# Interpreting Marginal Effects after the Probit Model:

Interpreting Marginal Effects after the Probit Model:

I have ran a Probit regression, then computed the Marginal Effects after Probit. I want to interpret the effects of men, education, and experience on the model. How does one make quantitative interpretations for this scenario? The output is below:

Marginal effects after probit
y = Pr(empst31) (predict)
= .78966013
------------------------------------------------------------------------------
variable | dy/dx Std. Err. z P>|z| [ 95% C.I. ] X
---------+--------------------------------------------------------------------
men*| .1729278 .00698 24.78 0.000 .159252 .186603 .467327
racebx*| -.0572455 .01134 -5.05 0.000 -.079465 -.035025 .151777
racethnx*| -.0409015 .0092 -4.45 0.000 -.058934 -.022869 .308472
education | .0510486 .00256 19.93 0.000 .046029 .056068 3.14991
experi~e | -.0016308 .00041 -3.93 0.000 -.002444 -.000818 30.5769
marry31x*| .0429733 .00777 5.53 0.000 .027744 .058202 .625885
region31 | -.0040225 .00373 -1.08 0.281 -.011339 .003294 2.76932
msa31*| .0291475 .01016 2.87 0.004 .009241 .049054 .830042
------------------------------------------------------------------------------
(*) dy/dx is for discrete change of dummy variable from 0 to 1
------------------------
Here is the way that I was interpreting the variables - I want to make sure it is correct I was not sure if you interpret as a percentage or not. The dependent variable in the model is employment

There were my answers I would like examined.

The probability of a man being employed is 17.29% higher then females, all else held constant. Education increases the probability of employment by 5.10% all else held constant. Experience decreases the probability of employment by -.163% all else held constant.
______________

Update
1=employed
0 = unemployed