ObjectiveThe association between farm production diversity and dietary diversity in rural smallholder households was recently analysed. Most existing studies build on household-level dietary diversity indicators calculated from 7d food consumption recalls. Herein, this association is revisited with individual-level 24 h recall data. The robustness of the results is tested by comparing household- and individual-level estimates. The role of other factors that may influence dietary diversity, such as market access and agricultural technology, is also analysed.DesignA survey of smallholder farm households was carried out in Malawi in 2014. Dietary diversity scores are calculated from 24 h recall data. Production diversity scores are calculated from farm production data covering a period of 12 months. Individual- and household-level regression models are developed and estimated.SettingData were collected in sixteen districts of central and southern Malawi.SubjectsSmallholder farm households (n 408), young children (n 519) and mothers (n 408).ResultsFarm production diversity is positively associated with dietary diversity. However, the estimated effects are small. Access to markets for buying food and selling farm produce and use of chemical fertilizers are shown to be more important for dietary diversity than diverse farm production. Results with household- and individual-level dietary data are very similar.ConclusionsFurther increasing production diversity may not be the most effective strategy to improve diets in smallholder farm households. Improving access to markets, productivity-enhancing inputs and technologies seems to be more promising.

Farm input subsidies are often criticised on economic and ecological grounds. The promotion of natural resource management (NRM) technologies is widely seen as more sustainable to increase agricultural productivity and food security. Relatively little is known about how input subsidies affect farmers’ decisions to adopt NRM technologies. There are concerns of incompatibility, because NRM technologies are one strategy to reduce the use of external inputs in intensive production systems. However, in smallholder systems of Africa, where the average use of external inputs is low, there may possibly be interesting complementarities. Here, we analyse the situation of Malawi's Farm Input Subsidy Program (FISP). Using panel data from smallholder farm households, we develop a multivariate probit model and examine how FISP participation affects farmers’ decisions to adopt various NRM technologies, such as intercropping of maize with legumes, use of organic manure, water conservation practices and vegetative strips. As expected, FISP increases the use of inorganic fertilizer and improved maize seeds. Yet, we also observe a positive association between FISP and the adoption of certain NRM technologies. For other NRM technologies, we find no significant effect. We conclude that input subsidies and the promotion of NRM technologies can be compatible strategies.

Farm production diversity and individual-level dietary diversity. Response to: 'Not all dietary diversity scores can legitimately be interpreted as proxies of diet quality' by Verger et al.

Response to: 'On the appropriate use and interpretation of dietary diversity scores' by Verger et al. Madam In their first letter, Verger et al. (1) challenged our original result (2) that the effect of farm production diversity on dietary quality is small, because we had worked with the twelve food groups that are often used for the household dietary diversity score (HDDS) to characterize individuallevel dietary quality. In fact, we had compared results with household-and individual-level data. Verger et al. (1) suggested that other food group classifications are better suited to characterize individual-level dietary quality for women and children. In our response (3) to that first letter, we provided further reasons for our approach but also carried out additional analyses with the alternative indicators suggested. These additional analyses confirmed the original results. We also showed that the different indicators are closely correlated, which means that the twelvefood-group classification is a valid proxy of dietary quality in this particular case. In their second letter, Verger et al. (4) emphasize that this correlation does not mean that the twelve-food-group indicator is a good proxy of individual dietary quality in general. We agree with this caveat. However, we want to stress that the results of our original study (2) , namely that the effect of farm production diversity is small and that market access is more important for household and individual dietary diversity and dietary quality, are robust to the different indicators used.

Determinants and livelihood impacts of natural resource management strategies among smallholder farmers in Malawi Dissertation zur Erlangung des Doktorgrades der Fakultät für Agrarwissenschaften der Georg-August-Universität Göttingen vorgelegt von Stefan Koppmair geboren in Dillingen a.d. Donau Göttingen, März 2016 I would like to thank the International Maize and Wheat Improvement Center (CIMMYT) for a rich collaboration, for giving me the chance to return to Malawi as a part of the Adoption Pathways Project, and for granting me access to secondary farm household data. I am especially thankful to Dr. Menale Kassie for his support and his valuable contributions to chapters 2 and 3. Thank you, Chalmers Mulwa and Geoffrey Muricho for enthusiastically sharing the Malawi experience with me, despite being so far from your families in Kenya. The financial support of the CGIAR Standing Panel on Impact Assessment (SPIA) through a grant to CIMMYT, which supported my participation, is also gratefully acknowledged.Many thanks go to Prof. Julius Mangisoni for the warm welcome at "The warm heart of Africa" and for introducing me to Bunda College of Agriculture. I would like to thank Sam Katengeza for being a great host at Bunda College, sharing his experience and knowledge, and for leading our field work in an outstanding manner; good success with completing your PhD my friend. I am also thankful to our team of field supervisors, enumerators and drivers who did a great job and stayed so motivated throughout the data collection; thanks for all the fun time too. Great many thanks also go to the village heads, farmers and extension officers in all the places we have visited, for their warm welcome and their amazing hospitality.