Anchoring is a well-known phenomenon [1]. However, I just ran an experiment where the effect seemingly backfired. Basically, I asked people to estimate the number of jelly beans in a jar. In my anchor condition, people were given an estimate produce by the Jelly Bean Counter 5000 computer system which was purported to be highly accurate, then they were allowed to see the jar. They were then asked to estimate the number of jelly beans.

People who were not anchored were allowed to see the jar of jelly beans first, then estimate the count. Then, they were showed Jelly Bean Counter 5000's estimate. They were then allowed to re-estimate the number of jelly beans in the jar.

In all cases, Jelly Bean Counter 5000 provided the exact number of jelly beans in the jar.

The results are shown in the figure below. Basically, people who were anchored with the Jelly Bean Counter 5000 number were about 62% accurate. People who made their own estimate first (self-anchoring) were initially about 46% accurate, but after seeing Jelly Bean Counter 5000's estimate, they increased to about 72% accurate.

I'm at a loss to explain why people exposed to an accurate estimate (Jelly Bean Counter 5000) ended up being less accurate than people who formed their own inaccurate opinion first. Can anybody provide any alternative explanations? Are there any theories that might be applicable?

[1] Tversky, A., & Kahneman, D. (1974). Judgment under Uncertainty: Heuristics and Biases. Science, 185(4157), 1124-1131.

  • 3
    $\begingroup$ I am very unclear about how the percentage accuracy rate statistic that you report is being computed. Is each estimate classified as either "accurate" or "inaccurate" and the accurate rate is the proportion of "accurate" estimates across people? What are the criteria for classifying estimates as accurate or inaccurate? Or is this "average accuracy rate" constructed in some entirely different way? If so, how? $\endgroup$ May 17, 2013 at 23:38
  • $\begingroup$ @JakeWestfall In my actual study, the decision was dichotomous, so people were either 100% right or 100% wrong for each judgment. Each individual's accuracy rate is the proportion of accurate responses over several judgments. So, if you got 3 out of 4 right, you had a 75% accuracy rate. Then, I compared the mean accuracy rates between the two groups. $\endgroup$
    – Jim
    May 20, 2013 at 22:03

1 Answer 1


I think there is a significant confound in your experimental design: The anchored subjects only make 1 estimate whereas the "no anchor" subjects make 2 estimates. Note that on the first estimate the anchored subjects were much closer to the anchor than the non anchored ones, which is entirely consistent with Tversky's heuristic.

Also, "In all cases, Jelly Bean Counter 5000 provided the exact number of jelly beans in the jar." means that this isn't really a standard anchoring task. Normally, anchoring values aren't correct (in fact much of the anchoring work involves very incorrect anchors). This task confounds being close to the anchor with being close to the correct answer.

You might have an interesting finding: more anchoring after a non-anchored guess, but without correcting for the 2-guess and perfect anchor confounds your results can't be properly interpreted.

  • $\begingroup$ True, it is not a classic anchoring scenario. $\endgroup$
    – Jim
    May 20, 2013 at 21:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.