The short answer

How OzPolitics handles missing evidence and uncertainty

We state what is known, what is inferred and what remains unsupported; identify the missing document or data; and avoid converting uncertainty into either certainty or automatic falsity.

This is a method for assessing public claims, not a verdict on a party or a direction on how to vote. Conclusions should change when the underlying law, data or implementation evidence changes.

The useful question is not only “what is the rule?” but also “who administers it, which document controls it, and when might it change?” That distinction prevents an accurate general explanation from becoming wrong advice in a particular election, chamber or policy setting.

Evidence review

Why acknowledging uncertainty is a core editorial principle

Political debate rewards certainty. Politicians assert, promises are stated as guarantees, and the 24-hour news cycle presses for clear answers on complex questions where the full evidence is not yet available. In that environment, it is tempting for any analysis site to fill gaps with assumptions, to treat absence of evidence as evidence of absence, or to present a confident narrative that smooths over the rough edges of what is actually known. OzPolitics takes a different approach. We treat uncertainty not as a failure of analysis but as a feature of honest political communication. When we cannot verify a claim, we say so. When evidence is incomplete, we describe what is missing. When two plausible interpretations of the same data exist, we present both and explain the basis for each. This approach means that some of our articles will end with more questions than answers, and that is intentional. A reader who understands what is not known is better equipped than one who has been given false closure. The Australian National Audit Office uses a methodology based on reasonable assurance and evidence standards that explicitly acknowledges the limits of audit findings, and this principle of openly stating the basis and limitations of evidence is one we adopt in our own work. Transparency about uncertainty is not a weakness; it is the foundation of analytical integrity.

Evidence review

The categories: known, inferred, unsupported, and missing

When we assess a political claim or a policy question, we classify information into four categories. The first is known: facts that are established by verifiable evidence from authoritative sources, such as official data from the Australian Bureau of Statistics, the text of legislation published on the Federal Register of Legislation, court judgments, parliamentary records, or official reports from bodies such as the Productivity Commission or the Australian National Audit Office. The second is inferred: conclusions that follow logically from known facts but have not been directly observed or confirmed. Inference is essential to analysis but must be flagged so the reader knows where the evidence ends and the reasoning begins. The third is unsupported: assertions for which no verifiable evidence has been provided or found. This does not mean the assertion is false. It means there is no basis on which a reasonable person could accept it as true without taking it on faith or deferring to the authority of the person making the claim. The fourth category is missing: evidence that should exist if a claim is true but cannot be found. For example, if a government claims a program has delivered a certain number of outcomes, the program's evaluation report, administrative data, or contract deliverables should exist. If they do not, or if they have been requested under freedom of information and not released, that gap is itself an important piece of information that we report.

Evidence review

How we identify what is missing and why it matters

Identifying missing evidence is not simply about noting that something cannot be found. It requires specifying what document, data set, or record would reasonably be expected to exist if the claim is true, and explaining why its absence is significant. For example, if a government claims that a new policy has reduced waiting times for a particular service, the expected evidence would include time-series administrative data on waiting times before and after the policy took effect, with a clear methodology for measurement and adequate controls for other factors that could affect waiting times. If no such data has been published, or if the published data covers only selected locations or time periods, or if the methodology is not documented, the claim is unverifiable regardless of how emphatically it is expressed. We will typically describe what the missing evidence would look like, note any requests we or others have made for it, and report any explanations offered by the relevant agency or government for its unavailability. This approach serves two purposes. It alerts the reader to the evidential gap, and it provides a clear standard by which the claim can later be assessed if the evidence does emerge. It also discourages the common political tactic of making a claim now and deferring the evidence to a review or evaluation that may never be completed or released. The Australian Electoral Commission's approach to addressing misinformation is instructive here: it identifies the specific claim, states whether it is accurate, and provides the authoritative factual basis for that conclusion. When the evidence is insufficient to make a definitive finding, it says so.

Evidence review

The difference between absence of evidence and evidence of absence

One of the most important logical distinctions in evidence assessment is the difference between not having found evidence for a claim and having found evidence that the claim is false. A claim that is not supported by publicly available evidence may nonetheless be true. Government decision-making often occurs behind closed doors, cabinet documents are protected by confidentiality conventions, and commercial-in-confidence provisions limit what can be disclosed about government contracts and negotiations. The fact that OzPolitics cannot verify a claim does not mean the claim is false. It means the evidence is not available to us on the terms required for verification, and we report that. Conversely, when we say a claim is false, we mean we have found authoritative evidence that contradicts it. For example, if a politician claims that a particular tax rate applies and the legislation or an Australian Taxation Office ruling clearly states a different rate, the claim is not merely unverified but positively contradicted. We distinguish these situations clearly in our articles, and we encourage readers to apply the same distinction in their own assessment of political claims. The default reaction to an unverifiable claim should not be disbelief but suspended judgment, accompanied by a clear understanding of what evidence would be needed to resolve the question and whether that evidence is realistically obtainable.

Evidence review

Our sources: what counts as authoritative and why

OzPolitics relies on a hierarchy of sources that reflects their reliability and verifiability. At the top are primary legal and official sources: legislation on the Federal Register of Legislation, ABS data, parliamentary records including Hansard and committee reports, court and tribunal decisions, budget papers, and official reports from agencies such as the Australian National Audit Office, the Productivity Commission, and the Commonwealth Ombudsman. These sources have institutional safeguards for accuracy and are publicly accessible, allowing any reader to verify our use of them. Next are secondary official sources such as Parliamentary Library research papers, which are non-partisan and produced with rigorous editorial standards. After that are academic sources, which carry the authority of peer review and institutional independence but may be contested within their disciplines. Media reports and political statements are treated as objects of analysis rather than sources of authority. When we cite a media report, we do so to note that a claim was made or an event occurred, not to endorse the accuracy of the claim. When we cite a politician's statement, we treat it as evidence of what was said, not as evidence that the statement's content is true. This sourcing discipline is essential to an evidence-based approach because it prevents the circular problem of citing one unverified claim to support another.

Evidence review

How we update articles when new evidence emerges

Most of our foundation articles describe stable institutional arrangements and are reviewed periodically rather than updated continuously. However, our review-desk articles and any article that assesses a specific claim or policy development may need updating when new evidence becomes available. When we update an article, we do not silently revise our earlier assessment. We note the date and nature of the update, describe the new evidence and its source, and explain whether and how it changes our previous assessment. If the new evidence contradicts an earlier conclusion we reached, we say so explicitly and explain why the earlier conclusion was reasonable on the evidence available at the time. This approach reflects the principle that analytical integrity means being willing to revise conclusions in light of better evidence, and being transparent about having done so. It also means that an article should not be read as a final and definitive statement. It is a statement of what the best available evidence supported at a particular point in time, with a clear indication of where the evidence was incomplete. Readers checking older articles are encouraged to look at the reviewedAt date and to consider whether events since that date may have changed the factual basis.

Common questions

Before you rely on the answer

Why does OzPolitics sometimes say it cannot verify a claim rather than calling it false?

Because these are different judgments with different implications. A claim is false when authoritative evidence contradicts it. A claim is unverifiable when the evidence needed to assess it is not publicly available, or does not exist, or exists but cannot be accessed on reasonable terms. Labelling an unverifiable claim as false would itself be inaccurate and would mislead readers. By distinguishing between what we have found to be false and what we cannot verify, we give readers a more precise understanding of the state of the evidence and avoid overclaiming.

What does OzPolitics do when official data is not published or is delayed?

We note the gap and its significance. If the Australian Bureau of Statistics has not released a data series that is normally published on a regular schedule, we check and report any official explanation for the delay. If a government program lacks published evaluation data despite having operated for a period when an evaluation would be expected, we report that absence and explain why it matters for assessing claims about the program's effectiveness. We may also note whether we or others have sought the data through freedom of information or other mechanisms, and what response was received.

How does OzPolitics decide which sources are reliable enough to use?

We prioritise sources that have institutional safeguards for accuracy, are publicly accessible so readers can verify our use of them, and are produced by bodies with a statutory or professional obligation of independence and rigour. These include ABS data, legislation from the Federal Register of Legislation, parliamentary records, budget papers, Australian National Audit Office reports, Productivity Commission reports, and Parliamentary Library research. We do not treat political statements, media reports, or advocacy group publications as authoritative sources of factual claims, although we may cite them as evidence of what was said or claimed. Where academic sources are used, we prefer peer-reviewed publications and note any significant competing interpretations.

Can a reader request that OzPolitics investigate a specific claim where evidence is missing?

Yes. OzPolitics welcomes requests from readers who have identified a political claim where the evidence appears to be missing or insufficient. When we receive such a request, we assess whether the claim falls within our scope and whether we have the resources to investigate it. We follow the same methodology described in this article: we identify the claim, specify what evidence would be expected if it were true, search for that evidence in authoritative sources, and report our findings, including what we could and could not verify. We do not undertake investigations that require legal advice, forensic accounting, or access to protected information, but we can assess claims against publicly available evidence using the transparent standards described here.

Source spine

Primary material used for this guide

Review trigger: Changes to OzPolitics editorial policy on evidence standards; significant methodological developments in evidence assessment; changes to the availability or structure of key official data sources; any finding by an external reviewer or auditor that OzPolitics has materially mishandled evidence or uncertainty in a way that requires correction of published articles.

Archive note: This article describes OzPolitics editorial methodology as it operates in mid-2026. It is a descriptive article about our own processes. It should be reviewed if our evidence-handling methodology changes materially, or if new categories of official data sources become available that change our ability to verify claims.

Primary links are provided without affiliate or tracking parameters. Confirm that the source still applies to the bill, sitting date, jurisdiction or reporting period before relying on it.