Suchergebnis

Titel des Experiments
Volunteering under Population Uncertainty: A Field Experiment in an Online Labor Market

Autoren
Hillenbrand, Adrian; Werner, Tobias; Winter, Fabian

Kurzbeschreibung des Experiments
In this project, we plan to investigate volunteering situations in a workplace environment. Specifically, we are interested in the influence that population uncertainty and the number of potential volunteers have on the volunteering decision of the subjects.
For this, we want to construct a Volunteer’s Dilemma situation (Diekmann, 1985) in an online labor market. The participants for this field experiment will be recruited using the online platform operated by clickworker.de, which is a German-based competitor to Amazon Mechanical Turk. Subjects participate as part of a standard job on Clickworker without being explicitly told that this job is part of a field experiment. Furthermore, all users of the Clickworker platform are free to work on this task.

The study consists of two main stages. In the first stage, participants will be matched in teams and work individually on a real effort task. For completing this task, they receive each a payoff of 0.90 €. After finishing this task, we will ask all subjects if they want to continue to work on the task. If they decide to keep working on the assignment, all members of their team receive an additional bonus of 0.90 €. All participants, who volunteered, enter the second stage, where they work again on the same task as in the first stage. If they do not volunteer, the task ends for the subjects after the first stage. One volunteer is sufficient to secure the additional benefit for the whole team. Thus, also participants who did not volunteer may receive the bonus, given there is another volunteer in their team. As a result, subjects face the trade-off between freeriding (not volunteering) and hoping that someone else will volunteer or securing a certain bonus for the whole team (volunteering).

The real effort task will be a standard assignment in an online labor market. We will show the participants comments from an online forum and a corresponding picture for each comment. The comments have been made with reference to the picture in previous studies. The theme of the picture is always related to migration, refugees or cultural differences. The pictures are comparable to those that can be found in the media about those topics.
The participants are asked to rate the comments regarding the expressed sentiment. For the first and the potential second stage, participants have to rate 100 comments each.
The comments have been collected in an earlier study as auxiliary data. The ratings of those comments or the comments themselves are not part of our research question. But, they offer a type of task which can be frequently found in online labor markets and thus this creates a natural environment for the participants. Also the ratings of the comments are used for the earlier study. Hence, the work is indeed meaningful and important.
Similar tasks can be frequently found in online labor markets. Participants will be familiar with the content and structure of the job. Thus, the task will not create any unusual risk or other inconveniences for the subjects. Furthermore, all participants are free to leave the task at any time.

Across the different treatments, we will vary the number of people that are matched to one team (3, 30 or 300 persons in expectation in one team) and the degree of population uncertainty (no population uncertainty or population uncertainty). Given population uncertainty, participants will only know an upper and lower bound for the possible size of their team (2 - 4, 20 - 30 or 200 - 400), but not the actual realization, which will be randomly drawn given those predefined bounds. In the treatments without population uncertainty the subjects know the fixed team size.
A minimum of 2400 participants will be recruited from the Clickworker platform and randomly assigned to one of the six treatment groups. Our aim is that 400 subjects will be matched to each treatment. As participants may enter the job simultaneously, it can occur that some treatment groups will consist of more subjects.

The study itself will be implemented in oTree and is expected to run for one week in January 2019. The payoff for the first stage and the possible bonus will be paid to the participants using the service provided by Clickworker once all participants finished the study.

Downloads

Institutional Review Board Certificate (English)