Household Dietary Diversity Score

Outcome

average number of food groups consumed by target households during the previous day and night

The Household Dietary Diversity Score (HDDS) measures the variety of food groups consumed by target households during the previous day and night. It is a proxy indicator of a household’s economic access to a variety of foods, including its ability to produce, purchase or otherwise obtain food. It does not measure the quantity of food consumed and therefore cannot show whether household energy needs are met; nor does it assess the nutritional adequacy of individual household members’ diets.

Collect the following data from a representative sample of households. In each sampled household, interview the person responsible for food preparation for the household on the previous day:

   

1) Check whether yesterday was a special day (religious festival or celebration) when unusually varied or limited diet was eaten - if so, do not proceed with collecting dietary data as it is likely that they will not reflect a typical diet. Instead, revisit the household on another day. If a replacement household is needed, select it using the survey’s pre-defined sampling procedure; do not let interviewers choose a convenient replacement.

  

2) List all foods and drinks consumed by household members during meals on the previous day and night in the Recording Meals Form (see link below).

  

3) Double check the meals' composition (e.g. porridge with or without milk).

  

4) Check for any snacks (including fruit) or drinks that were not mentioned.

  

5) Only then record in the questionnaire which food groups were eaten (a list of 12 food groups is provided in the FANTA guide below; additional guidance is provided in the FAO guide below). For any food groups not mentioned, ask whether anyone in the household consumed a food from that group (for example: "Did anyone in your household eat any eggs yesterday?"). If using FAO’s detailed 16-group questionnaire, aggregate the responses into the standard 12 HDDS food groups before calculating the score.

  

6) For each household, count the number of the 12 HDDS food groups consumed by any household member. Each food group contributes one point, giving an HDDS from 0 to 12.

     

7) To calculate the indicator's value, sum the HDDS values of all households with valid HDDS data and divide the result by the number of households with valid HDDS data.

This data can be disaggregated by geographic area, head of household, and household wealth. Only report subgroup estimates where the sample size is sufficient.

1) A household is defined as all persons living under the same roof who share meals.

2) HDDS reflects the variety of foods available to and consumed by the household as a whole during the recall period. A food group is counted if any household member consumed food from that group — not all members need to have eaten it. To assess individual diets, use other indicators, such as:

• Individual Dietary Diversity Score

• Minimum Dietary Diversity - Women

• Minimum Dietary Diversity - Children

3) There are no established universal HDDS cut-offs for classifying households as having adequate or inadequate, or low, medium or high dietary diversity. Use the mean score and, where useful, examine the distribution of scores and consumption of specific food groups.

4) Food security outcome indicators, including this one, are useful for assessing and monitoring food insecurity across populations, comparing groups and supporting geographic targeting. However, they should not be used as direct criteria for selecting individual households for food assistance. Results can be influenced by recall errors, temporary circumstances, seasonal and cultural practices, and how respondents understand or answer questions. They can also be difficult to verify and explain transparently to communities. Instead, household selection should use context-specific, transparent and verifiable vulnerability criteria, informed by community consultation and other relevant evidence.

  

5) HDDS is prone to seasonal differences. Where HDDS is compared over time, collect baseline and follow-up data at the same time of year where possible, because seasonal changes in food availability can otherwise affect the comparison. Avoid survey periods when a community-wide festival, extended fast or other event is likely to make food consumption atypical; if this cannot be avoided, document the event and interpret comparisons cautiously.

 

6) Do not include foods that were purchased and eaten outside the home. These foods are excluded because the respondent may not know the foods which other household members purchased and ate outside the home. You can include foods that were 1) prepared in the home and consumed in the home or outside the home; or 2) purchased or gathered outside and consumed in the home. For this reason, HDDS may be unsuitable where purchasing and eating food outside the home is very common, because it may underestimate household members’ total dietary diversity.

 

7) When training the data collectors, practice extensively which meals belong to which food group (allocate at least 3 hours full of examples and exercises). For example, while pumpkin flesh belongs to Vitamin A Rich Foods, pumpkin leaves belong to Dark Green Leafy Vegetables (see more examples in the FAO Guidelines below). If your questionnaire includes examples of different foods per each group, adjust them to the local context. Translate and adapt food examples to locally used names and mixed dishes, and pre-test them before data collection.

     

8) According to FAO guidance for household-level dietary diversity, no minimum quantity needs to be set; even small amounts of food are counted because HDDS is intended to reflect household economic access to food.

Also consider using the following indicators:

• Food Consumption Score (complements the 24-hour HDDS with a seven-day measure incorporating consumption frequency)

• Reduced Coping Strategy Index (captures short-term behavioural responses when households cannot access enough food)

• Household Hunger Scale (adds a measure of severe food deprivation where acute food insecurity is high)