Logistic Regression Issues

[Kyle Brees]

I am trying to perform a logistic regression but I am running into issues with the var. I keep getting the error "outcome does not vary". I tabbed my question

tab Q1_2

I was trying to use as the outcome in the variable I got

Q1_2	Freq
.v	194
0	117
1	29

I thought Stata would ignore the .v observations so I decided to

tabstat Q1_2

and got a mean of 2.5! I have no idea how the mean is so high unless the .v are holding values? Has anyone ran into this issue or how to remedy it so I can run the logistical regression?

[Kyle Brees]

I looked into the values and it seems that my 0 & 1 values are being stored with a value of 2 & 4.

[Marcos Almeida]

Welcome to the Stata Forum/ Statalist.

Please read the FAQ. There you’ll find out how to share command/output/data.

That said, the variable was probably ‘encoded’. Typing- codebook varname - will give the real value.

To exclude the (misleading) label, just use - label drop - command.

Hopefully that helps.