Evaluating a Free LED Lighting Program

Last registered on August 02, 2019

View Trial History

Pre-Trial

Trial Information

General Information

Title

Experimenter Effects in Experimental Program Evaluation

RCT ID

AEARCTR-0004476

Initial registration date

July 31, 2019

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

August 02, 2019, 3:30 PM EDT

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

Locations

Country

United States of America

Region

Primary Investigator

Name

Alec Brandon

Affiliation

University of Chicago

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Justin Holz

PI Affiliation

University of Chicago

Contact Investigator

Additional Trial Information

Status

On going

Start date

2019-07-13

End date

2019-11-30

Keywords

Environment & Energy, Other, Welfare

Additional Keywords

JEL code(s)

Secondary IDs

Abstract

Are experimental program evaluation studies robust to experimenter effects? Across different recruitment waves, we study this question in the context of a free LED lighting study.

External Link(s)

Registration Citation

Citation

Brandon, Alec and Justin Holz. 2019. "Experimenter Effects in Experimental Program Evaluation." AEA RCT Registry. August 02. https://doi.org/10.1257/rct.4476-1.0

Former Citation

Brandon, Alec and Justin Holz. 2019. "Experimenter Effects in Experimental Program Evaluation." AEA RCT Registry. August 02. https://www.socialscienceregistry.org/trials/4476/history/51082

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Recruitment intervention:

Recruitment wave 1: Households will be recruited for a free LED lighting study at their home. Recruitment will be randomly varied across two dimensions. First, whether participants receive a $10 incentive or not. Second, whether a household can opt-out of the study by checking a box on a warning given to them regarding the study.

Recruitment wave 2: Households will be recruited for a free LED lighting study on Amazon Mechanical Turk. Recruitment will be randomly varied across one dimension. Participants will either be told their participation would be a favor to the study or not.

Program intervention: A free LED lighting program that gives recipients 8 free LED light bulbs

Intervention Start Date

2019-07-13

Intervention End Date

2019-11-30

Primary Outcomes

Primary Outcomes (end points)

Recruitment wave 1: Door answer, study participation, and energy usage (all hours and evening hours, 6pm to Midnight)

Recruitment wave 2: Study participation and energy usage (all hours and evening hours, 6pm to Midnight)

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

See intervention description.

Experimental Design Details

Randomization Method

Computer

Randomization Unit

Recruitment wave 1: Blocks of 20-30 households will be assigned to different recruitment treatments. Within each recruitment treatment, households will be randomized at the household level for the free LED light bulbs.

Recruitment wave 2: Individual

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

Recruitment wave 1: Approach 5,000 households. Recruit 100 participants.

Recruitment wave 2: Recruit 500 participants.

Sample size: planned number of observations

Recruitment wave 1: Approach 5,000 households. Recruit 100 participants. Recruitment wave 2: Recruit 500 participants.

Sample size (or number of clusters) by treatment arms

Recruitment wave 1: Approach 5,000 households. Recruit 100 participants.

Recruitment wave 2: Recruit 500 participants.

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Recruitment wave 1 will be powered to observe the door answer effect observed in DellaVigna, List, and Malmendier (2012). Recruitment wave 2 will be powered to observe the strong demand effect in de quidt, Haushofer, and Roth (2018). Across both recruitment waves, simulations suggest this study will be powered to detect a 0.07 kWh effect of the LED lighting program during evening hours.

Supporting Documents and Materials

IRB