Combine Frequencies of Categorical Variables

[James Im]

Hi everyone! I have 3 variables denoted below. The end goal for me is to have one variable that combines the results as such:
Less than 10000 (numeric value 1) = 240 (25+114+101)
10000-19999 (numeric value 2) = 140 (6+82+52)
etc.

I tried using the stack function but that ended up removing all the other variables. Does anyone have any ideas on how to do this? Thank you in advance!

input double(A19_1 A19_2 A19_3)
6 . 8
2 1 1
1 1 2
4 4 4
1 10 5
. . .
1 3 1
1 1 1
10 9 .
. . .
. 7 .
. 3 9
. 2 10
1 1 1
1 . .
. 5 .
. . 7
. 2 2
1 1 6
1 1 4
. . .
. . 6
. 3 3
2 3 2
. . 4
. . .
. 1 .
. . .
. . .
. . .
8 7 .
. . 6
. . .
. . 7
2 . 1
. . 10
. . .
6 . 2
. 5 4
. 2 7
. 1 .
. . .
. 4 .
. 1 .
. . 6
. 4 3
. . .
. . .
. . 2
. . 3
. . .
. . 6
. . 2
. . .
1 2 3
. . 2
. . 3
. . .
. . .
. . .
. 2 .
. 4 3
. . 2
. . 4
. . .
. 1 2
. . 5
. 3 .
. . 3
. 3 2
. 2 .
. 2 .
. . .
. . 10
. 5 10
. 3 7
. . 3
. 1 7
. . 1
. 3 6
. 3 3
. . .
3 2 3
. 1 2
. 2 2
. 3 4
. 1 1
. . 3
. 2 7
2 3 9
. 3 4
. 1 2
. 4 10
. 3 3
. 2 2
. 10 10
. 3 .
. 7 4
. 2 .
. 2 4
end
label values A19_1 A19_1
label def A19_1 1 "less than 10,000", modify
label def A19_1 2 "10,000 - 19,999", modify
label def A19_1 3 "20,000 - 29,999", modify
label def A19_1 4 "30,000 - 39,999", modify
label def A19_1 6 "50,000 – 59,999", modify
label def A19_1 8 "70,000 – 79,999", modify
label def A19_1 10 "90,000 or more", modify
label values A19_2 A19_2
label def A19_2 1 "less than 10,000", modify
label def A19_2 2 "10,000 - 19,999", modify
label def A19_2 3 "20,000 - 29,999", modify
label def A19_2 4 "30,000 - 39,999", modify
label def A19_2 5 "40,000 – 49,999", modify
label def A19_2 7 "60,000 – 69,999", modify
label def A19_2 9 "80,000 – 89,999", modify
label def A19_2 10 "90,000 or more", modify
label values A19_3 A19_3
label def A19_3 1 "less than 10,000", modify
label def A19_3 2 "10,000 - 19,999", modify
label def A19_3 3 "20,000 - 29,999", modify
label def A19_3 4 "30,000 - 39,999", modify
label def A19_3 5 "40,000 – 49,999", modify
label def A19_3 6 "50,000 – 59,999", modify
label def A19_3 7 "60,000 – 69,999", modify
label def A19_3 8 "70,000 – 79,999", modify
label def A19_3 9 "80,000 – 89,999", modify
label def A19_3 10 "90,000 or more", modify

[Nick Cox]

You can apply the same value labels to one or more variables as is explained at

help label

Otherwise I don't confidently understand what the question is here. For example, even when all three variables are non-missing, the categories don't unequivocally imply a category for the sum. For example 1 1 1 could still mean that the sum was under 10,000 and yet again that it was almost 30,000. So the correct answer for combining 1 1 1 could be any of 1, 2 and 3.

stack is a command, not a function. You don't need doubles to hold integers up to 10 and missing.

That said, you may just want a combined table and tabm from tab_chi on SSC will do that.

. tabm A19_*, transpose

                  |             variable
           values |     A19_1      A19_2      A19_3 |     Total
------------------+---------------------------------+----------
 less than 10,000 |         9         14          7 |        30 
  10,000 - 19,999 |         4         13         14 |        31 
  20,000 - 29,999 |         1         14         12 |        27 
  30,000 - 39,999 |         1          5          9 |        15 
                5 |         0          3          2 |         5 
  50,000 – 59,999 |         2          0          6 |         8 
                7 |         0          3          6 |         9 
  70,000 – 79,999 |         1          0          1 |         2 
                9 |         0          1          2 |         3 
   90,000 or more |         1          2          6 |         9 
------------------+---------------------------------+----------
            Total |        19         55         65 |       139

Some of the value labels need to be fixed and I have no idea where your frequencies 240 140 ... come from unless it's a larger dataset you aren't showing us.

[James Im]

Hi Dr. Cox. Thank you for your reply! So each of the three variables represent income earned from a specific type of crop. I want to create one variable that represents income earned from any kind of crop (combines all observations from the three variables). When I use the "tab" function, I should get the cumulative frequency from combining all of the variables afterwards.

In regards to your last question, I only provided about 100 observations. They are part of a larger dataset.