Measuring Experimenter Demand

How to cite

Roth, Christopher; Quidt, Jonathan de; Haushofer, Johannes. 2016. "Measuring Experimenter Demand." AEA RCT Registry. AEARCTR-0001248. https://doi.org/10.1257/rct.1248-7.0

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

Christopher Roth* , Jonathan de Quidt , Johannes Haushofer

* Primary investigator

Registry ID
AEARCTR-0001248
DOI
https://doi.org/10.1257/rct.1248-7.0
First registered
Last updated
Status
Completed
Trial dates
2016-05-18 – 2016-05-20
Countries
United States of America ()
Registry record
http://www.socialscienceregistry.org/trials/1248

Abstract

Experimenter demand effects pose an important challenge to understand and interpret results from laboratory and field studies. In this paper, we manipulate participants’ beliefs about the intentions and desires of the experimenters. In particular, we are interested in how far people’s behavior in economic games is elastic to experimenter demand. In our experiment, people are randomly assigned to play either a dictator game, an investment game or to complete a convex time budget. We randomly assign whether people will be exposed to positive, negative or no experimenter demand and whether the choices involve real money or are hypothetical.

Experimental design

In all three different behavioral measures we either induce positive negative or no demand by telling them that they will do us a favor if, for example, they invest more in a project than they normally would. We will now outline the instructions for each of the different games:

Dictator game:
Positive Demand: You will do us a favor if you give more to the other participant than you normally would.
Negative Demand: You will do us a favor if you give less to the other participant than you normally would.
No Demand: We do not induce any demand.

Investment game:
- Positive Demand: You will do us a favor if you invest more in the project than you normally would.
- Negative Demand: You will do us a favor if you invest less in the project than you normally would.
- No Demand: We do not induce any demand.

Convex time budgets:
- Positive Demand: You will do us a favor if you choose to receive more money in seven days than you normally would.
- Negative Demand: You will do us a favor if you choose to receive less money in seven days than you normally would.
- No Demand: We do not induce any demand.

Design details

Randomization method
Randomization done by a computer.
Randomization unit
Individual
Clusters
4500
Observations
4500
Arms
1500 positive demand (750 positive demand with monetary incentives; 750 positive demand no monetary incentives) 1500 negative demand (750 negative demand with monetary incentives; 750 negative demand no monetary incentives) 1500 control (750 control with monetary incentives; 750 control no monetary incentives)
Primary outcomes
Giving in a dictator game; Points invested in an investment game; Choices in convex time budgets.

Cited by

No citation data available for this registration yet.

Keywords