Unadjusted group mean

[Amber Pong]

I have a dataset where twenty subjects with fever blisters were recruited and randomly assigned to either placebo or treatment. There are two outcomes: number of days to heal and whether blisters decrease or not. The variables are days(continuous) and reduced (binary; 1= yes, 0=no). The other variables are id (1-20), treatment (1=treatment, 0=placebo), history(1=prior history of blister, 0=no history), and age(1=20-30, 2=31-40, 3=41-50). I'm trying to figure out what command to use to find the unadjusted treatment group mean (as a decimal and in percent) for both outcomes?

[Emil Alnor]

bys treatment: days reduced

If this does not work, try to provide an example of your dataset using -dataex

[Amber Pong]

Emil Alnor That does not work. got the following when I tried that "command days is unrecognized"

This is an example of the dataset

* Example generated by -dataex-. For more info, type help dataex
clear
input int(id treatment reduced days history age)
 1 1 1  5 1 3
 2 0 1  6 0 3
 3 0 0  9 0 1
 4 0 0  9 1 3
 5 1 1  4 1 2
 6 1 1  4 1 1
 7 0 0 11 0 2
 8 1 0  8 1 3
 9 1 1  5 0 1
10 0 0  9 0 1
11 1 1  6 1 1
12 1 1  4 0 2
13 0 1  6 1 3
14 1 1  5 0 2
15 1 1  5 0 3
16 1 1  5 0 2
17 0 0  8 1 1
18 0 0 10 1 2
19 1 1  4 0 1
20 0 0  7 1 1
end

[Carlo Lazzaro]

Amber:
do you mean something along the following lines (that elaborates on Emil's suggestion)?:

. bys treatment: sum days reduced

------------------------------------------------------------------------------------------------------------------------------------------
-> treatment = 0

    Variable |        Obs        Mean    Std. dev.       Min        Max
-------------+---------------------------------------------------------
        days |          9    8.333333    1.732051          6         11
     reduced |          9    .2222222    .4409586          0          1

------------------------------------------------------------------------------------------------------------------------------------------
-> treatment = 1

    Variable |        Obs        Mean    Std. dev.       Min        Max
-------------+---------------------------------------------------------
        days |         11           5    1.183216          4          8
     reduced |         11    .9090909    .3015113          0          1


.

[Amber Pong]

Thanks, Carlo Lazzaro. I'm trying to get the unadjusted treatment (when treatment=1) group mean (as a decimal and %). I did the regression models for both outcomes adjusting for history and age and got the coefficients and standard error. But I also want the unadjusted treatment (when treatment=1) group mean. Is this something I can find in the model output?

[Carlo Lazzaro]

Amber:
sorry, but I do not follow you, especially as far as as a decimal and % are concerned (by the way, % makes me think about a log-linear regression model).
That said, what happens if you do not adjust your main predictors for history and age in your regression?

[Amber Pong]

Carlo, I think a simple descriptive of the treatment mean is what I'm looking for. Thanks for the guidance.

[Carlo Lazzaro]

Amber:
I would follow this approach:

. bys treatment: tabstat days, stat(count mean sd p25 p50 p75 min max)

------------------------------------------------------------------------------------------------------------------------------------------
-> treatment = 0

    Variable |         N      Mean        SD       p25       p50       p75       Min       Max
-------------+--------------------------------------------------------------------------------
        days |         9  8.333333  1.732051         7         9         9         6        11
----------------------------------------------------------------------------------------------

------------------------------------------------------------------------------------------------------------------------------------------
-> treatment = 1

    Variable |         N      Mean        SD       p25       p50       p75       Min       Max
-------------+--------------------------------------------------------------------------------
        days |        11         5  1.183216         4         5         5         4         8
----------------------------------------------------------------------------------------------

. bys treatment: tabstat reduced, stat(count mean sd p25 p50 p75 min max)

------------------------------------------------------------------------------------------------------------------------------------------
-> treatment = 0

    Variable |         N      Mean        SD       p25       p50       p75       Min       Max
-------------+--------------------------------------------------------------------------------
     reduced |         9  .2222222  .4409586         0         0         0         0         1
----------------------------------------------------------------------------------------------

------------------------------------------------------------------------------------------------------------------------------------------
-> treatment = 1

    Variable |         N      Mean        SD       p25       p50       p75       Min       Max
-------------+--------------------------------------------------------------------------------
     reduced |        11  .9090909  .3015113         1         1         1         0         1
----------------------------------------------------------------------------------------------

.