unsuccessful merge

[Barbora Pisecka]

Hello everyone!

I am trying to merge two data files from different sources using the command merge 1:1, based on a string variable Country. However, the merge is never successful, even though the names of countries in both data sets are identical.
I suspect this is because each of the data files has Country saved as a different type of variable (str42 and str30). When I try to merge, Stata tries to adjust the data (note: variable Country was str30, now str42 to accommodate using data's values) which, I believe, causes the merge to fail.

Is there any way to deal with this issue?

Any help is very appreciated.

Thanks in advance,
Barbora

[Rich Goldstein]

at a guess, you have spaces in one of the files - they may be between words (if a country name has more than one word) or at the beginning or at the end; see

help string functions

and search on "trim"

[Nick Cox]

See also

help compress

after applying trim().

[David Benson]

As Rich and Nick mentioned, it's probably extra spaces (probably at the end of country names) that are causing your problem.

To give you an example, imagine I have a list of countries, and I am trying to add in data on the land area to my master list:
(I just grabbed this data from here)

dataex id country pop  //  Data shared using Stata command dataex. To install: ssc install dataex
clear
input byte id str13 country double pop
 1 "China"         1420.062
 2 "India"         1368.738
 3 "United States"  329.093
 4 "Indonesia"      269.536
 5 "Brazil"         212.393
 6 "Pakistan"       204.596
 7 "Nigeria"        200.962
 8 "Bangladesh"     168.066
 9 "Russia"         143.896
10 "Mexico"         132.328
end

format pop %10.2fc
. list, noobs

  +-------------------------------+
  | id         country        pop |
  |-------------------------------|
  |  1           China   1,420.06 |
  |  2           India   1,368.74 |
  |  3   United States     329.09 |
  |  4       Indonesia     269.54 |
  |  5          Brazil     212.39 |
  |-------------------------------|
  |  6        Pakistan     204.60 |
  |  7         Nigeria     200.96 |
  |  8      Bangladesh     168.07 |
  |  9          Russia     143.90 |
  | 10          Mexico     132.33 |
  +-------------------------------+

*** This file will be the master file, I've called it "merge1 - country & pop.dta"

I've taken those same 10 countries and added some trailing spaces to the end of obs 1-5.

dataex id country land_area  //  Data shared using Stata command dataex. To install: ssc install dataex
clear
input byte id str15 country long land_area
 1 "China     "       9388211
 2 "India     "       2973190
 3 "United States  "  9147420
 4 "Indonesia  "      1811570
 5 "Brazil  "         8358140
 6 "Pakistan"          770880
 7 "Nigeria"           910770
 8 "Bangladesh"        130170
 9 "Russia"          16376870
10 "Mexico"           1943950
end

format land_area %12.0gc
label var land_area "Land area in square kilometers"

. list, noobs

  +-----------------------------------+
  | id           country    land_area |
  |-----------------------------------|
  |  1        China         9,388,211 |
  |  2        India         2,973,190 |
  |  3   United States      9,147,420 |
  |  4       Indonesia      1,811,570 |
  |  5          Brazil      8,358,140 |
  |-----------------------------------|
  |  6          Pakistan      770,880 |
  |  7           Nigeria      910,770 |
  |  8        Bangladesh      130,170 |
  |  9            Russia   16,376,870 |
  | 10            Mexico    1,943,950 |
  +-----------------------------------+

*** This will be the using file, I've called it "merge2 - country & land area.dta"

* When I try to merge the two files:
use "merge1 - country & pop.dta", clear

* Merging using the old syntax (someone can update)
merge country using "merge2 - country & land area.dta", unique sort nonotes nokeep  _merge(merge_land_area)
sort id

list, noobs

  +-------------------------------------------------------+
  | id         country        pop    land_area   merge_~a |
  |-------------------------------------------------------|
  |  1           China   1,420.06            .          1 |
  |  2           India   1,368.74            .          1 |
  |  3   United States     329.09            .          1 |
  |  4       Indonesia     269.54            .          1 |
  |  5          Brazil     212.39            .          1 |
  |-------------------------------------------------------|
  |  6        Pakistan     204.60      770,880          3 |
  |  7         Nigeria     200.96      910,770          3 |
  |  8      Bangladesh     168.07      130,170          3 |
  |  9          Russia     143.90   16,376,870          3 |
  | 10          Mexico     132.33    1,943,950          3 |
  +-------------------------------------------------------+

[Barbora Pisecka]

Thank you for the help, it worked!