Sources checked 30 October 2026. The work and home examples below are illustrations, not results from a study. The percentages come from the published papers listed at the end.
You are in a meeting, pitching a change. You say: "This removes thirty manual steps from the weekly report." Your manager nods politely and moves to the next item. A colleague pitches a nearly identical change a month later: "This gives the team back a morning a week for customer calls." It gets approved before she finishes the slide.
Same change. Same facts. The only thing that moved was the frame: what the sentence asked the listener to look at.
That is the framing effect, and it is one of the best-documented findings in the psychology of decisions. It is also one of the most misused. This article covers what the original study actually found, why frames move us, how big the effect really is, and where the line sits between framing honestly and manipulating.
What did Tversky and Kahneman actually find?
In 1981, the psychologists Amos Tversky and Daniel Kahneman published a short paper in Science with a now-famous problem. [1] Students at Stanford and the University of British Columbia were asked to imagine that the country was preparing for an unusual disease expected to kill 600 people. Two programmes to fight it had been proposed. Before reading on, make your own prediction.
Predict-then-reveal. The revealed figures are the published results of Tversky & Kahneman (1981), from 152 and 155 students.
One group read the options in terms of lives saved:
Programme A: 200 people will be saved.
Programme B: a one-third chance that all 600 are saved, and a two-thirds chance that nobody is saved.
72% of that group (152 people) chose Programme A, the sure thing.
A second group (155 people) read the very same options in terms of lives lost:
Programme C: 400 people will die.
Programme D: a one-third chance that nobody dies, and a two-thirds chance that all 600 die.
Now only 22% chose the sure option. 78% gambled.
Programme A and Programme C are the same plan. Out of 600 people, 200 live and 400 die. Nothing about the outcome changed. Only the description did, and the majority flipped.
Published choice shares from Tversky & Kahneman (1981). The dot grid shows the same outcome (200 live, 400 die) under both wordings.
Why does a frame move us at all?
Tversky and Kahneman's explanation came from their earlier work on prospect theory. [2] People do not judge outcomes in absolute terms. They judge them as gains or losses from a reference point, and the wording sets that point.
"200 will be saved" puts the reference point at "everyone dies", so the 200 feel like a gain. With gains, most people prefer certainty: a bird in the hand. "400 will die" puts the reference point at "everyone lives", so the 400 feel like a loss. With losses, most people will take a gamble to avoid a sure one. Losses also tend to weigh more than gains of the same size, which is why the grim wording stings.
Later researchers showed that frames work in quieter ways too. Irwin Levin and colleagues separated three kinds of framing. [5]
Risky-choice framing: the disease problem. The same gamble described as gains or losses.
Attribute framing: one feature described positively or negatively. In one study, people rated ground beef labelled "75% lean" as better tasting and less greasy than the same beef labelled "25% fat", and the gap shrank, but did not vanish, after they actually tasted it. [4]
Goal framing: the same action described by what you gain by doing it or what you lose by not doing it.
Medicine has its own famous example. In a 1982 study, patients, physicians and graduate students chose between surgery and radiation for lung cancer. When surgery's short-term outcome was described as a 90% survival rate, 18% preferred radiation. When the same outcome was described as a 10% mortality rate, 44% did. [3] Trained doctors were not immune.
Three published examples (Levin & Gaeth 1988; McNeil et al. 1982; Tversky & Kahneman 1981) and one illustrative work example.
How big is the effect, really?
Big enough to matter, smaller than the classic numbers suggest, and very sensitive to details.
A 1998 meta-analysis by Anton Kühberger pooled 136 empirical papers on risky-choice framing. [6] The effect was reliable, but on average small to moderate, and it varied a lot with how the problem was built. Problems that looked like the original disease problem produced the larger effects.
In 2014, the Many Labs project re-ran the disease problem, among other classic studies, in dozens of samples around the world. [7] The framing effect replicated. That puts it in better shape than many famous findings in social psychology.
There is a catch that matters for everyday life. In the original wording, "200 will be saved" quietly leaves out what happens to the other 400. When researchers spelled out both halves ("200 will be saved and 400 will not be saved"), the framing effect often shrank or disappeared. [8] Part of a frame's power comes from what it leaves unsaid.
How do you use a frame at work?
The honest use of framing is to point at the part of the truth that matters most to the listener. Most proposals are described from the presenter's side: what gets removed, cut, changed or replaced. Listeners care about what becomes possible.
"We will cut 30 manual steps from the weekly report" describes effort removed. "The team gets back about a morning a week, and we'll spend it on customer calls" describes what the effort turns into. Same change.
"We'll pause hiring for a quarter" is a loss frame. "We'll protect the current team's jobs through the slow quarter" is the same decision, framed by its purpose. Both are true; say both.
"There's a 5% chance this misses the deadline" and "This lands on time 19 times out of 20" describe one forecast. If you give one, consider giving the other.
A useful opening line: "Here's what this change would make possible." Then name it in concrete terms: hours, customers, decisions, not adjectives.
And at home?
Frames matter even more where nobody is keeping score. "No phones at dinner" is a rule about what gets taken away. "Thirty minutes of real family time every evening" is the same rule, described by what it gives. Children and partners usually respond better to the second, not because it hides anything but because it tells them why.
The same goes for budgets ("we're saving for the trip in June" rather than "we can't afford takeaway"), health ("a walk after dinner" rather than "less screen time") and chores ("Saturday mornings free" rather than "the chore chart").
Where is the line between framing and manipulation?
Every sentence has a frame; there is no neutral wording. So the question is not whether to frame, it is whether your frame helps the listener see the situation or prevents them from seeing it.
A frame turns into manipulation when it hides something the listener would need to make a good decision: a cost, a risk, a deadline, a catch. The research hints at a simple test. Because frames lose much of their pull when both halves are stated, [8] stating both halves is also a good definition of fair play. Lead with the benefit, then say the downside plainly.
A thinking aid based on the article's argument, not a validated test.
So, the one idea
Facts matter. So does the frame. The same truth told as a gain or as a loss can change what people choose, often by a lot. Use that power to point at what matters to the other person, never to hide what they'd want to know. Next time you pitch a change, start with: "Here's what this would make possible." Then tell them what it costs.
Sources
[1] Amos Tversky & Daniel Kahneman, "The framing of decisions and the psychology of choice", Science 211(4481), 453–458 (1981). Hypothetical choices by university students (152 and 155 in the two disease-problem groups). Not real decisions with real stakes.
[2] Daniel Kahneman & Amos Tversky, "Prospect theory: An analysis of decision under risk", Econometrica 47(2), 263–291 (1979). The theory behind the reference-point explanation, built mainly on hypothetical gambles.
[3] Barbara J. McNeil, Stephen G. Pauker, Harold C. Sox & Amos Tversky, "On the elicitation of preferences for alternative therapies", New England Journal of Medicine 306(21), 1259–1262 (1982). Hypothetical treatment choices by patients, physicians and students; percentages are for the whole sample.
[4] Irwin P. Levin & Gary J. Gaeth, "How consumers are affected by the framing of attribute information before and after consuming the product", Journal of Consumer Research 15(3), 374–378 (1988). A small study with students; ratings, not purchases.
[5] Irwin P. Levin, Sandra L. Schneider & Gary J. Gaeth, "All frames are not created equal: A typology and critical analysis of framing effects", Organizational Behavior and Human Decision Processes 76(2), 149–188 (1998). A review that separates risky-choice, attribute and goal framing.
[6] Anton Kühberger, "The influence of framing on risky decisions: A meta-analysis", Organizational Behavior and Human Decision Processes 75(1), 23–55 (1998). Pools 136 papers; the effect is reliable but varies strongly with task design.
[7] Richard A. Klein and many co-authors, "Investigating variation in replicability: A 'Many Labs' replication project", Social Psychology 45(3), 142–152 (2014). Multi-site replication of 13 classic effects, including the disease problem; mostly online and student samples.
[8] Anton Kühberger, "The framing of decisions: A new look at old problems", Organizational Behavior and Human Decision Processes 62(2), 230–240 (1995). Shows that the effect depends partly on incomplete descriptions; when the options are fully spelled out, it shrinks or disappears in these lab tasks.
Watch the episode
One email when we publish. Research, product decisions, and what teams report back.
