add one variable from one dataset to another one: error variable V2 does not uniquely identify observations in the master data

[Anna Petrova]

Hell everyone,

I have a dataset with V2 - country code

(in the codebook each country has a code so I believe the initial value of the V2 are codes):

tab V2 (master dataset)

I need to add to the dataset the GDP per capita variable. For this I have created the excel file with 2 columns: V2 and GDP per capita and saved this file as .dta.

As recommended in previous posts I have used

merge 1:1 V2 using "/.../GDP_per_capita.Wave_6.dta"

Problem: I am receiving the error variable V2 does not uniquely identify observations in the master data no matter if I am having in the GDP dataset country names (to match the V2 from master dataset) or codes correspond to each of the country.

With names:

With codes for each country:

Could you advise where can be a problem?

Thanks!

[Mike Lacy]

Look at -help merge- and particularly at the difference between 1:1, m:1, and 1:m. I think you will find that you want an m:1 merge.

[Anna Petrova]

Dear Mike,

Thank your your answer! It seems that it works now. But I am worried if I created the correct dataset (with GDP) which had to be merged. As now when I am tabulating the average GDP variable in the master dataset (after adding it) i receive the results:

I expected to have in the left part of the column: list with countries and where I have freq. now to have average GDP per capita. Or is it fine?

Can you confirm if I added it correctly to use for my future model ( GDP per capita as explanatory variable for my logic model:

logit charity_participation sex income_class education happiness averGDPpercapita

Thanks!

[Mike Lacy]

I'm sorry, I don't understand what you want. Perhaps someone else will.

[William Lisowski]

Perhaps the following will allow you to confirm that the merge worked as you expected it.

tabstat AverageGDPpercapita, by(country) statistics(n min mean max)

[Anna Petrova]

Thanks!