Creating a new variable that has a sum of the values of 4 other variables?

[Will Hall]

Hi everyone!

I have four variables in stata typenonap1_recode typenonap2_recode typenonap3_recode typenonap4_recode. For each observation, each variable has a value of 1-14. If I tab each variable separately I see the frequency for each value. I want to generate a variable that will show the combined frequency of all 14 values in all the 4 variables.

In other words, each variable says "What type of medication" and the answer is 1 antidepressant, 2 mood stabilizer... all the way up to 14 types. When I tab each variable separately I get the frequency of each value, so 1- antidepressant - freq 96, 2 mood stabilizer freq74. I can hand-sum the 4 variables, but can I just create a 5th variable that sums them?

thanks!
Will

[Hemanshu Kumar]

You don't actually want the new variable to sum up the values of these four variables -- if you did that, the new variable's values would run between (1 x 4 = ) 4 and (14 x 4 = ) 56, and would not make sense.

You can accomplish what you want by reshaping into a single variable:

Consider:

* made up data example
clear
input byte(id typenonap1_recode typenonap2_recode typenonap3_recode typenonap4_recode)
1   2   3   4   5
2   14  13  12  11
3   6   7   8   9
4   10  11  12  13
5   1   2   13  14
end

* something like this reshape command is all you need
reshape long typenonap@_recode, i(id) j(num)

so that now you can

. tab typenonap_recode

typenonap_r |
      ecode |      Freq.     Percent        Cum.
------------+-----------------------------------
          1 |          1        5.00        5.00
          2 |          2       10.00       15.00
          3 |          1        5.00       20.00
          4 |          1        5.00       25.00
          5 |          1        5.00       30.00
          6 |          1        5.00       35.00
          7 |          1        5.00       40.00
          8 |          1        5.00       45.00
          9 |          1        5.00       50.00
         10 |          1        5.00       55.00
         11 |          2       10.00       65.00
         12 |          2       10.00       75.00
         13 |          3       15.00       90.00
         14 |          2       10.00      100.00
------------+-----------------------------------
      Total |         20      100.00

[Nick Cox]

I imagine that you can do excellent tables with the new table framework, but tabm from the package tab_chi on SSC will do what I think you want.

This code is geared to your data example but note some key details.

1. A 4 rows x 14 columns display is likely to work less well than its transpose.

2. Concise but informative variable and value labels are crucial to a good table.

* made up data example
clear
input byte(id typenonap1_recode typenonap2_recode typenonap3_recode typenonap4_recode)
1   2   3   4   5
2   14  13  12  11
3   6   7   8   9
4   10  11  12  13
5   1   2   13  14
end

forval j = 1/4 { 
    label var typenonap`j'_recode "nap`j'"
}

tabm *_recode , transpose 

          |                  variable
    values |      nap1       nap2       nap3       nap4 |     Total
-----------+--------------------------------------------+----------
         1 |         1          0          0          0 |         1 
         2 |         1          1          0          0 |         2 
         3 |         0          1          0          0 |         1 
         4 |         0          0          1          0 |         1 
         5 |         0          0          0          1 |         1 
         6 |         1          0          0          0 |         1 
         7 |         0          1          0          0 |         1 
         8 |         0          0          1          0 |         1 
         9 |         0          0          0          1 |         1 
        10 |         1          0          0          0 |         1 
        11 |         0          1          0          1 |         2 
        12 |         0          0          2          0 |         2 
        13 |         0          1          1          1 |         3 
        14 |         1          0          0          1 |         2 
-----------+--------------------------------------------+----------
     Total |         5          5          5          5 |        20

.