Counts for multiple binary variables based on categorical grouping

[Lisa NiDhu]

Hi, I'm sure this is really basic but would appreciate any help!!

Using nutrition data, I've generated binary variables whereby participants are either classified as sufficient (0) or deficient (1) - I want to get counts (number of sufficient, number of deficient) for multiple nutrients (vit a, b6, b12, d etc etc) according to one categorical variable with three groups (ethnicity) but I can't figure out a way to do this without repeating tab!

Thank you,
Lisa

[Daniel Schaefer]

Hi Lisa,

You don't give a data example, but it sounds like your data look something like this:

clear
set obs 1000
gen ethnicity = runiformint(0, 2)
gen a = runiformint(0, 1)
gen b6 = runiformint(0, 1)
gen b12 = runiformint(0, 1)
gen d = runiformint(0, 1)

list if _n < 11, clean noobs

. list if _n < 11, clean noobs

    ethnic~y   a   b6   b12   d  
           0   1    1     1   1  
           0   1    0     1   0  
           0   1    1     1   1  
           2   1    1     0   1  
           0   1    1     0   1  
           2   0    1     1   1  
           0   1    1     1   1  
           2   1    1     1   1  
           2   1    1     0   0  
           0   1    0     1   1

Am I understanding that you want to find the number of 1s and 0s for each nutrient by ethnicity? I take it from your last sentence the problem is that you have to write the tab command over and over again (so your idea is to get the count using a cross tabulation) but you don't want to write the same command over and over again? If so, it sounds like you want a for loop.

foreach var in a b6 b12 d {
    tab ethnicity `var'
}

[Nick Cox]

Building on Daniel Schaefer 's very helpful reply:

I used his code to get a data example with also

set seed 314159

Check out official commands tab1 and tab2. For example

. tab2 eth a b6 b12 d, firstonly

-> tabulation of ethnicity by a  

           |           a
 ethnicity |         0          1 |     Total
-----------+----------------------+----------
         0 |       170        172 |       342 
         1 |       167        151 |       318 
         2 |       164        176 |       340 
-----------+----------------------+----------
     Total |       501        499 |     1,000 

-> tabulation of ethnicity by b6  

           |          b6
 ethnicity |         0          1 |     Total
-----------+----------------------+----------
         0 |       157        185 |       342 
         1 |       154        164 |       318 
         2 |       164        176 |       340 
-----------+----------------------+----------
     Total |       475        525 |     1,000 

-> tabulation of ethnicity by b12  

           |          b12
 ethnicity |         0          1 |     Total
-----------+----------------------+----------
         0 |       171        171 |       342 
         1 |       154        164 |       318 
         2 |       177        163 |       340 
-----------+----------------------+----------
     Total |       502        498 |     1,000 

-> tabulation of ethnicity by d  

           |           d
 ethnicity |         0          1 |     Total
-----------+----------------------+----------
         0 |       182        160 |       342 
         1 |       162        156 |       318 
         2 |       159        181 |       340 
-----------+----------------------+----------
     Total |       503        497 |     1,000

See also groups from the Stata Journal

. search st0496, entry

Search of official help files, FAQs, Examples, and Stata Journals

SJ-18-1 st0496_1  . . . . . . . . . . . . . . . . . Software update for groups
        (help groups if installed)  . . . . . . . . . . . . . . . .  N. J. Cox
        Q1/18   SJ 18(1):291
        groups exited with an error message if weights were specified;
        this has been corrected

SJ-17-3 st0496  . . . . .  Speaking Stata: Tables as lists: The groups command
        (help groups if installed)  . . . . . . . . . . . . . . . .  N. J. Cox
        Q3/17   SJ 17(3):760--773
        presents command for listing group frequencies and percents and
        cumulations thereof; for various subsetting and ordering by
        frequencies, percents, and so on; for reordering of columns;
        and for saving tabulated data to new datasets

 
. groups eth a b6 b12 d

  +-----------------------------------------------+
  | ethnic~y   a   b6   b12   d   Freq.   Percent |
  |-----------------------------------------------|
  |        0   0    0     0   0      17      1.70 |
  |        0   0    0     0   1      21      2.10 |
  |        0   0    0     1   0      24      2.40 |
  |        0   0    0     1   1      17      1.70 |
  |        0   0    1     0   0      33      3.30 |
  |-----------------------------------------------|
  |        0   0    1     0   1      19      1.90 |
  |        0   0    1     1   0      19      1.90 |
  |        0   0    1     1   1      20      2.00 |
  |        0   1    0     0   0      17      1.70 |
  |        0   1    0     0   1      18      1.80 |
  |-----------------------------------------------|
  |        0   1    0     1   0      26      2.60 |
  |        0   1    0     1   1      17      1.70 |
  |        0   1    1     0   0      21      2.10 |
  |        0   1    1     0   1      25      2.50 |
  |        0   1    1     1   0      25      2.50 |
  |-----------------------------------------------|
  |        0   1    1     1   1      23      2.30 |
  |        1   0    0     0   0      25      2.50 |
  |        1   0    0     0   1      17      1.70 |
  |        1   0    0     1   0      21      2.10 |
  |        1   0    0     1   1      16      1.60 |
  |-----------------------------------------------|
  |        1   0    1     0   0      23      2.30 |
  |        1   0    1     0   1      22      2.20 |
  |        1   0    1     1   0      22      2.20 |
  |        1   0    1     1   1      21      2.10 |
  |        1   1    0     0   0      15      1.50 |
  |-----------------------------------------------|
  |        1   1    0     0   1      20      2.00 |
  |        1   1    0     1   0      22      2.20 |
  |        1   1    0     1   1      18      1.80 |
  |        1   1    1     0   0      17      1.70 |
  |        1   1    1     0   1      15      1.50 |
  |-----------------------------------------------|
  |        1   1    1     1   0      17      1.70 |
  |        1   1    1     1   1      27      2.70 |
  |        2   0    0     0   0      17      1.70 |
  |        2   0    0     0   1      25      2.50 |
  |        2   0    0     1   0      18      1.80 |
  |-----------------------------------------------|
  |        2   0    0     1   1      24      2.40 |
  |        2   0    1     0   0      16      1.60 |
  |        2   0    1     0   1      25      2.50 |
  |        2   0    1     1   0      15      1.50 |
  |        2   0    1     1   1      24      2.40 |
  |-----------------------------------------------|
  |        2   1    0     0   0      21      2.10 |
  |        2   1    0     0   1      23      2.30 |
  |        2   1    0     1   0      23      2.30 |
  |        2   1    0     1   1      13      1.30 |
  |        2   1    1     0   0      30      3.00 |
  |-----------------------------------------------|
  |        2   1    1     0   1      20      2.00 |
  |        2   1    1     1   0      19      1.90 |
  |        2   1    1     1   1      27      2.70 |
  +-----------------------------------------------+

[Daniel Schaefer]

Thanks Nick Cox for the helpful addition! I was aware of tab1, but I didn't put 2 and 2 together (so to speak) and was unaware of tab2. Please note OP that the foreach loop in #2 is a general way to repeat the same line iteratively, but Stata often has useful built in features that let you avoid loops entirely, tab2 among them.