labmask error

[Stephen Okiya]

Hi,
I am using Stata 14 on Windows. I have tried implementing the labmask command and encountered the following error "x not constant within groups of y". Any work around on this?

Thanks in advance.

[Nick Cox]

Not in my book, because you would be trying to undermine the logic of labmask.

labmask (Stata Journal, as you are asked to explain; see FAQ Advice #12.1) asks that users sees the values (or optionally the value labels) of one variable rather than another in output: what you see is thus a "mask" on the face of a variable.

Consider this:

clear
input y str4 mask
1  "frog"
1  "frog"
1  "frog"
1  "toad"
2  "toad"
2  "toad"
end
labmask y, values(mask)

labmask fails because there isn't a one-to-one correspondence:

 
1 <-> "frog"
2 <-> "toad"

is violated in one observation. You can't program around this because labmask is just a wrapper that defines value labels for you and you can't have different value labels for the same value.

I see two possibilities:

1. What you want to do makes sense, but you can't do it this way.

2. labmask has found an inconsistency in your data.

http://www.stata-journal.com/sjpdf.h...iclenum=gr0034 may be of interest.

[Stephen Okiya]

Thanks so much Nick Cox I completely agree with you. I have been able to figure out the inconsistency in my data set. Much appreciated.

[Joe Mo]

Hi, I am experiencing an error with labmask in Stata 15. I have a banded income variable y01, which I want to generate and mask into income . I am using:

gen income = y01
labmask income, val(y01) decode

However, Stata produces the error message r;(110) - "label income already defined". It has worked fine when tried with other variables. Thanks.

[Stephen Okiya]

Hi Joe,

You are tyring to generate a label that already exists. Try using a different name like income2.

Thank you.