{"doc_desc":{"title":"Global Consumption Database 2010 (version 2014-03)","idno":"WLD_GCD_2010_v2014-03_M","producers":[{"name":"Development Data Group ","affiliation":"World Bank"}],"prod_date":"2022-02-20"},"study_desc":{"title_statement":{"idno":"WLD_GCD_2010_v2014-03_M","title":"Global Consumption Database 2010 (version 2014-03)","alternate_title":"GCD 2010","identifiers":[{"type":"DOI","identifier":"https:\/\/doi.org\/10.48529\/0s8s-xk76"}]},"authoring_entity":[{"name":"Development Data Group (DECDG)","affiliation":"World Bank"}],"production_statement":{"producers":[{"name":"Olivier Dupriez","affiliation":"World Bank, Development Data Group (DECDG)"},{"name":"Tefera Bekele","affiliation":"World Bank, Development Data Group (DECDG)"},{"name":"Yuri Dikhanov","affiliation":"World Bank, Development Data Group (DECDG)"}],"prod_date":"2014-03"},"distribution_statement":{"distributors":[{"name":"World Bank Microdata Library","affiliation":"World Bank Group","uri":"http:\/\/microdata.worldbank.org"}],"contact":[{"name":"Data helpdesk","email":"data@worldbank.org","uri":"https:\/\/data.worldbank.org\/about\/contact"}]},"version_statement":{"version":"GCD v2014-03","version_date":"2014-03"},"study_info":{"keywords":[{"keyword":"coicop"},{"keyword":"consumption"},{"keyword":"expenditure"},{"keyword":"consumption pattern"},{"keyword":"consumption profile"},{"keyword":"consumption share"},{"keyword":"icp basic heading"}],"abstract":"The Global Consumption Database (GCD) contains information on consumption patterns at the national level, by urban\/rural area, and by income level (4 categories: lowest, low, middle, higher with thresholds based on a global income distribution), for 92 low and middle-income countries, as of 2010. The data were extracted from national household surveys. The consumption is presented by category of products and services of the International Comparison Program (ICP) 2005, which mostly corresponds to COICOP. For three countries, sub-national data are also available (Brazil, India, and South Africa). Data on population estimates are also included.\n\n                  The data file can be used for the production of the following tables (by urban\/rural and income class\/consumption segment):\n                  - Sample Size by Country, Area and Consumption Segment (Number of Households)\n                  - Population 2010 by Country, Area and Consumption Segment\n                  - Population 2010 by Country, Area and Consumption Segment, as a Percentage of the National Population\n                  - Population 2010 by Country, Area and Consumption Segment, as a Percentage of the Area Population\n                  - Population 2010 by Country, Age Group, Sex and Consumption Segment\n                  - Household Consumption 2010 by Country, Sector, Area and Consumption Segment in Local Currency (Million)\n                  - Household Consumption 2010 by Country, Sector, Area and Consumption Segment in $PPP (Million)\n                  - Household Consumption 2010 by Country, Sector, Area and Consumption Segment in US$ (Million)\n                  - Household Consumption 2010 by Country, Category of Product\/Service, Area and Consumption Segment in Local Currency (Million)\n                  - Household Consumption 2010 by Country, Category of Product\/Service, Area and Consumption Segment in $PPP (Million)\n                  - Household Consumption 2010 by Country, Category of Product\/Service, Area and Consumption Segment in US$ (Million)\n                  - Household Consumption 2010 by Country, Product\/Service, Area and Consumption Segment in Local Currency (Million)\n                  - Household Consumption 2010 by Country, Product\/Service, Area and Consumption Segment in $PPP (Million)\n                  - Household Consumption 2010 by Country, Product\/Service, Area and Consumption Segment in US$ (Million)\n                  - Per Capita Consumption 2010 by Country, Sector, Area and Consumption Segment in Local Currency\n                  - Per Capita Consumption 2010 by Country, Sector, Area and Consumption Segment in US$\n                  - Per Capita Consumption 2010 by Country, Sector, Area and Consumption Segment in $PPP\n                  - Per Capita Consumption 2010 by Country, Category of Product\/Service, Area and Consumption Segment in Local Currency\n                  - Per Capita Consumption 2010 by Country, Category of Product\/Service, Area and Consumption Segment in US$\n                  - Per Capita Consumption 2010 by Country, Category of Product\/Service, Area and Consumption Segment in $PPP\n                  - Per Capita Consumption 2010 by Country, Product or Service, Area and Consumption Segment in Local Currency\n                  - Per Capita Consumption 2010 by Country, Product or Service, Area and Consumption Segment in US$\n                  - Per Capita Consumption 2010 by Country, Product or Service, Area and Consumption Segment in $PPP\n                  - Consumption Shares 2010 by Country, Sector, Area and Consumption Segment (Percent)\n                  - Consumption Shares 2010 by Country, Category of Products\/Services, Area and Consumption Segment (Percent)\n                  - Consumption Shares 2010 by Country, Product\/Service, Area and Consumption Segment (Percent)\n                  - Percentage of Households who Reported Having Consumed the Product or Service by Country, Consumption Segment and Area (as of Survey Year)","time_periods":[{"start":"2010","end":"2010"}],"nation":[{"name":"Afghanistan","abbreviation":"AFG"},{"name":"Albania","abbreviation":"ALB"},{"name":"Armenia","abbreviation":"ARM"},{"name":"Azerbaijan","abbreviation":"AZE"},{"name":"Bangladesh","abbreviation":"BGD"},{"name":"Belarus","abbreviation":"BLR"},{"name":"Benin","abbreviation":"BEN"},{"name":"Bhutan","abbreviation":"BTN"},{"name":"Bolivia","abbreviation":"BOL"},{"name":"Bosnia and Herzegovina","abbreviation":"BIH"},{"name":"Brazil","abbreviation":"BRA"},{"name":"Bulgaria","abbreviation":"BGR"},{"name":"Burkina Faso","abbreviation":"BFA"},{"name":"Burundi","abbreviation":"BDI"},{"name":"Cambodia","abbreviation":"KHM"},{"name":"Cameroon","abbreviation":"CMR"},{"name":"Cape Verde","abbreviation":"CPV"},{"name":"Chad","abbreviation":"TCD"},{"name":"China","abbreviation":"CHN"},{"name":"Colombia","abbreviation":"COL"},{"name":"Congo, Democratic Republic","abbreviation":"COD"},{"name":"Congo, Rep.","abbreviation":"COG"},{"name":"C\u00f4te d'Ivoire","abbreviation":"CIV"},{"name":"Djibouti","abbreviation":"DJI"},{"name":"Egypt","abbreviation":"EGY"},{"name":"El Salvador","abbreviation":"SLV"},{"name":"Ethiopia","abbreviation":"ETH"},{"name":"Fiji","abbreviation":"FJI"},{"name":"Gabon","abbreviation":"GAB"},{"name":"Gambia (The)","abbreviation":"GMB"},{"name":"Ghana","abbreviation":"GHA"},{"name":"Guatemala","abbreviation":"GTM"},{"name":"Guinea","abbreviation":"GIN"},{"name":"Honduras","abbreviation":"HND"},{"name":"India","abbreviation":"IND"},{"name":"Indonesia","abbreviation":"IDN"},{"name":"Iraq","abbreviation":"IRQ"},{"name":"Jamaica","abbreviation":"JAM"},{"name":"Jordan","abbreviation":"JOR"},{"name":"Kazakhstan","abbreviation":"KAZ"},{"name":"Kenya","abbreviation":"KEN"},{"name":"Kyrgyz Republic","abbreviation":"KGZ"},{"name":"Lao PDR","abbreviation":"LAO"},{"name":"Latvia","abbreviation":"LVA"},{"name":"Lesotho","abbreviation":"LSO"},{"name":"Liberia","abbreviation":"LBR"},{"name":"Lithuania","abbreviation":"LTU"},{"name":"Macedonia (FYROM)","abbreviation":"MKD"},{"name":"Madagascar","abbreviation":"MDG"},{"name":"Malawi","abbreviation":"MWI"},{"name":"Malaysia","abbreviation":"MYS"},{"name":"Maldives","abbreviation":"MDV"},{"name":"Mali","abbreviation":"MLI"},{"name":"Mauritania","abbreviation":"MRT"},{"name":"Mauritius","abbreviation":"MUS"},{"name":"Mexico","abbreviation":"MEX"},{"name":"Moldova","abbreviation":"MDA"},{"name":"Mongolia","abbreviation":"MNG"},{"name":"Montenegro","abbreviation":"MNE"},{"name":"Morocco","abbreviation":"MAR"},{"name":"Mozambique","abbreviation":"MOZ"},{"name":"Namibia","abbreviation":"NAM"},{"name":"Nepal","abbreviation":"NPL"},{"name":"Nicaragua","abbreviation":"NIC"},{"name":"Niger","abbreviation":"NER"},{"name":"Nigeria","abbreviation":"NGA"},{"name":"Pakistan","abbreviation":"PAK"},{"name":"Papua New Guinea","abbreviation":"PNG"},{"name":"Paraguay","abbreviation":"PRY"},{"name":"Peru","abbreviation":"PER"},{"name":"Philippines","abbreviation":"PHL"},{"name":"Romania","abbreviation":"ROU"},{"name":"Russia","abbreviation":"RUS"},{"name":"Rwanda","abbreviation":"RWA"},{"name":"Sao Tome and Principe","abbreviation":"STP"},{"name":"Senegal","abbreviation":"SEN"},{"name":"Serbia","abbreviation":"SRB"},{"name":"Sierra Leone","abbreviation":"SLE"},{"name":"South Africa","abbreviation":"ZAF"},{"name":"Sri Lanka","abbreviation":"LKA"},{"name":"Swaziland","abbreviation":"SWZ"},{"name":"Tajikistan","abbreviation":"TJK"},{"name":"Tanzania","abbreviation":"TZA"},{"name":"Thailand","abbreviation":"THA"},{"name":"Timor Leste","abbreviation":"TLS"},{"name":"Togo","abbreviation":"TGO"},{"name":"Turkiye","abbreviation":"TUR"},{"name":"Uganda","abbreviation":"UGA"},{"name":"Ukraine","abbreviation":"UKR"},{"name":"Viet Nam","abbreviation":"VNM"},{"name":"Yemen","abbreviation":"YEM"},{"name":"Zambia","abbreviation":"ZMB"}],"geog_coverage_notes":"For all countries, estimates are provided at the national level and at the urban\/rural levels.\n      For Brazil, India, and South Africa, data are also provided at the sub-national level (admin 1):\n      - Brazil: ACR, Alagoas, Amapa, Amazonas, Bahia, Ceara, Distrito Federal, Espirito Santo, Goias, Maranhao, Mato Grosso, Mato Grosso do Sul, Minas Gerais, Para, Paraiba, Parana, Pernambuco, Piaji, Rio de Janeiro, Rio Grande do Norte, Rio Grande do Sul, Rondonia, Roraima, Santa Catarina, Sao Paolo, Sergipe, Tocatins\n      - India: Andaman and Nicobar Islands, Andhra Pradesh, Arinachal Pradesh, Assam, Bihar, Chandigarh, Chattisgarh, Dadra and Nagar Haveli, Daman and Diu, Delhi, Goa, Gujarat, Haryana, Himachal Pradesh, Jammu and Kashmir, Jharkhand, Karnataka, Kerala, Lakshadweep, Madya Pradesh, Maharastra, Manipur, Meghalaya, Mizoram, Nagaland, Orissa, Pondicherry, Punjab, Rajasthan, Sikkim, Tamil Nadu, Tripura, Uttar Pradesh, Uttaranchal, West Bengal\n      - South Africa: Eastern Cape, Free State, Gauteng, Kwazulu Natal, Limpopo, Mpulamanga, Northern Cape, North West, Western Cape","data_kind":"Data derived from survey microdata","quality_statement":{"other_quality_statement":"All surveys used have a nationwide coverage. Their sample size ranges from less than 2,000 households to more than 100,000. The universe of each survey is composed of ordinary households only; institutional households (prisons, military barracks, hospitals, convents, and others) are not covered by household surveys. Homeless and nomadic populations and visitors present in a country during a survey are also excluded from the sample.\n\n        Few developing countries conduct household consumption or expenditure surveys on an annual basis. International organizations recommend conducting such surveys every three or four years. The surveys used in the database were conducted between 2000 and 2010 (except the one for Djibouti, which was conducted in 1996); most were conducted during the period 2007-10. All data presented in the Global Consumption Database are as of 2010. When based on a survey conducted before 2010, the estimates were obtained by extrapolation, as described in the notes on the standardization of data (see Step 4).\n\n        Household survey datasets are complemented by data on population, purchasing power parity (PPP) conversion factors, and average exchange rates obtained from the World Bank's World Development Indicators database.\n\n        Because of the diversity of methods and instruments used by the surveys, comparability across countries is limited. Survey questionnaires are provided below as an important metadata component. Links are also provided to the microdata when available.\n\n        Because household surveys differ across countries in design, methodology, and timing, there are limits to the extent to which household data can be standardized after they have been collected. Comparisons of household data across countries and over time must therefore be done with caution.\n\n        The Global Consumption Database uses multiple types of surveys, depending on data availability-including household budget surveys, living standards measurement surveys, and various kinds of country-specific socioeconomic surveys. All these surveys measure consumption or expenditure at the household (not individual) level. But because the surveys are designed for different purposes (such as to measure poverty or to update the consumption basket used to compile consumption price indices), they may differ substantially in design and methodology.\n\n        Key differences between surveys include these:\n\n        Duration of data collection. Data may be collected over a period of 12 months to account for seasonality or over a shorter period (a few weeks or a few months).\n        - Method for household reporting on consumption. Some surveys collect data on food and some nonfood consumption using diaries in which households or individuals report daily on what they spend. But most rely on the recall method, asking households to report what they recall spending over a certain period. The recall period varies across surveys and categories. For example, data might be collected on spending on food for the past 7 days, the past 2 weeks, or a typical month; on education for the past 12 months or the last academic year; on rent, outpatient health services, and clothing and footwear for the past month or the past 4 weeks; and on durable goods and hospitalization for the past 6 or 12 months. The choice of recall period may have a substantial effect on the levels of consumption reported. Longer recall periods for frequently purchased items typically produce lower levels of reported spending than do shorter recall periods.\n        - Level of detail. Some survey questionnaires include a long, detailed list of goods and services; others provide a shorter, more aggregate list. Longer lists with a finer breakdown of categories typically generate higher estimates of consumption.\n        - Method for estimating rental value of dwellings. In some countries, surveys ask households that own their home or occupy it for free to provide an estimate of the rental value of the dwelling. In others, surveys collect data on the characteristics of dwellings that can be used to impute the rental value of owner-occupied dwellings through hedonic regressions. And in still other countries it is not possible to measure the rental value of owner-occupied dwellings because the rental market is too limited. Because this rental value represents a substantial share of household expenditure, these differences have major implications for the calculation of household consumption aggregates and for the comparability of data across countries.\n        - Method for estimating value of durable goods. Some surveys collect data on household expenditures on durable goods such as musical instruments. Others attempt to estimate the annual use value of these goods. Estimating the use value of a good requires data on its price and date of purchase or on its resale value, data that are not available in all surveys. This too affects the calculation of household consumption aggregates and the cross-country comparability of data.\n\n        See file Draft_Technical note on standardization process.pdf and questionnaire coverage by country.xlsx"},"method":{"data_collection":{"sources":[{"name":"Afghanistan - National Risk and Vulnerability Survey 2007","origin":"Central Statistics Organisation (CSO)"},{"name":"Albania - Living Standards Measurement Survey 2008","origin":"Institute of Statistics (INSTAT)"},{"name":"Armenia - Integrated Living Conditions Survey 2009","origin":"National Statistical Service of the Republic of Armenia"},{"name":"Azerbaijan - Household and Targeted Social Assistance Monitoring and Evaluation Survey 2008","origin":"State Statistical Committee of the Republic of Azerbaijan"},{"name":"Bangladesh - Household Income and Expenditure Survey 2010","origin":"Bangladesh Bureau of Statistics (BBS)"},{"name":"Belarus - Belarus Household Survey 2010","origin":"National Statistical Committee of the Republic of Belarus"},{"name":"Benin - Questionnaire des Indicateurs de Base du Bien-\u00eatre 2003","origin":"Institut National de la Statistique et de l'Analyse \u00c9conomique (INSAE)"},{"name":"Bhutan - Living Standards Survey 2007","origin":"National Statistics Bureau (NSB)"},{"name":"Bolivia - Encuesta de Hogares 2007","origin":"Instituto Nacional de Estad\u00edstica (INE)"},{"name":"Bosnia and Herzegovina - Household Budget Survey 2007","origin":"Statistics of Bosnia and Herzegovina (BHAS), Federal Office of Statistics (FOS), and Republika Srpska Institute for Statistics (RSIS)"},{"name":"Brazil - Pesquisa de Or\u00e7amentos Familiares 2008-2009","origin":"Instituto Brasileiro de Geografia e Estat\u00edstica (IBGE)"},{"name":"Bulgaria - Household Budget Survey 2003","origin":"National Statistical Institute (NSI)"},{"name":"Burkina Faso - Enqu\u00eate burkinab\u00e9 sur les conditions de vie des m\u00e9nages 2009","origin":"Institut National de la Statistique et de la D\u00e9mographie (INSD)"},{"name":"Burundi - Questionnaire des Indicateurs de Base du Bien-Etre 2006","origin":"Institut de Statistique et d'\u00e9tudes Economiques du Burundi (ISTEEBU)"},{"name":"Cambodia - Socio-economic survey 2007-2008","origin":"National Institute of Statistics (NIS)"},{"name":"Cameroon - Enqu\u00eate Camerounaise aupr\u00e8s des M\u00e9nages III 2007","origin":"Institut National de la Statistique (INS)"},{"name":"Cabo Verde - Question\u00e1rio Unificado de Indicadores B\u00e1sicos de Bem-Estar 2007","origin":"Instituto Nacional de Estat\u00edstica (INE)"},{"name":"Chad - Enqu\u00eate sur la Consommation et le Secteur Informel au Tchad 2003","origin":"Institut National de la Statistique des Etudes Economiques et D\u00e9mographiques"},{"name":"China - Synthetic dataset"},{"name":"Colombia - Encuesta Nacional de Calidad de Vida 2010","origin":"Direcci\u00f3n de Metodolog\u00eda y Producci\u00f3n Estad\u00edstica (DIMPE)"},{"name":"Congo, Dem. Rep. - Enqu\u00eate 1-2-3 sur l'emploi, le secteur informel et les conditions de vie des m\u00e9nages 2004","origin":"Institut National de la Statistique (INS)"},{"name":"Congo, Rep. - Questionnaire des Indicateurs de Base du Bien-\u00eatre 2005","origin":"Centre National de la Statistique et des Etudes Economiques (CNSEE)"},{"name":"C\u00f4te d'Ivoire - Enqu\u00eate Niveau de vie des M\u00e9nages 2008","origin":"Institut National de la Statistique (INS)"},{"name":"Djibouti - Enqu\u00eate Djiboutienne aupr\u00e8s des m\u00e9nages 1996","origin":"Direction Nationale de la Statistique de Djibouti (DINAS)"},{"name":"Egypt - Household Expenditure and Consumption Survey 2009","origin":"Central Agency for Public Mobilization and Statistics (CAPMAS)"},{"name":"El Salvador - Encuesta de Hogares de Prop\u00f3sitos M\u00faltiples 2010","origin":"Direcci\u00f3n General de Estad\u00edstica y Censos"},{"name":"Ethiopia - Household Income Consumption and Expenditure 2004","origin":"Central Statistical Agency (CSA)"},{"name":"Fiji - Household Income and Expenditure Survey 2002","origin":"Bureau of Statistics (FIBoS)"},{"name":"Gabon - Enqu\u00eate Gabonaise pour l\u2019Evaluation et le suivi de la Pauvret\u00e9 2005","origin":"Direction G\u00e9n\u00e9rale de la Statistique et des Etudes Economiques (DGSEE)"},{"name":"Gambia (The) - Integrated Household Survey 2003","origin":"Gambia Bureau of Statistics (GBOS)"},{"name":"Ghana - Ghana Living Standards Survey 2006","origin":"Ghana Statistical Service (GSS)"},{"name":"Guatemala - Encuesta Nacional sobre Condiciones de Vida 2000","origin":"Instituto Nacional de Estad\u00edstica (INE)"},{"name":"Guinea - Enqu\u00eate L\u00e9g\u00e8re pour l'Evaluation de la Pauvret\u00e9 2007","origin":"Institut National de la Statistique (INS)"},{"name":"Honduras - Encuesta Nacional de Condiciones de Vida 2004","origin":"Instituto Nacional de Estadistica (INE)"},{"name":"India - National Sample Survey 66th Round 2009","origin":"National Sample Survey Organization (NSSO)"},{"name":"Indonesia - National Socio-Economic Survey 2010","origin":"Statistics Indonesia (BPS)"},{"name":"Iraq - Household Socio Economic Survey 2006","origin":"Central Organization for Statistics and Information Technology (COSIT) and Kurdistan Regional Statistics Office (KRSO)"},{"name":"Jamaica - Jamaica Survey of Living Conditions 2007","origin":"Statistical Institute of Jamaica (STATIN)"},{"name":"Jordan - Household Income and Expenditure Survey 2002","origin":"Department of Statistics (DoS)"},{"name":"Kazakhstan - Household Budget Survey 2003","origin":"The Agency of Statistics of the Republic of Kazakhstan"},{"name":"Kenya - Integrated Household Budget Survey 2005","origin":"Kenya National Bureau of Statistics (KNBS)"},{"name":"Kyrgyz Republic - Integrated Household Survey 2010","origin":"National Statistical Committee of the Kyrgyz Republic"},{"name":"Lao PDR - Household Expenditure and Consumption Survey 2007","origin":"Department of Statistics (DoS)"},{"name":"Latvia - Household Budget Survey 2009","origin":"Central Statistical Bureau of Latvia"},{"name":"Lesotho - Household Budget Survey 2002","origin":"Bureau of Statistics (BoS)"},{"name":"Liberia - Core Welfare Indicators Questionnaire 2007","origin":"Liberia Institute of Statistics and Geo_Information Services"},{"name":"Lithuania - Household Budget Survey 2008","origin":"Statistics Lithuania"},{"name":"Macedonia, FYR - Household Budget Survey 2003","origin":"Republic of Macedonia State Statistical Office"},{"name":"Madagascar - Enqu\u00eate Permanente Aupr\u00e8s des M\u00e9nages 2005","origin":"Institut National de la Statistique (INSTAT)"},{"name":"Malawi - Third Integrated Household Survey 2010-2011","origin":"National Statistical Office (NSO)"},{"name":"Malaysia - Household Expenditure Survey 2004","origin":"Department of Statistics"},{"name":"Maldives - Household Income and Expenditure Survey 2009-2010","origin":"Department of National Planning"},{"name":"Mali - Enqu\u00eate L\u00e9g\u00e8re Int\u00e9gr\u00e9e aupr\u00e8s des M\u00e9nages 2006","origin":"Direction Nationale de la Statistique et de l'Informatique (DNSI)"},{"name":"Mauritania - Enqu\u00eate Permanente sur les Conditions de Vie des M\u00e9nages 2004","origin":"Office National de la Statistique"},{"name":"Mauritius - Household Budget Survey 2006-2007","origin":"Statistics Mauritius"},{"name":"Mexico - Encuesta Nacional de Ingreso-Gasto de los Hogares 2010","origin":"Instituto Nacional de Estad\u00edstica y Geograf\u00eda (INEGI)"},{"name":"Moldova - Household Budget Survey 2009","origin":"National Bureau of Statistics of the Republic of Moldova"},{"name":"Mongolia - Household Income and Expenditure Survey 2007-2008","origin":"National Statistical Office (NSO)"},{"name":"Montenegro - Household Budget Survey 2009","origin":"Statistical Office of Montenegro"},{"name":"Morocco - Enqu\u00eate Nationale sur la Consommation et les D\u00e9pense des M\u00e9nage 2000","origin":"Statistics Department"},{"name":"Mozambique - Inqu\u00e9rito aos Agregados Familiares 2008","origin":"Direc\u00e7\u00e3o de Censos e Inqu\u00e9ritos"},{"name":"Namibia - Household Income Expenditure Survey 2009","origin":"Namibia Statistics Agency"},{"name":"Nepal - Living Standards Survey 2010","origin":"Central Bureau of Statistics (CBS)"},{"name":"Nicaragua - Encuesta Nacional de Hogares sobre Medici\u00f3n de Nivel de Vida 2005","origin":"National Institute of Statistics and Census"},{"name":"Niger - Enqu\u00eate Nationale sur le Budget et la Consommation des M\u00e9nages 2007","origin":"Institut National de la Statistique"},{"name":"Nigeria - Living Standards Survey 2009","origin":"National Bureau of Statistics (NBS)"},{"name":"Pakistan - Social and Living Standards Measurement Survey 2010","origin":"Federal Bureau of Statistics (FBS)"},{"name":"Papua New Guinea - Household Income and Expenditure Survey 2009","origin":"National Statistical Office (NSO)"},{"name":"Paraguay - Encuesta Integrada de Hogares 2000","origin":"Direcci\u00f3n General de Estad\u00edstica, Encuestas y Censos (DGEEC)"},{"name":"Peru - Enquesta Nacional de Hogares 2010","origin":"Instituto Nacional de Estad\u00edstica e Inform\u00e1tica (INEI)"},{"name":"Philippines - Family Income and Expenditure Survey 2006","origin":"National Statistics Office (NSO)"},{"name":"Romania - Household Budget Survey 2009","origin":"National Institute of Statistics (NIS)"},{"name":"Russia - Household Budget Survey 2008","origin":"Federal State Statistics Service of the Russian Federation"},{"name":"Rwanda - Enqu\u00eate Int\u00e9grale sur les Conditions de Vie des M\u00e9nages 2005","origin":"Direction de la Statistique (DS)"},{"name":"Sao Tome and Pr\u00edncipe - Enqu\u00eate sur les Conditions de Vie des M\u00e9nages 2000","origin":"Institut National de la Statistique (INE)"},{"name":"Senegal - Enqu\u00eate de Suivi de la Pauvret\u00e9 2005","origin":"Agence Nationale de la Statistique et de la D\u00e9mographie"},{"name":"Serbia - Living Standards Measurement Survey 2007","origin":"Statistical Office of the Republic of Serbia"},{"name":"Sierra Leone - Integrated Household Survey 2003","origin":"Statistics Sierra Leone (SSL)"},{"name":"South Africa - Income and Expenditure Survey 2010-2011","origin":"Statistics South Africa"},{"name":"Sri Lanka - Household Income and Expenditure Survey 2009","origin":"Department of Census and Statistics"},{"name":"Swaziland - Household Income and Expenditure Survey 2010","origin":"Central Statistical Office (CSO)"},{"name":"Tajikistan - Living Standards Measurement Survey 2009","origin":"State Statistical Agency"},{"name":"Tanzania - Household Budget Survey 2007","origin":"National Bureau of Statistics (NBS)"},{"name":"Thailand - Socio-Economic Survey 2009","origin":"National Statistical Office (NSO)"},{"name":"Timor-Leste - Survey of Living Standards 2010","origin":"National Statistics Directorate (NSD)"},{"name":"Togo - Questionnaire des Indicateurs de Base du Bien-\u00eatre 2006","origin":"Direction G\u00e9n\u00e9rale de la Statistique et de la Comptabilit\u00e9 Nationale (DGSCN)"},{"name":"Turkey - Household Income and Consumption Expenditures Survey 2009","origin":"State Institute of Statistics"},{"name":"Uganda - National Household Survey 2009","origin":"Uganda Bureau of Statistics (UBOS)"},{"name":"Ukraine - Household Budget Survey 2010","origin":"State Statistics Committe of Ukraine"},{"name":"Vietnam - Household Living Standard Survey 2008","origin":"Social and Environmental Statistics Department"},{"name":"Yemen, Rep. - Household Budget Survey 2005","origin":"Central Statistical Organization (CSO)"},{"name":"Zambia - Living Conditions Monitoring Survey VI 2010","origin":"Central Statistical Office (CSO)"}]},"data_processing":[{"type":"Annualizing consumption or expenditure data","description":"The Global Consumption Database draws on a diverse set of surveys. The data were standardized to the extent possible, through the six-step process described below.\n\n            Step 1: Annualizing consumption or expenditure data\n\n            The first step consisted of annualizing each household's consumption or expenditure data for each commodity (the data are nominal values in local currency; no regional price deflators were applied).\n\n            In some cases annualization is straightforward and consists simply of applying a multiplying factor to the data, which is determined by the recall period (the period on which households are asked to report by recalling their expenditure during that period). (For example, food data collected 'for the past 7 days' would be divided by 7, then multiplied by 365; monthly values would simply be multiplied by 12.) This is the method used for most purchased food products and regularly purchased nonfood products and services.\n\n            Annualization becomes more complex for home-produced and received goods and services, for which consumption values have to be calculated on the basis of data on quantities consumed and local (farm-gate or factory-gate) prices.\n\n            Another challenge in annualizing data is that quantities are not always reported in metric units and not all countries provide conversion factors for nonstandard measurement units.\n\n            Two consumption items are typically problematic: imputed rents and use value of durable goods.\n\n            Imputed rents. A rental value may be imputed for owner-occupied dwellings, through a process using hedonic regression models. But because the rental market is often very limited in developing countries, especially in rural areas, this process can be difficult and often produces unrealistic estimates.\n            Durable goods. Some surveys are intended to measure expenditure and collect data on purchases of durable goods with a recall period of one year. In these cases no annualization is needed. Many other surveys seek to measure consumption, not expenditure. In these cases an annual use value of the durable goods is calculated through the use of depreciation rates-a process that requires information, not always available, on the date and price of purchase and the estimated resale value (see Angus Deaton and Salam Zaidi, 'Guidelines for Constructing Consumption Aggregates for Welfare Analysis', Living Standards Measurement Study Working Paper 135, World Bank, Washington, DC, 2002)."},{"type":"Detecting and fixing outliers","description":"Step 2: Detecting and fixing outliers\n\n            The second step was to detect and fix outliers. All datasets obtained from countries contain outlying expenditure values.\n\n            There are two types of such outliers: those that indicate big spenders (rich households), which are valid values and should not be removed or changed, and those that are a result of errors in data collection (reporting), data coding, or data entry, which need to be fixed.\n\n            Expenditure data can be unrealistically high or low. Unrealistically low values, i.e. the bottom outliers (for example, a household consuming so little food that survival could not be possible) are difficult to detect and fix. Because a minimum level of consumption cannot be defined by commodity, there was no attempt to make imputations to compensate for low spending. Instead, the focus was on fixing the top outliers.\n\n            To fix top outliers resulting from data coding or data entry errors in variables related to quantities consumed (for example, the code for kg being applied rather than gram, which would lead to an overestimate of the true value by a factor of 1,000), the outlying values were replaced with the maximum of the valid positive values, calculated separately for urban and rural areas.\n\n            The detection and imputation rules used (described below) are conservative, and the proportion of outlying records found in the datasets was usually low. But fixing these outliers had a significant effect on the consumption distribution in some countries.\n\n            A value was flagged as being a potential outlier if it exceeded the average amount consumed in the third quartile plus 5 times the interquartile range, where the interquartile range is the difference between the first and third quartiles of the data. For some items (for example, food, transport services, and personal effects), the outliers were detected through the use of per capita values. For others (for example, rent or durable goods), they were detected through the use of per-household values.\n\n            Once records were flagged on the basis of these criteria, an additional confirmation step was run before imputations were made. It was assumed that if the values reported by a household for three or more nonfood items were flagged as being outliers, this might indicate a rich household. It was also assumed that relatively wealthy households (defined as those belonging to the top two consumption quintiles) might spend an unusually large share of their income on education or jewelry. Flags on the corresponding records were therefore removed. Outlier values were then replaced with the weighted mean of the nonextreme values for the consumption variable in question. Urban and rural means were calculated separately, by decile of population. Doing this resulted in an imputed value that is higher for wealthier households."},{"type":"Mapping commodities to the ICP\/COICOP classification","description":"Step 3: Mapping commodities to the ICP\/COICOP classification\n\n            The third step was to map commodities found in each survey dataset to a standard classification of products and services, and to aggregate these standard products and services into sectors and categories. This step used the International Comparison Program (ICP) classification, equivalent to the international Classification of Individual Consumption According to Purpose (COICOP). The ICP breaks down household consumption into 110 basic headings (107 of which can possibly be found in household surveys). A detailed description of the sectors, categories and products and services is provided here as an XLS document.\n\n            The design of some household survey questionnaires is based on the COICOP classification, or a national adaptation of it, which makes the mapping easier. But in many cases survey questionnaires do not provide sufficient detail in describing goods and services or do not cover all basic headings. Four situations can occur:\n\n            No data are available in the survey for a particular COICOP basic heading. The ability to measure true household consumption would require that a survey questionnaire cover all possible categories of products and services. But none of the surveys collected data on all 107 basic headings. Some exclusions in the coverage of commodities are predictable and justified, such as for pork and alcoholic beverages in Muslim countries and for rare items such as purchases of package holidays in low-income countries. But others are less justified and may be a result of poor design of the survey instrument. The extent of gaps in the coverage of commodities varies from country to country.\n            One item in the survey corresponds to one COICOP basic heading. In some cases there is a perfect match between an item in the survey questionnaire and a basic heading. The mapping is straightforward.\n            Multiple items correspond to one COICOP basic heading. This is most often the case for food items, such as fruits and vegetables. For example, in the dataset for Brazil (where the diary method was used to collect data on daily consumption), 274 different items are mapped to the basic heading 'fresh or chilled vegetables other than potatoes.' In the dataset for Bangladesh 22 items are mapped to this basic heading. Mapping is also straightforward in these cases; it involves mapping the multiple values reported by the household to the relevant basic heading.\n            One item in the survey corresponds to more than one COICOP basic heading. For example, a survey could ask respondents to report their expenditures on 'gas and electricity', while under the COICOP classification 'gas' and 'electricity' are two separate basic headings. Cases like these require splitting the value reported by the household between two or more basic headings. This was done by using national accounts data provided by the countries."},{"type":"Extrapolation to 2010","description":"Step 4: Extrapolation to 2010\n\n            In the fourth step, extrapolations were done to convert all consumption and population data to a common reference year, 2010.\n\n            For consumption data, the 2010 values were obtained by multiplying the survey values by the ratio of the household final consumption expenditure per capita (current) in 2010 to the corresponding value in the survey year. These data account both for inflation and for real growth in household consumption. The household final consumption data were obtained from the World Bank's World Development Indicators database: household final consumption expenditure per capita (current LCU), series code NE.CON.PRVT.CN (downloaded on October 2, 2012).\n\n            For Guinea, for example, the survey was conducted in 2007. The household final consumption expenditure per capita in current local currency units was 3,177,774 in 2010 and 1,547,012 in 2007 (the survey year). All survey values were thus multiplied by 3,177,774\/1,547,012 = 2.054137.\n\n            The consumption data were then converted from local currencies into U.S. dollars and into international dollars adjusted for purchasing power parity (PPP$), again by using data from the World Development Indicators database: official exchange rate (LCU per US$, period average), series code PA.NUS.FCRF (downloaded on October 3, 2012); and PPP conversion factor, private consumption (LCU per international $), series code PA.NUS.PRVT.PP (downloaded on October 2, 2012). Please note that these conversion factors are based on the 2005 round of the International Comparison Program (ICP) and not on the 2011 ICP round whose results have been published in May 2014).\n\n            For population data, the household weighting coefficients (sample weights) in the survey datasets were adapted by multiplying them by a factor that would guarantee that the extrapolated survey population (separated into urban and rural segments) would correspond to the 2010 population data published in the World Development Indicators database: population, total, rural, and urban, series code SP.POP.TOTL, SP.RUR.TOTL, and SP.URB.TOTL (downloaded on October 1, 2012)."},{"type":"Review and validation","description":"Step 5: Review and validation\n\n            The resulting data (particularly the mean and distribution of aggregate consumption) were compared with data from other sources, particularly the respective survey reports and the World Bank's online poverty database, Povcalnet . The Global Consumption Database is not an exact replication of national or Povcalnet estimates, because of differences in the methods used for annualization and for fixing outliers. In addition, the Global Consumption Database includes consumption items (particularly health expenditure) that would not be used in measuring poverty. The database therefore should not be used to produce poverty estimates."},{"type":"Production of summary tables and metadata","description":"Step 6: Production of summary tables and metadata\n\n            The sixth step was to generate a standard set of tables for each country showing consumption and demographic patterns across consumption segments (established by using global thresholds defined in PPP$ terms), by population category (e.g. lowest, low, middle, higher consumption segment), and for both urban and rural areas. Other information was added, including metadata on the survey and questionnaire design. Finally, an aggregate table summarizing the information from all country tables was produced."}]}}},"tags":[{"tag":"DOI"}],"schematype":"survey"}