Using a Mobile Application to Connect Schools and Farmers for Improved Meal Provisions: Experimental Evidence from a Pilot Study through Guatemala's School Feeding Program

Last registered on February 24, 2023


Trial Information

General Information

Using a Mobile Application to Connect Schools and Farmers for Improved Meal Provisions: Experimental Evidence from a Pilot Study through Guatemala's School Feeding Program
Initial registration date
January 25, 2023

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
January 30, 2023, 1:20 AM EST

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

Last updated
February 24, 2023, 12:51 PM EST

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



Primary Investigator

UC Berkeley

Other Primary Investigator(s)

PI Affiliation
World Bank
PI Affiliation
World Bank
PI Affiliation
World Food Programme
PI Affiliation
World Food Programme

Additional Trial Information

On going
Start date
End date
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
This study evaluates the impact of a mobile app designed to reduce frictions in agricultural output markets in the context of a government-led home-grown school feeding program in Guatemala. To prepare school meals, schools make purchase orders of ingredients using the app which are then shared for local farmers to make competitive offers. In coordination with the government and World Food Programme, we conduct a municipality-level random assignment where 30 municipalities are introduced to the mobile app (‘’treatment group”) and 29 municipalities are not (“control group”). The study leverages administrative data, high frequency phone survey, and in-person endline surveys to measure the impacts of introducing the mobile app on take-up, readiness, and procurement behaviors (e.g. adherence to the government-proposed menus, procurement efficiency, quantity and prices of food by types of sellers).
External Link(s)

Registration Citation

Heirman, Jonas et al. 2023. "Using a Mobile Application to Connect Schools and Farmers for Improved Meal Provisions: Experimental Evidence from a Pilot Study through Guatemala's School Feeding Program." AEA RCT Registry. February 24.
Sponsors & Partners

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

Request Information
Experimental Details


To encourage market participation from family farmers and to facilitate market transactions, the Guatemala WFP CO developed the School Feeding Management Application (SFMA or ’the app’), a smartphone application that connects schools and registered suppliers in the nearby area. Without the app, schools are often dependent on large-scale agricultural producers and/or intermediaries, who make it difficult for schools to procure directly from small-holder farmers. Introduced in 2017, the Guatemalan Ley de Alimentación Escolar (LAE) aims to both ensure that all Guatemalan primary-school children have access to nutritional school meals as well as increasing procurement from local small-holder farmers (suppliers) to 50% of purchases. The app thus assists in facilitating the connection between schools and local suppliers in support of the LAE, allowing schools and suppliers to easily submit orders and place bids, respectively, on this digital marketplace.

There are several goals this app aims to achieve. First, to strengthen the school procurement process by increasing the supply of locally produced nutritious food in thousands of schools in Guatemala where the app is implemented. Second, to increase its efficiency by reducing search and transaction costs. Finally, to improve the local economy by encouraging suppliers to sell food to schools.

The pilot will be implemented in six departments of Guatemala (Alta Verapaz, Chimaltenango, Chiquimula, El Progreso, Escuintla, and San Marcos). Training will be organized in schools with OPFs and suppliers to cover the importance of the LAE, Guatemala’s school feeding law, and compliance with the requirements of the law. MINEDUC and MAGA will also provide additional detailed training on the new procurement app to promote the use of the app among OPFs and suppliers.

This app can potentially support the LAE in developing agricultural markets in predominantly agricultural communities, which would positively impact local cooperatives and farmers. If the app improves the efficiency of market transactions, then it may also lead to increased quality and diversity of meals such that they adhere to the government-suggested menu.

While treatment and control status for schools and suppliers are decided by the treatment status of the municipality they belong to, both control and treatment schools and suppliers will receive training and be reminded on the importance of the LAE, compliance with the requirements of the law, and general procurement process. The only difference between the control and treatment groups is that treatment schools and suppliers will be trained on the use of the new SFMA app. Providing training to both control and treatment groups ensures that any changes in outcomes is attributable to the differences in the training content.
Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)
Service delivery; Food procurement; Supplier production and sales; App experience (treatment group only)
Primary Outcomes (explanation)
Service delivery (food) examines the outcomes related to the delivery of school meals.

Food procurement examines the way in which schools procure food items from local smallholder farmers, the efficiency and reliability of these purchases, and relationships between schools and suppliers.

Supplier production and sales examines the changes in supplier production quantity, revenues, and profits, as well as interaction with schools vs other buyers in the markets.

App experience (treatment group only) examines satisfaction/efficiency of the app itself for both schools and suppliers.

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The design of the app roll-out is a randomization at the municipality level of the 59 municipalities in the study, including 210 sampled schools and up to 535 MAGA registered suppliers in six departments of Guatemala.

Of the 59 municipalities, 30 municipalities were then randomly assigned to the treatment group and 29 to control. This corresponds to 210 schools as part of the impact evaluation sample, divided as follows: 108 schools in the treatment group and 102 schools in the control group. All 535 MAGA registered suppliers in the 59 municipalities will be targeted; 271 registered suppliers in the treatment group and 264 registered suppliers in the control group.
Experimental Design Details
Randomization Method
The sampling and randomization were conducted using Stata software.
Randomization Unit

Treatment group: 30 municipalities. 108 schools and up to 271 registered suppliers.

Control group: 29 municipalities. 102 schools and up to 264 registered suppliers
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
59 municipalities in Guatemala (340 municipalities total in Guatemala).
Sample size: planned number of observations
210 schools are sampled from the 59 municipalities (326 total schools in the 59 municipalities). 535 registered suppliers are sampled from the 59 municipalities (535 total registered suppliers in the 59 municipalities).
Sample size (or number of clusters) by treatment arms
Treatment: 108 schools and 271 suppliers in 30 municipalities

Control: 102 schools and 264 suppliers in 29 municipalities
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Power calculation is conducted using the administrative data on invoices from previous purchases. The MDE of this IE considers 80% power and a significance level of 5%. The share of schools that comply with the law (i.e., buy 50% or more from local farmers) is 26%. The IE will be powered to detect a 24-percentage point (p.p.) increase in that share (i.e., the share of compliers in the treatment group would have to be 50% or more). The share of purchases from local suppliers is 27%, and the MDE for this evaluation will be able to detect a 15 percentage points increase. The mean purchase value is estimated to be Q116,000. This IE is powered to detect an increase of 28% in the total value of purchases. However, a substantial increase in the total meal values is not expected, as the budget for school meals is fixed, and the number of children that receive school meals and/or take-home rations does not change. The Household Dietary Diversity Score (HDDS) is the sum of 12 food groups and is represented on a scale from 0 to 12. Each food group is assigned '1' if the school purchased at least one ingredient in that food group and '0' otherwise. The average HDDS of the most recent invoice by the school (defined by a combination of the log date and the largest invoice number) is 4.45 out of 12. The IE is powered to detect a 0.67-point change (that is less than one new food group per school) increase in score. A similar HDDS (4.87) and MDE (0.64) for the average HDDS of all invoices by school is observed.

Institutional Review Boards (IRBs)

IRB Name
Solutions IRB
IRB Approval Date
IRB Approval Number


Post Trial Information

Study Withdrawal

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

Request Information


Is the intervention completed?
Data Collection Complete
Data Publication

Data Publication

Is public data available?

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials