Merging multiple cross tabulations that use a common grouped variable

[Steve Patrick]

Greetings. I'm trying to merge 4-5 cross tabulations, that use a common group variable as one of the 2 variables in the cross-tabs, into one table. Additional context -assume the framework is variable "A" with 3 categories forming the columns, while 4-5 other variables, also with 3-4 subcategories, form rows. Can this be created by a commands similar to "tab2," or is this better addressed via command series associated with program such as "tabout"?

[Nick Cox]

You've yet to get an answer, and my guess is that while you know you want this is too abstract for many to understand. A concrete data example always beats a lengthy verbal explanation in which we're asked to imagine data. Please see the FAQ Advice, especially #12.

Why not translate into an example using nlswork, which has several categorical variables? Here distinct is from the Stata Journal.

. webuse nlswork
(National Longitudinal Survey.  Young Women 14-26 years of age in 1968)

. distinct

---------------------------------
          |     total   distinct
----------+----------------------
   idcode |     28534       4711
     year |     28534         15
 birth_yr |     28534         14
      age |     28510         33
     race |     28534          3
      msp |     28518          2
  nev_mar |     28518          2
    grade |     28532         19
 collgrad |     28534          2
 not_smsa |     28526          2
   c_city |     28526          2
    south |     28526          2
 ind_code |     28193         12
 occ_code |     28413         13
    union |     19238          2
   wks_ue |     22830         61
  ttl_exp |     28534       4744
   tenure |     28101        270
    hours |     28467         85
 wks_work |     27831        105
  ln_wage |     28534       8173
---------------------------------

groups (also Stata Journal) is a framework for cross-combinations of an arbitrary number of variables. https://www.statalist.org/forums/for...updated-on-ssc gives the flavour.

I am very likely misreading what you want, which brings us back to the request for a concrete example.

[Steve Patrick]

Thanks Professor! Upon second reading, I can see how the post is abstract to others. My apologies. However, I don't think you misread. The groups framework may do the trick. I will work with it and let you know how it goes.