«Minimum Dietary Diversity for Women A Guide to Measurement Minimum Dietary Diversity for Women A Guide to Measurement Published by the Food and ...»
In addition and optionally, a space can be left where enumerators can write down other recalled foods if they are uncertain where to classify the item on the questionnaire. If it will not be possible for supervisors to review forms with enumerators on an ongoing basis (as in many large-scale surveys), having enumerators write in “other” foods is not advised. In surveys with smaller sample sizes, allowing use of this space can contribute to ongoing enumerator training and data quality control, especially if there is timely review by supervisors and feedback to the enumerator.
The reason for including this category on the questionnaire is primarily to give a place for enumerators to mark these foods to avoid falsely classifying elsewhere.
Mixed dishes and food items with multiple ingredients Mixed dishes and food items with multiple ingredients present the most difficult challenge to implementation of food group recall surveys. It is not possible to provide comprehensive lists for classification. This guide aims to provide principles and some examples to aid in questionnaire adaptation and in training enumerators to record information about these dishes and food items.
One principle underlying many of the difficult choices reflected in the content of this guide is to err on the side of NOT falsely inflating food group diversity. This is particularly important when foods or ingredients are expensive and the poorest and most vulnerable women are more likely to consume trivial amounts.
The principle is applied to two distinct but related situations. The first involves “taking apart” mixed dishes and deciding which ingredients should be classified so that they can count in the MDD-W and which should be classified with “Condiments and seasonings”. These are decisions that need to
19 Section 2 Description of food groups
be made at the level of adapting the questionnaire with local foods. Enumerator instructions and training must also include clear guidance on probing for details of mixed dishes and on recording ingredients in their respective food groups and categories, including into the “Condiments and seasonings” category.
The second situation involves items that are known to contain multiple ingredients but should nevertheless be classified as a single food for purposes of the survey (e.g. bread). These should be categorised into a single food group or category during questionnaire adaptation and reviewed during enumerator training.
MIXED DISHESMixed dishes may contain some ingredients in large quantities (“main ingredients”) and others in smaller quantities to add flavour. The principle stated above (and also in Box 1) presents a rationale for excluding items likely to be consumed in very small quantities from counting in the MDD-W.
Box 1 also provides an argument for consistency across surveys. To support consistency, this guide provides a list of items that should not count and should instead be classified as “Condiments and seasonings”, as described above on page 19 and as listed in Appendix 2.
Survey designers can still choose to make different decisions, but they should do so only when they have access to nutrition experts with very good understanding of the principles of food group recalls and of variability in preparation of mixed dishes across the range of households in the survey. In addition, survey designers should realize that if different decisions are made for surveys in the same geographic area, survey results will not be comparable.
See Box 3 for some examples of classification for the following types of mixed dishes:
• Thin soups • Thick soups, stews and curries • Dark green leafy vegetables and other vegetable dishes • Sandwiches 20 Box 3. Mixed dishes – Examples for classifying ingredients into rows on the model questionnaire In the examples below, groups that “count” for Minimum Dietary Diversity for Women of Reproductive Age (MDD-W) are in bold font and underlined.
Thin and thick soups, stews and curries can all be served alone or served alongside or on top of a staple food. When relevant, enumerators should probe to determine which parts of the dish were consumed.
THIN SOUPS can include any combination of meat, fish, vegetables and seasonings boiled in liquid and may or may not include oil. Thin soups have a high water content, and individual ingredients can often be easily picked out and consumed or not consumed by individuals.
EXAMPLE OF THIN SOUP: CHICKEN SOUP – THIN BROTH [Respondent reports the soup contained chicken, water, onion, garlic and herbs] Respondent reports the broth only Mark under “Condiments and seasonings” (for the garlic and herbs) and “Other beverages and foods” (for the broth).
all parts of the Mark under the two groups above and also mark “Meat and poultry” (for the chicken) and “Other vegetables” (for the soup onion).
THICK SOUPS AND STEWS have the same types of ingredients as thin soups but are served with thicker consistency because of long, slow cooking and sometimes as a result of adding thickeners (starch). In thick stews, there may be little or no broth. In thinner stews, some items are dissolved in the broth, but it may still be possible to pick out some items, particularly when meat is included.
EXAMPLE OF THICK SOUP OR STEW: KIDNEY BEAN STEWRespondent reports the stew contained Mark under “Pulses (beans, peas and lentils)” (for the kidney beans), “Condiments and seasonings” (for the garlic and kidney beans, water, oil, garlic and spices. spices) and “Other oils and fats” (for the oil).
CURRIES are similar to stews and can contain meat, fish or vegetables. Curries are usually characterised by use of many spices and seasonings.
EXAMPLE OF A CURRY: EGGPLANT AND ONION CURRYRespondent reports the curry contained Mark under “Other vegetables” (for the eggplants, onions and tomatoes) and “Condiments and seasonings” (for the eggplants, onions, tomatoes, garlic, ginger, garlic, ginger, chilies, cumin seed, coriander seed and cilantro).
chilies, cumin seed, coriander seed and cilantro (coriander leaf).
DARK GREEN LEAFY VEGETABLES (“GREENS”) AND OTHER VEGETABLES can be included in soups, stews or curries, or they can be the main ingredient in dishes. In many cuisines, dishes where vegetables are the main ingredient contain no other ingredients or only very small amounts of other ingredients.
EXAMPLE OF A VEGETABLE DISH: CASSAVA LEAVES WITH FISH POWDERRespondent reports the dish was made Mark under “Dark green leafy vegetables” (for the cassava leaves) and “Condiments and seasonings” (for the salt and with pounded cassava leaves, water, salt the fish powder).
and fish powder.
SANDWICHES are mixed dishes with meat, cheese, vegetables and/or spreads served on bread, with or without toppings or condiments that are added mainly for flavour.
Section 2 Description of food groups Respondent reports she had a cheese Mark under “Grains, white roots and tubers, and plantains” (for the bread), “Dairy” (for the cheese) and “Condiments
Food items with multiple ingredients Some food items present a situation exactly opposite that of mixed dishes. These are items that typically have more than one ingredient, but that are usually dominated by one ingredient. For these items, it is not necessary to probe for ingredients, as the item can be classified into one food group or category based on the main ingredient. See Box 4 for a list of examples of these items. This list is not exhaustive but aims to provide sufficient information so that survey designers can identify similar local items that should be treated in the same way.
This section provides two elements that comprise the MDD-W questionnaire. The first element is a block of standard text (a “script”) to adapt and use in guiding the respondent through an open recall of foods and beverages consumed the previous day and night. The text also includes statements to guide the enumerator in recording information. The second element is a model questionnaire form, which needs to be adapted with local foods (see Section 4 for guidance on translation and adaptation of the text and questionnaire).
The guidance to enumerators for recording information is provided as an example and is not meant to be prescriptive, as different surveys employ different standard procedures (circling versus underlining or “ticking” foods; codes for “yes” and “no” answers; etc.).
Forms may also optionally provide a space for enumerators to record each food or ingredient as mentioned. A grid structure for morning, mid-day and evening may be helpful. This type of questionnaire structure has been used and found to be helpful in a number of small- to mediumsized surveys; however, it is not reflected in the instruction text below. This approach requires the enumerator to first write the foods/ingredients, then to code each food or ingredient item into its respective row on the questionnaire at the end of each interview. This two-step process may not be feasible in larger surveys.
Increasingly, survey teams may be carrying out Computer-Assisted Personal Interviewing (CAPI) using tablets or other devices to capture data. The methods for marking responses will differ when using CAPI as compared with paper forms.
23 Section 3 Model questionnaire
Guiding the respondent through an open recall and recording information The following text can be provided to enumerators on a job aid/guidance sheet or included on the
questionnaire form; the italics indicate the example script that would be spoken to the respondent:
Now I’d like to ask you to describe everything that you ate or drank yesterday during the day or night, whether you ate it at home or anywhere else. Please include all foods and drinks, any snacks or small meals, as well as any main meals. Remember to include all foods you may have eaten while preparing meals or preparing food for others. Please also include food you ate even if it was eaten elsewhere, away from your home. Let’s start with the first food or drink consumed yesterday.
Did you have anything to eat or drink when you woke? If yes, what? Anything else?* Did you have anything to eat or drink later in the morning? If yes, what? Anything else?* Did you eat or drink anything at mid-day? If yes, what? Anything else?* Did you have anything to eat or drink during the afternoon? If yes, what? Anything else?* Did you have anything to eat in the evening? If yes, what? Anything else?* Did you have anything else to eat or drink in the evening before going to bed or during the night?
If yes, what? Anything else?* * For each eating episode, after the respondent mentions foods and drinks, probe to ask if she ate or drank anything else. Continue probing until she says “no, nothing else”. If the respondent mentions a mixed dish like a soup or stew, ask for all the ingredients in the mixed dish. For mixed dishes where it is possible to pick out ingredients or consume only broth, ask if she herself ate each ingredient or if she only had the broth. Continue to probe about ingredients until she says “nothing else”.
INSTRUCTIONS FOR RECORDING INFORMATIONThe following text can be provided to enumerators on a job aid or included on the questionnaire
As the respondent recalls foods and drinks, mark the corresponding item in the “Description/ examples to be adapted” column and mark ‘1’ in the response column for that row on the questionnaire. If more than one item in a row is mentioned, mark each item. If the same food or drink is mentioned more than once, you do not need to mark it again after the first time.
[Optionally: If the food is not listed in any of the rows on the questionnaire, write the food in the bottom row labelled “Other beverages and foods”.] In some surveys, it may be possible for the enumerator to review the foods that have been reported by repeating them to the respondent and making a final probe (“anything else?”), but this will depend on the specific survey context. Immediately after completing the recall, the enumerator should mark “no”16 for rows where the respondent did not report consuming any items.
Again, the method for marking “no” will vary depending on standard practice preferred by the survey designers; it 16 can be entered as a code “0” in a blank space (|0|) or codes on the form may be circled or ticked.
Model questionnaire The model questionnaire on the next two pages provides a few examples of food items for each row in the questionnaire. During questionnaire adaptation, these examples of food items need to be replaced by lists of common foods in the local context that fall into the row (see Section 4).
Table 3. MDD-W model questionnaire
If rows O, P, Q, R, S and/or T are not included, examples for the “Other beverages and foods” category must be expanded a to include these types of items.
The final two rows (“Condiments and seasonings” and “Other beverages and foods”) should always be included on the questionnaire.
Before measuring MDD-W in a new setting, it is important to prepare the data collection materials to reflect the foods and dietary habits of the target population. This section covers a series of steps to be carried out by survey designers for preparing a linguistically and culturally adapted MDD-W questionnaire.
In locations where previous food group diversity surveys have been implemented, existing questionnaires can be useful inputs to this process. If it is known that previous surveys were prepared following a thorough process (i.e. steps similar to those described in this guide), questionnaire adaptation can be much shorter, as food items on previous questionnaires can be used to populate the adapted MDD-W questionnaire.