中文

Cognitive biases · Judgement

Anchoring effect

The first number we see tends to drag our later estimates toward it.

What it is

Anchoring is the tendency for a number we have just seen to pull our estimates toward it. We start from that “anchor” and adjust, but usually not far enough. The anchor can be useful information, like an asking price, or completely irrelevant, like a random number.

An everyday example

Maya is selling a used bike. A friend asks how much she wants, and before she answers, the friend jokes, “Surely not 900?” Maya had been thinking of about 250. Somehow she ends up listing it at 400, and the first buyer offers 350, which now feels low.

Nothing about the bike changed. The joking number simply became the starting point for everyone’s thinking.

Why it happens

When we do not know an answer, we need somewhere to begin. A number that is already in mind is a convenient starting point. From there we adjust, but the adjustment is typically too small, so the final answer stays closer to the anchor than it should (Tversky & Kahneman, 1974).

How to counter it

  • Make your own estimate first. Write down a number before you look at someone else’s offer, quote or forecast.
  • Argue the other side. Ask yourself why the true value might be far from the anchor, in either direction.
  • Use outside reference points. Compare several independent prices or base figures instead of reacting to one.
  • Notice who set the anchor. If you are negotiating, consider putting a well-researched number on the table first.

What the research shows

In a classic demonstration, people watched a wheel of fortune stop at a number, then estimated the percentage of African countries in the United Nations. Those who saw a higher number gave higher estimates (Tversky & Kahneman, 1974).

Basic anchoring with numerical questions has held up well. It replicated across many labs and countries in a large coordinated project (Klein et al., 2014).

Some variants are less solid. Studies claiming that arbitrary anchors, such as a randomly chosen price, strongly change how people value goods or experiences have not always held up: one replication of a well-known study of this kind found a much weaker effect than the original (Maniadis et al., 2014). A fair summary: relevant-seeming anchors reliably sway judgment, while the size of effects from truly random numbers is still debated.

Related biases

  • Framing effect — The same facts can lead to different choices depending on how they are worded.
  • Overconfidence — We often think we know more, or do better, than we actually do.
  • Availability heuristic — We judge how likely something is by how easily examples spring to mind.

Sources

  1. Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124–1131. https://doi.org/10.1126/science.185.4157.1124
  2. Klein, R. A., Ratliff, K. A., Vianello, M., et al. (2014). Investigating variation in replicability: A "Many Labs" replication project. Social Psychology, 45(3), 142–152. https://doi.org/10.1027/1864-9335/a000178
  3. Maniadis, Z., Tufano, F., & List, J. A. (2014). One swallow doesn't make a summer: New evidence on anchoring effects. American Economic Review, 104(1), 277–290. https://doi.org/10.1257/aer.104.1.277

All biases