The Gambia Poverty and Gender Assessment 2022 - Appendix 1: Construction of the Welfare Aggregate

Jul 19, 2026

Upload by Asher

Appendix 1 documents how The Gambia’s poverty and gender assessment constructs a monetary welfare indicator through a detailed consumption aggregate. It specifies which market and non-market transactions are included, how “lumpy” purchases are handled via annual use values, and how the aggregate is divided into food, education, frequent non-food (including housing), and infrequent non-food durable goods. The appendix also explains annualization methods for food and health spending and the imputation of geographic prices and value outliers.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages25

Reading Time1 h 3 min

Appendix 1: Construction of the welfare aggregate

Construction of nominal total consumption

Components of the consumption aggregate

Food consumption

Health expenditure

Education expenditure

Frequent non-food expenditures

Housing imputed rent estimation

Public Disclosure Authorized Public Disclosure Authorized Appendix 1: Construction of the welfare aggregate Construction of nominal total consumption Following common practice in The Gambia, as well as other low and lower-middle countries, consumption is used as the main indicator of monetary welfare because it is likely to be a more reliable and accurate measure of living standards than income in these contexts.1 Several theoretical and practical considerations guided the construction of the consumption aggregate. First, both market and non-market transactions should be included; this meant ensuring that own production of food products and food gifts (including eating with family/friends) were captured. Second, while the survey collects data on consumption for many food and non-food items, consumption on "lumpy" purchases such as durable goods and owner-occupied dwellings was not directly included. In these cases, additional analysis was performed to compute the "annual use value" of these assets. The consumption aggregate comprises four main components: food, education, frequent non-food items (including housing), and infrequent non-food items (durable goods) . Table A.1 indicates the sections in the IHS 2020 survey questionnaire that provide the source data for these components. Table 1: Component of Consumption Aggregate | Consumption Aggregate | Questionnaire Sections | | --- | --- | | Food consumption | PART B Section 1A PART B Section 1B | | health expenditure | PART A Section 2 | | Education expenditure | PART A Section 3 | | Non-food frequent | PART A Section 8 PART B Section 2A PART B Section 2B PART B Section 2C PART B Section 2D | | Durable goods | PART A Section 9 | Food consumption The food consumption aggregate was broken down into two sub-aggregates. These are consumption outside the household such as in restaurants, and consumption within the household. For the subaggregate of food consumption outside the household, data were available at the household level on total expenditure for 18 items (based on type of food and type of food service provider) with a recall period of 7 days. For these expenditures, we annualized consumption by multiplying weekly expenditures by 52.4. 1 See Deaton and Zaidi (2002) and Haughton and Khandker (2009) . For the sub-aggregate of food consumption within the household, data was available at household level on the total quantity of over 200 items consumed over the past 7 days. Prices of items were obtained from actual household purchases and therefore geographic unit prices were used to impute values for own consumption and gifts. When there are missing prices, we started by assigning average Enumeration area (EA) prices within the EA. If these were missing, we moved to the next higher level (in the following order: district, rural/urban and national) . The use of national prices was extremely rare.2 To correct for value outliers after price imputation, we first identified outliers as values with a Z score greater than 3. In these cases, outliers were imputed using LGA (Local Government Area), item and type of unit. Finally, the total food (purchased, own consumption, gifts) by the household was then annualized. health expenditure Household expenditures on health and education services for all household members were collected. The aggregate does not include hospitalization expenses, on the grounds that they are not welfare-enhancing. It includes consultation fees, payments for prescribed medications purchased within or outside of the health care provider, and fees for tests and medical supplies over the past 4 weeks. It also includes dental expenses. Special annualization factors were used for most of the health expenditure components. For example, Xray, scanning, ambulance, childbirth services are all two-week recall expenses in the questionnaire but cannot be annualized, because they are not regular expenses. In such cases, in conjunction wi

FAQs

Why does the assessment use consumption as the main indicator of monetary welfare in The Gambia?

Consumption is used because it is likely to be more reliable and accurate for living standards than income in low and lower-middle income contexts.

How are “lumpy” purchases like durable goods and owner-occupied dwellings included?

They are not directly included in the basic consumption reporting; instead, additional analysis computes the “annual use value” of these assets.

What methods are used to impute values when geographic or unit prices are missing?

The appendix uses geographic unit prices imputation starting from average Enumeration Area prices, then district, rural/urban, and rarely national prices. It also corrects value outliers using Z-scores greater than 3 with imputation based on LGA, item, and unit type.

Jul 19, 2026

Upload by Asher