Too Much or Too Little Faith? Behavioral Factors and Opportunity Entrepreneurship in Colombia

Last registered on March 10, 2022

Pre-Trial

Trial Information

General Information

Title
Too Much or Too Little Faith? Behavioral Factors and Opportunity Entrepreneurship in Colombia
RCT ID
AEARCTR-0008835
Initial registration date
March 09, 2022

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 10, 2022, 9:03 PM EST

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

Locations

Region

Primary Investigator

Affiliation
Stanford University

Other Primary Investigator(s)

PI Affiliation
Stanford University

Additional Trial Information

Status
On going
Start date
2021-11-23
End date
2022-06-30
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
We conduct an online lab-in-the-field experiment in Colombia to study behavioral explanations for the high failure rates of opportunity entrepreneurs. We study the decision-making of entrepreneurs by simulating the entrepreneurial entry decision through offering the option to participate in an entrepreneurship game.
External Link(s)

Registration Citation

Citation
Rodriguez Martinez, Andres Felipe and Reka Zempleni. 2022. "Too Much or Too Little Faith? Behavioral Factors and Opportunity Entrepreneurship in Colombia." AEA RCT Registry. March 10. https://doi.org/10.1257/rct.8835-1.0
Sponsors & Partners

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

Request Information
Experimental Details

Interventions

Intervention(s)
Intervention (Hidden)
This study is not an impact evaluation. It is an online lab-in-the-field experiment. Thus, there is no specific intervention; instead, we use various treatments and randomizations described below in the experimental design section.
Intervention Start Date
2021-11-23
Intervention End Date
2022-06-30

Primary Outcomes

Primary Outcomes (end points)
Our primary outcome variables are the binary decisions of playing the game with stakes or choosing the certain lower payment.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Incentive compatible beliefs about success and expected performance
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Participants complete the experiment online via Qualtrics.
Experimental Design Details
Participants complete an online lab in the field experiment. The experiment is implemented via a Qualtrics survey. The main part of the experiment is a game. The game is an entrepreneurship-themed multi-round multi-armed bandit game. The game is implemented such that we can vary the importance of randomness (luck) on the game outcome (relative to skill), as either high or low. Participants get randomly assigned to one of these two game versions: high or low luck importance.
Before playing the game, participants make a series of decisions about whether or not they want to play the game with or without stakes. They decide whether to take a certain COP 5.000 ($1.24) payment regardless of game performance or try winning the high prize, the COP 20.000 payment, conditional on scoring above a specific threshold and receiving no payment otherwise.
Participants make this binary decision 3 or 4 times. At the time of each binary decision, participants also share their non-incentivized beliefs about their likelihood of scoring above the threshold. In-between the decisions, additional information is revealed about their expected likelihood of scoring high enough to win the high prize. These information pieces come from the game's design and our analysis of pilot participants comparable to the experiment participants.
We randomize whether participants have 3 or 4 decisions to eliminate concerns about order effects. Otherwise, the order of the additional information pieces shared is the same. The high or low luck treatment is cross randomized.
Participants who make four decisions are only given basic information about the choice before making their first decision. Before their second decision, they are told that on average, we expect 20% of participants to score high enough to win the COP 20.000 prize. This information is provided for the first decision of participants making only three decisions.
Next, we share with participants their Entrepreneurial Preparedness (EP) score that we have calculated for them based on their answers during the earlier part of the survey. Then, we share how we expect them to score in the game compared with other participants based on their EP score and our analysis of 80 pilot participants' data. (The pilot participants were also Colombian entrepreneurs, and they too had their EP score calculated the same way and played the same game.) At this point, we elicit the participants' incentive-compatible belief about their performance in the game compared to other participants'.
Finally, we share with participants the information about the role of luck in the game and how we expect their performance to range based on their EP score and luck during the game in the top or bottom 10% of all participants' luck.
After participants have made these decisions and are aware of all of the above information, we elicit their incentive-compatible expected likelihood of scoring high enough to win the prize.
Then, they play the game.
Randomization Method
Stratified randomization implemented by Qualtrics. The randomization is stratified based on participants' connection to entrepreneurship (past, present, or aspiring) and their EP score.
Randomization Unit
Individual participant
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
Treatment is not clustered.
Sample size: planned number of observations
As many participants as we can get, up to 500 and no less than 200. We have a very specific target population (Colombian entrepreneurs) who also tend to be busy, making it challenging to guarantee a specific sample size. We will send out at least 2000 invitations to participate in the study; however, we cannot guarantee the exact number of participants.
Sample size (or number of clusters) by treatment arms
We cross randomize participants into the four treatment arms (2 by 2 design). (One-quarter of the total sample falls into each arm)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Administrative Panel on Human Subjects in Non-Medical Research of Stanford University
IRB Approval Date
2020-08-21
IRB Approval Number
52202
Analysis Plan

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

Request Information

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