Living Standards Measurement Study
- Introduced by
- World Bank (Living Standards Measurement Study programme)
- Year
- 1980
- Type
- Multi-topic integrated household survey
- Family
- Household welfare survey
What does it do?
The Living Standards Measurement Study (LSMS) is a multi-topic integrated household survey programme launched by the World Bank in 1980 to improve the quality of household data for measuring and analysing welfare in developing countries.
Built around a modular questionnaire that links a detailed household interview to community and price questionnaires, the LSMS measures living standards through consumption expenditure rather than income, and connects welfare outcomes to their determinants — employment, education, health, agriculture, and access to services — within a single, internally consistent dataset.
The intuition behind the method
Rather than running separate single-purpose surveys for poverty, health, and agriculture, the LSMS interviews each household once but in depth, asking about everything that shapes its standard of living and recording how much it actually consumes. Because consumption is smoother and harder to misreport than income, a carefully built consumption aggregate gives a more reliable picture of welfare. Pairing the household interview with a community questionnaire (what services and infrastructure exist locally) and a price questionnaire (what things cost) lets analysts deflate spending into comparable real terms and trace how local conditions translate into household outcomes.
History
The World Bank established the Living Standards Measurement Study in 1980 to address the poor quality and limited policy relevance of existing household data in developing countries. Early surveys in Côte d'Ivoire and Peru in the mid-1980s established the integrated, consumption-based, multi-topic template. The 2000 two-volume guide edited by Margaret Grosh and Paul Glewwe consolidated 15 years of design lessons, and Angus Deaton's 1997 monograph codified the microeconometric analysis of the resulting data. The programme later spawned the LSMS-Integrated Surveys on Agriculture (LSMS-ISA) and has informed household survey design across dozens of countries.
When to use it
Use the LSMS approach when you need a single, multi-topic dataset that measures household welfare through consumption and links it to its socio-economic determinants for poverty analysis, policy design, and impact evaluation. It is well suited to producing nationally representative poverty and inequality estimates and to studying the joint determination of welfare, schooling, health, and labour. It is less appropriate when you need rapid, low-cost monitoring, very large samples for small-area precision, or a narrow single-topic instrument; its depth makes fieldwork costly and slow, and the consumption aggregate demands careful, expert construction.
Limitations and considerations
- The detailed consumption interview is long and costly, constraining sample sizes and limiting frequency of data collection.
- Constructing a defensible consumption aggregate (imputing own production, durables, and housing flows) requires considerable expertise and judgment.
- Recall-based consumption is sensitive to the length of the recall period and to telescoping and recall error, which can bias welfare estimates.
- Modest sample sizes limit reliable disaggregation to small geographic areas without combining the data with census-based small-area estimation.
Source conditions do not replace independent scientific review or validation by an analysis engine.
Strengths
- Measures welfare through consumption, which is smoother over time and generally more reliably reported than income in low-income settings.
- The integrated multi-topic design links welfare to its determinants in one internally consistent dataset, enabling multivariate analysis of poverty.
- Community and price questionnaires allow spatial price deflation and contextual analysis that single household surveys cannot support.
- Decades of methodological documentation and World Bank support make the design well understood, comparable across countries, and widely replicated.
Applications
- Producing official national poverty headcounts, poverty gaps, and inequality measures from a consumption-based welfare aggregate.
- Estimating the determinants of welfare, schooling, and health through multivariate analysis of the integrated modules.
- Evaluating the welfare impact of policy reforms and programmes such as subsidy changes, social transfers, or service expansion.
- Analysing agricultural production, input use, and farm-household welfare through the LSMS-ISA agricultural modules.
- Providing the household-survey input for small-area estimation that combines LSMS welfare data with census covariates.
How it works
How it works
The survey is built from interlocking modules — a household questionnaire covering consumption, education, health, employment, housing, and assets; a community questionnaire on local infrastructure, services, and shocks; and a price questionnaire recording prices of common goods in local markets. Modules are designed so that consumption can be aggregated and so that welfare can be related to its determinants, with the questionnaire customised to country context while preserving cross-survey comparability.
A multi-stage, stratified, clustered probability sample is drawn, typically selecting primary sampling units (census enumeration areas) with probability proportional to size and then a fixed number of households per cluster. Sampling weights are constructed as the inverse of selection probabilities and adjusted for non-response, so that estimates are representative at the national and major sub-national levels. Sample sizes (commonly a few thousand households) trade statistical precision against the heavy per-household interview burden.
Trained enumerators administer the consumption modules in face-to-face interviews, recording food consumed (from purchases, own production, and gifts) over a recent recall period along with non-food expenditure and the use value of durables and housing. Frequent visits, detailed item lists, and recording of the value of home-produced food are used to capture the full consumption basket, including the large non-monetary component typical of poor rural households.
A real consumption aggregate is constructed by summing food and non-food consumption, imputing the value of own production and the flow of services from durables and housing, and deflating with spatial and temporal price indices built from the price questionnaire. The aggregate is usually expressed per adult equivalent or per capita and is the basis for poverty lines, poverty headcounts, inequality measures, and welfare-determinant regressions.
Common pitfalls
- Omitting the value of home-produced food and consumption in kind, which understates the welfare of subsistence-oriented rural households.
- Failing to apply spatial and seasonal price deflators, so that nominal consumption differences are mistaken for real welfare differences.
- Ignoring or mis-specifying sampling weights and the clustered design, producing biased point estimates and understated standard errors.
- Comparing consumption aggregates across surveys whose recall periods, item lists, or imputation rules differ, yielding spurious trends in poverty.
Frequently asked questions
Why does the LSMS measure welfare by consumption rather than income?
In low-income economies a large share of livelihoods comes from self-employment, agriculture, and the informal sector, where income is irregular, seasonal, and hard to report accurately. Consumption is smoother over the year, better reflects permanent income and households' own assessment of their living standard, and is generally reported more reliably. The LSMS therefore builds a comprehensive real consumption aggregate — including the imputed value of home production and the service flow from durables and housing — as its core welfare measure, though it also collects income data.
What are the three questionnaires in a standard LSMS survey?
A household questionnaire administered to each sampled household (consumption, education, health, employment, housing, assets); a community questionnaire collected once per cluster on local infrastructure, services, prices of services, and shocks; and a price questionnaire recording market prices of common goods in or near each community. Together they let analysts build spatial and temporal price deflators and relate household welfare to the local environment.
How is the consumption aggregate used to measure poverty?
The real per-adult-equivalent or per-capita consumption aggregate is compared with a poverty line, usually anchored on the cost of a minimum food basket plus an allowance for non-food needs. Households below the line are counted as poor, yielding the headcount ratio, poverty gap, and squared poverty gap, and the same aggregate supports inequality measures such as the Gini coefficient and distributional analysis.
Related methods
Sources
Grosh, M., & Glewwe, P. (Eds.). (2000). Designing Household Survey Questionnaires for Developing Countries: Lessons from 15 Years of the Living Standards Measurement Study. Washington, DC: World Bank.
ISBN: 9780821345283Copied.Deaton, A. (1997). The Analysis of Household Surveys: A Microeconometric Approach to Development Policy. Baltimore: Johns Hopkins University Press for the World Bank.
ISBN: 9780801852541Copied.