Impact of Information on Domestic water saving among urban households in Tanzania

Last registered on March 16, 2023

View Trial History

Pre-Trial

Trial Information

General Information

Title

Impact of Information on Domestic water saving among urban households in Tanzania

RCT ID

AEARCTR-0010285

Initial registration date

October 27, 2022

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

October 31, 2022, 3:56 PM EDT

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

Last updated

March 16, 2023, 5:48 AM EDT

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country

Tanzania, United Republic of

Region

Dar es Salaam

Primary Investigator

Name

JULIETH TIBANYWANA

Affiliation

UNIVERSITY OF CAPE TOWN

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

EDWIN MUCHAPONDWA

PI Affiliation

UNIVERSITY OF CAPE TOWN

Contact Investigator

PI Name

ELIZABETH ROBINSON

PI Affiliation

LONDON SCHOOL OF ECONOMICS AND POLITICAL SCIENCE

Contact Investigator

Additional Trial Information

Status

On going

Start date

2022-11-24

End date

2023-10-24

Keywords

Behavior, Other

Additional Keywords

JEL code(s)

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

Many parts of the world have been experiencing poor quality water, which has not been sufficient, especially at the domestic level where water usage keeps on increasing over time. It is expected that the demand for water will be greater than the supply by 40% as a result of a fast-growing population and poor water management. Tanzania is among the developing countries that have been facing water scarcity for households living in urban areas and for those that have access to it, its availability is unreliable, irregular, and highly erratic, and as a result, households do not consider water from authorities as the main source. This scarcity of water and its unreliability calls for the need to conserve water as an important natural resource. Therefore, this study proposes to use randomized control trials to investigate whether the provision of information on water usage can influence urban domestic households water saving behavior. This study will disseminate knowledge to local authorities and the public about the importance of being accountable in the use and management of water resources in line with the Tanzania’s development vision of 2025 on effective utilization and management of natural resources. This will guide policy decisions that aim at promoting efficient use of water. The paper will use baseline data from the government water authority and survey data which will be collected in Dar es Salaam city, Tanzania. A sample size of 900 urban households will be surveyed while during an intervention, 1600 households will be used.

External Link(s)

Registration Citation

Citation

TIBANYWANA, JULIETH, EDWIN MUCHAPONDWA and ELIZABETH ROBINSON. 2023. "Impact of Information on Domestic water saving among urban households in Tanzania." AEA RCT Registry. March 16. https://doi.org/10.1257/rct.10285-4.0

Sponsors & Partners

Interventions

Intervention(s)

The experiment will involve sending monthly information to households for six months (January, February, March, April, May and June 2023). Wards from Dar es Salaam will be randomly assigned into treatment and control group. We will have one treatment and one control group. In the treatment group, households will receive information regarding their own water consumption for that month in a comparison with the median average neighborhood water consumption in the ward that the household is living. The control group will only continue to receive their normal monthly water bills without extra information.

Intervention Start Date

2023-01-24

Intervention End Date

2023-06-24

Primary Outcomes

Primary Outcomes (end points)

The household average monthly water consumption.

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

The household average monthly electricity consumption.

Secondary Outcomes (explanation)

Experimental Design

The targeted population for the study is high water users who are connected with the water authority. A total of 256,661 households are connected to water excluding inactive and disconnected customers. We use the median value to obtain high water users. Any household that is above the median water consumption will be considered a high user. Households with bulk meters such as flats and institutional apartments are treated as outliers and hence are not part of the study. To avoid spillover effects, randomization will be done at the cluster level. For this case, wards will be used as clusters. In Dar es Salaam, we have more than 50 wards, hence a minimum number of clusters is guaranteed.
To check for the orthogonality assumption, balance tests will be used. The variables that will be considered for balance tests include average monthly household water consumption, median monthly water consumption, collection value.

Experimental Design Details

The main data source to be used for analysis will be obtained from the water authority (DAWASA) before and after the intervention. We will collect follow-up survey data from the water authority for a total of six months which also includes the intervention period( January 2023, February 2023, March 2023, April 2023, May 2023 and June 2023). If time allows, we will continue intervention in August, September, and October. Baseline data for some households will be used as an addition to test the balance and use covariates in the regression framework to increase the precision of our estimates.
Randomization will be done at the ward level to minimize spillover effects among urban households, hence power calculation with clusters is used to detect the minimum impact and the minimum sample required for the study. According to Khandker et. al (2010), the minimum number of clusters is 30 per arm. Data from the water authority indicates there are more than 100 wards in total.

Randomization Method

Randomization will be done using STATA software in the office by computer. We have a list of all households connected to piped water from government water authority named DAWASA( Dar es Salaam Water and Sewerage Authority). The lists has variables like name of the household head, contact information, geographical location, monthly water consumption in cubic meters, billing and collection value to mention few.

Randomization Unit

Unit of analysis is the Household.
Randomization will be done at cluster levels (wards).
I

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

The planned number of clusters is 80 wards in Dar es Salaam city.

Sample size: planned number of observations

The planned number of observations is 1600 households. In addition, we expect to survey 900 households out of 1600 to gather more information.

Sample size (or number of clusters) by treatment arms

We expect to have 40 wards in control group and 40 wards in treatment group.

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

We used a sample proportion of 0.6 for sample 1 and 0.7 for sample 2. We assumed a significance level of 5% and statistical power of 80% to be able to detect the impact. Given the average cluster size of 12, coefficient of variation of cluster size of 0.9, and intra-cluster correlation of 0.01, the total number of clusters was calculated to be 37 per arm. The minimum detectable impact for our main outcome variable is expected to be around 2 to 5% reduction in water consumption.

Supporting Documents and Materials

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

IRB