Using Cardinality Matching to Design Balanced and Representative Samples for Observational Studies

TL;DR

Cardinality matching is a computational technique that optimizes the selection of matched pairs in observational studies to ensure balance, maximize sample size, and enhance representativeness of the target population.

Contribution

This paper introduces and explains the process of cardinality matching, a novel method for improving sample quality in observational research.

Findings

Maximizes matched sample size while maintaining balance.

Ensures the matched sample is representative of the target population.

Addresses multiple concerns simultaneously in observational study design.

Abstract

Cardinality matching is a computational method for finding the largest possible number of matched pairs of exposed and unexposed individuals from an observational dataset, with specified patterns of baseline characteristics that represent a target population for analysis. This article explains the process of cardinality matching and how it simultaneously addresses the concerns of balance, sample size, and representativeness of matched samples in observational studies.