Voices of the Hungry A worldwide survey of

slide1. Voices of the Hungry A worldwide survey of people’s experience with food insecurity<br>
slide2. Background & Motivation FAO has the mandate to monitor progress towards the achievement of the hunger target of MDG 1 and the WFS target
Prevalence of Undernourishment (PoU) as a measure of the extent of chronic food deprivation (Official MDG Ind. 1.9)
FAO food security monitoring: a long history ...
World Food Surveys (1946, 1952, 1963, 1977, 1987, 1996)
The State of Food Insecurity in the World (1999 - 2013)
... always debated
FAO International Symposium 2002
CFS Round Table 2011
Yet another FAO International Symposium 2012<br>
slide3. Background & Motivation Monitoring food security worldwide is a challenge:
Food security is a multidimensional phenomenon: multiple indicators required
Lack of a global standard: over 200 indicators proposed
Trade off between “best” operational definition of food insecurity and feasibility/sustainability of the data collection
Global monitoring: need for comparable data across countries on an annual basis with limited time lag
National monitoring: need for disaggregated data to identify vulnerable groups and vulnerable areas as well as possible determinants in order to support policy interventions (annual monitoring not needed and too costly)<br>
slide4. MDG 1.9 = Proportion of population below minimum level of dietary energy consumption
Probabilistic model referred to a representative consumer: Probability that a randomly selected individual regularly consumes an amount of calories insufficient for conducting an active and healthy life
Main limitations: very complex to calculate; narrow concept of inadequate food access (calories); not measuring short-term crises; only available at national level
Not based on a headcount as the poverty rate: individual energy requirements are virtually unobservable
Mistakenly described as an indicator of food supply only: combining information on food availability and food access at the macro and micro level The Prevalence of Undernourishment (PoU)<br>
slide5. Background & Motivation Other indicators are also problematic:
Children underweight/stunted
Multiple causes, not only food insecurity (e.g. diseases and infections)
South East Asia paradox (prevalence higher than in SS Africa)
Information only on one segment of the population
Similar problems of timeliness
Additional problems of coverage
IFPRI Global Hunger Index:
It is plagued by the same problems of the individual indicators on which it is based (PoU, Child underweight, Child mortality)
Aggregates different indicators with no clear rationale for the aggregation function<br>
slide6. Background & Motivation Need to improve our ability to monitor food insecurity at the global and national level
Food security and nutrition has taken center stage in the Post 2015 development agenda
The UN S.G. launched already the “Zero Hunger” challenge (5 targets)
FAO response
Revise its methodology for estimating the PoU
Develop a suite of indicators to monitor the 4 dimensions of food security (availability, access, utilization, stability) => CFS mandate
The Voices of the Hungry project: a new metric system for monitoring food access, one of the dimensions of food security for which indicators are largely missing<br>
slide7. Measure the inability to access adequate food in terms of quantity and quality (and the associated “welfare”)
Identify various degrees of food insecurity experiences (mild, moderate, or severe);
Assess food insecurity experiences at the individual level, thus allowing proper analysis of gender related food insecurity disparities;
Indicator based on sound methodology (Item-Response Theory)
Ensure timely monitoring of the prevalence of people experiencing food insecurity for all developing countries in the world on an annual basis
A leading indicator of potential malnutrition outcomes (link to the SUN initiative) VoH project: Main Benefits<br>
slide8. A new approach started in the 1990s Dr. Kathy L. Radimer’s Ph.D. dissertation: “Understanding Hunger and Developing Indicators to Assess It”, Cornell University, August 1990
“The lack of an operational definition for hunger has been frequently cited as a barrier to progress in addressing the problem.”
“Three scales, one each for household, women’s, and children’s hunger, emerged and were found to be valid and reliable indicators for measuring hunger directly”
(Radimer et al., 1992)
Establishes the concept of food insecurity as an experiential construct<br>
slide9. Quite a history since….<br>
slide10. The food insecurity experience scale (FIES) The severity of the constraints in accessing food is reflected in typical experiences:
People start being worried about how to get food and thinking of giving up other expenses;
Then limit the variety/quality;
Then will start reducing quantities (cutting portions, skipping meals)
Eventually will reach the point of being exposed to hunger
Different levels of severity have different consequences:
Mild food security - > welfare reduction due to reduction of other consumption (education, health insurance, etc)
Moderate food insecurity - > bad quality of food, unbalanced diets, various forms of malnutrition, including obesity and other NCD
Severe food insecurity - > undernourishment and hunger<br>
slide11. The food insecurity experience scale (FIES) Food security Food insecurity mild moderate severe<br>
slide12. The (current version of the) FIES “During the last 12 months, was there a time when, because of lack of money or other resources:
You were worried you could not get enough food to eat?
You were unable to eat healthy and nutritious food?
You ate only a few kinds of foods?
You had to skip a meal?
You ate less than you thought you should?
Your household ran out of food?
You were hungry but did not eat?
You went without eating for a whole day?”<br>
slide13. The (current version of the) FIES “During the last 12 months, was there a time when, because of lack of money or other resources:
You were worried you could not get enough food to eat?
You were unable to eat healthy and nutritious food?
You ate only a few kinds of foods?
You had to skip a meal?
You ate less than you thought you should?
Your household ran out of food?
You were hungry but did not eat?
You went without eating for a whole day?” The qualifier is important, to make sure experiences refer to food insecurity andnot to other possible reasons(health, culture, normal seasonality, etc.)<br>
slide14. The (current version of the) FIES “During the last 12 months, was there a time when, because of lack of money or other resources:
You were worried you could not get enough food to eat?
You were unable to eat healthy and nutritious food?
You ate only a few kinds of foods?
You had to skip a meal?
You ate less than you thought you should?
Your household ran out of food?
You were hungry but did not eat?
You went without eating for a whole day?” Can be adapted to shorter reference periods, when the survey is repeated over different seasons/periods to explore shorter term dynamics in food insecurity.<br>
slide15. The (current version of the) FIES “During the last 12 months, was there a time when, because of lack of money or other resources:
You were worried you could not get enough food to eat?
You were unable to eat healthy and nutritious food?
You ate only a few kinds of foods?
You had to skip a meal?
You ate less than you thought you should?
Your household ran out of food?
You were hungry but did not eat?
You went without eating for a whole day?” When households are sampled, instead of individuals, it can be adapted to refer to the household (ex. “you or any other member of your household”).
Care should be given to select the most appropriate respondent.<br>
slide16. The (current version of the) FIES “During the last 12 months, was there a time when, because of lack of money or other resources:
You were worried you could not get enough food to  eat?
You were unable to eat healthy and nutritious food?
You ate only a few kinds of foods?
You had to skip a meal?
You ate less than you thought you should?
Your household ran out of food?
You were hungry but did not eat?
You went without eating for a whole day?” Even though some of the questions include possibly subjective elements, the scale aims at constructing an objective measure of the severity of the constraints faced in accessing food.<br>
slide17. The tool It is based on the concepts that:
Experiences can be ranked in terms of severity from the least severe (just being worried) to the most severe (suffering from hunger)
The more food insecure one is, the more likely it is that he or she will report having suffered from the “worst” experiences
It uses a long established probabilistic model (the Rasch measurement model) to estimate the severity of the situation revealed by each experience, and by each respondent
The individual measure of severity depends on the entire pattern of responses (not on any single answer considered in isolation!)<br>
slide18. The analytics Given a set of responses (Yes = 1, No = 0) to the eight questions, the item severity parameters and the person parameters can be estimated by Maximum Likelihood procedures.
The person parameters are then used to classify respondents into one of three classes: food secure, moderately food insecure and severely food insecure
The relative frequency of each class in the sample is used as an estimator of the prevalence of that class of food insecurity severity in the population
When data are available from different countries or different samples, the scores can be standardized to ensure comparability of the measurement<br>
slide19. Empirical testing of the results The theory allows for testing whether or not each of the question is contributing to the measurement
Once items are ranked by increasing severity, expected patterns are of the form {1,..,1, 0, …, 0} (affirming one item implies affirming also all the less severe ones)
By comparing the probability associated with each observed pattern with the expected probability of the theoretically consistent pattern for the same raw score, an ‘error’ or ‘residual’ can be estimated
Methods developed in the educational testing practice allow to map the scores obtained with different tests (sets of items) on the same, common underlying scale of severity
Measures of severity obtained in different countries can be made comparable even if not all items have the same associated severity in each country
Solution to the problem found by Coates et al. for the HFIAS<br>
slide20. VoH: the implementation Pilot tested in Angola, Ethiopia, Malawi and Niger in 2013
Validate the feasibility of the data collection tool
Evaluate the robustness of the scale items and refine some of them
Defining the analytic protocol for scale calibration and equalization and to set the thresholds
FIES included as a module in the Gallup World Poll (GWP) starting from 2014, during a calendar covering March – November
The GWP is a worldwide survey of nationally representative samples of the adult population (15+) conducted annually in 150+ countries
The first set of results will be available in August, allowing us to define the global standard reference for scale calibration and to set the thresholds for classification
Micro data from the FIES questionnaire will be made publicly available, together with a number of socio-demographic covariates
Licenses to the entire GWP dataset will be awarded to partner research institution to conduct research on Food Security against many other social and political conditions in the countries<br>
slide21. VoH: Objectives To establish a worldwide valid standard for measuring the severity of food insecurity
To estimate the prevalence of moderate and of severe food insecurity in 150+ countries in 2014 and 2015, to set a benchmark against which to monitor progress at national level
To make available to any interested user linguistically and culturally adapted versions of the questionnaire in more than 200 languages
To promote adoption of the FIES in national food security monitoring systems (inclusion of the module in national household surveys)<br>
slide22. VoH: Current partnership Resource partners
DfID, Government of Belgium
Scientific Advisory group
Angus Deaton, Lawrence Haddad, Romulo Paes de Sousa, Mark Nord, Hugo Melgar-Quinonez
ESS Team
Pietro Gennari, Carlo Cafiero, Terri Ballard, Mauro Del Grossi, Anne Kepple, Sara Viviani, Aymeric Songy, Verena Wilke
Friends/Collaborators
Ana Maria Segal-Correa, Rafael Perez-Escamilla, Jennifer Coates<br>
slide23. VoH: possible partnerships Collaborating with professionals from all over the world to conduct linguistic and cultural adaptation of the FIES
Collaboration with National Statistical Offices to promote use of the FIES in any survey that covers households or individuals in the country
Introduction of the FIES into any national and international Food Security Information System
Explore synergies with other existing food security monitoring tools
ELCSA, EBIA, CSI
Promote the use of the FIES as a tool for impact analysis of policies that are intended to promote food security and nutrition
Promote use of FAO’s VoH indicators in the Post 2015 Development Agenda new monitoring framework<br>
slide24. Need for further support Current plan is to collect data through the GWP for five years
Overall budget of about 18 million USD. Resources have been mobilized to cover the first two years
Need to mobilize additional funding to cover 2016-2018
Complete the project’s staffing
Deploying the capacity development activity to promote wide use of the tool<br>
slide25. Selected references Michelle Hackett, Hugo Melgar-Quiñonez, Christopher A. Taylor, Martha Cecilia Álvarez Uribe. Factors associated with household food security of participants of the MANA food supplement program in Colombia. Archivos latinoamericanos de nutrición 2010; 6(1) 42-47
Hackett M, Melgar-Quiñonez H, Álvarez MC. Household food insecurity associated with stunting and underweight among preschool children in Antioquia, Colombia. Rev Panam Salud Pública. 2009;25(6):506-510.
Álvarez, MC, Estrada Alejandro. Seguridad alimentaria de los hogares colombianos según localización geográfica y algunas condiciones sociodemográficos. Revista Perspectiva de Nutrición Humana. 2008:10(1):37-50
Hackett M, Melgar-Quinonez H, Alvares MC. Internal validity of a household security scale es consistent among diverse populations participating in a food supplement program in Colombia. BMC Public Health. 2008, 8:175
Pérez-Escamilla, R, Melgar-Quiñonez, H, Nord, M, Álvarez, MC, Segall-Correa, AM. Análisis comparativo entre ítems de las escalas de seguridad alimentaria usadas en Brasil (PNAD 2004) y Colombia (ENSIN 2005).
Perspectivas de Nutrición Humana. Separata Memorias de la 1 Conferencia en América Latina y el Caribe sobre la medición de la seguridad alimentaria. 2007: 103-12.
Álvarez, MC. La experiencia colombiana en la adaptación y validación de la Escala de Seguridad Alimentaria. Perspectivas de Nutrición Humana. Separata Memorias de la 1 Conferencia en América Latina y el Caribe sobre la medición de la seguridad alimentaria. 2007: 77-87.
Álvarez, MC. Seguridad Alimentaria en los hogares colombianos. En: Encuesta nacional sobre la situación nutricional en Colombia. Bogotá 2007:319-335. ISBN 978-958-623-087-2.
Álvarez, MC, Estrada Alejandro, Melgar-Quiñonez. Validación de escala de seguridad alimentaria doméstica en Antioquia, Colombia. Rev Salud Pública de México. 2006;48:474-481.<br>
slide26. “Voices of the Hungry” Thank you for your attention! Pietro Gennari, ESS
Voices-of-the-Hungry@FAO.org<br>