Audit-Protected Performance Measurement

In Simple Words

Imagine a classroom exam where students know that random answer sheets will be double-checked by independent outside inspectors; unpredictable audits prevent students from cheating or making fake arrangements with graders.

Research question

Public sector performance evaluation based on beneficiary surveys or worker self-reports risks metric manipulation, output inflation, and worker-client collusion.

Participants

N = 2,009 Community Health Workers and primary female caregivers across 10% of households in 372 Peripheral Health Units in Sierra Leone.

Study Design

Randomized field trial incorporating weekly unannounced back-checks by independent field monitors across a random 25% subset of surveyed households via phone or in-person visits.

Results

Unannounced third-party auditing prevented data fabrication, yielding verified performance data showing no statistically significant difference in reported visit rates between socially connected households (friends/family) and non-connected households (p = 0.439).

Takeaway

Implement mandatory unannounced third-party audit back-checks on a random 25% sample of performance evaluation data to eliminate collusion and reporting bias.

Read alongside the research

Plain-English glossary

Original explanations, informed by the references below. Examples are illustrations, not data from the studies above. The research text stays unchanged.

Statistically significant

A label used when a result passes a chosen statistical threshold, often p < 0.05. It does not show how large or useful an effect is. A result that is not statistically significant does not prove there is no effect.

ExampleA study reports p = 0.03 against a 0.05 threshold. It passes that test, but the effect could still be too small to matter.

American Statistical Association · Statement on p-values ↗Read more about statistically significant →

Related findings

• Administrative Social Security Data Linkage for Earnings Measurement →Economics (general)

  Instead of asking individuals to remember and report exact yearly income numbers from a decade ago, researchers obtain signed consent to link survey profiles directly to official government tax records. This creates an unalterable, objective earnings ledger covering years before and after a financial event.

• Agnostic Machine Learning Heterogeneity Detection in Field Trials →Economics (general)

  Evaluating a program using only average results is like measuring the average temperature of a room with one frozen wall and one burning wall and concluding the room is comfortable. Machine learning acts like a thermal camera, splitting the room into sections to reveal the extreme cold and extreme heat hiding behind the average.

• Asymmetric Wealth Effect of Subsidized Index Contracts →Economics (general)

  Imagine buying a lottery ticket that costs $1, but pays out $1,000 if a major storm hits. Paying $1 during normal weather is so cheap that it does not change weekly spending or force extra work hours to make up for the cost.

• Bayesian Downward Belief Revision of Infectious Risk →Economics (general)

  Imagine living in a town where everyone assumes a severe storm hits every afternoon, leading everyone to wear heavy raincoat gear daily. If neighbors start checking barometers and announcing that the sky is clear, residents realize the true weather risk is far lower than assumed and adjust expectations accordingly.