Wrong values when generating an equal variable

[HaiPhong Nguyen]

Hi,

This is a sample of my data:

id year code code2 test check
2 2015
2 2016 85182100 85182100 85182096 4
3 2012
3 2013
3 2014
3 2015
3 2016 85182100 85182100 85182096 4
4 2012
4 2013
4 2014
4 2015
4 2016 85182100 85182100 85182096 4
5 2012
5 2013
5 2014
5 2015
5 2016 85182100 85182100 85182096 4
.......

Variable "code" is in string form.

This is what I did:

gen code2 = code1
destring code2, replace
gen test = code2
gen check = code2-test

I'm pretty sure I literally just copied variable code2 into the column for variable test, however, the values for some of the observations seem to be different from what it should be. The incorrect values are always = code2 +/- {1,2,3,4}. There are others were correctly copied.

When I browse the dataset, instead of showing test = 85182096, it shows 8.52e+07. I reformatted it so that variable test shows all the digit as shown above. I don't think this would cause the above problem, however?

I don't really know why or how this would be the case? It seems really bizarre to me. Thank you for your help.

[Rich Goldstein]

please use -dataex- in the future when showing data examples - see the FAQ for help on this

when making exact copies of a variable, please use the -clonevar- command

in particular, your result from -destring- is a double, while you result from "gen test = code2" is a float, so you have a precision problem; see

help precision

[HaiPhong Nguyen]

please use -dataex- in the future when showing data examples - see the FAQ for help on this

when making exact copies of a variable, please use the -clonevar- command

in particular, your result from -destring- is a double, while you result from "gen test = code2" is a float, so you have a precision problem; see

help precision

Thank you! I just found out that it was the type mismatch issue and fixed it, so it's all good now. Thank you for the suggestion, however. I will keep this in mind when I post another question here. Thanks!