NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
Her Time: a time use study of women participating in livelihoods programs in Ethiopia
Initial registration date
June 03, 2020
July 30, 2020 10:49 AM EDT
University of Georgia
Other Primary Investigator(s)
Additional Trial Information
This pre-analysis plan outlines four groups of pre-specified hypotheses for analyzing data collected via an umbrella project referred to as ‘Her Time.’ The primary research questions relate to methodologically validating high-frequency phone surveys vis-a-vis single instance in-person surveys relying on 24-hour or 7-day recall periods that are standard in the literature. We use the cases of time use, dietary diversity and eating practices, sanitation and hygiene, and financial transactions. In addition, we test the implications for women’s time burden of a simple information treatment designed to update men’s beliefs about their peers’ acceptance of male participation in household chores. Finally, we compare women’s responses to sensitive gender norms and empowerment questions elicited via different collection methods. Data collection is scheduled in Amhara, Ethiopia from February 2020 - January 2021. Due to COVID-19, data collection was suspended in March 2020.
McCullough, Ellen and Tamara McGavock. 2020. "Her Time: a time use study of women participating in livelihoods programs in Ethiopia." AEA RCT Registry. July 30.
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
For the methodological experiment addressing time use, our primary outcomes relate to women's time use share by activity and participation (minutes per day) by household members in unpaid household chores.
For the methodological experiment addressing diets, our primary outcomes relate to women's diet diversity score, household diet diversity score, prevalence of food consumption away from home, prevalence of fasting, hand-washing frequency and use of soap, drinking water treatment, drinking water storage, prevalence of diarrhea, and the total number and value of household financial transactions by type. For the gender norm's experiment, our primary outcomes relate to men's intentions to help with household chores and men's participation in household chores.
For the women's life experience methodological experiment, our primary outcomes are a physical violence norms index, emotional and physical well-being norms index, and freedom of movement norms index.
Primary Outcomes (explanation)
We explain how we will construct each outcome variable in detail in our pre-analysis plan document.
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Our research focuses on a population of poor women in Ethiopia who are participating in multifaceted graduation-from-poverty programs. Within each graduation-from-poverty village level group (VESA) selected randomly from the implementing NGOs' program administration records, we recruit one woman from each unique household in the VESA, regardless of whether the woman is a VESA participant herself or not. The recruited female respondents will be randomly assigned to one of three possible treatment arms, which will determine which questions will be asked of her via a series of phone calls and which questions will be asked of her via in-person interview (both types of data
collection, in-person and phone, occur during the 7-day period following the conclusion of baseline interviews within each VESA). Within each treatment arm, we further assign each household to a behavioral intervention addressing gender norms. Each household will also be randomly assigned an interview time slot for both the baseline and validation visits and a survey method for a module addressing women’s life experiences.
1. Phone Time Use Treatment (PT): Respondents will be issued cell phones and support for charging them. Enumerators will call 5 times per day over each respondent’s survey period, which will last 8-12 days. During each brief call, enumerators will ask about primary and secondary time use activities. During the in-person visit, enumerators will use traditional survey recall instruments to collect data on food consumption (as well as financial transactions, sanitation and hygiene) via traditional recall survey instruments.
2. Phone Diets Treatment (PD): Respondents will receive cell phones and support for charging them. Enumerators will call 4 times per day over the respondent’s survey period, which will last 8-12 days. During each brief call, enumerators will ask about either food consumption,
financial transactions, or sanitation and hygiene. During the in-person visit, enumerators will ask about time use via traditional recall survey instruments.
3. "In Person" group (CC for "control"): Respondents will not receive cell phones at the beginning of the study, nor will they receive high frequency interviews over the phone. During an in-person visit, enumerators will ask about time use via traditional recall survey. Enumerators will also ask in person about food consumption, financial transactions, sanitation and hygiene. For the sake of equality, women assigned to the CC group will be issued a phone after they complete the in person validation interview and will continue to be surveyed (for
a separate study) at low frequency in the months following the validation visit.
Experimental Design Details
We randomize the following:
1. For the survey wave, we randomly assign each VESA group to one of eight survey waves (in office, by computer). 2. For the methodological experiment, we use VESA administrative lists to randomly assign each VESA member household to a data treatment (PT, PD or CC) (in office, by computer). We also randomly assign each member household to a day of week for the validation interview (in office, by computer, orthogonal to PT/PD/CC treatment). If additional members are present who are not on administrative lists, we first confirm the membership and then assign additional members in the field according to a pre-selected random order for each village. Finally we randomly assign each phone interview (for women who receive high frequency phone calls) to a 24-minute slot within the period of high frequency data collection. 3. For the gender norms experiment, we randomly assign each woman's male partner to the gender norms treatment randomly (in office, by computer), with treatment orthogonal to the woman's data treatment. 4. For the women's life experience experiment, we randomly assign each woman to a survey method for the WLE module (male enumerator in person interview, male enumerator phone interview, female enumerator phone interview) orthogonal to the data treatment. Additional details are provided in the pre-analysis plan.
Randomization is at the individual level for all treatments except for survey wave, which is at the VESA level.
Was the treatment clustered?
Sample size: planned number of clusters
Sample size: planned number of observations
Sample size (or number of clusters) by treatment arms
For the methodological experiment, 300 women are assigned to phone time use, 300 women to phone diets, 300 women to "control" (in person).
For the gender norms experiment, 450 women's male partners are assigned to the gender norms treatment, and 450 are assigned to the gender norms control treatment. For the women's life experience experiment, 300 women receive the women's life experience questionnaire by phone from a female enumerator, 300 receive it by phone from a male enumerator, and 300 receive it in person from a male enumerator.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
See pre-analysis plan.
INSTITUTIONAL REVIEW BOARDS (IRBs)
University of Georgia
IRB Approval Date