Measuring the prevalence of hunger in the Philippines directly from food consumption data

• Main Authors: Arcilla, Rechel Guino, Albacea, Zita VJ, David, Isidoro P.
• Format: text
• Published: Animo Repository 2008
• Subjects: Hunger > Philippines; Hunger > Measurement; Food consumption > Philippines; Poverty > Philippines; Statistics and Probability
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/11991
• Institution: De La Salle University

Two methodologies for measuring prevalence of hunger, defined as the consumption of a diet inadequate to sustain good health and normal activity, growth and development, were developed using the 2003 Food Consumption Survey. The first methodology compared per capita energy consumption using the divisor total consumption unit TCU with per capita hunger threshold. The second methodology compared household's total dietary energy consumption with household hunger threshold.Prevalence of hunger was estimated using the design-based combined ration estimator rg. It is a ratio of two counts : total number of hungry households (persons) and total number of households (persons). Properties of the design-based estimates of hunger prevalence were determined and evaluated in terms of their bias, standard error, and coefficient of variation (CV).Results showed that 60.95% of Filipino households and 65.05% of Filipinos were hungry using the first methodology, higher than the estimates obtained using the second methodology at 55.99% and 60.26%, respectively. Although, both methodologies gave estimates that were accurate with negligible biases relative to its standard error, precise having standard errors lower than 0.012, and reliable with CVs at most 1.92%, the second methodology was recommended for it avoids the difficulty of choosing the divisor for per capita calculations.