Create new dummy variable

[Theresia Verena]

Hi everyone,

I am trying to create a new dummy variable named DIS2 which defines a participant as having 1 or more definitions of disability. In my data set there are 5 dummy variables for each definition of disability, namely:
hearing: equal to 1 if the participant has difficulty in hearing, other equal to 0.
visual: which is equal to 1 if the participant has difficulty seeing, the other is equal to 0.
speaking: equal to 1 if the participant has difficulty in speaking, others equal to 0.
physical: equal to 1 if the participant has physical limitations, others equal to 0.
mental: which is equal to 1 if the participant has mental limitations, others equal to 0.

For example, if a participant has difficulty in speaking and listening, then in DIS2 it will be equal to 1. On the other hand, it will be equal to 0 when the participant only has 1 type of disability.
Any ideas? Thanks in advance!

[Rich Goldstein]

you did not provide a data example (please use -dataex-; read the FAQ to find out more) but you probably want the -egen- command with the "anyvalue" function; see

help egen

[Nick Cox]

Consider

gen dis = hearing + visual + speaking + physical + mental

It sounds as if you want

gen multdis = dis > 1 if dis < .

which will be 1 if dis is 2 or more and 0 if dis is 0 or 1 and missing if dis is missing.

If you want missing values to be treated differently, you may want to work with

egen dis2 = rowtotal(hearing  visual speaking  physical  mental)

which will ignore missings.