Average mean of variables

[Temilolu Bamiwuye]

Please help
in a situation where I have 6 variables (Proportion A - F)about various proportion.
I want to create a new variable (a composite variable) on the average proportion of Proportion A - F.
what do i do?

eg.

Proportion A (100%)	Proportion B (100%)	Proportion C (100%)	Proportion D (100%)	Proportion E (100%)	Proportion F (100%)	Average proportion X (100%)
20	0	0	60	0	20
0	50	10	10	10	20
80	10	0	5	5	0
10	10	10	30	20	20

[Rich Goldstein]

those cannot be your real variable names as Stata does not allow spaces in variable names; please read and follow the FAQ advice on how to post; meanwhile, try

egen avgprop=rowmean(propA, etc.)

where you replace avgprop with your chosen name and, more importantly, replace what is in parentheses with the actual names of the variables; see

h egen

note that I have assumed your data are in wide format as implied above but that might not be true (another reason to post as per the FAQ)

[Nick Cox]

Rich Goldstein is spot on as usual. In addition, I note that if the row total is always 100% then the row mean is necessarily 100/6% to whatever number of decimal places you prefer. There would or could be exceptions whenever A to F are not the only constituents; whenever there are errors in the data; and possibly as a matter of rounding.

[Temilolu Bamiwuye]

thanks Rich Goldstein and Nick Cox.
The data was reshaped from long format to wide format.
What I'm trying to do is to create a composite variable to represent the average proportion per household id

eg. this is the data after reshaping

prop_a1	prop_a2	prop_a3	prop_a4	prop_a5	prop_a6
5	10	0	0	0	0
15	0	0	0	0	0
5	10	0	0	0	0
0	15	0	0	0	0
0	15	0	0	0	0
10	5	0	0	0	0
0	0	10	5	0	0
10	5	0	0	0	0
5	5	5	0	0	0
15	0	0	0	0	0

[Nick Cox]

#4 Fine by me. I was guided by your data example in which the row total is always 100%.

I recommend this usage:

proportions add to 1

percents add to 100.