Psychology research · Reference
What is anchoring bias?
Anchoring bias is the tendency to rely too heavily on the first piece of information encountered — the anchor — when making estimates or decisions, so that later judgements stay too close to that initial value.
Definition
Anchoring describes how an initial reference point distorts numerical and evaluative judgements. People appear to start from the anchor and then adjust, but the adjustment is usually insufficient, leaving the answer pulled towards the anchor even when it is arbitrary or irrelevant. Tversky and Kahneman demonstrated the effect with a now-classic experiment in which participants spun a wheel of fortune rigged to land on 10 or 65, then estimated the percentage of African nations in the United Nations. Those shown the higher number gave markedly higher estimates, despite the spin being obviously random.
How it works
Two mechanisms are usually proposed. In anchoring-and-adjustment, people consciously move away from the anchor but stop too soon. In a second account, the anchor selectively brings to mind information consistent with it, biasing the judgement without deliberate adjustment.
The effect is robust and surprisingly hard to avoid. It influences both novices and experts, persists when people are warned about it, and operates even when the anchor is plainly uninformative. This resistance is one reason anchoring is considered a fundamental feature of how estimates are formed.
Examples and research relevance
Anchoring shapes everyday judgements: an initial asking price frames negotiation, a first salary offer frames later pay, and a suggested donation amount frames giving. In research, anchoring can bias survey responses and expert estimates — for instance, the wording or numerical examples in a question can pull respondents' answers towards them. This is a key reason questionnaire design pays close attention to the order and framing of items and to providing neutral, non-leading reference points.
Significance for methods
Awareness of anchoring informs the design of surveys, elicitation procedures, and decision processes. Mitigations include avoiding arbitrary starting figures, generating estimates independently before comparing them, considering reasons the anchor might be wrong, and using multiple independent judges. In measurement and forecasting, recognising anchoring helps keep instruments and protocols from inadvertently steering the very responses they aim to record.
Key facts
At a glance
- Type: cognitive bias in estimation and judgement
- Anchor: an initial value that biases later judgements
- Identified by: Amos Tversky and Daniel Kahneman
- Classic study: the rigged wheel-of-fortune UN estimate
- Mechanism: insufficient adjustment away from the anchor
- Notably robust: persists even with irrelevant anchors and warnings
Common questions
FAQ
What is an example of anchoring bias?+
In negotiation, the first price mentioned often anchors the discussion: a high opening offer tends to lead to a higher final price, and a low one to a lower price, because both parties adjust from that starting figure rather than judging value independently.
Who discovered anchoring bias?+
Anchoring was identified by Amos Tversky and Daniel Kahneman as one of the heuristics in their work on judgement under uncertainty, famously demonstrated with a rigged wheel-of-fortune experiment described in their 1974 Science paper.
Can anchoring bias be avoided?+
It is difficult to eliminate but can be reduced. Useful strategies include making estimates independently before seeing others' figures, deliberately considering why the anchor might be wrong, and avoiding arbitrary reference numbers in survey and decision processes.
The step most authors miss
Doing CRediT right? Don’t stop at the statement.
A CRediT statement credits you inside one paper. The recognition CRediT was built for happens when those roles are tied to you, persistently. Sign in with your ORCID — free — and claim your CRediT contributions on casrai.org, the home of the standard. They become a verified, portable part of your identity, not a line that disappears into one PDF.
Free: claim your contributions, then export a journal-ready CRediT statement, schema.org structured data, JATS XML, CSV or BibTeX — and preview your public profile. A membership publishes that profile publicly and verifies the journals you serve.
