perceptions to numerical data to research on development issues

In the discipline of food science and technology a range of methods has been developed in order to translate qualitative perceptions into data with numerical value. These sensory evaluation methods are widely used both by the food industry and academics. The translation of qualitative information into quantitative data allows statistical analysis and more accurate monitoring of changes. There are various models that have been developed which can be selected to fit the goals of research. This paper gives some suggestions, based on our own experience, of how different sensory evaluation models can be used in development-based research. A model for consumer preference testing was applied in rural Tanzania, to assess acceptability of newly developed products. The paper explains how the tests were conducted in a participatory way, and outlines important adaptations such as training of interviewers and a simplified design. Further it discusses how the outcomes related to the findings obtained by conventional qualitative assessments. A model using a semi-expert trained panel was applied to monitor changes in estimated market value of products. It out lines the importance of priming of semi-trained panellists, and the findings are compared to qualitative appraisals at the markets. Line scales were used as a tool during stakeholder workshops. This technique assisted in the prioritisation of different development alternatives. The use of line-scales is a powerful tool because it is visual for