Science checking due diligence: a new approach to scientific accountability

Governments, regulators and businesses increasingly rely on scientific advice to make decisions affecting public safety, environmental outcomes, economic productivity and major investments. Yet unlike financial, legal or safety decisions, scientific advice is rarely subjected to structured due diligence or independent accountability review.

Gaps in science checking

Governments, regulators and businesses increasingly rely on scientific advice to make decisions affecting public safety, environmental outcomes, economic productivity and major investments. Parliamentary inquiries, courts, review bodies and regulatory agencies also depend upon scientific evidence when evaluating risks and determining outcomes.

Yet unlike financial, legal or safety decisions, scientific advice is rarely subjected to structured due diligence or independent accountability review.

Financial decisions may be audited. Safety systems may undergo independent review. Governance arrangements are routinely assessed against recognised standards. However, the assumptions, evidence, forecasts and implementation arrangements underpinning major science based decisions often receive far less scrutiny.

This gap applies both to decisions already implemented and to proposals being considered for future adoption.

The attached Science checking due diligence and audit checklist has been developed to help address this challenge. The framework provides a structured process for reviewing policies, regulations, programs, scientific recommendations and operational decisions. It is intended for Auditor Generals, Productivity Commissions, parliamentary inquiries, regulatory review bodies, scientific review panels, government agencies, courts, tribunals and organisations seeking to strengthen accountability and decision quality.

Objectives of science checking

The primary purpose of science checking is to improve transparency, strengthen confidence in evidence based decisions and reduce the risks associated with weak evidence, flawed assumptions or implementation failures.

The consequences of scientific error can be significant. Impacts may extend to public safety, environmental management, economic productivity, employment, government expenditure, industry competitiveness and community wellbeing. Even relatively small errors can have substantial consequences when decisions affect large populations, critical infrastructure or long term policy settings.

Science checking due diligence and audit checklist

The attached checklist framework begins by examining the foundations of evidence used to support decisions. Reviewers assess the quality and strength of research, the availability of independent verification, the treatment of uncertainty and whether stronger or contradictory evidence exists. Questions also examine distinctions between causation and correlation and whether conclusions remain proportionate to the available evidence.

The framework evaluates scientific methodology by examining whether methods are documented, findings can be replicated, limitations are disclosed and alternative explanations have been considered. It also focuses on assumptions and uncertainty, recognising that many decisions depend on assumptions about future conditions, economic behaviour, environmental responses or technological performance. Critical assumptions, confidence levels and knowledge gaps are assessed to determine how robust conclusions remain when subjected to challenge.

Forecasting and modelling receive similar scrutiny. The framework asks whether models have been independently reviewed, validated against observations, subjected to sensitivity testing and compared with alternative approaches. It also introduces the concept of prediction accountability by examining whether forecasts ultimately align with real world outcomes.

A distinctive feature of science checking is its emphasis on reality testing. Conclusions may appear persuasive in reports, simulations or academic literature, but practical implementation often produces different outcomes. The framework examines whether field observations, historical experience, practitioner knowledge, examples of success and examples of failure have been adequately considered. This helps bridge the gap between theory and operational reality.