Testing the elicitation procedure of the Minimum Acceptable Probability

Last registered on October 25, 2021

View Trial History

Pre-Trial

Trial Information

General Information

Title

Testing the elicitation procedure of the Minimum Acceptable Probability

RCT ID

AEARCTR-0007776

Initial registration date

June 16, 2021

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

June 16, 2021, 10:54 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated

October 25, 2021, 11:25 AM EDT

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country

United Kingdom of Great Britain and Northern Ireland

Region

Primary Investigator

Name

Maria Polipciuc

Affiliation

Maastricht University

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Martin Strobel

PI Affiliation

Maastricht University

Contact Investigator

Additional Trial Information

Status

Completed

Start date

2021-06-17

End date

2021-07-01

Keywords

Behavior

Additional Keywords

minimum acceptable probability, betrayal aversion, risk, risk aversion, online experiment

JEL code(s)

D81, C72, C99

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

We examine in how far the elicitation procedure of the Minimum Acceptable Probability (MAP) used by Bohnet and Zeckhauser (2004) is influenced by the presentation of the probability distribution of the winning probability. Should the distribution have an influence on participants’ MAP, this is evidence that was has been called ‘betrayal aversion’ in previous studies is (partly) due to factors other than an anticipated disutility of betrayal.

External Link(s)

Registration Citation

Citation

Polipciuc, Maria and Martin Strobel. 2021. "Testing the elicitation procedure of the Minimum Acceptable Probability." AEA RCT Registry. October 25. https://doi.org/10.1257/rct.7776-1.1

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

We compare minimum acceptable probabilities (MAPs) for which a participant prefers a binary lottery to a sure payoff across treatments within and across individuals. Individuals have to decide in randomized order in three scenarios (treatments). The treatments keep the sure payoff and the payoffs of the lottery fixed, but vary the distribution of the winning probability of the lottery.

Intervention Start Date

2021-06-17

Intervention End Date

2021-07-01

Primary Outcomes

Primary Outcomes (end points)

Minimum acceptable probabilities (MAPs)

Primary Outcomes (explanation)

The MAPs are elicited as thresholds for preferring a lottery to a sure payoff.

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

We elicit MAPs in an online experiment in a lottery similar to the Decision Problem in Bohnet and Zeckhauser (2004). Bohnet and Zeckhauser (2004) find that participants require a premium for being willing to trust someone compared to taking an equally risky bet with the same payoff externalities for an uninvolved person. They attribute this premium to betrayal aversion—an anticipatory disutility from being betrayed.
In this study, we experimentally test a confounding explanation for this premium. We test whether participants’ MAPs vary systematically as a result of the underlying distribution from which the probability of success of the lottery is drawn. We exogenously manipulate participants’ probabilistic beliefs about this distribution by varying the different objective distributions of the winning probability.
We then compare the resulting MAPs across treatments. We also compare the magnitude of the treatment effects with the effects in Bohnet and Zeckhauser (2004); Bohnet et al. (2008).

Experimental Design Details

Randomization Method

Computer

Randomization Unit

Individual

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

450

Sample size: planned number of observations

450

Sample size (or number of clusters) by treatment arms

450 individuals. It is a within-subject design, so all individuals go through the three treatments. The sequence in which the go through the treatments is randomized (there are six possibilities in total).

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

With 400 valid responses, we have 80% power to detect a statistically significant effect at 5% level of 0.062 (for MAPs in the [0,1] interval) with no order effects. Since in this study MAPs are integers between 0 and 15, this translates into a minimum detectable effect size of 0.93. If we assume second and third decision are `sticky' i.e. stay close to the first decision, the we have 80% power to detect a statistically significant effect at 5% level of 0.05 on for MAPs on a scale from 0 to 1, or 0.75 for MAPs between 0 and 15.

Supporting Documents and Materials

IRB