Group 1 DISC 323 – Research Project

DISC – 323 DECISION BEAHVIOR

RESEARCH REPORT
THE RELATION BETWEEN DEBIASING & EDUCATION

PREPARED BY:
Saad Ahmad (2016-11-0246)
Usman Khurshid (2016-11-0092)
Hashim Masood (2016-11-0229)
Mehrunisa Riaz (2016-11-0007)

1
Group 1 DISC 323 – Research Project

CONTENTS

Introduction

Literature Review

Analysis

Conclusion

Limitations

2
Group 1 DISC 323 – Research Project

INTRODUCTION
Research Question:

"Does increasing the level of education decrease the vulnerability to cognitive biases?"

Method:

We used primary research for data collection i.e. it was done primarily through
cautiously prepared questionnaires. The questions are formulated in a manner which covers
most of the aspects of behavior and biases. The following steps were taken:

1. A literature review was done to see what we’re adding to the existing body of
knowledge

2. Prepared a standard questionnaire with a battery of cognitive biases

3. Included the following group of respondents:

a. Higher Education (Undergraduate, Masters, PhD)

b. Intermediate Level (A/O-Level students. & F.Sc/F.A students)

c. Illiterate people.

4. The sample sized for each group varied between 25-80

5. We tested the respondents on various biases – Availability, Misconception of
Chance, Anchoring, Bounded Awareness, Framing, Self-Serving, Regret
Avoidance, Affect Heuristic, and Escalation of Commitment.
6. Analyzed the results and drew a conclusion.

3
Group 1 DISC 323 – Research Project

LITERATURE REVIEW
“Critical Thinking and Debiasing” by Kenyon and Guillaume:

In the article “Critical Thinking and Debiasing” the authors state that the Intuitive
approach to teaching debasing may be dubious since it teaches us information about biases and
expects us to remember the situational relevance of that information in our daily lives, which
may be filled with biases. Yet most of us fail to apply such knowledge even after studying it
under the intuitive approach. They say that there should be practical guidance on how to apply
this knowledge rather than passively learning it.

We propose that by increasing the level of education, a person gets more practice in
applying this knowledge about Biases either through formal Decision Behavior courses or via
street smarts/folk lore. He gets more skilled in identifying biases in real life and avoiding them.

“Debiasing” by Richard. P. Larrick:

This article on debiasing talks about various strategies to help biased individuals in
making better decisions and the sole purpose of these strategies is to eliminate de-biasing. Some
of these strategies are as follows:

Motivational Strategies:

1) Incentives: Motivation and Incentive can lead to better less-biased decision making
provided individuals have the cognitive capital. The application of these strategies can lead
to rather inferior decision making if people do not have the required cognitive capital
2) Accountability: We can make people accountable and make them expect that they will have
to give explanations for their decisions. Thus, people start identifying and correcting flaws
in their own argument. The main problem is, that if the audience’s preferences are known,
it leads to further biases.

Cognitive Strategies:

1) Considering the opposite: This involves asking oneself “Why is my judgement wrong?”
Thus, individuals evaluate their judgements and consider alternatives
2) Training: The purpose of training is to replace inferior strategies by new ones and help
people to apply the basic knowledge that they have about economics, statistics and so on.
Formal strategies also transmit the necessary cognitive strategies.

4
Group 1 DISC 323 – Research Project

Technical Strategies:

1) Group Decision Making: This poses both benefits and drawbacks for the decision makers.
Where people pool in Information to reach a conclusion, on the other hand, unshared
information is omitted from discussions which only enhances the initial bias.

As mentioned above, these strategies can prove to be quite effective. However, the
problem lies in the fact that under certain conditions, the use of all these, and a few other
techniques can potentially lead to even worse outcomes.

The purpose of our research is to fill this gap. We intend to justify that the root cause
of bias formation which in our case is illiteracy, should be eliminated. Education will help to
provide individuals the necessary cognitive capital. While education in itself, as per our
research, leads to debiasing, the application of various strategies on literate individuals will
lead to even more superior decision making.

However, there’s also a similarity between our research and this article. We intend to
use education, and this article mentions training, as a means to transmit cognitive capital which
helps in debiasing.

"Center for Applied Rationality: Practical Rationality Training” by Stephan

Schubert:

The article revolves around the decision making process of humans. It highlights how
we, as humans, think we are more rational than we actually are. The research identifies the
factors that trigger irrational decision making and talks about ways to stop the triggering
process. One important way to stay away from biases is to change the environment one is in
while making an important decision, or to change oneself completely. Humans weigh future
benefits irrationally low in comparison to today’s benefits. The article states that relevant
education can help de bias people. Whereas formal (usual) education is more or less ineffective.
So in order to de bias people, it is important for people to join a program that is specifically
designed for this purpose. A Californian nonprofit start up “Centre for Applied Rationality”
has been designed to help people make more rational decisions. This institution focuses on
fixing the areas where people are more likely to fall prey to biases. One of their classes “Staff
Reference Class” focuses on the mistakes people make because they underestimate the time it
will take to implement various plans. Instead of focusing on future events, it is more rational

5
Group 1 DISC 323 – Research Project

to estimate the time duration of such events by looking at the time such events took in the past.
CFAR also teaches ways to overcome the brain’s system 1’s impulses. They say that the best
way to achieve goals is to go about the implementation intention and specify one’s “if-then”
plan. This means that a person is more likely to implement a plan if he has specified exactly
when he would perform that action. Once a relation is built between the trigger and the action,
it is more likely that one will be able to accomplish those actions. The classes at CFAR are
very rewarding so this factor reinforces participant’s motivation to be rational.

Hence it seems there is some merit to the effect of debiasing training/education on

debiasing itself, however we intend to see whether formal (usual) education has any effect of
not.

“Costs and Benefits of Judgment Errors” by Hal R. Arkes:

The author has grouped the errors that human beings commit in decision making and
judgment into three broad categories and then he has given some strategies to eliminate them.

Those three categories are:

i) Strategy based errors
ii) Association based errors
iii) Psychophysically based errors

He defines strategy based errors as the “errors which occur when subjects use a
suboptimal strategy, the extra effort required to use a more sophisticated strategy is a cost that
often overweighs the potential benefits of enhanced accuracy. Hence, the decision makers
remain satisfied in using suboptimal strategy in low stake decisions.” (Arkes)

“Association based errors are the cost of otherwise highly adaptive system of
associations within semantic memory. The automaticity of such associations, generally of
enormous benefit, becomes a cost when judgmentally irrelevant or counterproductive semantic
associations are brought to bear on the decision of judgment.” (Arkes)

“Psychophysically based errors result from nonlinear mapping of physical stimuli onto
psychological responses. Such errors represent costs incurred in less frequent stimulus ranges
where very high and low stimulus magnitudes are located. These costs are more than offset by
sensitivity gains in the more frequent stimulus ranges located in central portion of the stimulus
spectrum.” (Arkes)

6
Group 1 DISC 323 – Research Project

Debiasing:

Strategy based errors:

The way to improve judgment within this category is to raise the cost of using
suboptimal judgmental strategy. Typically this results in the judge’s utilization of currently
available data in a much more thorough way, an obviously superior strategy. If a subject in a
low stakes judgment environment were using only a subset of the available data, it would be
obvious how to improve one’s judgment should the stakes increase: Use more date.

When subjects are made accountable for their decisions and predictions, they consider
every aspect in a less superficial way.

Association based errors:

Some authors said earlier that incentives are good to control it but Arkes said that
incentives are not effective in debiasing this type of errors because motivated subjects will
merely perform the sub optimal behavior with more enthusiasm. Instead it will be more helpful
to instruct the subjects to use a behavior that will add or alter the associations.

Daily feedback on appropriateness on one’s confidence is a debiasing technique against

over confidence bias. Subjects must be instructed to look for the information that is counter to
their judgments and decisions.

Psychophysically based errors:

Framing can be used. You can change the concentration of related items in the options
from which the decision is to be made.

All the above mentioned ways can be achieved or improved with getting education and
this supports our research. In the end, author, deliberately, says that training improves your
decision making which supports our research that education improves decision making. He
says that accounting training, statistical training and other trainings provide people with tools
that help them to reach at the correct decisions and answers.

However, there wasn’t any empirical done which explicitly highlighted the role or
impact of education on de-biasing or rationality; hence our research is the first of its kind.

7
Group 1 DISC 323 – Research Project

ANALYSIS
Note: Many of our questions are taken straight from the book (cited at the end) or have been
modified slightly so as to make them more applicable to our context.

Availability Heuristic - Ease of Recall:

This is a bias which emanates from the Availability bias. Basically it says the more
vivid stories are more available via various sources such as the media and they skew our
perception of the frequency of various causes of an event. We overestimate vivid causes of an
event and underestimate less vivid causes.

We included the following question on this bias in our questionnaire:

In Pakistan, which of the following do you think has caused an increase in the death rate?

A) Car/Air Accidents

B) Poor health or diseases

C) Terrorist Attacks

In this case the actual statistics tell us that poor health or diseases are the main causes
of increase in the death in Pakistan but it is not a very vivid alternate and tends to not be reported
in the media as much as the other two options. So for a respondent to be trapped in the Ease of
Recall bias, he/she would have to select any option other than B since they are more vivid and
they would be more available in the memory. The results showed:

AVAILABILITY BIAS - EASE OF RECALL

67%

46.70%
48%

ILLITRATE INTERMEDIATE HIGHER EDUCATION

8
Group 1 DISC 323 – Research Project

Results:

The results confirm our hypothesis, with the frequency of biased responses going down
from Illiterates (67%) to Intermediate (48%) and finally to Higher Education (46.7%). Indeed
an increase in the educational level makes a person not rely on just vivid causes present in the
memory but rather to use intellect to choose the most likely cause for an event, even if it is not
so vivid.

Misconception of Chance:

People that are trapped in this bias expect probabilities to even out. When for example
on a toss of a coin (a random event) “heads” comes five times, then we say to ourselves that
a tail has to come this time , after all what are the chances of a “heads” appearing sixth time
in a row. We have a misconception of connectedness. Even though the chances of that
happening are not dependent on a past occurrence of an event, the odds are still 50-50.
This gives rise to many other common fallacies such as the belief in the “hot hand” in
sports.

We included the following question related to misconception of chance in our questionnaire:

You have 5 children, and all of them are girls. Now, you are expecting another child. Is the
probability of having a boy this time high?

A) True

B) False

In this question, the irrational or trapped respondents would say “True” to this question,
wrongly believing that it makes intuitive sense that having 6 girls in a row is highly unlikely.
Whereas those not under the influence of the Gambler’s Fallacy would choose “False” since
gender determination of each new baby is a chance event and it does not depend on many other
girls have been born previously.

9
Group 1 DISC 323 – Research Project

Results:

MISCONCEPTION OF CHANCE

61%

17%

6.70%
ILLITRATE INTERMEDIATE HIGHER EDUCATION

The results are consistent with our hypothesis. The percentages of biased responses is
going down as the education levels increases, from Illiterates (61%) to Intermediate (17%) and
finally the Higher Education (6.7%).

Anchoring Bias:

Susceptibility to anchoring decreases as the education level increases. Education gives

us knowledge about the world and we become better in knowing the worth of different things.
When you have knowledge about something or know the worth of something, you are less
likely to get anchored to the false irrelevant information about that thing. Moreover, anchoring
affect is being taught at university level so it further decreases the susceptibility to get affected
by it.

We included the following question in our questionnaire:

Assume that you’re a teacher in a school which categorizes students by ability at an early age.
Do you expect less from students in the lowest group, and higher from students in the top
group? (Anchoring bias – Confirmation Heuristic)
A) Yes
B) No

10
Group 1 DISC 323 – Research Project

Results:

The results confirmed our hypothesis. Ranging from Illiterate to higher education, the number
of people caught by the Anchoring Bias fell by approximately 15%. As the level of education increases,
people tend to understand realities and become less prone to biases.

ANCHORING
61%

51%

46.70%
ILLITRATE INTERMEDIATE HIGHER EDUCATION

Bounded Awareness:

Without lifting your pencil (or pen) from the paper, draw four (and only four) straight lines
that connect all nine dots shown here:

11
Group 1 DISC 323 – Research Project

BOUNDED AWARENESS

100%

100%

100%
ILLITRATE INTERMEDIATE HIGHER EDUCATION

As far as bounded awareness is concerned, it was found in all of our groups. It’s a
“Universal Bias” and no one in our sample from all of the categories was able to solve this
question, hence proving that he/she did not suffer from bounded awareness. The major reason
is that in Pakistan, enough emphasis is not put on creativity. People do not think out of the box.
We are just told to be creative or think out of the box but we are never trained or taught how to
be more creative and how we can think out of the box. That’s why this problem exists in
everyone, ranging from illiterate to higher education, in Pakistan.

Framing:
Framing is a concept whereby a person changes his/her preferences of two objectively
same alternatives bases solely on how the two options are worded. Research tells us that we
are risk averse when it comes to benefits and risk seeking in case of losses. A rational person
must have consistent risk preferences and should not change them going from one alternative
to the next, however due to the framing bias people end to show inconsistent risk preferences.
We included two situations (4 questions) on framing in our questionnaire:

Q7. Imagine that Pakistan is preparing for the outbreak of an unusual disease that is expected
to kill 600 people. Two alternate programs to combat the disease have been proposed. Assume
that the exact scientific estimates of the consequence of the program are as follows, which
would you choose?

12
Group 1 DISC 323 – Research Project

Program A: If Program A is adopted, 200 people will be saved.

Program B: If Program B is adopted, there is a one-third probability that 600 people will be
saved and a two-thirds probability that no people will be saved.

Q8: Imagine that Pakistan is preparing for the outbreak of an unusual disease that is
expected to kill 600 people. Two alternate programs to combat the disease have been proposed.
Assume that the exact scientific estimates of the consequence of the program are as follows,
which would you choose?

Program C: If Program C is adopted, 400 people will die.

Program D: If Program D is adopted, there is a one-third probability that no one will die
and a two-thirds probability that 600 people will die.

In Question.7, option A is the sure, riskless outcome that should be selected by risk
averse people while option D is the risky alternate which should be selected by risk seeking
subjects. Similarly in Question.8 option C is the riskless alternative and option D is the risky
one. However do note that the expected value in both the cases is the same:

Option A: 200 people will be saved.

Option B: (1 / 3 * 600) + (2 / 3 * 0) = 200 people will be saved.

Option C: 400 people die (conversely 200 people will be saved).

Option D: (1 / 3 * 0) + (2 / 3 * 600) = 400 people die (conversely 200 people will

be saved).

The only difference between the two questions is in the wording and this is where the
Framing bias comes into play. Question.7 uses “people saved” and most people view it
similarly to benefits and exhibit risk averse preferences and choose the sure, riskless
alternative. Whereas Question.8 uses the terms “people die” and most people view it similar to
losses and exhibit risk seeking preferences and go with the risky uncertain outcome in hopes
that the odds will be in their favor and less people die.

A rational person should choose one alternative either risky or riskless in Question.7
and then stick with that risk type alternative for Question.8.

13
Group 1 DISC 323 – Research Project

According to the context above, in order to get trapped in the framing bias, the irrational
respondents would have to choose option A in Question.7 and then switch to option D in
Question.8.

Results:

FRAMING EFFECT
39%

13%
11%

ILLITRATE INTERMEDIATE UNDERGRADUATE

The results were not as expected. The percentage of irrational responses was greatest
for illiterates (39%), then it went down for Intermediate respondents (11%) and then it again
arose to (13%). Although not a very sharp rise, but a rise nonetheless whereas it should have
been a decrease.

We can say that an increase in education decreases susceptibility to framing bias but
the drop is most significant for Intermediate education and then it starts to level off. Thus small
rise may just be due to the small sample sizes and it may disappear by increase in the sample
size from 30 to 50+ per category.

Regret Avoidance:
Q10) Imagine you’re in a meeting, out of town. The meeting runs late and you then you head
to the airport to catch the last flight back home. Your flight is scheduled for 8:30pm. If you
miss the flight, you’ll have to stay overnight and miss an important meeting the next day. You
reach the Airport at 8:57pm. When you arrive you found out that:
A – The flight departed at 8:30pm
B – You see the plane depart, having left at 8:55pm.
Which is more upsetting, (a) or (b)?
A) The first situation
B) The second situation

14
Group 1 DISC 323 – Research Project

Results:

REGRET AVOIDANCE

89%

87.70%

86.70%
ILLITRATE INTERMEDIATE HIGHER EDUCATION

Perhaps Regret Avoidance is the only bias which does seem to decrease, but there isn’t
a significant difference. Basically, people who are susceptible to this bias are caught by
‘counterfactual thoughts’ and come to believe that a slight change in their past actions could
prevent an undesired outcome. Both Illiterate and Literate respondents seem to have more or
less the same emotions in such situations.

SELF-SERVING BIAS
83%

67%
39%

ILLITRATE INTERMEDIATE HIGHER EDUCATION

Self-Serving Bias:
As per our research, there isn’t any connection between education and the self-serving bias.
Rather, the literate seem to be more inclined towards it.

15
Group 1 DISC 323 – Research Project

Affect Heuristic:
You're about to move to a new city, and you have to ship an antique grandfather clock. In the
first case, the grandfather clock was a gift from your grandparents on your 5th birthday. In
the second case, the clock was a gift from a remote relative and you have no special feelings
for it. How much would you pay more for an insurance policy that paid out $100 if the clock
were lost in shipping (In both conditions?) – Affect Heuristic
Ans: a) ____________ b) _____________
Results:

AFFECT HEURISTIC
72%

48%

47%

ILLITRATE INTERMEDIATE HIGHER EDUCATION

As the graph above suggest, people with higher education were less susceptible to this
bias. The reason being, illiterate people usually have weaker decision making skills and rely
on emotion. In this case, most of them (72%) were willing to pay more in the first condition.
On the other hand, education enables us to think more rationally and take into account the
realities of this world. People with higher education were significantly able to avoid it, only
47% reported that they would be willing to pay more in the first condition. Affect and Cognition
are related. Affect refers to emotions and Cognition is how the information enters or leaves the
memory. The relationship between the two can take different forms, depending on who’s the
decision maker. For the illiterate, feelings (Affect) shape thinking (Cognition) whereas as the
level of education increase, thoughts start to shape feelings. Thus, respondents holding higher
education were able to think practically and used their thought to shape the feeling that no, they
won’t be willing to pay more for the insurance policy even if it was given to them by a close
relative. They would either pay nothing, or equal amounts for both conditions.

16
Group 1 DISC 323 – Research Project

Escalation of Commitment:

You accept a prestigious position in a firm, believing that that it’s an excellent career
opportunity and will allow you to grow. Two years later, you have not progressed as rapidly
as you expected. Anxious to demonstrate your worth to the company, you invest large amounts
of unpaid overtime. You still don’t get the recognition that t you think you deserve. By now,
you have been with the organization for several years and will lose many benefits if you leave.
You are in your late thirties, and think that you have invested your best years with this company,
Will you quit? (Escalation of Commitment)

A) Yes
B) No
Results:

ESCLATION OF COMMITMENT
94%

41%

33.30%

ILLITRATE INTERMEDIATE HIGHER EDUCATION

The results for this Question again confirmed our hypothesis. People who decide to
continue this Job, basically they are risk seeking, since the situation in this Question is
negatively framed. Also, the illiterate usually only see the positive aspects of their past decision
and want to create the impression that their previous decision wasn’t a mistake. The percentage
of illiterate who were willing to continue the Job was 94%. This figure fell significantly to 33%
as the level of education increased. Literate people are able to keep aside their past decisions
and non-rational thinking, and are able to isolate the rational decide of various decisions.

17
Group 1 DISC 323 – Research Project

CONCLUSION

In conclusion, our hypothesis was true. We tested the respondents on various biases –
Availability, Misconception of Chance, Anchoring, Bounded Awareness, Framing, Self-
Serving, Regret Avoidance, Affect Heuristic, and Escalation of Commitment. With the
exception of one or two biases, all others tended to decrease as the level of education increased.
The most significant gap was between illiteracy and intermediate education, which shows that
even basic levels of education can make humans less prone to various biases.

In this project, we intended to focus on uncovering the root cause of bias formation,
and find ways to eliminate it. Thus, the reason we identified is lack of literacy, thus lack of
cognitive capital which makes us more inclined towards biases. The remedy is to provide at
least basic levels of education.

There are a number of strategies for debiasing, which we have identified in our
Literature Review, but the problem is, that if individuals lack the “Cognitive Capital”, the use
of these techniques will either give no results, or may even lead to the construction of further
biases and hence inferior decision making. The role education, as per our research, is to build
this Cognitive Capital, which in itself leads to debiasing. And if there’s still any gap, then these
strategies for debiasing will be more effective.

Our literature review also showed that one major aspect of debiasing is to “Modify the
Person” – meaning that they should be equipped with the necessary skills and also should be
taught how to apply them in practical life. Education undoubtedly does make things better and
assists in making debiased judgements, but it will make even a greater difference if individuals
are given training so that they do understand under what situation exactly they can apply their
knowledge and also, how to do that.

LIMITATIONS

There was one slight limitation in our research – we did not take into account gender
differences. In other words, most of the illiterate respondents were females, and the literate
respondents were usually males. Hence, the gender difference could also contribute towards
the downward trend of biases, and our research does not take this into account.

Moreover, in the past, no extensive research has been done in this area so we do not
have much historic evidence to support our claims. Although yes, our results do show that
educations helps to become less susceptible to biases, but the sample size is still not enough to
generalize our findings across the globe.

18
Group 1 DISC 323 – Research Project

Works Cited

Arkes, Hal R. "Costs and Benefits of Judgment Errors: Implications for Debiasing." Psychological
Bulletin (1991). Web. 17 April 2015.
<http://www.researchgate.net/profile/Hal_Arkes/publication/232547304_Costs_and_benef
its_of_judgment_errors_Implications_for_debiasing/links/53ff57390cf212df056d65f4.pdf>.

Bazerman, M. H. and D. A. Moore. Judgement in Managerial Decision Making. Wiley, 2009. Print.

Kenyon, Tim & Beaulac, Guillaume. "Critical Thinking and Debiasing." Informal Logic 34.4 (2014):
341-363. Web. 3 May 2015.
<http://ojs.uwindsor.ca/ojs/leddy/index.php/informal_logic/article/viewFile/4203/3339>.

Larrick, Richard P. "Debiasing." Blackwell Handbook of Judgement and Decision Making (2004). Web.
25 May 2015.

<http://onlinelibrary.wiley.com/doi/10.1002/9780470752937.ch16/pdf>.

Schubert, Stephan. "Center for Applied Rationality: Practical Rationality Training". n.d. Web. 10 May
2015.

<http://rationality.org/>.

19