issue with reshaping wide to long

[Tina Samuel]

Hello Statlist community,
I have the following data that I would like to reshape to long:
Findid sex_98 age_98 qwealth_98 sex_06 age_06 qwealth_06
12345 Male 32 70 Male 40 50
I would like to reshape it so that it appears as follows
Findid year sex age qwealth
12345 1998 Male 32 70
12345 2006 Male 40 50

when I use the following code: reshape long sex age crempstp mnthwgAllJob yrschl crstablp crsectrp qwealth, i(Findid) j(year)
I get an error message saying that: "variable year contains all missing values"

I thought there was an issue with Stata trying to identify the year variable so created one, and when I tried reshaping the data from wide to long, I also got an error message saying that the data is already in long format even though it's not.
Do you have any idea what's wring and if so how I can fix it?

Thank you.

[Nick Cox]

reshape long sex age crempstp mnthwgAllJob yrschl crstablp crsectrp qwealth, i(Findid) j(work) string 
gen year = cond(work == "_98", 1998, 2006)

[Clyde Schechter]

When you write -reshape long sex age, i(findid) j(year)- [I shortened the variable list to simplify the presentation but the same problem arises with all of your variables.] Stata looks to find variables named sex# and age#, where # is any number. But you have no such variables. You have sex_98 and sex_06, but _98 and _06 are not numbers. So you have a few ways around this.

You can rename all your variables to eliminate the underscores ( _ ) that separate their "stem" from the year number and then use your -reshape- command the way you originally wrote it.

Or, you can do it as

reshape long sex age, i(findid) j(year) string
destring year, ignore("_"), replace

Added: Crossed with #2 which shows yet another way.

[Tina Samuel]

Thank you a lot, it worked perfectly.