The propensity score method is frequently used to deal with bias from standard regression in observational studies. The propensity score method involves calculating the conditional probability (propensity) of being in the treated group (of the exposure) given a set of covariates, weighting (or sampling) the data based on these propensity scores, and then analyzing the outcome using the weighted data. I first review methods of allocation of weights for propensity score analysis and then introduce weighting within strata and proportional weighting within strata as alternative weighting methods. These new methods are compared to existing ones using empirical analysis and using a data set on whether sending patients to a respite unit prevents readmission or death within ninety days. Simulations are then described and discussed to compare the existing and new methods.