These files contain the data from 3 surveys conducted during the course of the Integrated Soil and Fertility Management Project (ISFM) in Malawi which ran between 2014 and 2019. 

Introduction Project: 

The official name of the project was: The effect of demonstration plots and the warehouse receipt system on ISFM adoption, yield and income of smallholder farmers: a study from Malawi’s Anchor Farms. But we nicknamed it ISFM-Malawi. The original principal investigators include: Ephraim Chirwa, Wezi Mhango, Hope Michelson, Chris Barrett, Cheryl Palm and Annenmie Maertens. 

In this project, we aim to support the Clinton Development Initiative (CDI) in their scale-up of the Anchor Farm Project (AFP). The AFP aims to increase smallholder farmers’ incomes and productivity through the adoption of Integrated Soil Fertility and Management Practices (ISFM), a group of techniques designed to increase the fertility of soils. ISFM includes application of mineral fertilizers, incorporation of organic matter, adoption of agroforestry, crop rotation and intercropping with legumes (such as soy), and use of conservation agriculture practices. ISFM provides potential solutions to increasing smallholder productivity, but to date, little information exists on how to implement these solutions.

CDI approaches the problem of low adoption as follows: first, CDI disseminates production knowledge through the use of demonstration plots, farmer clubs, lead farmers and farmer field days; and second, CDI improves farmers’ access to output markets. When we started this research study in 2014 CDI was operating in three districts in central Malawi. In the coming years, CDI plans to further expand their activities. In this research project we aimed to support this scale-up by providing critical evidence as to the effectiveness of these various components.
Our evaluation ran from 2014 until 2019, using a quasi-Randomized Controlled Trial (RCT), in which some villages, randomly selected, receive certain components of the AFP, while others do not. Using baseline, midline and endline surveys, we established the short- and medium-term impacts of these various AFP components. Specifically, we use the random assignment of 250 villages into (i) a control group and (ii) a treatment group. The villages in the treatment group were requested to form farmer clubs, CDI’s preferred delivery method, and invited to participate in CDI’s extension program activities. While all treatment clubs were invited to farmer field-days, only a subset of clubs were selected to set up farmer demonstration plots. This program ran for two years. In the fourth year of our evaluation, CDI rolled out their marketing treatment. Again, we created a control group and a treatment group, orthogonal to the other two treatment groups, and invited only farmers in the treatment group to participate.
We collected three rounds of data, in 2014, 2015 and in 2018. A unique feature of our panel dataset is that we not only follow households over time, but we also follow fields over time, i.e., each field received a unique ID at baseline, and we followed up on each field to monitor perceived and actual soil fertility, as well as inputs and outputs over time. Using the baseline and midline data, we find that farmers who participated in farmer-led demonstration plots learn about the production processes of ISFM technologies, including the type and amount of pesticide to be used on soy, and this learning leads to increased planned adoption. Farmers invited to attend field-days learn considerably less, and what they learn is conditional on the degree to which they are credit-constrained.
Using the baseline and endline data, we note that the marketing program did not have the intended effects, largely due to the fact that the uptake of the program remained very low. However, we present some evidence of unintended positive effects: as farmers await the implementation of the marketing program, they postpone sales, and fetch a higher price.

Data provided: 

The data packages are in Stata (v13) format, and all related questionnaires can be found under in a seperate folder under the ISFM Malawi account. 

Please note that this data package has been de-identified. All identifying information such as respondent name, village name, GPS coordinates of fields have been removed. Each respondent has been allocated a unique numerical identifier that is consistent through out the dataset.

There are three folders: 

1. Baseline Survey (Fall, 2014): household composition, landholding and assets, beliefs regarding yields, group membership and networks, time preferences, input-output modules per plot, marketing and past agronomic practices (with a focus on ISFM). Please refer codebook_baseline for more information. 

2. Midline Survey (Fall, 2015): updated household composition, networks, knowledge about ISFM, participation in program activities, adoption plans, beliefs regarding yields, input-output modules per plot, and marketing. Please refer codebook_midline for more information.

3. Endline Survey (Fall, 2018): household composition, networks, knowledge about ISFM, participation in the program activities, adoption plans, beliefs regarding yields, input-out modules per plot, marketing, dietary diversity, and landholding and assets. Please refer codebook_endline for more information.

Reports provided: 

We have also made the various project reports available under the same ISFM-Malawi account.