AGORA

Talking politics

How do you actually change someone's mind about politics?

By the Agora desk · Published · How we count player splits

Slowly, and mostly by listening. Facts do help, and people rarely dig in when corrected, but facts alone seldom shift deeply held views. What works better is a calm conversation where you ask questions, listen properly, share a personal experience, and make your case in terms of values the other person already holds. Expect small shifts, not conversions.

In short

  • Aim for a small shift, not a conversion: even the best-tested methods move views a little at a time.
  • Corrections rarely backfire: in large experiments, people mostly moved towards the facts.
  • Listening and swapping personal stories changed minds in field studies where arguments alone didn't.
  • Frame your case around their values, not yours.
  • Asking someone to explain how a policy would work can soften their certainty, and yours.

Practise on today's story · 22 November 2026

Lobby Market Winner: Fully release the Epstein files with only victims names redacted

Should the DOJ release all Epstein files with only victims' names redacted?

▶ Play today's story

Read the briefing first

Start with realistic expectations

Political views are tied to identity, friendships and values, so they rarely flip in one conversation. Even the most successful approaches in the research move people a little, not all the way. That's not failure: a small shift that lasts can matter, and it often opens the door to more.

It also helps to be clear about your goal. Are you trying to change their vote, correct one fact, or get them to see why someone might disagree? The last is often the most realistic, and the most useful for the relationship.

Facts still matter, and they rarely backfire

You may have heard of the "backfire effect": the idea that correcting someone makes them believe the false thing more strongly. Political scientists Thomas Wood and Ethan Porter tested this across five experiments with more than 10,000 people and 52 issues, including highly charged ones. They found no correction that triggered backfire. On the whole, people moved towards the accurate information, even when it cut against their side.

The catch is that accepting a fact isn't the same as changing a view. Someone can accept a statistic and still hold their position, because the disagreement was really about values: what matters most, who should decide, what's fair. So get the facts straight, but don't expect them to do all the work. It often helps to agree on the facts first ("Can we agree this is what the numbers say?") and then talk about what should be done.

Listen first: what deep canvassing found

Some of the strongest evidence comes from "deep canvassing", where volunteers have long, two-way conversations on people's doorsteps instead of reciting a script. In three field experiments, political scientists Joshua Kalla and David Broockman studied 230 canvassers talking with 6,869 voters across seven US locations about attitudes towards unauthorised immigrants and transgender people. Conversations that only made arguments had no detectable effect. Conversations that also included a non-judgemental exchange of personal stories shifted views by a modest amount that lasted at least four months.

The technique isn't tied to any one cause. Roughly, the conversation goes like this:

  • Ask where they stand, and why, without arguing: "On a scale of one to ten, where are you on this? What makes you say that?"
  • Ask for a personal connection: "Has this ever touched your life, or someone you know?"
  • Listen, and reflect back what you heard before responding.
  • Share your own story about why it matters to you.
  • Ask again: "Having talked it through, where are you now?" Then let it rest, whatever the answer.

Argue from their values, not yours

Researchers Matthew Feinberg and Robb Willer found across six studies that people tend to argue using their own moral values, which is exactly why their arguments don't land with the other side. Arguments reframed around the values of the person being persuaded were usually more persuasive, on issues from health care to military spending.

In practice, that means asking what the other person cares about (fairness, freedom, security, loyalty, tradition, preventing harm) and connecting your case to it. Two illustrations of our own (not from the study): someone arguing for a bigger welfare budget might talk to a sceptic about rewarding hard work and helping families stand on their own feet. Someone arguing for lower taxes might talk to a sceptic about fairness to working people who keep less of what they earn. Either way, you're not hiding your view; you're translating it.

Ask them to explain, not to justify

"Why do you believe that?" and "How would that work, step by step?" sound similar but have different effects. In a 2013 study by Philip Fernbach and colleagues, people asked to explain how policies such as a national flat tax would actually work realised they understood them less well than they'd thought, and their views became more moderate, across the political spectrum. People asked only to list their reasons didn't soften.

Use it gently, and on yourself too: "I'm not sure how it would work, actually. How do you see it playing out?" works better than a quiz. Personal stories help here as well: a 2021 set of 15 studies found that opponents saw people who shared relevant experiences as more rational and respectable than people who cited statistics.

Things that reliably make it worse

Most failed political conversations fail for the same few reasons. Avoid these and you're ahead of most arguments online.

  • Mocking, sarcasm, or "How can you possibly believe that?"
  • Arguing in front of an audience, where backing down means losing face.
  • Firing off ten points at once instead of exploring one.
  • "You people": treating them as a stand-in for the worst of their side.
  • Demanding they admit they were wrong on the spot. Give them room to think it over and change their mind later, privately.

Common questions

Isn't using these techniques manipulative?

Not if you're honest about your view and open to theirs. Listening, telling true stories and explaining your case in terms the other person cares about are just good communication. It becomes manipulation if you fake interest, invent stories or hide what you're arguing for.

What if they won't budge at all?

That's common, and fine. Views often shift weeks later, after people have had time to think without an audience. If you've listened well, you've also made the next conversation easier, and you've probably learned something about why they think what they do.

Should I just send them an article?

Rarely on its own. An unsolicited link can feel like an attack, especially from an outlet they distrust. If you do share something, choose a source they're likely to respect, say why it made you think, and ask what they make of it.

What if I'm the one who's wrong?

Then the same methods work in reverse. Good persuasion is a two-way conversation, so go in willing to update. If you'd like to practise weighing both cases, Agora's guide to term limits lays out each side in its supporters' own terms. And see how to form your own political opinions.

Sources and further reading

Put it into practice: latest stories

Every story

Written by the Agora desk from the sources above. See our editorial standards. More Learn articles.