CODEWITHSUNDEEP
statacategorization
aeiz
aeiz

Reputation: 31

Stata - Link observations if at least one of several variables have same value

I have the following data structure (Picture of structure of my data), where each row represents a household and the variable "group1" identifies the classroom of child 1 in the household, "group2" the classroom of child 2, and so on. It is worth noting that there are around 3000 groups in total, as these are households all over the country. I need to categorize households as belonging to the same group if at least one of the "group" variables have the same value (i.e., if at least one of their children go to the same class). This can happen if, for two households, "group1" = "group1", but also if "group1" = "group2", or "group3", etc.

I have experimented using inlistand looping through all "group values", but haven't arrived anywhere.

I will be extremely grateful for any help you can offer.

Upvotes: 0

Views: 442

Answers (1)

Robert Picard
Robert Picard

Reputation: 1051

This is easier to do with data in a long layout, with one observation per child. You can then group households with children in the same class (as identified in the group variable) using group_id (from SSC):

* Example generated by -dataex-. To install: ssc install dataex
clear
input long household float(group1 group2 group3 group4)
101 15 16  .  .
102 13 14 15 17
103 11 17  .  .
104 33 34 35  .
105 34 37  .  .
end

reshape long group, i(household) j(child)
drop if mi(group)

clonevar hhgroup = household
group_id hhgroup, matchby(group)

reshape wide group, i(household) j(child)

list

and the results

. list

     +--------------------------------------------------------+
     | househ~d   group1   group2   group3   group4   hhgroup |
     |--------------------------------------------------------|
  1. |      101       15       16        .        .       101 |
  2. |      102       13       14       15       17       101 |
  3. |      103       11       17        .        .       101 |
  4. |      104       33       34       35        .       104 |
  5. |      105       34       37        .        .       104 |
     +--------------------------------------------------------+

.

Upvotes: 1

Related Questions