A Manual for Planning and Implementing the LSMS Survey

Foreword

The following is from:
Grosh, M. & Muñoz, J. 1996. A Manual for Planning and Implementing the Living Standards Measurement Study Surveys. LSMS Working Paper #126, The World Bank.

In making sound policy decisions, governments need to know how those decisions affect the populations in their countries. Answers to some of the important questions can come only from household survey data. For example who is poor and who is rich and why? Who uses government services such as schools, clinics, agricultural extension offices, welfare programs, and old-age pensions? Are those not using government services able to get services in the private sector? How do households change their decisions about who works and how much, whether and where to send their children to school, and how many children to have? To answer these questions requires household survey data that cover many aspects of household welfare. Until a few years ago, such surveys were very rare in developing countries. The Living Standards Measurement Study (LSMS) program was launched in 1980 to help foster the collection of good data from household surveys and improve its subsequent use in policymaking. The first LSMS surveys were implemented in Côte d'Ivoire in 1985 and in Peru in 1985/86. Since then, over forty LSMS surveys have been conducted in two dozen countries and new LSMS surveys are currently in the field or being planned in nine additional countries.

LSMS surveys provide high quality, timely, and comprehensive data on most aspects of household welfare (consumption, income from activities in the labor market, household enterprises or agriculture, asset ownership, migration, health, education, nutrition, fertility, and anthropometrics). LSMS surveys have become a powerful tool for understanding household economic decisions and the effects of social and economic policies. The use of LSMS data in poverty assessments helps to ensure that the development community's efforts to reduce poverty can be guided by quantitative information on levels, causes, and consequences of poverty. The data have been used by governments in various direct and indirect ways. In Bolivia, LSMS data were used to help the government evaluate its public employment program. In Jamaica, the government used data from its LSMS survey to reformulate the food stamps program. In South Africa, the government used the data in designing their tax reform program.

LSMS surveys have evolved over time. Originally they were motivated primarily to support research; now they are much more often driven by policy needs. The contents of the questionnaires have accordingly changed over the years and from country to country. The modular design of the LSMS questionnaires has facilitated this flexibility and country-specificity. The surveys have also benefitted from developments in computer technology. The LSMS program had to design its own software to lay out the first questionnaires, but now such software is available commercially. In the first surveys in Côte d'Ivoire and Peru, it was novel to carry out data entry on personal computers in regional field offices, with electricity often provided by gas-fueled generators. In the Nepal LSMS now in the field, data entry is carried out in the field on notebook computers powered by portable solar panels. In 1995, Tanzania became the first country to allow data from its LSMS survey to be put on the Internet for easy access by scholars worldwide.

The interest in conducting and analyzing surveys like the LSMS has grown markedly since the early days of the project. Such surveys are now being done in many more places than the LSMS division of the World Bank can work. Since many of those now involved in the implementation of new LSMS-type surveys have little familiarity with the old surveys, it is important to ensure that the lessons from the first ten years of LSMS field work are widely available. This manual is one of a series of efforts to compile and disseminate the lessons of LSMS experience. Here the focus is on the planning of the survey and the conduct of the field work. A comprehensive review of the content of the questionnaires and the way in which the various modules can be combined is currently underway, and the documentation and dissemination of data sets from surveys already fielded have recently been upgraded.

Lyn Squire, Director Policy Research Department

Revised 10/09/97