DNA fingerprinting and varietal adoption: getting the full picture
DNA fingerprinting allows researchers to identify a crop variety genetically. While survey teams can ask farmers what crop varieties they are growing, an evaluation method that combines survey data with DNA fingerprinting can identify crop varieties more accurately, which is important in adoption and impact studies.
To assess the impact of adopting improved crop varieties, researchers usually ask farmers what is in their fields. But when farmers are not aware of the exact names or origin of the varieties that they grow, the adoption rate can be underestimated or exaggerated. To get a more accurate picture, researchers working with RTB crops have combined DNA fingerprinting identification with questionnaires.
DNA fingerprinting to estimate varietal adoption was pioneered by Mywish Maredia at Michigan State University, Byron Reyes at CIAT and others for wheat in Ethiopia, rice in Bolivia, cassava in Ghana and beans in Zambia. All researchers found a mismatch between what farmers said they were growing and the varieties identified with DNA fingerprinting.
Ricardo Labarta, agricultural economist, explains that in 2015 CIAT researchers collected cassava stakes and applied a questionnaire to 307 households in Cauca, Colombia, as described in 2018 in the Journal of Agricultural Economics. In over half the cases, the variety name given by farmers was not the one found by DNA fingerprinting. The farmers reported growing improved varieties on 25% of their acreage, while the DNA analysis raised this estimate to 29%.
The falling costs of DNA fingerprinting have made it easier to combine with questionnaires, for example on the IITA Cassava Monitoring Survey in Nigeria. As Tahirou Abdoulaye explains, the survey starts with a questionnaire, asking farmers what crop varieties they grow, and if these are improved or local. The survey-takers then visit the farmers’ fields and collect some cassava leaf tissue in a vial with a barcode label to avoid confusion. Abdoulaye and colleagues found that the adoption of improved cassava varieties would have been under- reported if the questionnaire data had been accepted at face value. According to the farmers, 54% of them were growing improved cassava varieties, while DNA fingerprinting gave a much higher figure, 68%.
Frédéric Kosmowski of the CGIAR Standing Panel on Impact Assessment, and colleagues at CIP, the University of Canberra, the World Bank and the Central Statistical Agency of Ethiopia compared what Ethiopian sweetpotato farmers said they were growing with varieties identified with DNA fingerprinting; 20% of farmers identified a variety as improved when it was really local, and 19% said a variety was local when it was improved.
In Vietnam, different methods used to assess cassava adoption also gave different results. Labarta and Dung Phuong Le of CIAT and Maredia of Michigan State University asked panels of experts to estimate the land area in Vietnam devoted to improved cassava varieties. While the expert panels estimated single varieties with high adoption rates, a follow-up DNA fingerprinting study with Vietnamese farmers found 85 different improved varieties being extensively used in over 90% of production areas.
Combining DNA fingerprinting with questionnaires allows collaborative learning between impact assessment teams working with different crops and in different regions, to better refine and standardize the method. The results from the studies help to reduce identification bias, identify material that is uncommon or is less familiar to farmers, and shed light on the distribution and potential flow of crop varieties.