Hi-
Any thoughts on efficient ways to accomplish the following recoding
task?
I've got six variables, wherein respondents ranked the importance of
each of six strategies. For the vast majority of cases, respondents
assigned one rank of "1", one of "2" and so on.
Some respondents, however, ranked more than one strategy with the same
rank. This takes many forms: some gave two ranks of "1", then one of
"3", "4", "5", and "6". Others gave two ranks of "1", then one of "2"
through "5" but no "6". Others gave more than two of the same rank to
different items, etc.
We want to do the following to produce a fully consistent set of
rankings:
For multiple identical rankings, randomly order them, and then rank
subsequent ranks sequentially from there. So, for example, if the data
look like this for two cases:
S1 S2 S3 S4 S5 S6
1 1 2 3 4 5
3 1 1 4 5 6
I would want to change it to:
S1 S2 S3 S4 S5 S6
1 2 3 4 5 6
3 1 2 4 5 6
That is, for the first case, S1 was randomly chosen to be kept as one
and S2 as two; and the remaining ranks ordered from there. Similarly
for the second case.
Any thoughts?
I should say, in closing, that I did not decide on this approach, so
while I would certainly enjoy any commentary on why it's a bad idea, I
will need to implement it regardless...
Thanks
Nick W.
-----------------------------------------------------------
Nicholas Winter, Ph.D. P 202.939.5343
Policy Studies Associates F 202.939.5732
1718 Connecticut Avenue, NW [email protected]
Washington, DC 20009-1148 www.policystudies.com
-----------------------------------------------------------
<<winmail.dat>>
