NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
CSR and Real Effort Coordination
Initial registration date
March 01, 2018
March 02, 2018 3:03 PM EST
University of Toronto, Rotman School of Management
Other Primary Investigator(s)
Additional Trial Information
To what degree do prosocial incentives interact with the necessary effort (real or chosen) needed to complete a task? In this companion experiment to "CSR and Coordination", I investigate the impact that real effort provision may have on individual and group coordination outcomes following provision of prosocial (or extrinsic) incentives. Prior research has mostly measured coordination outcomes through "effort choice", where individual choice of effort does not require any significant physical or mental exertion. This study examines the impact of real effort on coordination outcomes through adapting a summation task from Niederle and Vesterlund (2007) and paying participants based on the minimum correct summations performed in their group.
There will be two interventions in this experiment, one for each of the two experimental periods. The first period will involve participants playing a chosen effort "weakest link" game, with earnings either going entirely to charity (the Canadian Red Cross) or entirely to themselves. The second period intervention will vary whether participants play a chosen effort game (i.e., the same weakest link game as in the first period) or a real effort game (i.e., the summation task from Niederle and Vesterlund (2007)). These interventions allow for comparison between the chosen effort results of "CSR and Coordination" with respect to the effort provision type used in the second period.
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
Key outcome variables include (1) number of groups coordinating at the Pareto optimal point by condition, (2) average number choices by individuals by condition, (3) the variation in number choices by condition, (4) differences in the number choices of prosocial and non-prosocial individuals by condition, and (5) any spillover effects in period two occurring for outcomes (1-4).
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Within the second period, I will measure real-time responses of individuals to the displayed "minimum summations correct" based on their current "number correct", which will allow me to have a dynamic understanding of individual coordination behaviors based on ability.
Secondary Outcomes (explanation)
Taking cues from prior studies and practices in the field of behavioral economics and psychology, each of these experimental conditions will target roughly 50 participants, for a total of around 100 participants in the study (Peysakhovich & Rand, 2015; Simmons, Nelson, & Simonsohn, 2013). Randomization will occur at the session level. All experiments in this paper will be performed using zTree software (Fischbacher, 2007).
Participants will be randomly divided into groups of three and the experiment will last approximately one hour. Participants will complete a manipulation check which tests their knowledge of the payoffs of the weakest link game. The first period will consist of ten rounds of the weakest link game, with participants either playing for the Canadian Red Cross or their own earnings. After rounds 1 and 6, participants will be asked what numbers they think were chosen by the other two participants in their group. In the second period, the treatment group will play seven minutes of the summation game and the control group will play another 10 rounds of the same weakest link game. Participants will remain in the same groups as period 1, and all participants will be playing for their own earnings. Participants in the chosen effort (weakest link game) condition will undergo the exact same procedures as were described in period 1. Participants in the summation condition will be asked to sum randomly generated sets of five 2-digit numbers for seven minutes. Participants will be able to view their individual correct number of summations and the minimum number of summations in their group. These values will update whenever the participant clicks "Submit", and there will be no penalty for incorrect or blank submissions. The payoffs from this task will be $0.30 times the minimum number of correct summations in the participant's group. After playing the two experimental rounds of playing the games, individuals will complete a questionnaire that gathers information on demographics, prosociality, risk aversion, Big 5 personality traits, and tendency towards systematizing or empathizing modes of thought (Baron-Cohen, Knickmeyer, & Belmonte, 2005).
Experimental Design Details
Randomization into sessions will be done in an office by a random number generator on a computer.
The unit of randomization will be at the experimental session level.
Was the treatment clustered?
Sample size: planned number of clusters
The planned number of clusters will be around 2 experimental sessions per each of the two conditions.
Sample size: planned number of observations
The planned number of observations is 2,000 (20 number selections for 50 individuals in the control condition and roughly 10 number selections + 10 summations per 50 individuals in the treatment condition).
Sample size (or number of clusters) by treatment arms
There will be around 50 participants in the control and 50 participants in the treatment.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
INSTITUTIONAL REVIEW BOARDS (IRBs)
University of Toronto Research Ethics Board
IRB Approval Date
IRB Approval Number
Post Trial Information
Is the intervention completed?
Is data collection complete?