Coordination and Information Acquisition: An Experiment

Last registered on May 17, 2019

Pre-Trial

Trial Information

General Information

Title
Coordination and Information Acquisition: An Experiment
RCT ID
AEARCTR-0004039
Initial registration date
April 18, 2019

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
May 16, 2019, 4:39 PM EDT

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

Last updated
May 17, 2019, 5:13 AM EDT

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

Locations

Region

Primary Investigator

Affiliation
Nazarbayev University

Other Primary Investigator(s)

PI Affiliation
Lund University

Additional Trial Information

Status
In development
Start date
2019-05-01
End date
2019-10-01
Secondary IDs
Abstract
There are numerous instances where people can only achieve their goals by coordinating their actions. Examples range from friends choosing a place for an evening out, through traders on financial and foreign exchange markets, to protesters against a political regime. The resulting coordination game that is often used to model such situations has multiple equilibria under complete information. Yet, information is rarely complete and the structure of the game is rarely a common knowledge. Even if a trader knows the fundamental value of a stock he can never be sure that it is also known to other traders, and a citizen going to an anti-government march can only guess how many people will gather on the streets. Moreover, in many of those situations information is not exogenous to the decision, and agents can choose to learn about fundamentals.
This leads to the question of how information acquisition affects game equilibria, and ultimately the agents' behaviour. Morris and Yang (2016) argue that uniqueness of the equilibrium is tightly connected to the technology of information acquisition. They show that a coordination game has a unique equilibrium whenever it is sufficiently easier to distinguish states that are further apart than nearby ones. Our goal is to test their insight experimentally.
External Link(s)

Registration Citation

Citation
Goryunov, Maxim and Alexandros Rigos. 2019. "Coordination and Information Acquisition: An Experiment." AEA RCT Registry. May 17. https://doi.org/10.1257/rct.4039-2.0
Former Citation
Goryunov, Maxim and Alexandros Rigos. 2019. "Coordination and Information Acquisition: An Experiment." AEA RCT Registry. May 17. https://www.socialscienceregistry.org/trials/4039/history/46671
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2019-05-01
Intervention End Date
2019-10-01

Primary Outcomes

Primary Outcomes (end points)
Probability of choosing the risky action in a coordination game conditional on the state of the world. The state of the world at which the probability of risky action exceeds 1/2.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Participants in the experiment are randomly matched into pairs to play a coordination game. Each of them has a choice of two actions: either participate (P) in the project or abstain (A). Abstention generates fixed payoff (riskless action). Payoff from participation (risky action) depends on the unknown state of the world and the other participant's choice. Treatment vary in the way in which participants are provided with information about the unknown, payoff-relevant state of the world. Specifically, in the Continuous Treatment (CT), information provision induces continuous probability of choice of the risky action. In the Discontinuous Treatment (DT), information provision allows for discontinuity in the choice probability of the risky action. The main interest is in the differences in behavior between CT and DT.
Experimental Design Details
In the Continuous Treatment (CT), on the screen, participants see a rectangle that represents the uniform distribution of the state of the world over interval [0, 3000]. At the left-most and right-most points of the rectangle the values of the extreme coordinates (0 and 3000) are shown. No other coordinate is visible. A random point is drawn from within the rectangle. The horizontal coordinate of the drawn point is the realization of the state of the world. The drawn point is displayed on the screen for a limited amount of time. Then, it disappears, and experimental subjects must choose their action while observing the empty rectangle.

If the dot’s Position is larger than or equal to 2000 and an experimental participant decides to participate, he/she receives 0 points. If the dot’s Position is larger than or equal to 1000 and less than 2000 and the experimental participant decides to participate, he/she receives (2000 - dot's position) points if his/her partner decides to participate, and receives 0 points if his/her partner decides to not participate. If the dot’s Position is smaller than 1000 and the experimental participant decides to participate, he/she receives (2000 - dot's position) points if his/her partner decides to participate, and receives (1000 - dot's position) points if his/her partner decides to not participate. The participant receives 300 points if he/she decides to not participate, independently of the dot's position.

In the Discontinuous Treatment (DT), the payoff structure is the same. Subjects observe a similar box. However, in this treatment they have a vertical line that they can place anywhere inside the box. Effectively, the line is expected to serve as a choice of an exact threshold. In view of the fact that the discontinuous treatment is closer to full information environment studied in the literature, we view the participants in the DT as our control group and the participants in the CT as our treatment group.
Randomization Method
Randomization is done by a computer in the experimental laboratory
Randomization Unit
First, experiment participants are randomized into sessions. Secondly, within sessions, participants randomized into a specific treatment are randomly split into two groups, within which randomized pair matching takes place.
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
28 groups
Sample size: planned number of observations
196 participants
Sample size (or number of clusters) by treatment arms
14 groups (98 participants) in each of the two treatments (Continuous and Discontinuous).
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
The sample size was choosen to have minimal detectable effect at 0.5 standard deviation of the threshold at which subjects choose risky action with probability more than 1/2. The estimates of the standard deviation of the treshold were taken from previous studies of coordination behavior.
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials