Propensity scoring was developed as a statistical technique for adjusting for selection bias in causal estimates of treatment effects in observational studies. Unlike randomized experiments, in observational studies researchers have no control over treatment assignment, and, as a result, individuals who receive different treatments may be very different in terms of their observed covariates. These differences, if left unadjusted, can lead to biased estimates of treatment effects. For example, if smokers tend to be older than nonsmo-kers, then comparisons of smokers and nonsmokers will be confounded with age. Propensity scores can be used to adjust for this observed selection bias.

Survey researchers have used propensity scores to adjust for nonresponse bias, which arises when respondents and nonrespondents differ systematically in terms of observed covariates, and to ...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles