Matchit July 2009

matchit@lists.gking.harvard.edu

9 participants
8 discussions

Re: [matchit] can a variable be used as a co-variate and a dependent variable?

by Michael Foster

hi all-- the verdict is really out on the role of those baseline, pre-treatment measures of the dependent variable Chris Winship and his collaborators have a paper on the topic that I saw presented and they (the variables, not Chris!) certainly seem like colliders if there's autocorrelation i know from a econometric perspective including lagged dependent variables as a covariate can be a problem I've grown to feel a bit like this is an area where "damned if you do, damned if you don't" applies /m > > Stuart, Elizabeth A. wrote: > > Hi Casey, > Matching on the outcome is certainly problematic....imagine if you got > perfect balance (which you would hope to do), then that would also imply a 0 > treatment effect! > > What wouldn’t be problematic (and in fact would be good) is if they were > actually matching on a pre-treatment measure of the outcome. E.g., in > education studies there is often pre-treatment test scores available, where > those same tests form the outcome value as well (but then measured > post-treatment). This kind of matching on the outcome variables (but > measured before the treatment) can be very effective. > > Thanks > Liz > > > On 7/27/09 2:56 PM, "Casey Klofstad" <klofstad(a)gmail.com> wrote: > > Hey all- > > In a paper I'm reviewing one of the covariates used in a 1-to-1 > matching procedure ends up also being used as the dependent variable > in the final analysis. From a theoretical point of view this seems > odd, since the variable is being treated as both a pre- and > post-treatment variable. But, is there any empirical reason to be > worried about this? > > Thanks much, > > -c > > -- > Casey A. Klofstad > University of Miami > Department of Political Science > Coral Gables, FL > > klofstad(a)gmail.com > http://www.as.miami.edu/personal/cklofstad/ > > -- > MatchIt mailing list served by Harvard-MIT Data Center > List Address: matchit(a)lists.gking.harvard.edu > Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=matchit > MatchIt Software and Documentation: http://gking.harvard.edu/matchit/ > Browse/Search List Archive: http://lists.hmdc.harvard.edu/lists/matchit/ > > > -- MatchIt mailing list served by Harvard-MIT Data Center List Address: matchit(a)lists.gking.harvard.edu Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=matchit MatchIt Software and Documentation: http://gking.harvard.edu/matchit/ Browse/Search List Archive: http://lists.hmdc.harvard.edu/lists/matchit/

14 years, 9 months

can a variable be used as a co-variate and a dependent variable?

by Casey Klofstad

Hey all- In a paper I'm reviewing one of the covariates used in a 1-to-1 matching procedure ends up also being used as the dependent variable in the final analysis. From a theoretical point of view this seems odd, since the variable is being treated as both a pre- and post-treatment variable. But, is there any empirical reason to be worried about this? Thanks much, -c -- Casey A. Klofstad University of Miami Department of Political Science Coral Gables, FL klofstad(a)gmail.com http://www.as.miami.edu/personal/cklofstad/ -- MatchIt mailing list served by Harvard-MIT Data Center List Address: matchit(a)lists.gking.harvard.edu Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=matchit MatchIt Software and Documentation: http://gking.harvard.edu/matchit/ Browse/Search List Archive: http://lists.hmdc.harvard.edu/lists/matchit/

14 years, 9 months

match.data undefined columns selected error

by Toby Edison

I am having a problem getting the match.data function to work on my data. Just getting started with matchit, but cannot seem to get the match.data function to work on my data. I am using a Mac. Here is the code I use: all<-read.table("file:///all.txt", header = TRUE, sep = "\t", quote="\"", dec=".") m.out <- matchit(wins03 ~ firstcontractyear + foundingyear + employees + top_no + sub03 + prime03 + tns2002 + ts2002+ phiiawards + phiawards + cai, data = all, method = "cem") m.data <- match.data(m.out) z.out1<-zelig(tns05 ~ wins03+ tns2002 + ts2002 + tns03 + foundingyear + employees + sub03 + prime03 + phiiawards + phiawards + cai, data = m.data, model="ls") resulting in : Error in `[.data.frame`(d, , all.vars(as.expression(formula))) : undefined columns selected match.data works fine with the lalonde data set but I cannot get it to work with my imported data. I've saved my data from an excel spreadsheet into a tab delimited text file. I've tried using .csv imports as well. Any ideas. Toby Edison

14 years, 9 months

MatchIt stops responding

by Minh-thien Vu

Hi everyone, I am using the latest MatchIt version 2.4-11 and R 2.9.1 to match students in the attachment. I used the following commands for my job. > library(MatchIt) > mse<-read.csv('E:\\Matching\\Gr4\\Data\\Prob\\MSE.csv', header=TRUE) >attach(mse) > mse.out<-matchit(rankcfa~tcap08 + READ08 + SCI08, data=mse, method="full", max.controls=1) When I executed the last command, there was a pop up window with title "Important license information regrading optmatch pakage). I did not get any responses from the program after running for three hours. Yet MatchIt only took me under five minutes to match another file with the same size and same options. I have eliminated null values as well any blank spaces/ double commas in the file and reapplied the above routines. Unforunately, all my attempts failed. I am not sure what else I can try. If any of you have an ideas, may I seek for your advice? Thank you! Minh

14 years, 9 months

match.data problem

by Alyssa Deckert

Hello, I'm attempting to extract a data set. I enter: 'mdata <- match.data(m.out)' after assigning 'm.out <- matchit(treat ~ x1...)' 'summary(mdata)' yeilds the following: distance.Length distance.Class distance.Mode NA.Length NA.Class NA.Mode weights 1 -none- function 1 -none- numeric Min. :1 1 -none- function 1 -none- numeric 1st Qu.:1 1 -none- function 1 -none- numeric Median :1 1 -none- function 1 -none- numeric Mean :1 1 -none- function 1 -none- numeric 3rd Qu.:1 1 -none- function 1 -none- numeric Max. :1 1 -none- function 1 -none- numeric In addition, I am unable to extract the dataset: write.dta(mdata, "mdata.dta") Error in write.dta(mdata, "mdata.dta") : unknown data type I am, however, able to view the results of the matching analysis, use the plot function, etc. Any idea what's going wrong? I am really just looking to observe which control observations were matched to my treatment observations. Is there an easier way to do this within R? If the manual helps on this point I am too dense to understand it because I've tried... Thanks, Alyssa PS: The commands work fine when I run the lalonde data.

14 years, 9 months

Re: MatchIt stops responding.

by Daniel Ho

Hi Minh. Unfortunately, it's very difficult to assess this problem without knowing any of the details of the dataset. Can you provide a way to replicate the problem? I'm CC'ing the MatchIt listserv to which you should direct your queries. Best, Dan On Fri, Jul 17, 2009 at 4:51 PM, Minh-thien Vu<mthien(a)stanford.edu> wrote: > Dear Dr. Ho, > > I am professor Suppes's data analyst, and thank you so much for responding > to our question the other day. I have another problem with MatchIt or R, > may I ask for your help? > > I am using the latest MatchIt version 2.4-11 and R 2.9.1. I successfully > load MatchIt library and read in my csv dataset. After I ran the MatchIt > with full matching method and min.controls=1, R stopped responding. I do not > believe there is a problem with the syntax because I have been using the > same syntax for other datasets (of course different paths and datafiles). > One thing I am not sure about is whether this datafile itself causes the > problem. However I don't understand why so because I don't have any > problems with the others which was created the same way as this one. I am > puzzeling about this and hope to seek your experty. > > > Thank you so much in advance. > > > Best, > > Minh > > > -- MatchIt mailing list served by Harvard-MIT Data Center List Address: matchit(a)lists.gking.harvard.edu Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=matchit MatchIt Software and Documentation: http://gking.harvard.edu/matchit/ Browse/Search List Archive: http://lists.hmdc.harvard.edu/lists/matchit/

14 years, 9 months

pdf font issue

by Gary King

14 years, 9 months

can you match with "post-treatment" variables?

by Casey Klofstad

Hey all- I'm diving into the ANES panel data, and I find myself wanting to match with variables that were collected in a wave that was collected AFTER the treatment variable was collected. Thoughts? Is this contingent on whether you can argue that the treatment (theoretically) had no effect on the post-treatment covariate? Best, -c -- Casey A. Klofstad University of Miami Department of Political Science Coral Gables, FL klofstad(a)gmail.com http://www.as.miami.edu/personal/cklofstad/ -- MatchIt mailing list served by Harvard-MIT Data Center List Address: matchit(a)lists.gking.harvard.edu Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=matchit MatchIt Software and Documentation: http://gking.harvard.edu/matchit/ Browse/Search List Archive: http://lists.hmdc.harvard.edu/lists/matchit/

14 years, 9 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Matchit July 2009