Decision Making With Many Options Tibor Besedes Cary Deck Sudipta Sarangi Mikhael Shor October 2007
Decision Making With Many Options
Tibor Besedes
Cary Deck
Sudipta Sarangi
Mikhael Shor
October 2007
2
Motivation
Life is full of choices Many important life decisions are made from an often
overwhelming number of options Mathematical truism:
Psychological perspective: Information Overload People “give up” when facing too many options
Cognitive perspective: Brain “processing power” is limited
)max()max( YXX
3
Evidence on Information Overload
Fewer people join a 401(k) retirement plan when more savings options are presented Iyengar, Jiang and Huberman 2004
Physicians are less likely to prescribe any drug when more drugs are available Redelmeier and Shafir 1995, Roswarski and Murray 2006
Total amount of recycling decreases when people are offered multiple recycling options.
Greater “choice satisfaction” when choosing among six Godiva chocolates than among 30 Iyengar and Lepper 2000
I don’t like long restaurant menus
4
Limitations of Prior Studies
Past studies examine either satisfaction with choice or whether a choice was madeFewer choices made does not imply that
the average choice is worseSatisfaction with choice does not imply
objectively good decision-making
We want to know whether a choice is optimal
Research Hypotheses
When faced with a large set of options, individuals make inefficient and suboptimal decisions.
Older individuals, will suffer a greater deterioration of decision accuracy as decision complexity increases.
Reducing the complexity of the task makes decision-making more efficient.
5
Medicare Part D
Private insurers offer prescription drug plans
A person may see
as many as 140 competing plans
6
Motivating Example
7
Motivating Example
Motivating Example
8
Motivating Example
10
Motivating Example
11
Nature of Decisions
Physician Office Visit
Preventive Care
Urgent Care Service
Emergency Room Service
Hospital Expenses (inpatient)
Hospital Expenses (outpatient)
Diagnostic Services
Available Plans
A B C D
Nature of Decisions
There exist unknown future states of nature I’ll be healthy or sick. I’ll need what drug?
States have associated probabilities Options “cover” some states but not others Choice is a maximization over states
Simplified:Exactly one state is realizedNo “cost” of options If chosen option covers the state that occurs,
subject receives payment
12
Experimental Decision Problem
13
Experimental Design
2 x 2 x 2 (+ 1) within-subject design Number of states: Either 6 or 10 Number of options: Either 4 or 13 options Probability distribution
10 state problems equivalent to 6 state problems Options “expanded” (all check marks preserved)
For probability distribution 1: All states are rather likely Going from 4 options to 13 by introducing suboptimal options
For probability distribution 2: Several states have very low probability 4 to 13: one new option is much better (96% v. 71%)
14
Methodology
Random order of Decision problems Options States
125 subjects recruited online Paid $1 for every successful state, plus $3 Collected demographics:
Age, sex, education Dependent variables:
Frequency of optimal decisions Efficiency of decisions (how suboptimal is suboptimal)
15
Results
Selection of optimal option4 options 13 options
PDF 1(all states likely, new options all bad)
6 attributes 41% 29%10 attributes 45% 24%
PDF 2 (low prob. states, new option better)
6 attributes 50% 50%10 attributes 52% 36%
16
Increasing options reduces frequency of optimal choice
Results
How suboptimal are the choices?
42% of all choices were the optimal option
66% of all choices were within 10% of optimal
Average efficiency loss was 13%
Subjects were half-way between optimal choice and random choice
17
Impact of Age: First Order Effects
18
Optimal decision-making decreases with age
Age Group18-40 41-60 61+
Frequency of Optimal Choice
52% 42% 31%
Nearly Optimal Choice (within 10%)
72% 66% 59%
Improvement over random choice
61% 49% 36%
Frequency of “dominated” options
0% 5% 18%
Impact of Age: Second Order Effects
19
Decision complexity interacts with age
Age Group18-40 41-60 61+
Relative Frequency 82%
18%
66%
34%
47%
53%
decision)complex (least choice optimal %
decision)complex (most choice optimal %
Results
20
Regressions
Chance of selecting optimal option:
Decreases with age
Increases with education
Does not depend on sex
Parameter magnitudes
11 years of age offsets an education category
3 education categories offsets having more options
Implications
With reasonable a priori knowledge about optimal options, presenting fewer options is better AT&T knows this, government does not
For older people, fewer may be better even without any a priori knowledge Best of any 4 better than random of 13
Future investigation of choice presentations Default “suggested” options Break up big decisions into smaller ones Recommender systems
21