Tournaments in Non-Routine Analytical Tasks

How to cite

Schindler, David; Englmaier, Florian; Grimm, Stefan; Grothe, Dominik; Schudy, Simeon. 2018. "Tournaments in Non-Routine Analytical Tasks." AEA RCT Registry. AEARCTR-0002916. https://doi.org/10.1257/rct.2916-2.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

David Schindler* , Florian Englmaier , Stefan Grimm , Dominik Grothe , Simeon Schudy

* Primary investigator

Registry ID
AEARCTR-0002916
DOI
https://doi.org/10.1257/rct.2916-2.0
First registered
Last updated
Status
In development
Trial dates
2018-04-23 – 2018-07-13
Countries
Private
JEL codes
C92, C93, J33, D03, M52
Registry record
http://www.socialscienceregistry.org/trials/2916

Abstract

I this study, we investigate the effect of rankings and tournament incentives on team performance in non-routine, analytical team tasks. We use a field setup in which groups of teams work on a non-routine analytical team under varying incentives. The task makes teams face complex and novel problems, it is analytical and cognitively demanding as teams need to collect and recombine information which requires to think outside the box to succeed, and it is interactive as team members have to collaborate, discuss possible actions, and develop ideas jointly with their team members.

Intervention

We conduct four treatments: Treatment T1 serves as a control condition in which we do not intervene, but simply record data. T2 in which teams are asked to choose a name for their team only. T3 in which we post teams' relative performance within the week to Facebook. T4 in which the team with the highest relative performance within the week receives a monetary reward.

Experimental design

We conduct the experiment with our partner in a setting that resembles non-routine and analytical team tasks. Subjects are not aware that they participate in an experiment. Treatments are randomized by week over the sampling period. We record completion time, several background characteristics and variables related to the task.

Design details

Randomization method
Block randomization (by week) using random number generator (computer)
Randomization unit
Treatments will be randomized by week.
Clusters
Approximately 8 weeks. See explanation wrt planned number of observations
Observations
360-400 teams of participants (approximately 1,200 participants, the number of participants is not controlled by us, as we record observations on the team level and teams form endogenously). We schedule data collection aiming at 360 observations (taking into account that some participants may not show up). We stop scheduling new data collection after reaching 360 observations and only collect the data for which we have already scheduled collection.
Arms
90-100 teams in T1, 90-100 teams in T2, 90-100 teams in T3, 90-100 teams in T4
Primary outcomes
Time needed until completion of task
Secondary outcomes
Hints taken during the task

Cited by

No citation data available for this registration yet.

Keywords