Effects of Female-Targeted Cash Transfers on Experimental Measures of Agency in Household Decision-Making

Last registered on October 04, 2018

Pre-Trial

Trial Information

General Information

Title
Effects of Female-Targeted Cash Transfers on Experimental Measures of Agency in Household Decision-Making
RCT ID
AEARCTR-0003360
Initial registration date
September 27, 2018

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 04, 2018, 10:09 PM EDT

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

Locations

Region

Primary Investigator

Affiliation
International Food Policy Research Institute

Other Primary Investigator(s)

PI Affiliation
World Bank - Africa Gender Innovation Lab
PI Affiliation
University of Maryland
PI Affiliation
World Bank - Africa Gender Innovation Lab
PI Affiliation
Stanford University
PI Affiliation
World Bank - Africa Gender Innovation Lab

Additional Trial Information

Status
Completed
Start date
2015-04-01
End date
2018-07-29
Secondary IDs
Abstract
We detail the design of lab-in-the-field experiments that have been carried out on married couples in two rural locations of northern Nigeria. We measure women’s (as well as men’s) agency in household decision-making in different decision domains which include the choice of gender-specific privately consumed goods and household public goods as well as money allocation decisions between self and spouse. We also consider the choice of food and drinks for immediate consumption within the experimental session. Our main experimental measure of agency is the propensity to not defer decision-making to one’s spouse (in the absence of any communication). We also investigate other potential measures of agency: propensity to make a suggestion to one’s spouse that is closer to own preferences, the likelihood to not revise earlier stated allocation decision after receiving communication from spouses, etc. By varying whether participants’ decisions (such as those mentioned above) can be observed by their spouses, we measure whether there is a demand for agency in the absence of spousal observation versus an ability to actually exercise one’s agency in household decision-making when decisions are observed by one’s spouse. Combined with data from a 15-month-long unconditional cash transfer (UCT) program which was targeted at women and carried out as a Randomized Control Trial, these lab experiments in the field, additionally, can provide evidence on the potential impact of receiving an UCT on women’s agency in household decision-making, one year after the UCT program ended. We hypothesize that UCT-receiving women will exhibit higher agency according to these experimental measures than non-recipients. Some likely channels include greater experience dealing with cash as well as higher involvement with income generating activities.
External Link(s)

Registration Citation

Citation
Bakhtiar, M. Mehrab et al. 2018. "Effects of Female-Targeted Cash Transfers on Experimental Measures of Agency in Household Decision-Making." AEA RCT Registry. October 04. https://doi.org/10.1257/rct.3360-1.0
Former Citation
Bakhtiar, M. Mehrab et al. 2018. "Effects of Female-Targeted Cash Transfers on Experimental Measures of Agency in Household Decision-Making." AEA RCT Registry. October 04. https://www.socialscienceregistry.org/trials/3360/history/35276
Experimental Details

Interventions

Intervention(s)
Villages were randomly assigned to become either a (pure) UCT village or a (Program+UCT) village. In both types of villages, eligible households were randomized into receiving or not receiving the UCT. However, in (Program+UCT) villages, all eligible households had access to most of the program components, which included nutrition messages, savings groups, farmer’s groups, etc.

We carried out lab experiments in the field on a subset of this impact evaluation sample which consists of 503 households, with 251 households that did not receive a UCT and 252 households that did.
Intervention Start Date
2015-09-01
Intervention End Date
2017-03-01

Primary Outcomes

Primary Outcomes (end points)
Our main experimental measure of agency is the propensity to not defer decision-making to one’s spouse (in the absence of any communication). We also investigate how likely women are to not revise their earlier stated allocation decision after receiving communication from spouses, as another potential measure of their agency.
Primary Outcomes (explanation)
Mean number of deferred decisions to spouse

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Villages were randomly assigned to become either a (pure) UCT village or a (Program+UCT) village. In both types of villages, eligible households were randomized into receiving or not receiving the UCT. However, in (Program+UCT) villages, all eligible households had access to most of the program components, which included nutrition messages, savings groups, farmer’s groups, etc.

We carried out lab experiments in the field on a subset of the impact evaluation sample which consists of 503 households, with 251 households that did not receive a UCT and 252 households that did. We targeted villages that had a school nearby where we could invite married couples from the evaluation sample. We could have up to 14 married couples per session. Depending on the sample size and availability of respondents in the nearby study village(s), we conducted one or two sessions in every experimental site. In total, 38 experimental sessions were carried out in 22 different sites. Participants from 27 villages attended the experimental sessions. Households were invited around four to five days before the day of the experimental session. Only married couples were invited to participate.
Experimental Design Details
Randomization Method
Public lottery
Randomization Unit
Household level
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
In total, 38 experimental sessions were carried out in 22 different sites. Participants from 27 villages attended the experimental sessions.
Sample size: planned number of observations
503
Sample size (or number of clusters) by treatment arms
Non-UCT/Control HHs: 251
UCT HHs: 252

Lab-experimental treatments:
Secret: 255 HHs
No-Secret: 248 HHs
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
To achieve a statistical power of 90% to detect an effect of 10% on the mean of deferring decision-making to one’s husband (assuming a 79% deferral rate for control and 69% in treatment - which is a more conservative assumption than the pilot estimates) and with a SD of 0.297 (which is again calculated from the pilot sample), we need a sample of 372 - half in the UCT/treatment and the other half in non-UCT/control households. We targeted a higher sample size because of our plans to carry out “Secret” and “No-Secret” experimental treatments.
IRB

Institutional Review Boards (IRBs)

IRB Name
The University of Maryland College Park (UMCP) IRB
IRB Approval Date
2018-04-02
IRB Approval Number
1112913-2
Analysis Plan

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

Request Information

Post-Trial

Post Trial Information

Study Withdrawal

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

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials