I am using a dataset in which there is a mistake in the way the answers for a given country are coded. Let's call my first variable my.data$country_year and the second one my.data$attitude. A table of the two gives the following output:
table(my.data$country_year, my.data$attitude)
(1) Very Suitable (2) Suitable (3) Somewhat Suitable (4) Not Suitable
Yemen.2006 101 142 1192 0
Lebanon.2007 13 14 60 1063
Yemen.2007 49 113 122 248
Palestine.2008 131 653 387 2093
The table in question has a lot of countries which I omitted in this example.What I want to do is to ask R to do the following in my original dataset while keeping the observations for all other countries:
for my.data$country="Yemen.2006" & my.data$attitude="(3) Somewhat Suitable", "(4) Not Suitable.
for my.data$country="Yemen.2006" & my.data$attitude="(2) Suitable", "(3) Somewhat Suitable"
In other words, I would like to move to the right the second and third values of the attitude variable for the survey conducted in Yemen in 2006, without creating a new variable. I would like the result to be the following
(1) Very Suitable (2) Suitable (3) Somewhat Suitable (4) Not Suitable
Yemen.2006 101 0 142 1192
Lebanon.2007 13 14 60 1063
Yemen.2007 49 113 122 248
Palestine.2008 131 653 387 2093
Ok, so thank you everybody, I based myself on your answers in order to find the solution. I created a subset of my.data with less variables and this time the within function worked without errors. Only it's r2evans' script that worked best, as follows
my.data2 <- within(my.data2, {attitude[country_year=="Yemen.2006" &
attitude=="(3) Somewhat Suitable"] <- "(4) Not Suitable At All" })
my.data2 <- within(my.data2, {attitude[country_year=="Yemen.2006" &
attitude=="(2) Suitable"] <- "(3) Somewhat Suitable" })
Cheers and thanks for your time and help!