Skip to main content

Table 2 Calculation and interpretation of propensity weights

From: Youth vaping and smoking and parental vaping: a panel survey

Estimand

How predicted probabilities of exposure are used to calculate weightsa:

Re-weighting of confounding characteristics in exposure groups:

Estimated effect for youth vaping applies to …

Estimated effect for Parental Vaping applies to …

Numerator

Denominator

Unexposed

Exposed

  

Average Treatment Effect

(ATE)

P

P^

Resemble sample characteristics

Resemble sample characteristics

All youth

All youth

Average Treatment Effect among the Treated

(ATT)

1 if exposed,

1-P^ if unexposed

1 if exposed,

P^ if unexposed

Resemble Exposed group

Unchanged

Youth who do vape

Youth whose parents did vape

  1. P=Overall, unadjusted probability of individual’s observed exposure level
  2. P^=Predicted probability of individual’s observed exposure level conditional on confounders
  3. aWhen investigating effects of parental vaping within strata of parental smoking, both P and P^ were additionally conditional on parental smoking