won't do variances as far as I know but would do background variables.