Anchoring effect

A respondent sees the question "How much are you willing to pay for the product?" and first reads: "The average market price is $50." That first number becomes an "anchor": subsequent estimates gravitate toward it, even if it's irrelevant. This is the anchoring effect: the first piece of information or number a respondent sees influences all subsequent estimates and judgments. The anchoring effect is one type of bias and a subtype of response bias.

The anchoring effect is especially noticeable in questions involving numbers, prices, ratings, and comparisons. Respondents "anchor" to the first value and adjust their answers relative to it rather than evaluating independently. This can distort the results of surveys about prices, quality, importance, and other numerical estimates.

What the anchoring effect is in plain terms

The anchoring effect (Anchoring Effect) is a cognitive bias in which the first piece of information or number a respondent sees becomes an "anchor" and influences all subsequent estimates and judgments. Respondents "anchor" to the first value and adjust their answers relative to it, even if the anchor is irrelevant or random. The anchoring effect is especially noticeable in questions involving numbers, prices, ratings, and comparisons. It is a subtype of response bias related to the order in which information is presented.

Put simply: the anchoring effect is when the first number or piece of information "gets stuck" in your head and pulls all subsequent estimates toward it. If a high price is mentioned at the start of a survey, all subsequent cost estimates will be shifted upward; if a low one, downward.

Why the anchoring effect occurs

The anchoring-and-adjustment heuristic. Respondents use a heuristic: they first "anchor" to the first value (the anchor), then adjust their estimate, but the adjustment is usually insufficient. The anchor stays in the mind and influences the final answer.

Insufficient adjustment. Even when respondents understand that the anchor is irrelevant, they don't adjust their estimate enough. The anchor continues to exert influence because it has already "lodged" in working memory.

Uncertainty. In questions where respondents are unsure of the answer (a product's price, the importance of a feature, a quality rating), the anchor becomes a reference point for the estimate. The greater the uncertainty, the stronger the anchor's influence.

Order of presentation. The first piece of information gets more attention and is remembered better. If the anchor appears at the start of a question or at the start of a survey, its influence is stronger.

Context and comparison. The anchor creates a context for comparison. Respondents evaluate not in absolute terms but relative to the anchor: "higher" or "lower" than the first value.

When the anchoring effect is stronger

Questions with numbers and prices. "How much are you willing to pay?", "How would you rate the importance on a scale of 1-10?" — questions with numerical answers are especially prone to the anchoring effect. The first number becomes a reference point.

Uncertain topics. In questions where respondents have no clear opinion or experience, the anchor becomes the only reference point. For example, estimating the value of a new product or the importance of an unfamiliar feature.

An anchor at the start of a question. If the anchor appears within the question itself ("The average price is $50. How much are you willing to pay?"), its influence is stronger than if it is mentioned separately or at the end.

Extreme anchors. Very high or very low anchors influence estimates more strongly than moderate ones. Respondents "anchor" to the extreme value and adjust relative to it.

The first question in a survey. If the anchor appears in the very first question of a survey, its influence carries over to subsequent questions, creating an overall context for evaluation.

Examples of the anchoring effect

Estimating a product's price. The question "The average market price is $50. How much are you willing to pay for our product?" — the $50 anchor shifts estimates upward. If the anchor is $20, estimates shift downward.

Estimating the importance of features. The question "The importance of product quality is 9 out of 10. Rate the importance of price" — the "9" anchor shifts the price rating upward. Respondents compare it with the first value rather than evaluating independently.

Estimating service quality. If "excellent service" or "high quality" is mentioned at the start of a survey, subsequent ratings shift upward. The anchor creates a positive context.

Comparing options. The question "Option A costs $10. How much are you willing to pay for Option B?" — the $10 anchor influences the rating of Option B, even if they are incomparable.

How to minimize the anchoring effect

Avoid anchors in questions. Don't mention numbers, prices, or ratings within the question itself if they could become an anchor. Instead of "The average price is $50. How much are you willing to pay?" — simply "How much are you willing to pay for the product?".

Randomizing the order of questions. Use randomization of question order so that different respondents see questions in a different order. This reduces the influence of an anchor from the first question on subsequent ones.

Neutral wording. Avoid leading questions and mentions of specific values that could become anchors. Neutral wording without numbers and ratings reduces the anchoring effect.

Independent questions. Word questions so they are independent and don't refer to previous answers or values. Each question should be evaluated separately.

Piloting and testing. In a pilot study, check whether the wording or order of questions creates unwanted anchors. Reword or reorder if necessary.

Splitting into blocks. Divide the survey into thematic blocks with breaks or instructions between them. This helps "reset" the anchor from the previous block.

Relationship to other effects

The anchoring effect is related to other order and context effects:

• The primacy effect. The first piece of information is remembered better and influences subsequent estimates — this overlaps with the anchoring effect.
• The recency effect. The last piece of information can also become an anchor for evaluating previous items.
• Leading questions. Leading questions often contain anchors that push toward a particular answer.

It's important to distinguish these effects and apply the appropriate minimization measures. For more on the influence of order and context, see the article on response bias.

Common mistakes

Mentioning numbers in questions. Including average values, prices, or ratings directly in a question without realizing they will become an anchor and shift answers.

Ignoring question order. Failing to account for the fact that the first question with a number can become an anchor for all subsequent estimates.

Referring to previous answers. Wording questions so they refer to previous values ("Quality importance is 9. Rate the price"), creating anchors.

Not testing the wording. Not running a pilot to check whether the wording creates unwanted anchors.

How this looks in SurveyNinja

In SurveyNinja you can use randomization of question order to reduce the influence of anchors from the first question on subsequent ones. You can set up neutral wording without mentioning specific numbers or ratings. Logic jumps help split the survey into blocks and "reset" anchors between topics. When creating surveys, you can use ready-made survey templates that incorporate best practices for minimizing the anchoring effect.

Practical recommendations

Avoid numbers in questions. Don't mention average values, prices, or ratings within the question itself if they could become an anchor. Use neutral wording.

Randomize the order. Use randomization of question order, especially if the first questions contain numbers or ratings that could become an anchor.

Test the wording. In a pilot, check whether the wording or order of questions creates unwanted anchors. Reword if necessary.

Split into blocks. Divide the survey into thematic blocks with breaks or instructions between them to "reset" anchors from previous blocks.

Account for it in interpretation. When analyzing results, account for the possible influence of the anchoring effect, especially if the survey mentioned numbers or ratings in the questions.

What to write in the report. If randomization of question order was used, you can briefly note in the methodology: "To minimize the anchoring effect, question order was randomized for each respondent".

The anchoring effect is the influence of the first piece of information or number on all subsequent estimates. It is minimized by avoiding anchors in questions, randomizing question order, using neutral wording, splitting into blocks, and piloting to check the wording.