Back to History

Fields Changed

Registration

Field Before After
Trial Title It’s Showtime! Improving Agricultural Productivity through TV Programming and Improved Input Access Improving Agricultural Productivity through TV Programming and Improved Input Access
Investigator Jonathan Robinson Berber Kramer
Last Published December 06, 2023 09:02 AM February 19, 2025 05:56 AM
Intervention (Public) We will run an RCT in Zambia in which we simultaneously provide farmers with agricultural information and access to inputs. Agricultural information will be provided through the TV show, with farmers in the treatment arm being shown the show via screenings in local villages. Households with TVs will also be encouraged to watch the show on their own, and reminded via text message. In addition, households with smartphone access will be sent YouTube links to the show. Access to inputs will be provided through “roadshows”, in which input and service providers will offer their products for sale at accessible locations in the vicinity of study villages. We will rely on a 2 by 2 cross-randomized study design, for a total of four treatment arms, with the village as the unit of randomization. To implement the roadshow treatment, we need large enough numbers of participants to make participation worthwhile for the private sector. We will therefore identify accessible locations among a given cluster of roughly 8 study villages and invite the 4 in the treatment group to attend the road show. Specifically, we will randomly select the administrative unit of "camp" to create these clusters. We will run an RCT in Zambia in which we simultaneously provide farmers with agricultural information and access to inputs. Agricultural information will be provided through the TV show, with farmers in the treatment arm being shown the show via screenings in local villages. Households with TVs will also be encouraged to watch the show on their own, and reminded via text message. In addition, households with smartphone access will be sent YouTube links to the show. Access to inputs will be provided through “roadshows”, in which input and service providers will offer their products for sale at accessible locations in the vicinity of study villages. We will rely on a 2 by 2 cross-randomized study design, for a total of four treatment arms, with the village as the unit of randomization for the TV show intervention. To implement the roadshow treatment, we need large enough numbers of participants to make participation worthwhile for the private sector, and it will be difficult to prevent participation from farmers in neighboring villages. We will therefore randomly select 10 of the 20 "camps" (the lowest administrative unit above villages), in which we identify accessible locations to organize the roadshow. All farmers in these 10 roadshow treatment camps will be invited to attend the roadshow in their camp.
Experimental Design (Public) We will identify 160 villages in 5 districts in Eastern and Central provinces in Zambia. We will randomize these villages into one of the 4 treatment groups (information only, market access only, information + market access, and control), stratified by district. We will identify 160 villages from 20 camps (8 villages per camp) in 5 districts in Eastern and Central provinces in Zambia. We will randomize assign these villages to the information (TV show) treatment, stratified by district. In addition, we randomly assign two camps per district to the market access (roadshow) treatment, again stratified by district. This results in 4 treatment arms (information only, market access only, information + market access, and control).
Randomization Unit The market access and information interventions will be randomized at the village level. The information interventions will be randomized at the village level. The market access intervention will be randomized at the camp level.
Planned Number of Clusters 160 villages 160 villages and 20 camps
Planned Number of Observations We plan on 10-16 surveys per village, i.e. 1600-2560 surveys total. We sampled 8 villages per camp, and plan on 10-16 surveys per village, i.e. 1600-2560 surveys total across 20 camps.
Sample size (or number of clusters) by treatment arms 40 villages in each treatment group (market access + information, information only, market access only, control) Control: 40 villages from the 10 camps without roadshows (market access treatment) Information only: 40 villages from the 10 camps without roadshows Market access only: 40 villages from the 10 camps with roadshows Market access + information: 40 villages from the 10 camps with roadshows
Public locations No Yes
Back to top

Sponsors

Field Before After
Public No Yes
Back to top
Field Before After
Public No Yes
Back to top