Mixed Strategies and Context: Evidence from a lab experiment

Last registered on November 08, 2021

View Trial History

Pre-Trial

Trial Information

General Information

Title

Mixed Strategies and Context: Evidence from a lab experiment

RCT ID

AEARCTR-0005550

Initial registration date

March 12, 2020

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

March 12, 2020, 6:59 PM EDT

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

Last updated

November 08, 2021, 4:16 PM EST

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

Husnain Ahmad

Affiliation

The University of the South

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Sheheryar Banuri

PI Affiliation

University of East Anglia

Contact Investigator

Additional Trial Information

Status

On going

Start date

2020-03-12

End date

2021-11-30

Keywords

Other

Additional Keywords

Mixed Strategies, Lab experiment

JEL code(s)

Secondary IDs

Prior work

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

Abstract

Due to COVID 19, the lab experiment needed to be halted part way through. We resumed data collection in Nov 2021.

We study the effect of a game’s contextual framing on how subjects play the game in a lab. Subjects play two strategically equivalent games across treatments, namely; hide and seek, and penalty kicks. The games have identical unique Nash equilibria in mixed strategies, with the only difference between them being the context in which the game is presented. We test whether subjects play is closer to theoretical prediction in the penalty kick game, which is a more familiar context for subjects and is closer to games played in the field. Data from past experiments show that while subjects play in accordance with theory in the field, they fail to do so in the lab. Our experiment attempts to determine whether this discrepancy is due to the difference in context between lab and field games. It can also be seen as a test of the underlining assumption of strategic equivalence in games with unique mixed strategy equilibria.

External Link(s)

Registration Citation

Citation

Ahmad, Husnain and Sheheryar Banuri. 2021. "Mixed Strategies and Context: Evidence from a lab experiment." AEA RCT Registry. November 08. https://doi.org/10.1257/rct.5550-1.3000000000000003

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Variation in the level of context in a game.

Full context involves a video game like PK environment, partial context involves descriptions drawn from football, and no context uses de-contextualised hide and seek.

Intervention Start Date

2020-03-12

Intervention End Date

2021-11-30

Primary Outcomes

Primary Outcomes (end points)

- Player choices in each round.
- Player set "frequencies" for computer to play multiple rounds on their behalf.

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

The context of the game will be varied with three levels: no context, written context and visual context.

Experimental Design Details

Randomization Method

In lab with lottery.

Randomization Unit

Individual randomisation for treatment.
Within treatment random matching of players into pairs by computer.

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

n/a

Sample size: planned number of observations

150 choice of pure strategies per individual. Additionally 4 computer "frequencies" (or mixed strategies) per individual.

Sample size (or number of clusters) by treatment arms

Targeted sample size of 200 individuals (100 pairs).

EDIT: Dure to COVID related closure of our lab mid-way, we have increased the sample size.

88 (44 pairs) - Pre-COVID.

212 (106 pairs) - Post-COVID

As the experiment remains unchanged, we expect to pool the sample for a total size of 300 (150 pairs, 50 in each treatment).

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

Supporting Documents and Materials

IRB