Creating Tables with results on one row

[Leonard Scott]

Hello: I'm not sure this is going to be picked up on an old thread that had no replies, so I would like to re-address if able?

https://www.statalist.org/forums/for...oc#post1539751

I am hoping someone has this answer perhaps? I too am trying to figure out creating simple tables where the frequency and percentage appear in one row for each category. I cannot find how to do this as the table command seems to only allow the percentages and frequencies to be on different rows?

For example using the example table presented, I would like to create a table in similar formatting that says:

	single	married	total
white	487 (60.57%)	1150 (79.75%)	1637

Instead of the original:

	singe	married	total
white (frequency)	487	1150	1637
white (percent)	60.57	79.75

Please advise and thanks!

[Andrew Musau]

ssc install estout, replace

sysuse nlsw88, clear
estpost tab race married
esttab, aux(colpct) wide keep(white) unstack nonumb mlab(none) compress nonotes

Res.:

. esttab, aux(colpct) wide keep(white) unstack nonumb mlab(none) compress nonotes

-------------------------------------------------------------------------------
              single                married                  Total             
-------------------------------------------------------------------------------
white            487      (60.57)      1150      (79.75)      1637      (72.89)
-------------------------------------------------------------------------------
N               2246                                                           
-------------------------------------------------------------------------------

[Leonard Scott]

Thank you!!

Andrew Musau,
I do wonder--without needing to use this additional package, are the inherent table/tab features in Stata unable to perform this sort of customization of layout?

Thanks again for help.

[Rich Goldstein]

yes, internal commands can be used; you might want to start with Chuck Huber's 7-part blog intro; here is a link to the first of the seven: https://blog.stata.com/2021/06/07/cu...table-command/

[Leonard Scott]

Thank you both and I will review the materials.

I do have a follow up question regarding tables.
This may involve the -command- option embedded within the table feature, but I do not know how to use properly. Examples I've seen only describe adding regressions using this feature. Or perhaps there is a different way to approach:

How does one take data populated from a table and apply an operation to it?
Say I have a table that I generated using command

table (widget) (broken_status)

Cost by Widget	Broken (1)	Not Broken (0)	Total Annual Cost
W1	200.00	800.00	1000.00
W2	1000.00	1000.00	2000.00
W3	300.00	200.00	500.00
W4	50.00	50.00	100.0

How could I apply simple arithmetic to the OUTPUT of the table rather then have to go to the original data set to create an additional table column that shows "cost per month of broken widgets", which would == broken/12:

Cost by Widget	Broken (1)	Not Broken (0)	Total Annual Cost	Monthly Broken Cost
W1	200.00	800.00	1000.00	16.666
W2	1000.00	1000.00	2000.00	83.333
W3	300.00	200.00	500.00	25
W4	50.00	50.00	100.0	4.166

Please advise and thanks again!

[William Lisowski]

Post #5 was subsequently reposted as a new topic, with subsequent replies, at

https://www.statalist.org/forums/for...ns-or-commands