---
title: Evidence and credibility
url: https://propagandaarchive.org/write/evidence/
description: 'Readers ask two questions of every political piece: is this true? and why should I believe you? Evidence answers the first.'
section: Write
---

![Illustration: a dossier collage of an intelligence briefing room with case files and behavioral-analysis notes on a chalkboard.](https://propagandaarchive.org/static/generated/hero-learn.jpg)

[Write](https://propagandaarchive.org/write/) · Handbook

# Evidence and credibility

On this page

Readers ask two questions of every political piece: *is this true?* and *why should I believe you?* Evidence answers the first. Credibility, Aristotle's *ethos*, answers the second. Honest persuasion earns both, and it holds up when someone checks.

## Do this

1. **Say who you are.** Put your name, and your organisation and funder if you have them, on everything. It's the single biggest difference between advocacy and propaganda ([Fair, grey or deception: where each archive technique sits](https://propagandaarchive.org/write/the-line/)).
2. **One claim, one source.** For every factual claim, have one named source a reader can find. Primary documents (the bill text, the budget, the court record) beat summaries.
3. **Give numbers their denominator.** "300 complaints" means nothing without "out of 40,000 journeys". Give the date and the comparison.
4. **Use real, named endorsers.** A named local doctor or business owner who actually supports you, quoted accurately. Say if they're paid.
5. **Name the strongest objection and answer it.** Don't just ignore the other side's best point.
6. **Admit what you don't know.** A stated limit makes the rest of your case more believable.

## What the research says

**Credible sources persuade more, but the effect can fade.** Hovland & Weiss (1951) found that the same message changed opinions more when attributed to a trusted source. Over weeks, the gap narrowed as people remembered the message but forgot the source: the "sleeper effect" (see [Sleeper effect](https://propagandaarchive.org/techniques/sleeper-effect/) for how that is exploited).

**Two-sided messages that refute the other side persuade best.** O'Keefe's (1999) meta-analysis found that messages which acknowledged opposing arguments *and refuted them* were more persuasive than one-sided messages. Messages that mentioned opposing arguments without refuting them were less persuasive than one-sided messages. So: name the objection, then answer it.

**Social proof works when it's real.** People follow what others like them do ([Social proof](https://propagandaarchive.org/learn/social-proof/), [Conformity](https://propagandaarchive.org/learn/conformity/)). Goldstein, Cialdini & Griskevicius (2008) found hotel guests reused towels more when told most previous guests *in that same room* had done so. True, specific norms ("most families on this street already recycle") persuade. Fake ones are [Bandwagon](https://propagandaarchive.org/techniques/bandwagon/) or [Manufactured consensus](https://propagandaarchive.org/techniques/manufactured-consensus/).

**Authority persuades, even when it shouldn't.** Classic obedience research shows how strongly people defer to perceived authority ([Obedience to authority](https://propagandaarchive.org/learn/obedience-to-authority/)). Use real expertise relevant to the claim, not borrowed prestige.

**Corrections don't fully erase false claims.** Misinformation keeps influencing reasoning after it has been corrected: the continued influence effect (Lewandowsky et al., 2012; [Continued influence effect](https://propagandaarchive.org/learn/continued-influence-effect/)). Get it right the first time; if you get something wrong, correct it prominently.

For how FRAME grades the strength of psychology findings, see [How we rate the evidence](https://propagandaarchive.org/learn/evidence-ratings/).

## Worked examples: real campaigns

- **Ross Perot, independent, US, 1992.** Perot bought half-hour prime-time slots and walked viewers through charts of the federal deficit. Evidence was the message.
- **Al Gore, *An Inconvenient Truth*, US, 2006.** A slide lecture built on temperature and CO2 charts made climate evidence the centrepiece of an advocacy film. In 2007 a UK High Court judge found the film broadly accurate and its four main scientific hypotheses well supported, and ruled it could be shown in schools, but required guidance notes on nine points where it departed from the scientific mainstream (Dimmock v Secretary of State for Education and Skills). The episode shows why each figure needs its source.
- **The "Brexit bus", UK, 2016.** Vote Leave's claim that "We send the EU £350 million a week" used a gross figure without deducting the UK rebate. The UK Statistics Authority said in May 2016 that it was "misleading and undermines trust in official statistics". The episode is a case study in a missing denominator ([Card stacking](https://propagandaarchive.org/techniques/card-stacking/)), and in how a disputed number can still dominate a debate because arguing about it repeats it.
- **"Project Fear", UK, 2014 and 2016.** Opponents used the label for pro-Union and pro-Remain economic warnings. The Treasury's 2016 short-term forecast of an immediate recession after a Leave vote did not come to pass, and critics cited it as a warning stated without its limits. A House of Commons Treasury Committee report in May 2016 criticised figures from both campaigns, including the £350 million claim and the Treasury's own per-household estimates.

## Worked example: you applying it

*Invented case: Riverton's Measure 12, a half-cent sales tax for buses every ten minutes.*

**For**, handling the strongest objection:

> Opponents say shoppers will cross the county line. The price gap on a $40 shop is 20 cents. The council's own study found most Main Street customers live within a mile, which is exactly the area where buses would run every ten minutes. *(Each figure must come from a named document.)*

**Against**, handling the strongest objection:

> Supporters say buses will bring shoppers to Main Street. They might, but the council's study doesn't measure that, and the three routes that pass Main Street already run every 20 minutes. The county grant could improve those routes without a new tax. *(Same rule: every figure from a named document.)*

## Where it crosses the line

- "Experts say" with no named expert or data is [Testimonial](https://propagandaarchive.org/techniques/testimonial/) in the archive's sense.
- Picking only the numbers that help and hiding the one that reverses your case is [Card stacking](https://propagandaarchive.org/techniques/card-stacking/).
- Borrowing prestige that has nothing to do with the claim (a flag, a lab coat, a celebrity) is [Transfer](https://propagandaarchive.org/techniques/transfer/).
- Many seemingly independent voices who all trace back to you is [Manufactured consensus](https://propagandaarchive.org/techniques/manufactured-consensus/) or [Astroturfing](https://propagandaarchive.org/techniques/astroturf/).
- A funder named only in a footer nobody sees is [Disclosure theater](https://propagandaarchive.org/techniques/disclosure-theater/).

## Go deeper

- [Social proof](https://propagandaarchive.org/learn/social-proof/), [Conformity](https://propagandaarchive.org/learn/conformity/), [Base-rate neglect](https://propagandaarchive.org/learn/base-rate-neglect/), [Continued influence effect](https://propagandaarchive.org/learn/continued-influence-effect/), [How we rate the evidence](https://propagandaarchive.org/learn/evidence-ratings/)
- Before you publish: [Before you publish: the fair-play check](https://propagandaarchive.org/write/before-you-publish/)

## Sources

- Aristotle, *Rhetoric*, Book I, ch. 2 (ethos).
- Hovland, C. I. & Weiss, W. (1951). The influence of source credibility on communication effectiveness. *Public Opinion Quarterly*, 15(4), 635–650.
- O'Keefe, D. J. (1999). How to handle opposing arguments in persuasive messages: A meta-analytic review of the effects of one-sided and two-sided messages. *Annals of the International Communication Association* (Communication Yearbook), 22(1), 209–249.
- Goldstein, N. J., Cialdini, R. B. & Griskevicius, V. (2008). A room with a viewpoint: Using social norms to motivate environmental conservation in hotels. *Journal of Consumer Research*, 35(3), 472–482.
- Lewandowsky, S., Ecker, U. K. H., Seifert, C. M., Schwarz, N. & Cook, J. (2012). Misinformation and its correction: Continued influence and successful debiasing. *Psychological Science in the Public Interest*, 13(3), 106–131.
- UK Statistics Authority (2016). Letter from Sir Andrew Dilnot to Norman Lamb MP, 27 May 2016, on the £350 million figure.
- *Dimmock v Secretary of State for Education and Skills* [2007] EWHC 2288 (Admin).
- House of Commons Treasury Committee (2016). *The economic and political arguments for and against UK membership of the European Union*, HC 122, May 2016.
- HM Treasury (2016). *The immediate economic impact of leaving the EU*, May 2016.
- Cialdini, R. B. (2021). *Influence: The Psychology of Persuasion*, new and expanded ed. Harper Business.

## Techniques

[Testimonial](https://propagandaarchive.org/techniques/testimonial/) [Card stacking](https://propagandaarchive.org/techniques/card-stacking/) [Bandwagon](https://propagandaarchive.org/techniques/bandwagon/) [Transfer](https://propagandaarchive.org/techniques/transfer/)

## Related

- [Emotional appeal](https://propagandaarchive.org/write/emotion/)
- [Before you publish: the fair-play check](https://propagandaarchive.org/write/before-you-publish/)
- [Social proof](https://propagandaarchive.org/learn/social-proof/)
- [How we rate the evidence](https://propagandaarchive.org/learn/evidence-ratings/)
- [Obedience to authority](https://propagandaarchive.org/learn/obedience-to-authority/)