Nutrition surveys are often a key component for measuring household food security or food insecurity. The SMART Methodology (Standardized Monitoring and Assessment of Relief and Transitions), an inter-agency initiative launched by a network of organizations and humanitarian practitioners, is a widely used survey method based on two basic public health indicators for assessing the magnitude and severity of a humanitarian crisis: 1. Nutrition status of children under five, and 2. Mortality rate of the population (SMART Methodology).
ENA (Emergency Nutrition Assessment) software is a user-friendly analytical program recommended by SMART. This article provides a sample workflow for implementing a nutrition survey in SurveyCTO with an output that can be directly pasted into ENA software for analysis.
Furthermore, the Household Nutrition Survey utilizes a series of calculations developed by UNHCR and partners to detect and flag children at risk of malnutrition during surveying, allowing enumerators to refer households for treatment in real time.
Populate a Local Events Calendar to be used for identifying child age in months.
Fill a Household Roster form with biodata for each member of selected households that is also used to identify children eligible for the nutrition survey.
Complete a Household Nutrition Survey with automatic in-form evaluations of collected anthropometric measurements to determine at-risk children in real-time.
Access ENA-ready data for immediate nutrition assessments.
Complete the Household nutrition survey for each household with eligible children.
To be able to effectively test this workflow you can follow these outlined tasks:
Open the Local events calendar form and fill it out. Complete a new survey for each unique event.Note: There are 60 months included in the Local events calendar dataset, corresponding to the number of months of a 5-year period. The goal is to fill out events for as many months as possible to populate the calendar; however, it is not required for each month to have an associated event.
Open and fill out the Household roster form. Include at least 2 children between the ages of 6 months and 5 years in the household roster to effectively test the workflow (e.g.: child1 is 1 year old and child2 is 2 years old.)Press “Add group” to add another household member until all household members are listed. When the listing is complete, press “Do not add” to continue.
Note: the “recall period” must be determined before the start of the survey by the implementing organization, and there are several factors to consider based on the survey context. There are advantages and disadvantages of shorter or longer recall periods depending on the survey needs. TheSMART Manual offers detailed guidance on choosing an appropriate recall period.
After finalizing and submitting each Household roster form, open the Nutrition data for ENA software server dataset from the Design tab. Note that the household information for any eligible child has been published into the dataset.
Open the Household nutrition survey form and fill it out by selecting a location and household.Note: the “Cluster number” will default to the information previously submitted in the Household Roster form for the selected location. This function will help reduce discrepancies in data entry but still allows for the option to make a different selection if necessary (for example, if there was a submission error in the Household roster form).
In this form, there are two methods for entering the child’s age. If birth records are available, enter the exact birth date. If these records are not available, you will be prompted to use the Local events calendar to determine the correct age of the child (in months). Test the workflow by entering the exact birthdate for child1, while selecting an event that corresponds with the child’s birth for child2.
This is how the calendar of local events will appear in the survey:
Next, enter the anthropometric data for each child.
Note: data flags are built into the survey to catch potential data entry errors and to flag children at risk of moderate or severe malnutrition. To test this function, enter an improbable measurement for child1 to trigger a second set of measurement questions and enter the corrected measurements. If a child has been measured twice, only the second measurement will be recorded in the final dataset.
In the example above, the data was flagged because the height and weight of child1 do not correspond with plausible measurements for a 12-month-old child.
Once the Household nutrition survey form has been filled and submitted, open the Nutrition data for ENA software server dataset again. You should now see all fields filled for child1 and child2 including the “completed” column indicating that “this child has been measured”. You can now download the Nutrition data for ENA software dataset data from the Design tab or Export tab. The dataset will be in .csv format:
The columns starting with “date_end” through “muac” can be copied and pasted directly into the “Data Entry Anthropometry” tab of the ENA software.
Note: the ENA software has specific format requirements for inputting data into the program and therefore the data published into the Nutrition data for ENA software dataset follows these requirements:
SURVDATE must be in short format
CLUSTER, TEAM, ID, AND HH are numerical
SEX is either “m” or “f”
BIRTHDAT must be short format
MONTHS is numerical
WEIGHT is recorded in kg and HEIGHT is recorded in cm
This workflow has three forms, three server datasets, and one enumerator dataset. Below, you can find a diagram showing the relationship between these components, as well as a non-comprehensive list of all components, the corresponding files, and main purposes.
Workflow Component Files Click on the section to access the files.
A calendar of local events helps respondents estimate a child’s age in completed months in the absence of birth documentation. The calendar is populated by respondents (such as a focus group) prior to the launching of the nutrition survey. The calendar should be tailored to the local context with notable events that would be common knowledge to the respondent population.
This form uses repeat groups to collect biodata for each household member of selected households, which includes information that can be used for mortality rate calculations. The data collected in this form feeds into the Nutrition data for ENA software server dataset.
This form is used to collect anthropometric measurements for children between 6 and 59 months. The data collected in this form is published into the Nutrition data for ENA software server dataset.
SurveyCTO has endless possibilities, so this workflow is only the beginning. Feel free to modify and expand on this workflow to fit your own needs. Here are some tips on what you can add:
Adapt the ENA – Example locations server dataset: Replace the data in the server dataset example provided with the administrative-level data for your survey location, using as many administrative levels as necessary. Once populated, this dataset should be attached to the Household Roster form and will be used to dynamically populate the survey location choice list.
Adapt the ENA – Enumerators dataset: This workflow uses the enumerator management functionality to help enumerators identify themselves in the forms and automatically store their team number. Based on your context specifics, you will need to include your list of enumerators, enumerator IDs, and other enumerator-related information in the enumerator dataset.
Adapt the household identification approach: There are many ways to generate unique IDs in SurveyCTO and use them to identify your households or children. Depending on your context, it might make sense to adapt our approach. Take a look at Enforcing unique study IDs – how to avoid duplicate IDs for some inspiration!
Manage your households in a separate households dataset: While we are collecting household data, we are recording and tracking child data in our Nutrition data for ENA software dataset. If you would like, create a server dataset to store and list all households, which can be used as auxiliary data in your forms to easily identify the correct household and pre-load other helpful information.
Integrate the Locating households in fieldwork workflow: Use this workflow to help enumerators locate the households where the ENA – Household Nutrition Survey should be implemented based on different data points that can be incorporated into the ENA – Household Roster form.
Enable theAdvanced Offline features to make the best use of this workflow. These features are key if you are collecting data from multiple forms and have no reliable internet connection in between them to ensure your form(s) are using the latest data. You can further adapt this workflow to a case management system where you track children as cases and, in this context, the advanced offline features allow you to transfer cases between enumerators.
Build snowball sampling into the survey design: Include a module in the Household roster form that identifies a household as either a “seed” or a “referral” household. If it is a seed household, ask the respondent the number of referrals they can provide, which will inform the number of times a repeat group should repeat. Within the repeat group, include fields such as the household name, contact information (such as a phone number), and any other information that may help identify the household. The contact information gathered from each “seed” respondent is published into a server dataset for enumerators to use to contact referral households for surveying. Referral households contacted and surveyed become seed households, providing additional contacts.
When conducting household nutrition surveys, snowball sampling can be particularly useful in the context of hard-to-reach populations, such as refugee camps. Rather than starting from a randomly selected household within the cluster, selection starts with a “seed” household or households from the target population who then provide the contact details of other households within the target population (for example, other refugee households within the cluster area). Visit UNHCR’s guidance on conducting nutrition surveys for refugee populations for more information on sampling methodologies.
