Before you write a word, decide who you are trying to move and what you want them to do. Most political writing fails because it speaks to people who already agree, or to "everyone", which means no one in particular.
Do this
- Name one audience. Not "voters" but "renters under 35 in the east wards who voted last time". If you have two audiences, write two pieces.
- Place them on the ladder. Supporters need a task. Undecided people need a reason. Opponents rarely switch in one reading; aim to make them doubt one assumption, not to convert them.
- Find what they already value. Write down two or three things this audience cares about in their own words. Build the argument from those values, not yours (see "moral reframing" below).
- Pick one action. Vote, sign, attend, share, call. One piece, one action.
- Check how much attention they'll give you. A flyer in a letterbox gets seconds; an op-ed reader has chosen to spend minutes. Match length and evidence to that.
What the research says
People process persuasion in two modes. The Elaboration Likelihood Model (Petty & Cacioppo, 1986) finds that motivated, able readers weigh the quality of arguments, and attitude change from that route lasts longer. Distracted or uninvolved readers lean on cues such as who is speaking, how many agree, and how confident the message sounds. Write for the mode your audience will actually be in.
People defend what they already believe. Motivated reasoning (Kunda, 1990) means people scrutinise unwelcome evidence harder than welcome evidence. See Motivated reasoning and Confirmation bias. Opening with a direct attack on someone's identity or side usually hardens them.
Different groups weight different moral values. Moral Foundations Theory (Graham, Haidt & Nosek, 2009) found that US liberals and conservatives weight care, fairness, loyalty, authority and purity differently. See Moral foundations. Feinberg & Willer (2015) found that arguments reframed into the audience's moral values were more persuasive than arguments in the writer's own values, in both directions: liberal positions reframed in terms of loyalty and purity moved conservatives, and conservative positions reframed in terms of fairness and care moved liberals.
Group identity shapes what feels credible. Messages from someone seen as "one of us" get a hearing that outsiders don't (Social identity, In-group favoritism). This is legitimate when the speaker really is part of the group, and becomes Plain folks or Identity fusion when it's a costume or a gag on questions.
Expect small effects. A meta-analysis of 49 field experiments by Kalla & Broockman (2018) found the average persuasive effect of campaign contact and advertising in US general elections was close to zero, with larger effects when campaigns contacted persuadable voters early or on issues where the parties took unusual positions. In that US evidence, honest persuasion is slow. A single piece rarely flips anyone; it moves a few people a little, and it mobilises people who already agree.
Worked example: a real campaign
Deep canvassing (US, 2015 onwards). Broockman & Kalla (2016) tested a canvassing method developed by the Los Angeles LGBT Center, in which canvassers asked voters about their own experiences and exchanged personal stories instead of reciting arguments. Ten-minute conversations produced attitude change that was still measurable three months later. Later studies by the same authors applied it to attitudes toward unauthorised immigrants (Kalla & Broockman, 2020). Both published field tests measured attitudes that progressive campaigns were trying to shift. The method itself (listen first, exchange stories) doesn't depend on the position being argued, but FRAME has not yet found a published field test of it for a conservative cause. What made it work was listening first and letting people reach a conclusion from their own values.
Worked example: you applying it
Invented case: Riverton's Measure 12 would add a half-cent sales tax to fund buses every ten minutes on six routes. Here is one audience, written up both for and against.
| For Measure 12 | Against Measure 12 | |
|---|---|---|
| Audience | Parents in the north wards who drive children to school. | Small shop owners on Main Street. |
| Where they are | Undecided. Like the idea of less driving, worried about cost. | Leaning no. Worried about customers shopping out of town. |
| Their values, in their words | "Time with my kids." "Not being a taxi service." | "Keeping the doors open." "Being treated fairly by the council." |
| One action | Vote yes. | Come to the council meeting on the 14th. |
| Attention | A flyer at the school gate: 10 seconds. | A letter in the Chamber newsletter: 3 minutes. |
Neither column says the other side is stupid. Each starts from what its audience already cares about.
Where it crosses the line
- Targeting people with messages they can't see were aimed at them, from a source that hides who it is, is Astroturfing or Narrative laundering, not audience research.
- Implying that anyone in the group who disagrees is a traitor is Identity fusion.
- Inventing support to create pressure is Bandwagon with fake numbers, or Manufactured consensus.
Go deeper
- Motivated reasoning, Moral foundations, Social identity, Group polarization
- Next chapter: Framing
Sources
- Petty, R. E. & Cacioppo, J. T. (1986). The Elaboration Likelihood Model of persuasion. Advances in Experimental Social Psychology, 19, 123–205.
- Kunda, Z. (1990). The case for motivated reasoning. Psychological Bulletin, 108(3), 480–498.
- Graham, J., Haidt, J. & Nosek, B. A. (2009). Liberals and conservatives rely on different sets of moral foundations. Journal of Personality and Social Psychology, 96(5), 1029–1046.
- Feinberg, M. & Willer, R. (2015). From gulf to bridge: When do moral arguments facilitate political influence? Personality and Social Psychology Bulletin, 41(12), 1665–1681.
- Kalla, J. L. & Broockman, D. E. (2018). The minimal persuasive effects of campaign contact in general elections: Evidence from 49 field experiments. American Political Science Review, 112(1), 148–166.
- Broockman, D. & Kalla, J. (2016). Durably reducing transphobia: A field experiment on door-to-door canvassing. Science, 352(6282), 220–224.
- Kalla, J. L. & Broockman, D. E. (2020). Reducing exclusionary attitudes through interpersonal conversation: Evidence from three field experiments. American Political Science Review, 114(2), 410–425.