topic Re: Random assignment of pairs to one of two treatment groups in SAS Procedures

Random assignment of pairs to one of two treatment groups

MMozel — Mon, 25 Jun 2018 16:18:53 GMT

Hi SAS experts,

I have a listing of paired subjects that need to be randomly assigned to one of two treatments groups. I'm sure this is a pretty simple task but PROC PLAN doesn't seem to account for the paired structure of my data. To be clear, for each pair (A1/A2. B1/B2, etc...) I need to assign one to treatment group #1 and the other to treatment group #2. Please see attached pdf for a visual description of what I'm trying to do. I'm running SAS 9.4

Any help will be greatly appreciated!

Michelle

Re: Random assignment of pairs to one of two treatment groups

ballardw — Mon, 25 Jun 2018 20:14:23 GMT

The image in your PDF isn't very helpful. It appears as if you have every pair assigned to both treatments.

Provide some starting data in the form of a data step so we can see what actual variables you have, their names and types, and what you might expect as output.

Things that are important are do you have a single variable that identifies the "pairs"? or a single variable that we have to guess which "pair" is made from or two or more variables?

Re: Random assignment of pairs to one of two treatment groups

MMozel — Mon, 25 Jun 2018 21:13:52 GMT

I have 13 experimental pairs and need to generate a randomization scheme to assign one member of each pair to either the intervention group or control group.

Currently all I have in my dataset:

Substratum	Participant
A	A1
	A2
B	B1
	B2
C	C1
	C2
	…
M	M1
	M2

The output should be a third column with randomly generated intervention or control assignment.

thanks!