| QUICK ANSWER |
| Anchoring is a tendency for an initial reference value to influence a later estimate or decision, sometimes disproportionately to its relevance. Experiments show that even arbitrary numerical anchors can affect judgments. Several mechanisms have been proposed, including insufficient adjustment and selective accessibility. The effect is not inevitable. |
Table of Contents
Imagine that someone asks you to estimate a population after mentioning an unrelated number. A higher starting number may influence the estimate compared with a lower one, even when the number is irrelevant. This is an illustration of anchoring, not a guarantee about any individual judgment.
The Research Evidence
Anchoring bias describes how a starting reference can influence an estimate, even when that reference is not informative. Tversky and Kahneman’s (1974) anchoring research used a wheel of fortune rigged to stop at either 10 or 65. Participants then estimated the percentage of African countries in the United Nations. Those who saw 65 estimated higher percentages than those who saw 10, despite knowing the wheel result was random and irrelevant. The absurdity of the influence was the point: even obviously irrelevant numbers distort subsequent numerical judgment.
Selective accessibility is one proposed explanation: considering an anchor may make information consistent with it easier to retrieve. Insufficient adjustment from a starting value is another account. The relative importance of these processes depends on the task.
| RESEARCH NOTE: Anchoring in Professional Judgments |
| Englich, Mussweiler, and Strack (2006) found anchoring in legal sentencing decisions. Experienced legalprofessionals who received randomly determined sentencing requests from dice rolls subsequently producedsentences correlated with the dice value. Prosecutors who rolled higher numbers produced higher sentencingrecommendations. The research established that domain expertise does not eliminate anchoring, directlychallenging assumptions that professional training overcomes systematic biases. |
Anchoring In Negotiations, Pricing, And Salary
In 2001 negotiation experiments, Galinsky and Mussweiler found that opening offers influenced final outcomes, though considering information inconsistent with the initial offer could counter that influence. Opening first does not guarantee an advantage in every negotiation.
Salary discussions and advertised list prices can provide reference points. Independent market data and actual alternatives help determine whether the starting figure is relevant or misleading.
| KEY TAKEAWAY |
| An initial reference can shape a later judgment. Use independent benchmarks, consider contrary evidence and assess the relevance of the starting value. First offers can influence negotiations, but their effects depend on the information and circumstances. |
Frequently Asked Questions
How Can I Protect Against Anchoring In Negotiations?
Research by Galinsky and Mussweiler suggests actively focusing on counter-anchor evidence, specifically on information that supports the value you are pursuing. This selective attention to disconfirming evidence produces better adjustment from the anchor than simply trying to ignore it.
Does Anchoring Apply To Non-Numerical Judgments?
Yes, Research documents anchoring effects on attitude formation, personality judgments, performance evaluations, and probability estimates. The mechanism generalises to any judgment that starts from an initial piece of information.
Should You Make The First Offer In A Salary Negotiation?
Research supports yes, if you have done your research and can make a high but plausible first offer. A well-researched aggressive first offer anchors the range to your advantage. An implausible offer risks rejection and reactance.
Do Experts Show Anchoring?
Research consistently finds anchoring in domain experts including judges, physicians, and financial analysts. The effect may be slightly smaller in experts than novices for domain-specific judgments but persists even in experienced professionals.
How Is Anchoring Different From Framing?
Framing is how the same information presented differently influences decisions, typically through gain-loss framing. Anchoring is specifically about initial numerical or categorical information influencing subsequent estimates. They are related but describe different mechanisms.
Why An Anchor Can Feel Like Evidence
An anchor does not need to be accurate to become psychologically influential. Imagine entering a shop and seeing a jacket labeled $400, now $220. The first figure may make $220 seem attractive even when comparable jackets usually cost $160. The important question is not whether the reduction is mathematically correct, but whether the original price represents a meaningful market comparison. A high reference number can change what feels reasonable before the buyer checks independent evidence. For a related explanation, see decision fatigue.
Anchoring is especially troublesome when a decision has no obvious correct answer. People are asked to estimate a home’s value, a project’s duration or a fair salary without knowing the full distribution of plausible outcomes. In that uncertainty, a number supplied by another person can become a convenient reference point. This is not proof that people are irrational in every purchase. It is a reminder that the order in which information appears can affect judgment.
What The Classic Wheel Experiment Did And Did Not Show
In the classic work by Amos Tversky and Daniel Kahneman, participants first encountered a wheel that produced either 10 or 65, then answered whether the percentage of African countries among United Nations members was higher or lower than that number. Their subsequent estimates differed by condition. The experiment was striking because the wheel’s number had no informational connection to the question. The published paper introduced anchoring and adjustment alongside other judgment heuristics.
The demonstration does not mean every arbitrary number always changes every estimate, or that the first number has an identical effect across people and settings. Experimental estimates can depend on the task, the participant’s knowledge, the plausibility of the anchor and how it is presented. The article’s title uses ‘first number’ as a memorable shorthand, not a universal law.
Two Routes To Anchoring
One account emphasizes adjustment. People begin with a reference point and move away from it, but the movement may stop too soon. If someone hears that a renovation should cost $50,000, they may adjust downward after learning the work is modest, yet remain closer to $50,000 than a fully independent estimate would justify.
Another account emphasizes selective accessibility. When someone considers whether a house is worth $900,000, they may spontaneously retrieve evidence consistent with a high valuation, such as nearby luxury sales, while neglecting less flattering comparables. Both processes have research support in different settings. It is too strong to claim that all anchoring operates through one mechanism or that awareness alone can never help.
Anchoring At Work: Salaries, Deadlines And Budgets
A salary range printed in a job advertisement can establish an expectation before either side discusses skills or local market rates. A candidate may negotiate within that range without asking whether the position is misclassified or the range is outdated. Conversely, a candidate who proposes an unsupported figure may weaken credibility. The useful principle is to establish independent benchmarks, not to throw out the largest possible number. This also connects with car buying psychology.
Project teams face a similar risk. A manager says a redesign will take three weeks, and subsequent planning revolves around defending or adjusting that estimate. An independent breakdown of research, design, testing and revisions may reveal a different schedule. When the first estimate comes from a respected colleague, the social pressure to agree can reinforce the numerical anchor. Written estimates collected separately before a meeting may help reveal disagreement.
Why The First Offer Matters, And When It Does Not
Galinsky and Mussweiler studied first offers in negotiations and found an advantage for the party making the opening offer in their experimental settings. Importantly, they also examined ways for the receiving party to counter the effect by considering information inconsistent with the opening number. This finding is more useful than the simplistic advice to always speak first.
A first offer can be useful when the person making it understands the market and has a defensible target. It can backfire when information is poor, the figure is implausible, or the other side has a credible alternative. Negotiation also involves trust, relationships, fairness and nonfinancial priorities. A single laboratory effect does not supply a universal script for salary conversations or high-stakes commercial contracts.
How Experts Can Still Be Influenced
Englich, Mussweiler and Strack investigated anchoring in judicial sentencing judgments. Their experiments found that irrelevant sentencing demands, including randomly generated numbers, could affect decisions made by participating legal professionals. That is an important warning about judgment under uncertainty, but it does not establish that every judge or every real-world sentence is similarly influenced.
Professional experience can improve knowledge of relevant evidence while leaving some cognitive vulnerabilities intact. The practical response is not to distrust every expert. It is to build procedures that reduce reliance on an unsupported starting figure, such as independent assessments, transparent criteria and documented reasons for decisions. A useful comparison is sunk cost fallacy.
A Practical Method For Resisting An Anchor
Begin by writing down what you would need to know to make a good decision before inspecting the seller’s price or the other party’s offer. For a used car, that might include condition, mileage, service history and comparable transactions. For a salary, it might include role scope, location, seniority and reliable compensation data. Independent information gives your judgment a foundation outside the anchor.
Next, generate at least two plausible alternatives. Ask what evidence supports a materially lower or higher figure. Consider what you would conclude if the opening number had never been mentioned. In negotiations, decide on a target, an acceptable range and a walk-away condition based on your actual alternatives. These are practical safeguards, not guarantees of bias elimination.
Finally, slow down decisions when the stakes are high. Request written terms, compare offers on the same basis and check whether fees or exclusions make a headline price misleading. Where possible, obtain estimates from several independent sources without telling each source the previous estimate. The aim is to replace a single salient number with a defensible range.
What Anchoring Is Not
Anchoring is not the same as the framing effect. Framing concerns how equivalent information is described, such as emphasizing a survival rate rather than a mortality rate. Anchoring concerns the influence of a starting reference on a later judgment. A price comparison can involve both effects, but the concepts are not interchangeable.
Nor is every response to an initial number evidence of bias. If an experienced engineer supplies a credible estimate supported by measurements, taking it seriously may be rational. The problem arises when a number receives more weight than its evidential quality deserves. Distinguishing useful reference information from irrelevant anchors is more valuable than trying to ignore every number encountered.
Further Questions Readers Ask
Can you eliminate anchoring by learning its name? Knowledge may encourage scrutiny, but it is not a reliable universal cure. Specific preparation and independent benchmarks are more actionable than confidence in one’s immunity. For a complementary perspective, consider psychology of spending.
Does anchoring happen online? Digital storefronts can display list prices, countdown offers or comparison tiers that create reference points. Whether a particular presentation is deceptive depends on the facts and applicable rules; the psychological possibility alone does not prove wrongdoing.
Can an anchor be helpful? Yes. A credible reference range can make a difficult estimate more manageable. The question is whether the starting value is relevant, well supported and appropriately weighted.
What should I do after I realize I was anchored? Reconstruct the decision from independent data. Ask someone who has not seen the original figure for an assessment, and compare the outcome with your own revised range. If the decision is consequential, a qualified professional can help assess domain-specific evidence.
Use Independent Evidence
The first number in a conversation is rarely just a number. It can become a mental reference that shapes what seems expensive, reasonable or possible. Yet the effect is neither magical nor inevitable. The strongest protection is not a claim to perfect objectivity, but a disciplined habit of checking independent evidence before treating an opening figure as a fact.
The goal is to notice when an estimate has acquired authority simply because it arrived first. A useful anchor is a starting point for inquiry. An unexamined anchor quietly becomes the answer.
