"Outcome does not vary" error

[Kiana Delavarkasmaei]

Hi everyone

I am trying to run a logit regression: logit employed firmsize

and I receive this error:
outcome does not vary; remember:
0 = negative outcome,
all other nonmissing values = positive outcome

I dropped the missing values in my independent variable (firmsize) using the code: drop if firmsize== .

but I am still receiving this error. Does anyone know how I can fix this?

Thank you in advance

* Example generated by -dataex-. For more info, type help dataex
clear
input float employed int firmsize
1 4
1 2
1 4
1 2
1 4
1 4
1 3
1 2
1 1
1 2
1 3
1 4
1 4
1 4
1 1
1 4
1 2
1 4
1 1
1 1
1 4
1 4
1 2
1 3
1 4
1 4
1 4
1 4
1 3
1 2
1 4
1 3
1 2
1 4
1 1
1 4
1 4
1 4
1 4
1 4
1 4
1 4
1 2
1 1
1 4
1 3
1 3
1 2
1 1
1 4
1 2
1 4
1 4
1 1
1 4
1 4
1 1
1 1
1 1
1 4
1 4
1 4
1 4
1 4
1 1
1 2
1 3
1 1
1 4
1 4
1 4
1 4
1 3
1 4
1 3
1 4
1 4
1 4
1 1
1 4
1 2
1 3
1 4
1 4
1 1
1 4
1 4
1 4
1 4
1 4
1 4
1 2
1 4
1 3
1 4
1 4
1 4
1 1
1 4
1 4
end
label values firmsize firmsize
label def firmsize 1 "Less than 20 employees", modify
label def firmsize 2 "20 to 99 employees", modify
label def firmsize 3 "100 to 500 employees", modify
label def firmsize 4 "More than 500 employees", modify

[Nick Cox]

According to your data example everyone is 1 for employed. You need some values of 0 for the logit regression to make sense.

[Kiana Delavarkasmaei]

According to your data example everyone is 1 for employed. You need some values of 0 for the logit regression to make sense.

Hi Nick. Thank you for your response. I realized that when I drop the missing values for firmsize (indep variable), the 1 values for employed (dependent variable) will be dropped automatically! I don't understand why this happens?

[Nick Cox]

Not so. In your data example, and outside it too, observations with 1 for employed and non-missing firmsize will not be dropped. But this is nothing to do with why your logit regression didn't work.

Observations with missing firmsize will just be ignored by such a regression. The problem lies in the outcome variable, which is employed.

[Kiana Delavarkasmaei]

Not so. In your data example, and outside it too, observations with 1 for employed and non-missing firmsize will not be dropped. But this is nothing to do with why your logit regression didn't work.

Observations with missing firmsize will just be ignored by such a regression. The problem lies in the outcome variable, which is employed.

I am very sorry for my typo. What I meant was, when I drop missing values for firmsize (indep), the 0 values for employed (dep) are dropped automatically so that all it shows are 1 values. Also, the employed variable is a dummy variable which I created using the following codes:


generate employed=1 if inlist(lfsstat,1,2)
replace employed=0 if inlist(lfsstat,3,4)

I realized that when I drop missing values for the independent variable (firmsize), the values for lfsstat will change, accordingly the values for employed variable change as well!

I am not sure which part of the codes I should fix?

[Nick Cox]

OK; but if everyone is employed if the firm size is known, there is no variation to explain.