@article {Miratrix:2013-03-01T00:00:00:1369-7412:369,
author = "and and",
title = "Adjusting treatment effect estimates by poststratification in randomized experiments",
journal = "Journal of the Royal Statistical Society: Series B (Statistical Methodology)",
volume = "75",
number = "2",
year = "2013-03-01T00:00:00",
abstract = "
Summary. Experimenters often use poststratification to adjust estimates. Poststratification is akin to blocking, except that the number of treated units in each stratum is a random variable because stratification occurs after treatment assignment. We analyse
both poststratification and blocking under the NeymanRubin model and compare the efficiency of these designs. We derive the variances for a poststratified estimator and a simple differenceinmeans estimator under different randomization schemes. Poststratification
is nearly as efficient as blocking: the difference in their variances is of the order of 1/n
2, with a constant depending on treatment proportion. Poststratification is therefore a reasonable alternative to blocking when blocking is not feasible. However, in finite
samples, poststratification can increase variance if the number of strata is large and the strata are poorly chosen. To examine why the estimators variances are different, we extend our results by conditioning on the observed number of treated units in each stratum. Conditioning
also provides more accurate variance estimates because it takes into account how close (or far) a realized random sample is from a comparable blocked experiment. We then show that the practical substance of our results remains under an infinite population sampling model. Finally, we provide
an analysis of an actual experiment to illustrate our analytical results.",
pages = "369-396",
url = "http://www.ingentaconnect.com/content/bpl/rssb/2013/00000075/00000002/art00008",
doi = "doi:10.1111/j.1467-9868.2012.01048.x"
}