Understanding the Role of Knowledge in Peak-avoidance Behaviour of Electricity Consumption: Experimental Evidence from Residential Energy Use

How to cite

Wang, Yu; Goette, Lorenz; Xiao, Erte; Wang, Yacan. 2024. "Understanding the Role of Knowledge in Peak-avoidance Behaviour of Electricity Consumption: Experimental Evidence from Residential Energy Use." AEA RCT Registry. AEARCTR-0013098. https://doi.org/10.1257/rct.13098-1.0

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

Yu Wang* , Lorenz Goette , Erte Xiao , Yacan Wang

* Primary investigator

Registry ID
AEARCTR-0013098
DOI
https://doi.org/10.1257/rct.13098-1.0
First registered
Last updated
Status
On going
Trial dates
2024-02-12 – 2024-12-31
Countries
Private
JEL codes
C93, D12, D83, Q2
Registry record
http://www.socialscienceregistry.org/trials/13098

Abstract

Lack of information about energy usage inhibits efficiency in energy conservation but can be overcome by providing simple, low-cost information. This study uses a randomized control trial to test the effect of knowledge information on the peak-avoidance behaviour of electricity consumption, with a design of various rounds of household-level message sending to test the mechanism for persistence in behaviour and habit formation process. We anticipate that our design and findings will help develop low-cost and long-lasting interventions for guiding households to choose peak-avoidance behaviour.

Intervention

1. Control group: Participants receive weather forecast information through SMS every Monday lasting for 8 weeks;
2. Knowledge groups(there are 4 groups with different treatment cycles): Participants receive knowledge information through SMS. The knowledge information includes revealing the peak/off-peak periods of electricity consumption, negative externalities of excessive electricity consumption during peak periods, and suggestions for how to peak avoidance. Message sending lasts for 8 weeks, but different groups have different sending frequencies. In weeks of no treatment, households will receive weather forecast information.

Experimental design

1 Treatment groups
This experiment has five treatment groups, and each group is planned to have 2000 subjects.
Control group
Knowledge group - frequency 1
Knowledge group - frequency 2
Knowledge group - frequency 3
Knowledge group - frequency 4

2 Treatment design
This experiment plans to send SMS messages to all 5 treatment groups every Monday lasting for 8 weeks. Different groups have different sending cycles.

3 Randomization
The data used for randomization is from 10000 households, covering 15 Jan to 21 Jan 2024, including household ID and the electricity consumption in each hour.

Design details

Randomization method
Stratified randomization by STATA
Randomization unit
Household
Clusters
10000 Households
Observations
10000 Households
Arms
2000 Control group 2000 Knowledge group - frequency 1 2000 Knowledge group - frequency 2 2000 Knowledge group - frequency 3 2000 Knowledge group - frequency 4
Primary outcomes
1. Household Electricity Usage - During the Experiment 2. Household Electricity Usage - After the Experiment 3. Peak Avoidance Behaviour - During the Experiment

Cited by

No citation data available for this registration yet.

Keywords