Count number of households with multiple IDs

[Anne Todd]

Hi all, I have some data below where individuals are nested within households.

I know that, to isolate the number of unique households, I can simply do:

by household, sort: gen nvals = _n == 1
tab nvals

This tells me there are 31 unique households in my data. However, the question I really want to know the answer to is: "What percent of households have multiple individuals"? This would just be the inverse to what I did above if it were the case that all multi-individual households had only 2 individuals, but as you can see that is not the case with my data, some households have 3 or 4.

Does anyone have a concise way of figuring this out? Thanks.

* Example generated by -dataex-. For more info, type help dataex
clear
input int(household individual)
432 111
433 112
145 113
145 114
145 115
145 116
603 117
439 118
440 119
441 120
442 121
443 122
444 123
784 124
785 125
447 126
448 127
449 128
449 129
449 130
452 131
453 132
454 133
455 134
455 135
457 136
134 137
459 138
460 139
460 140
643 141
345 142
465 143
674 144
658 145
255 146
468 147
336 148
end

[Clyde Schechter]

by household (individual), sort: gen byte multiple_members = _N > 1 if _n == 1
tab multiple_members