Matching cases

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Torvon
Posts: 34
Joined: Sun Nov 13, 2011 8:41 pm

Matching cases

Postby Torvon » Thu Jun 14, 2012 5:45 pm

Hello.

I have a huge dataset (N=20000), and would like to match people with a 1 on variable A to people with a 0 on variable A, according to variables B C D E F, to get pairs.

E.g. Susan is married (A=1). She is my experimental group. I want to find a control for Susan now that is unmarried (A=0), but has same age (B) gender (C) SES (D) etc.

Is there any way to do this in SPSS (or any other software you would know of) other than by doing it manually?
Thanks
T
WirelessK
Posts: 3
Joined: Thu Jun 14, 2012 8:03 pm

Re: Matching cases

Postby WirelessK » Fri Jun 15, 2012 4:46 pm

Hello,
You may have had your questions long answered by now, but...

Many people use propensity score matching -- both R and Stata have routines have I used. I can give you more info if you're about and want additional details.

If you want to use propensity score matching you will need to match the people on variables collected prior to the treatment. Kathy :D
Torvon
Posts: 34
Joined: Sun Nov 13, 2011 8:41 pm

Re: Matching cases

Postby Torvon » Fri Jun 15, 2012 9:09 pm

Kathy,

I haven't used these kind of procedures before and would love to get additional information. I currently don't have access to Stata, but am proficient with R.

My goal is: imagine N=20000, 10% of them with variable A=1. I want to match people with A=0 individually now, making sure that both groups have the same means on demographic variables B C D E.

Thank you
-T

Who is online

Users browsing this forum: No registered users and 1 guest

cron