Multiple observations and individuals in database for merge

[Sam Fe]

Hey everyone !

I'm attempting to merge panel surveys database ( LSMS ). The problem I encounter is that in some database I have more than one observations per individual. exemple I'm studying the crop harvested per households and for each households I have more than one line ( see below ):

Id Crop
1 123
1 124
1 125
2 127
2 123
3 209

And in other base I have just one observation per household. What I would like to is redesign the database to have only one observation per household in attempt to merge all the database.
I thought about creating dummie variables refering to the type of crop harvested, and to the others variable who have multiple observations but I can't imagine the time needed to do something like that for each database.
I may be not really clear on my explanations but if you know a shorter way to do that let me know.

Best regards

[Carlo Lazzaro]

Sam:
welcome to the list.
I would start merging from the multiple lines database:

input Id Crop
1 123
 1 124
 1 125
 2 127
 2 123
 3 209
end
save "C:\Users\user\Desktop\Sam_1.dta"
drop _all
set obs 3
g Id=_n
g Nation=_n
label define Nation 1 "Argentina" 2 "Australia" 3 "Brazil"
label val Nation Nation
save "C:\Users\user\Desktop\Sam_2.dta", replace
use "C:\Users\user\Desktop\Sam_1.dta", clear
merge m:1 Id using "C:\Users\user\Desktop\Sam_2.dta", keepusing(Nation)
save "C:\Users\user\Desktop\Sam_1.dta", replace

Then you can decide which observations to -keep- (or -drop-) according to your research goals.

[Sam Fe]

Thanks for your answer your code was helpful.

Kind regards,
Sam