Within Statistics
Why Better Than Chance Is Not Proof
A statistically unusual remote-viewing result can flag an anomaly without proving a psychic mechanism or usable hidden information.
On this page
- What a low p value can and cannot say
- Why extraordinary claims need more than significance
- How remote viewing claims can overread statistics
Page outline Jump by section
Introduction
A low p-value in a remote-viewing experiment can indicate that the observed results would be unlikely if the study’s specific chance model were correct. It does not demonstrate that psychic ability exists, identify a paranormal mechanism, or prove that the information obtained is accurate enough to be useful in practice. This distinction lies at the heart of the debate over remote viewing. Supporters argue that repeated above-chance findings deserve serious scientific attention, while critics argue that statistical significance alone cannot distinguish a genuine psychic effect from methodological artefacts, bias, or an incomplete understanding of how the data were produced.[CIA]cia.govAN EVALUATION OF THE REMOTE VIEWING PROGRAMUtts and Hyman were asked to prepare independent reports based on their review. In this rev…
What a low p-value can and cannot say
A p-value answers a narrow statistical question. It estimates how unusual the observed data would be if the null hypothesis—typically that performance is no better than chance under the study’s design—were true. It is not the probability that psychic ability exists, nor the probability that the study’s conclusion is correct.
The American Statistical Association has emphasised that p-values do not measure the size of an effect, the importance of a finding, or the truth of a scientific hypothesis. Statistical significance is only one piece of evidence and should not be treated as a substitute for scientific judgement. Those general cautions apply especially strongly to claims about remote viewing because the proposed mechanism remains unestablished within mainstream science.[CIA]cia.govAN EVALUATION OF THE REMOTE VIEWING PROGRAMUtts and Hyman were asked to prepare independent reports based on their review. In this rev…
In remote-viewing research, an experiment might produce a statistically significant result because participants collectively ranked correct targets above chance. That finding means the data departed from the chosen chance model under the experimental conditions. It does not identify why the departure occurred.
Several explanations remain possible, including:
- an unidentified psychological or sensory influence;
- subtle methodological weaknesses;
- statistical or analytical choices that unintentionally favour positive results;
- publication bias across a research field;
- or, as proponents argue, a genuine anomalous information process.
A low p-value alone cannot distinguish among these competing explanations.
Why extraordinary claims need more than significance
Remote viewing proposes that people can obtain information without known sensory channels. Because this would challenge well-established scientific understanding, researchers generally expect evidence stronger than a single statistically significant experiment.
In most areas of science, confidence grows through several independent lines of evidence rather than one statistical result. These include successful independent replication, transparent methodology, pre-registered analyses, agreement across different laboratories, plausible theoretical explanations, and practical predictive performance.
The debate surrounding the 1995 review of the U.S. government’s remote-viewing programme illustrates this point. Statistician Jessica Utts argued that the accumulated statistical evidence supported the existence of an anomalous effect beyond chance. Psychologist Ray Hyman accepted that some experimental findings appeared statistically unusual but argued that this did not justify concluding that psychic functioning had been demonstrated, particularly without stronger independent replication and a convincing theoretical framework.[CIA]cia.govAN EVALUATION OF THE REMOTE VIEWING PROGRAMUtts and Hyman were asked to prepare independent reports based on their review. In this rev…
Importantly, these experts were not disagreeing over the arithmetic of p-values alone. They differed over what level and type of evidence should be required before accepting an extraordinary scientific claim.
How remote-viewing claims can overread statistics
A common mistake is to treat “better than chance” as equivalent to “psychic ability proven”. These are different claims.
Several statistical realities help explain why:
- Small effects can become statistically significant. If enough trials are collected, even modest deviations from chance may produce very small p-values. Statistical significance therefore says little about whether the effect is large enough to matter in practice.[UC Irvine Bren School]ics.uci.eduUtts Stat PsiProfessor Utts find extremely small p-values, but the estimate of the overall effect size is still small. As…Read more…
- Significance does not guarantee reliability. A viewer who performs slightly above chance across hundreds of trials may still produce individual sessions that are vague, inconsistent, or unusable for real-world decision-making.[CIA]cia.govAN EVALUATION OF THE REMOTE VIEWING PROGRAMUtts and Hyman were asked to prepare independent reports based on their review. In this rev…
- Different analyses can produce different impressions. Choices about scoring, judging procedures, exclusion criteria, or handling outliers can influence reported significance, making transparent protocols and independent replication especially important.
- Replication matters more than isolated findings. A result that repeatedly appears under independently controlled conditions carries more scientific weight than a single experiment reporting a very small p-value.
For these reasons, scientists increasingly evaluate evidence using combinations of effect size, confidence intervals, replication history, study quality and prior plausibility rather than relying on statistical significance alone.
Why practical proof is a higher standard
Even if a statistical anomaly were consistently demonstrated, another question would remain: can remote viewing produce information that is dependable enough for practical use?
The review commissioned after the U.S. government’s Stargate programme separated these questions. Some reviewers acknowledged statistically interesting laboratory findings while concluding that the available evidence did not demonstrate reliable operational value for intelligence work. Reports were often considered too vague, inconsistent or difficult to interpret for dependable decision-making.[CIA]cia.govAN EVALUATION OF THE REMOTE VIEWING PROGRAMUtts and Hyman were asked to prepare independent reports based on their review. In this rev…
This distinction is easy to overlook. A phenomenon may produce measurable statistical departures from chance without providing predictions or information accurate enough to support real-world decisions.
The key misunderstanding to avoid
The strongest interpretation supported by a low p-value is that the observed data appear inconsistent with a specified chance model under the conditions of a particular experiment. It is not proof that psychic perception exists.
To move from statistical anomaly to scientific acceptance would require evidence that survives independent replication, rules out conventional explanations, demonstrates reliable predictive performance, and fits within a broader explanatory framework. Until those standards are met, low p-values in remote-viewing studies remain evidence that is open to competing interpretations rather than definitive proof of psychic ability.[CIA]cia.govAN EVALUATION OF THE REMOTE VIEWING PROGRAMUtts and Hyman were asked to prepare independent reports based on their review. In this rev…
Amazon book picks
Further Reading
Books and field guides related to Why Better Than Chance Is Not Proof. Use these as the next step if you want deeper reading beyond the article.
Thinking, Fast and Slow
Explains cognitive errors that affect evidence interpretation.
Statistics Done Wrong
First published 2015. Subjects: Missing observations (Statistics), Methodology, Statistics.
Naked Statistics
First published 2013. Subjects: Statistics, Statistics as Topic, MATHEMATICS/Statistics/, Statistique, M ethodes statistiques.
eBay marketplace picks
Marketplace Samples
Live-tested eBay searches with available results related to this page.
Endnotes
1.
Source: cia.gov
Link:https://www.cia.gov/readingroom/docs/CIA-RDP96-00791R000200180005-5.pdf
Source snippet
AN EVALUATION OF THE REMOTE VIEWING PROGRAMUtts and Hyman were asked to prepare independent reports based on their review. In this rev...
2.
Source: cia.gov
Link:https://www.cia.gov/readingroom/docs/CIA-RDP96-00791R000200180006-4.pdf
3.
Source: ics.uci.edu
Title: Utts Stat Psi
Link:https://www.ics.uci.edu/~jutts/UttsStatPsi.pdf
Source snippet
Professor Utts find extremely small p-values, but the estimate of the overall effect size is still small. As...Read more...
Additional References
4.
Source: researchgate.net
Link:https://www.researchgate.net/publication/333228024_An_Assessment_of_the_Evidence_for_Psychic_Functioning
5.
Source: reddit.com
Link:https://www.reddit.com/r/remoteviewing/comments/1dazs09/creation_of_study_on_statistical_evidence_of/
Source snippet
Neither were granted clearance, so 98+% of Project Stargate's material was off...Read more...
6.
Source: ucdavis.edu
Title: [psychic spying]({{ ‘psychic-spying/’ | relative_url }}) research produces credible evidence
Link:https://www.ucdavis.edu/news/psychic-spying-research-produces-credible-evidence
Source snippet
'Psychic Spying' Research Produces Credible Evidence28 Nov 1995 — The main psychic ability tested in the research program is called "remo...
7.
Source: irp.cdn-website.com
Title: The+Seventh+Sense+ The+Secrets+of+Remote+Viewing
Link:https://irp.cdn-website.com/6b820530/files/uploaded/The%2BSeventh%2BSense%2B-The%2BSecrets%2Bof%2BRemote%2BViewing.pdf
Source snippet
Seventh Sense - The Secrets of Remote ViewingThis story concerned remote viewing, the wondrous ability to view persons, places, and thing...
8.
Source: youtube.com
Title: The FPR50: a simple, but rough, solution to the p values war (?)
Link:http://www.youtube.com/watch?v=Kp-jpQCLL1o
Source snippet
"Statistical significance" "p-value" "pseudoscience" Significance (The Body of Evidence) The Body of Evidence Podcast...
9.
Source: youtube.com
Link:http://www.youtube.com/watch?v=MVHoE9V_X5g
Source snippet
The limitations to the p-value - David Colquhoun...
10.
Source: youtube.com
Title: Better Science
Link:http://www.youtube.com/watch?v=gMYRuihe9KU
Source snippet
The FPR50: a simple, but rough, solution to the p values war (?)...
11.
Source: pmc.ncbi.nlm.nih.gov
Link:https://pmc.ncbi.nlm.nih.gov/articles/PMC10275521/
Source snippet
Central Intelligence Agency's (CIA...by Á Escolà‐Gascón · 2023 · Cited by 10 — Since 1972, the U.S. Central Intelligence Agency (CIA) co...
12.
Source: youtube.com
Title: The limitations to the p-value
Link:http://www.youtube.com/watch?v=-FHIr_LQJfI
Source snippet
Better Science - Fisher's tests of significance III (misinterpretations)...
13.
Source: youtube.com
Title: Significance (The Body of Evidence)
Link:http://www.youtube.com/watch?v=wlKWwCoA3PU
Source snippet
Deborah Mayo | [Statistics]({{ 'statistics/' | relative_url }}) & Severe Testing vs Pseudoscience | Philosophy of Data Science...
Topic Tree