Anchoring Bias Results Writeup

Anchoring Results & Write Up
Decision Theory Project

Nolan McMillen
Data Collection
I collected my data by having students complete surveys that I had created for this project. I had two different
surveys each with the same questions and style. The only difference between the two surveys is that one
survey included low anchors while the other survey included high anchors. Survey A1 contained the low
anchors while Survey A2 contained the high anchors. Before asking the individual to give their own estimate to
each question, I first asked whether they thought the estimate would be lower or higher than one of my
anchors. Then I asked for the student’s estimate. For example, the first questions in Survey A1 were…
1A. Do you believe the annual rainfall in the Amazon Rainforest exceeds or is less than 60 inches?
(Student’s Answer choice of higher or lower)
1B. What is your estimate of the annual rainfall in the Amazon Rainforest in inches?
(Student’s estimate)
I had a total of 26 students that took Survey A1 and a total of 22 students that took Survey A2. These surveys
were linked to my Google Drive in which the answers were automatically recorded and organized within
Google Sheets. The list of questions and anchors presented to the students can be found below.
Question A1 Anchor A2 Anchor

What's your estimate of the annual rainfall in the Amazon 60 inches 200 inches
Rainforest in inches?
What is your estimate of Taylor Swift’s net worth? $400 million $900 million
What's your estimate of the height of the Eiffel Tower in feet? 857 feet 1,200 feet
How long, on average, do you think a lion lives? 10 years 30 years
What do you think the actual average that students pay for tuition $8,000 $23,000
annually at Mars Hill University is?
Process
I began by putting all my data collected into an excel sheet. I had two different sets of columns within this
sheet. One set of columns represented answers to each question from Question Set 1 (the low anchor
question set). The other set of columns represented answers to each question from Question Set 2 (the high
anchor question set). I then used this data to find my anchoring index for each question.
Finding the Anchoring Index

1. Find the averages of each column
2. Find the difference between Anchors from Set 1 and Set 2
3. Find the differences between the averages from Set 1 and Set 2
4. Divide the difference between the average by the difference between anchors for each column (or
question set)
Results from Anchoring
Q1B Q2B Q3B Q4B Q5B

A1 87.3461538 846153846 838.961538 15.0384615 13634.6154
A2 313.909091 758074582 1511.18182 23.7727273 23159.0909
Differences Between Anchors 140 500,000,000 343 20 15,000
Difference Between Mean Estimates 226.562937 -88079263.8 672.22028 8.73426573 9524.47552
Anchoring Index 162% -18% 196% 44% 63%
After analyzing my results, I realized that I had some odd percentages. I went back into my data and discovered
that I did have some outliers that could be swaying my results. I then went back into Excel, found my outliers
for each column, and got rid of them in a new sheet. I then did the anchoring index process over again and
came up with the results below.
Results from Anchoring (Excluding Outliers)
120 100000000 850 12 12000

100 500000000 976 17 15000
115 900000000 700 8 5000
100 600000000 984 12 15000
48 150000000 1100 14 15000
76 700000000 900 20 15000
35 350000000 900 8 12000
Anchoring Index According to Kahneman

The anchoring index aims to measure the effect that an anchor has on one’s ability to estimate a number. An
anchoring effect of 100% would mean that individuals simply adopted the anchor as their own answer
whereas an anchor of 0% would mean that the individuals were able to completely ignore the anchor
altogether. In Kahneman’s own research, he had multiple studies that showed somewhere around a 50%
anchoring effect. To him, this was a significant result. In another study, the effects of 41% and 48% were also
considered to be significant. In using these significant numbers, I analyzed my anchoring indexes. Also, I will be
focusing on the results that excluded outliers as I feel this makes the data a bit more reliable. Some of the
students with extreme answers clearly threw off the indexes when comparing the results before and after the
exclusion of outliers.
Conclusions
 Q1 recorded a 112% effect. This means that the anchors displayed had a very large effect on the
answers that were being given by the students. According to the index, the students were heavily
relying on the anchor to come up with their estimate.
 Q2 recorded a -13% meaning the anchors were not very effective on the students estimates of Taylor
Swift’s net worth. This could be the case in that Taylor Swift has become increasingly more popular over
recent years and has been seen in the news a lot recently due to her record-breaking numbers. It may
be easier for students to disregard the anchors that were given.
 Q3 and Q5 recorded very similar effects, 71% and 75%. According to Kahneman this would mean that
the anchors had a very significant effect on most of our participants. If the students were not familiar
with the height of the Eifel Tower or tuition paid on average at Mars Hill University, they were more
than likely swayed by the anchors when giving their own answer.
 Q4 recorded a 39% effect which closely relates to results that Kahneman received when conducting his
own experiments. Once again, this would mean that the anchors did play some part in affecting the
individual’s own answer to the question.
Why Does This Matter?

The anchoring bias is a thinking pattern that happens when we rely too heavily on the first piece of information
we receive about something. We use this initial piece of information, called the "anchor," as a reference point
when making decisions. Even if the anchor is irrelevant or random, it can affect how we think and make
choices. This bias often stops us from updating our decisions as we should.
Steps For Defending Anchoring Bias

1. Be Aware: The first step in defending against anchoring bias is to be aware that it exists. Knowing that
our first piece of information can influence our decisions is essential.
2. Take Your Time: When making important decisions, slow down and think it through carefully. Don't
rush. Consider different perspectives and gather more information.
3. Question the Anchor: If you recognize an anchor, ask yourself if it's relevant or based on facts. Is it a
fair reference for your decision?
4. Seek Multiple Opinions: Talk to others, seek advice, and get multiple opinions. This can help you break
free from a single anchor and see the bigger picture.
5. Consider Alternatives: Before settling on a decision, think about different options and their pros and
cons. This can help you avoid getting stuck on one idea.
6. Stay Flexible: Remember that it's okay to change your mind if you get better information. Don't feel
locked into your initial choice

Anchoring Bias Results Writeup

Uploaded by

Anchoring Bias Results Writeup

Uploaded by

Anchoring Results & Write Up

Decision Theory Project

Question A1 Anchor A2 Anchor

How long, on average, do you think a lion lives? 10 years 30 years

Finding the Anchoring Index

Q1B Q2B Q3B Q4B Q5B

Results from Anchoring (Excluding Outliers)

120 100000000 850 12 12000

Anchoring Index According to Kahneman

Why Does This Matter?

Steps For Defending Anchoring Bias

You might also like