Steinke et al. (2019): Prioritizing options for multi-objective agricultural development through the Positive Deviance approach. PLoS ONE 14(2): e0212926.
The wired door gives a soft groan as Mariam Ally enters her brickwork chicken coop in the village of Mchengamoto, to present her invention. Like many smallholder farmers in the Tunduru district of Tanzania, not far from the Mozambican border, the 33-year-old lady makes some money by selling poultry to the traders who pass her village. But unlike other farmers, Ms. Ally had a unique idea to make the most of her small-scale production: Having discovered that artificial light keeps the chickens active beyond sunset, she installed a small solar panel to the coop’s roof. A light bulb now extends the light of day for the chicken, causing them to eat more and fatten more quickly. The increased sales soon repaid her initial investment.
Mariam Ally is a “positive deviant”, part of a recent study led by Bioversity International and funded through SAIRLA, which investigated innovative farm-related behaviours in Tanzania. The study addressed a challenge faced by agricultural development organisations around the world: Which practices should be promoted to local farmers to support sustainable development? Often, promising innovations “imported” from other places see only weak adoption, and can have unexpected downsides in new target context.
Positive deviants: Some farmers get more out of their farms than others
The Positive Deviance approach takes a different perspective: Working with particularly innovative individuals, researchers can directly identify practices that contribute to intensified production under local constraints and opportunities. The concept, originally developed to identify good child care practices in Asia, is simple:
The grass isn’t always greener on the other side. And yet, somebody in the village has the greenest grass, despite facing the same challenges as others. So what can be learnt from that high performer?
In all rural communities, there are some farmers who are creative and apply innovative practices, leading to better outcomes. These practices, just like Ms. Ally’s solar coop lamp, could be smart options for other local farmers, too. In practice, however, researchers cannot detect innovative practice directly, as this would require large-scale surveillance. More easily, research can quantify the performance of many rural households and then identify the outliers – farmers whose performance seems to stick out from the crowd. With these individuals, there is a high chance to observe locally viable, innovative practice.
Agricultural development must address multiple concurrent objectives
In the new SAIRLA-funded study, researchers used a quantitative survey of over 500 farming households in Southern Tanzania to identify households with outstanding performance. Then, they re-visited some of these positive deviants for extensive farm observations and interviews, to identify uncommon, innovative behaviours.
But the complex reality of smallholder agriculture precludes too straightforward analysis: Farming is a whole way of life, involving interconnected decisions about scarce household resources and competing goals. If Positive Deviance focuses narrowly on the most productive households, it might lead to practices with adverse implications, for example, for household income or the environment.
To achieve Sustainable Agricultural Intensification, households must deal successfully with inevitable trade-offs, for example, between productivity and diversity, or between productivity and environmental stewardship.
Thus, the recent study defined positive deviants by exceptionally strong “overall” performance in five key dimensions: Food security – Income – Nutrition – Environmental sustainability – Social equity. For each household in the survey, there were five performance indicators. These included, for example, annual household income, a dietary diversity score (for Nutrition), and a measure of gender equity (for Social equity).
Visits to diverse positive deviants identified a range of locally viable SAI practices
But does this approach not risk simply identifying the wealthiest farmers, with largest farms or best market access, as positive deviants?
By accounting for individual household resources, the researchers put the performance indicators into perspective: Every household’s performance was assessed against the “expected” performance, given their land and livestock holdings, family size, region, and market potential. As a result, the selection criterion for being a positive deviant was performing “better than expected” – which may include any farm size, at any place. The study then identified positive deviants by the concept of Pareto-optimality. That is, households qualified as positive deviants if they achieved stronger outcomes than other households with equivalent resource levels, balancing the five concurrent development objectives.
Mariam Ally, the soft-spoken, but entrepreneurial farmer from Mchengamoto, was but one of 15 rural household leaders visited by the research team. All in all, in ten days of fieldwork, the interviews and farm visits revealed a list of 14 interesting practices that plausibly contributed to Sustainable Intensification. Positive deviants engaged in a large diversity of practices: From agronomic improvements, such as cereal-pulse intercropping, to on-farm businesses, such as a tree nursery, to off-farm enterprises, such as a private transport business or classic wage labour.
In their study, ”Prioritizing options for multi-objective agricultural development through the Positive Deviance approach”, published in PLoS ONE, the authors draw three main conclusions:
- Positive deviants were diverse with respect to resource endowments and livelihood activities. Exploring the behaviours of these heterogeneous positive deviants can inform an ample range of locally viable household-level practices that support Sustainable Agricultural Intensification in diverse contexts.
- Identifying such locally viable practices for SAI does not require complex econometric modelling nor extensive qualitative fieldwork. Rather, a resource-efficient combination of quantitative and qualitative research methods makes the Positive Deviance approach accessible for local development agencies.
- The performance differences between positive deviants and other households can indicate most-promising domains for development interventions. In the case of Southern Tanzania, the study suggests highest potential for improvements in food security and income. Improvements in nutrition and social equity, for example, seem harder to achieve without repercussions for other dimensions.