Posted by: jasonpoulos | January 5, 2010

Intertemporal Choice

A short intertemporal choice survey was submitted to 110 workers, mostly from the United States (51%) and India (42%) (40% female; mean age = 33). After discarding incomplete responses and those completed under 50 seconds, the usable sample size was n = 94.

Workers were tasked with the “French dinner problem” described in Loewenstein and Prelec (LP; 1993), which tests violations of independence in the context of intertemporal choice.

LP AMT
1.  Which would you prefer if both were free? n = 95 n = 94
A.  Dinner at a fancy French restaurant 86% 64%
B.  Dinner at a local Greek restaurant 14% 36%
For those who prefer French:
2.  Which would you prefer? n = 82 n = 60
C.  Dinner at the French restaurant on Friday in 1 month 80% 65%
D.  Dinner at the French restaurant on Friday in 2 months 20% 35%
3.  Which would you prefer? n = 82 n = 60
E.  Dinner at the French restaurant on Friday in 43% 78%
1 month and dinner at the Greek restaurant
on Friday in 2 months
F.  Dinner at the Greek restaurant on Friday in 57% 22%
1 month and dinner at the French restaurant on
Friday in 2 months

Results from LP (1993; left column) and our study (right column) are posted in the table above. Among those who prefer a French dinner, our results show a relatively stronger preference for delayed consumption with the single-outcome event (i.e., Question 2). While LP finds a preference for sequential improvement in Question 3, our workers maintain a positive discount rate and do not opt for the sequential improvement.

We also included the intertemporal choice survey used in Kirby and Marakovic (KM; 1996) to test the independence of delay-discounting rate and monetary reward size. Workers were asked to choose between the following hypothetical pairings, ranked in increasing order by hyperbolic discounting parameter values (i.e. those values at which immediate and delayed rewards are equalized):

Choice Trials and Their Associated Discounting Parameter Values
Order Choice Trial Hyperbolic %Ss (KM) %Ss (AMT)
4 $34 tonight or $35 in 43 days 0.0007 12 7
15 $53 tonight or $55 in 55 days 0.0007 12 17
7 $83 tonight or $85 in 35 days 0.0007 12 15
20 $27 tonight or $30 in 35 days 0.0032 17 18
9 $48 tonight or $55 in 45 days 0.0032 34 32
12 $65 tonight or $75 in 50 days 0.0031 44 28
8 $21 tonight or $30 in 75 days 0.0057 36 37
16 $47 tonight or $60 in 50 days 0.0055 57 49
14 $30 tonight or $35 in 20 days 0.0083 44 36
10 $40 tonight or $65 in 70 days 0.0089 67 49
3 $67 tonight or $85 in 35 days 0.0077 70 54
18 $50 tonight or $80 in 70 days 0.0086 74 54
11 $25 tonight or $35 in 25 days 0.016 68 53
2 $40 tonight or $55 in 25 days 0.015 71 52
19 $45 tonight or $70 in 35 days 0.0159 90 73
21 $16 tonight or $30 in 35 days 0.025 86 66
6 $32 tonight or $55 in 20 days 0.0359 94 77
17 $40 tonight or $70 in 20 days 0.0375 97 85
5 $15 tonight or $35 in 10 days 0.1333 99 86
13 $24 tonight or $55 in 10 days 0.1292 99 87
1 $30 tonight or $85 in 14 days 0.131 99 N/A

The last two columns show the percentage of subjects choosing the delayed reward on each trial. Our results show a relatively steeper discount function than KM, with an estimated median hyperbolic parameter value of 0.012 compared with 0.007 in KM. The 0.012 value indicates the median worker switched from choosing the immediate outcome to the delayed outcome between the 12th and 13th trial ranked in the table above.

References

Loewenstein, G., Prelec, D. (1993). Preferences for Sequences of Outcomes. Psychological Review 100(1), 91-108.

Kirby, K., Marakovic, N. (1996). Delay-discounting probabilistic rewards: Rates decrease as amounts increase. Psychonomic Bulletin & Review 3(1), 100-104.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Categories

%d bloggers like this: