Score Design for Multi-Criteria Incentivization

TL;DR

This paper introduces a framework for designing multi-criteria scores that accurately reflect performance metrics, ensuring improvements in scores correspond to metric improvements and vice versa, inspired by hospital rating systems.

Contribution

The paper proposes a novel score design framework with algorithms that produce minimal-dimensional scores aligned with multiple performance objectives.

Findings

Algorithms for score design are provably minimal under mild assumptions.

Scores improve all metrics when scores improve, ensuring alignment.

Framework is motivated by real-world hospital rating practices.

Abstract

We present a framework for designing scores to summarize performance metrics. Our design has two multi-criteria objectives: (1) improving on scores should improve all performance metrics, and (2) achieving pareto-optimal scores should achieve pareto-optimal metrics. We formulate our design to minimize the dimensionality of scores while satisfying the objectives. We give algorithms to design scores, which are provably minimal under mild assumptions on the structure of performance metrics. This framework draws motivation from real-world practices in hospital rating systems, where misaligned scores and performance metrics lead to unintended consequences.