Breaking down variables by multiple other variables

[John Wilkinson]

I have already broken down two variables (sat and colgpa) by whether or not the student is an athlete or not. However, I am trying to further break down the analysis by gender. That is, I would like to report the same statistics by athlete status separately for both genders. I have successfully done this using two separate tables; however, am wondering if it is possible to accomplish this in one table.


Here is my first table, breaking down the two variables by athlete status.

Here are my two tables (one for male, one for female) breaking down the two variables by athlete status.

What I cannot figure out is whether there is a way to condense these two tables down into one table, as the two tables are not as visually descriptive.

[Clyde Schechter]

The layout will not be quite the same, but you can do this with the -table- command:

table athlete female, c(mean sat sd sat mean colgpa sd colgpa)

[Nick Cox]

A simple strategy is just to create your summary variables first and then reach for a basic "show me" command. Here I have echoed your example of mean and SD for two numeric variables for a cross-combination of binary variables, but the strategy is much more general.

I use a dataset all can use; conversely using your own data and showing results as images isn't the best way to give examples for us to answer. Please do read and act on FAQ Advice #12.

. sysuse auto, clear
(1978 Automobile Data)

.
. gen himpg = mpg > 30

.
. foreach s in mean sd {
  2.    foreach v in weight price {
  3.        egen `v'_`s' = `s'(`v'), by(foreign himpg)
  4.            label var `v'_`s' "`s' `v'"
  5.            format `v'_`s' %2.1f
  6.     }
  7. }
(1 missing value generated)
(1 missing value generated)

.
. egen group = group(foreign himpg), label

. tab group

group(forei |
  gn himpg) |      Freq.     Percent        Cum.
------------+-----------------------------------
 Domestic 0 |         51       68.92       68.92
 Domestic 1 |          1        1.35       70.27
  Foreign 0 |         18       24.32       94.59
  Foreign 1 |          4        5.41      100.00
------------+-----------------------------------
      Total |         74      100.00

. label def group 1 "Domestic Low" 2 "Domestic High" 3 "Foreign Low" 4 "Foreign High",
>  replace

.
. tabdisp group, c(weight_* price_*)

------------------------------------------------------------------
group(foreign |
himpg)        | mean weight    sd weight   mean price     sd price
--------------+---------------------------------------------------
 Domestic Low |      3346.9        668.0       6104.7       3119.1
Domestic High |      1800.0                    4425.0            
  Foreign Low |      2368.9        462.7       6829.5       2689.4
 Foreign High |      2077.5         82.6       4383.0        778.0
------------------------------------------------------------------

.
. * groups should be installed from Stata Journal website
. groups group weight_* price_*, show(f) missing

  +-------------------------------------------------------------------+
  |         group   weight~n   weight~d   price_~n   price_sd   Freq. |
  |-------------------------------------------------------------------|
  |  Domestic Low     3346.9      668.0     6104.7     3119.1      51 |
  | Domestic High     1800.0          .     4425.0          .       1 |
  |   Foreign Low     2368.9      462.7     6829.5     2689.4      18 |
  |  Foreign High     2077.5       82.6     4383.0      778.0       4 |
  +-------------------------------------------------------------------+

.  
. groups group weight_* price_*, show(f) missing colorder(1 6)

  +-------------------------------------------------------------------+
  |         group   Freq.   weight~n   weight~d   price_~n   price_sd |
  |-------------------------------------------------------------------|
  |  Domestic Low      51     3346.9      668.0     6104.7     3119.1 |
  | Domestic High       1     1800.0          .     4425.0          . |
  |   Foreign Low      18     2368.9      462.7     6829.5     2689.4 |
  |  Foreign High       4     2077.5       82.6     4383.0      778.0 |
  +-------------------------------------------------------------------+

PS Graphs are good too.

On groups see

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

or https://www.statalist.org/forums/for...updated-on-ssc