informant interviews To understand the role of key gatekeepers or decision-makers in infant and young child feeding within the family or broader community. Same as in-depth interviews with women. Trials of Improved Practices To test actual new or modified practices in order to gain additional insight into how programs and efforts to support them can best be designed. Dicken K, Griffiths M, Piwoz E. Designing by Dialogue: A Program Planner’s Guide to Consultative Research for Improved Young Child Feeding. Support for Analysis and Research in Africa Project, 1997: www.manoffgroup.com/ resources/Designing%20by%20Dialogue.pdf. Market surveys To obtain information on the availability and costs of different foods or potential food options in the diet. Pan American Health Organization. ProPAN: Process for the Promotion of Child Feeding, 2003: www.paho.org/English/ AD/FCH/NU/ProPAN-Index.htm. Dietary analysis To obtain information about the nutritional adequacy of the diet, or to estimate adequacy of a particular nutrient. Pan American Health Organization. ProPAN: Process for the Promotion of Child Feeding, 2003: www.paho.org/English/ AD/FCH/NU/ProPAN-Index.htm. Dicken K, Griffiths M, Piwoz E. Designing by Dialogue: A Program Planner’s Guide to Consultative Research for Improved Young Child Feeding. Support for Analysis and Research in Africa Project, 1997: www.manoffgroup.com/ resources/Designing%20by%20Dialogue.pdf. Linkages Project. Formative Research: Skills and Practice for Infant and Young Child Feeding and Maternal Nutrition. Linkages/India, January 2003. Recipe trials To obtain information about foods available at home and how mothers combine and prepare foods. Dicken K, Griffiths M, Piwoz E. Designing by Dialogue: A Program Planner’s Guide to Consultative Research for Improved Young Child Feeding. Support for Analysis and Research in Africa Project, 1997: www.manoffgroup.com/ resources/Designing%20by%20Dialogue.pdf. Some of the more common missteps to avoid in matching information needs with collection methods include: · Using a method based on what the researcher is familiar with rather than choosing one (or more) based on the information needed/research questions. 9 · Conducting focus group discussions to gather information about daily child feeding practices. Focus groups do not allow for ‘honest’ answers about practices, as participants are often reluctant to describe what they do in front of people they know, or they will mimic what others say. Focus groups are good methods for discussing notions of child raising, beliefs about the properties of foods, what usually happens in the community, and ideas about what might or might not be acceptable to change and why. · Asking key informants such as nurses and village leaders to provide information on caregiver practices. They cannot speak for caregivers about what they do and why; their answers would be speculative and uninformative. · Thinking that baseline or quantitative surveys in which information is collected on prevalence of certain practices or beliefs provide insight into practices. These surveys lack insight into why, or how, precisely certain practices are followed. · Assuming that defining current behaviors and determinants will lead to answers about the feasibility of potential changes or new practices. The determinants of current behaviors are not necessarily determinants of new behaviors (see Box 6). Rather, they are merely a starting point for defining what and how a practice might be modified. Box 6. Current behaviors do not always determine new behaviors In many countries, the vast majority of women delay the initiation of breastfeeding. They say they delay offering the breast because it is a tradition and the recommendation of elders and birth attendants to not give their babies colostrum. They also cite many disadvantages to giving colostrum. It may seem a daunting task to confront such determinants of delayed initiation of breastfeeding; however, experience has shown that when mothers are asked as part of Trials of Improved Practices research to try to breastfeed immediately and to offer colostrum, and are given compelling reasons to do so, they are willing to try and virtually all of them make the change. The determinants of the new behavior in this case are: (1) ‘new’ information about the increased chance of survival that immediate breastfeeding offers, and (2) respect for the advice of a health care professional encourages immediate breastfeeding. For this practice, ncovering the cogent benefits communicated by a trusted source for a new behavior demonstrates the relative ease of changing this behavior. While this is not always the case, by exploring determinants of new behaviors (in addition to understanding current behaviors), the possibility of change becomes clearer, albeit sometimes more complex, particularly if the determinant is beyond the control of the family or community. Research plans A rapid assessment of the various participant groups that might be included, along with the questions that need to be answered and the variety of methods that could be used, would lead to the conclusion that there are many ways to structure a formative research activity. There is no one correct way; formative research is context specific and dependent on the time and budget available. Technically, what is most important is that the right people are selected and the right method is used to obtain the most valid or true information regarding attitudes and practices and their influences/determinants. 10 Box 7 provides two examples of research plans for exploring infant and young child feeding practices in Nepal and Nigeria. Box 7. Examples of research plans Literature Review Key informant interviews: Health centers Community health workers Local shop owners In-depth interviews with caregivers: - well children 0–24 months - children 0–24 months with diarrhea Trials of Improved Practices: Hygiene behaviors and care of sick children Focus group discussions with caregivers and mothers-inlaw outside of immediate research area Literature Review Market