[R] Propensity score matching with MatchIt

[Holger Steinmetz]

Hi there,

I try to create two similar cross-cultural sub samples with MatchIt and it
does not work. Perhaps, someone
can lead me in the right direction.

I have two samples that differ on some occupational data (mainly pay and job
experience in years).
However using MatchIt gives warning I cannot understand:

Warnmeldungen:
1: glm.fit: fitted probabilities numerically 0 or 1 occurred 
2: In matchit2nearest(c(0L, 0L, 0L, 0L, 0L, 1L, 1L, 0L, 0L, 0L, 0L,  :
  Fewer control than treated units and matching without replacement.  Not
all treated units will receive a match.  Treated units will be matched in
the order specified by m.order: largest


The summary is:


Summary of balance for all data:
         Means Treated Means Control SD Control Mean Diff eQQ Med eQQ Mean
eQQ Max
distance        0.7093        0.3053     0.2914    0.4040  0.4380   0.4043 
0.6404
pay             2.1186        6.5809     8.4825   -4.4623  2.8806   4.5098
61.3719
jexp           10.4596       17.2035    12.3805   -6.7439  6.6000   6.8369
19.5000
hours02        44.1827       48.2222    10.3129   -4.0395  5.0000   4.7020
10.0000


Summary of balance for matched data:
         Means Treated Means Control SD Control Mean Diff eQQ Med eQQ Mean
eQQ Max
distance        0.7354        0.3053     0.2914    0.4301  0.4707   0.4318 
0.6576
pay             1.9540        6.5809     8.4825   -4.6269  2.9806   4.6529
65.9736
jexp            9.9540       17.2035    12.3805   -7.2495  6.8000   7.2495
19.0000
hours02        44.3990       48.2222    10.3129   -3.8232  5.0000   4.6414
10.0000

Percent Balance Improvement:
         Mean Diff. eQQ Med eQQ Mean eQQ Max
distance    -6.4478 -7.4715  -6.7965 -2.6742
pay         -3.6892 -3.4715  -3.1715 -7.4980
jexp        -7.4968 -3.0303  -6.0353  2.5641
hours02      5.3545  0.0000   1.2889  0.0000

Sample sizes:
          Control Treated
All               198     208
Matched       198     198
Unmatched       0      10
Discarded         0       0


Thanks in advance,
Holger
-- 
View this message in context: http://r.789695.n4.nabble.com/Propensity-score-matching-with-MatchIt-tp3015626p3015626.html
Sent from the R help mailing list archive at Nabble.com.