In 2003 national heads of African states, including Zambia, met in Maputo and pledged to increase financial resources to the agriculture sector, up to 10 % of their national budgets, in order to meet growth targets. Given the need to increase investments in agriculture, it follows that impact assessment studies have become handy as funding agencies demand better accountability and empirical justification for further investment. However, experts have reasoned that the quantity of resources is as important as the quality of spending in that, if resources are allocated efficiently, more could be achieved with the same level of resources. The lack of an effective strategy and basis upon which investments in agriculture crop research ought to be prioritised in order to improve agricultural productivity is the main concern in Zambia. This study sought to illustrate the use of the Dynamic Research Evaluation for Management (DREAM) model to assess the economic returns of investing in agriculture technologies and to set priorities for crop-based research activities in Zambia. The study hypothesised that the financial outlay allocated to agricultural crop research is not efficiently allocated so as to achieve the nation’s agricultural production potential, and that agricultural crop research investment influences the distribution of welfare effects on producers and consumers. The DREAM model is conceptually based on the economic surplus theory and is designed for research priority setting and ex ante evaluations. It computes the net present value (NPV) of benefits for both producers and consumers as a result of investing in agriculture technology. The findings from this study reveal that investment in agriculture crop research in Zambia is worthwhile as positive net present values were obtained for all crops under consideration in this study. Maize research gives the highest return to both large-scale and smallholder producers. Maize also yielded the highest returns for consumers in Zambia. In order of priority, maize is followed by soya bean, groundnuts, cotton, millet, sunflower and sorghum. In spite of this, the order of priority in terms of financial expenditure on crop research is maize, cotton, sorghum, soya bean, groundnuts, sunflower and millet. Therefore, the allocation of financial resources towards crop research is not efficient for all crops except maize since some crops such as sorghum receiving high financial expenditure in research did not necessarily generate high returns. This is because the Government still conducts the bulk of research in Zambia, and as such, other social objectives such as equity and food security considerations play a major role in determining investment patterns. The study further establishes that the choice of crop research expenditure influences the distribution of welfare benefits on different producer groups; and that smallholder farmers in Central, Eastern and Southern province are among the group that received the highest proportion of benefits even for crops such as maize for which financial resources were efficiently allocated. Therefore, the efficiency objective may not necessarily leave smallholder farmers worse off as long as they have access to complementary infrastructure and institutions for agriculture production and marketing. As such the study recommends that the Government, private sector and other development partners must focus on raising agriculture productivity by expanding investments in crop science-based technologies; and also recommends re-allocation of financial resources between crops in favour of crops with high returns because this benefits both large scale and smallholder farmers. This must be accompanied by further investment in complementary infrastructure and good governance. Copyright / Dissertation (MScAgric)--University of Pretoria, 2010. / Agricultural Economics, Extension and Rural Development / unrestricted