The impact of data collection for farmer organisations

CTA has been working with farmer organisations in six countries in sub-Saharan Africa to implement its Data4Ag project. The aim is to investigate how the collection and effective analysis of farmer data can be used by farmer organisations to improve the livelihoods of their members. This article focuses on the impact of collecting data, and how this information could be systematically incorporated within the monitoring and evaluation systems of farmer organisations.

Impact assessment was based on the example of the Igara Growers Tea Factory Limited (IGTF) in Uganda one of the partners of the Data4Ag project. Before 2018, IGTF used to store farmer member information in an old database, and process member data manually into a spreadsheet. Collecting farmers’ raw tea leaves and weighing them accurately using analogue scales was challenging; farmers were often cheated at collection centres and then received delayed payments. In partnership with CTA, IGTF launched a digital profiling project in December 2018 to help solve these challenges. The project involved compiling geo-referenced information about the tea farmers and their land using GPS-enabled tablets. Extension officers then uploaded the data onto a dedicated online platform (the ONA Platform/Open Data Kit), and subsequently onto the IGTF’s new QGIS database. The profile database is now linked to a financial and accounting system, allowing IGTF to track records of transactions with member farmers. They also connected Internet of Things (IoT) digital weighing scales of at each collection point to this system to reduce fraud. “Now, my green leaf cannot be cheated by leaf collection clerks, I can have my payment when I need it and I can have system reports about my supply and credit statements for months,” says tea farmer Mwenderehi Eliphaz.

Qualitative and quantitative approaches

To measure the impacts of farmer profiling for the IGTF members, CTA’s research team used two approaches; the first was a qualitative assessment approach using an innovation framework inclusion tool. The second was a quantitative approach involving the use of machine learning to create a model to determine the best algorithm to predict the yield of tea leaves, and the use of Statistical Package for the Social Sciences (SPSS) to analyse statistical data.

The qualitative framework inclusion tool was used to describe the processes involved in value creation for the beneficiaries, and outlines five main criteria essential for digital projects: the extent of collaboration, the value to be created, the involvement of users or local community in the design, the digital readiness of the ecosystem, and the availability of human and financial resources. In this case, the value created by the project is increased yields for profiled farmers, which is supported by the findings of the SPSS.

From the analysis of the machine learning data, it was clear that more extensive data were needed to improve the model to optimise yield predictions. There were some missing values in data variables, such as farm size, for example. There was however enough data to conclude that the profiled farmers experienced an increase in yield, both in the first quarter of 2018 and 2019 (data for the other seasons was also not available). The machine learning analysis was also able to investigate the relationship between the variables of credit access and the yield of tea leaves.

From the SPSS analysis, yield differences were compared between profiled farmers and non-profiled farmers, and between farmers with and without access to credit. The results showed that the mean yield for farmers who were digitally profiled and who had access to credit was significantly higher (10%) than farmers who were not profiled and had no financial access. The findings, supported by the results from the machine learning analysis, showed that access to credit was dependent on profiling, with the majority of the profiled farmers gaining access to credit. This can be explained by the fact that profiling made it easier for credit cooperatives to access farmer data and indicate their creditworthiness. Available farmer data also enabled farmers to be targeted with the right agronomic advice and input for production. From the predictive model, farmers with access to credit had higher yields than farmers without access to credit. It was also evident from the analysis that the Data4Ag project employed a gender inclusive approach, as male and female farmers had equal chances of been profiled.

An internal assessment

An internal impact review tool was also employed by the project to provide a practical set of questions that allowed IGTF to methodically assess the current effectiveness of its operations, in regards to impact, cost effectiveness and significant reach. Based on the results of this tool, the profiling can be considered to be cost-effective, but has a low reach (33%) – reach is seen as the ability to scale up to at least 50,000 farmers, while IGTF only has 7,468 members. IGTF is now implementing the farmer profiling model at other farmer-owned cooperatives in the tea sector which will increase the reach.

Building up local expertise to improve systematic analyses of all data collected can support organisations like IGTF to prove their impact and embed systematic analysis within their monitoring and evaluation (M&E) practices. Another important capacity that needs critical attention is data management. This is the first step towards better M&E. Data management involves optimising data collection and organisation processes to ensure the quality, reliability, and timeliness of data. This is vital to derive the correct insights from data to drive better decision-making.

IGTF embraced data analysis and will improve the yield prediction model by continuing to add farmer data, as well as build an algorithm to predict how much nitrogen tea farmers need to apply to their soil. The introduction of farmer profiling connected with digital scales and the financial system has enabled IGTF farmers to increase their competitiveness with multinationals through well-managed farmer data, improved quality control, access to finance and a faster payment process for raw material delivered.