Thomas G. Stewart Assistant Professor
This lecture is part 9 of the Propensity Scores and Related Methods Series presented and organized by Robert Greevy within Vanderbilt University's Center for Health Services Research.
I reserve the right for these notes to be wrong, mistaken, or incomplete.
These notes will continue to be updated as I
Please feel free to provide content and comments.
StataCorp. 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP.
E. Leuven and B. Sianesi. 2003. PSMATCH2: Stata module to perform full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. (link). Version 4.0.11. To install in STATA, use command:
ssc install psmatch2
ssc install table1
To motivate the propensity score matching, I'll use the cattaneo2
dataset, a STATA example dataset. It can be loaded with the following command:
webuse cattaneo2
The data in cattaneo2
is a subset of data that was analysed in the following journal articles:
Almond, D., Chay, K.Y., Lee, D.S., 2005. The costs of low birth weight, Quarterly Journal of Economics 120, 1031-1083.
Cattaneo, M.D. 2010. Efficient semiparametric estimation of multi-valued treatment effects under ignorability, Journal of Econometrics, 155(2), 138-154.
The dataset included information about infant/mother/father characteristics from singleton births in Pennsylvania between 1989 and 1991. The original dataset included nearly 500,000 births. The STATA example dataset includes 4642 births.
RESEARCH QUESTION: What is the effect of maternal smoking during pregnancy on the infant's birthweight?
The dataset includes the following set of variables:
Infant | Mother | Father |
---|---|---|
birth weight (grams) | married/not married | |
birth month | hispanic (yes/no) | hispanic (yes/no) |
race (white/not white) | race (white/not white) | |
age | age | |
education | education | |
foreign born (yes/no) | ||
birth number | ||
months since last birth | ||
infant of previous births died (yes/no) | ||
number of prenatal care visits | ||
trimester of first prenatal care visit | ||
alcohol during pregnancy (yes/no) | ||
smoking during pregnancy (0, 1-5, 6-10, 11+ daily) | ||
smoking during pregnancy (yes/no) |
You can access the complete codebook with the command codebook
after loading the data.
Randomized Clinical Trial | Observational Study | |
---|---|---|
Treatment Assignment: | Investigators generate a treatment schedule prior to patient enrollment. The schedule is constructed based on the design of the study, which includes randomization in some fashion. Physicians (who may be blind to treatment as well) assign treatments/exposures to study participants following the sequence in the schedule. |
|
CONSEQUENTLY | ||
Probability of Treatment: | Known | Unknown, may be 0 or 1 |
Covariate Balance: | Relationship between covariates and treatment assignment are known from study design. Usually the study is designed so that there is no relationship between treatment assignment and covariates. | Relationship between covariates and treatment assignment is unknown. There may be covariate imbalance. |
Differences in outcomes between the treated and untreated (or exposed and unexposed) may be the consequence of confounding variables and not the treatment (or exposure).
Dataset may include sub-groups for which a treatment effect should not be calculated because
There are several methods for estimating a treatment effect with observational data. In this lecture series, you have been exposed (not randomly) to a family of methods which use the propensity score. The primary focus has been on propensity score matching.
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。