Young Lives Peru (Niños del Milenio) recognised by the World Bank for innovative data methodology. 10 April 2008

More than 150 programmes from 20 countries participated in this Latin America and Caribbean region competition, which aimed to recognise the importance of statistics in development and promote innovation in this area.

The Grupo de Analisis para el Desarrollo (GRADE) Young Lives team in Peru won through to the finals in the 'Survey' category against stiff competition. This category had the highest number of original entrants and the 12 other semi-finalists included national statistical offices, ministries, non-governmental organisations and universities, many with much larger teams than GRADE.

GRADE was one of only 3 NGOs to win through to the second round overall. They were 1 of 31 semi-finalist programmes from a total of 12 countries (Argentina, Brazil, Colombia, Costa Rica, Honduras, Mexico, Panama, Paraguay, Peru, Santa Lucia, Trinidad and Tobago and Uruguay).

As a finalist in the competition GRADE will have the opportunity to disseminate and promote their experiences at the regional and international level and their work will be publicised to a broad audience of international experts and donors, key government officials, academics as well as other important stakeholders of the statistical community. They will be included in an on-line inventory of regional best practices in statistical development.

See the Niños del Milenio website for more details (in Spanish and English): http://www.ninosdelmilenio.org/

Competition Background:
The World Bank, in collaboration with the UK Department for International Development, made awards for the design, implementation, and evaluation of priority public policies in three categories: Censuses, Surveys, and Administrative Records.

The statistical programs/ activities of use had to be:

• of high overall quality
• clearly innovative in at least one particular process related to institutional aspects, data production and/or its usage
• and implemented between January 1996 and December 2006 by any statistics-producing entity from a World Bank member country in the Latin America and the Caribbean region.

A Selection Committee reviewed the applications and evaluated them for: innovation (type and impact), data quality, and use and usefulness of the data.

The selected programmes were asked to submit a technical brief. These second round submission documents were reviewed, evaluated and certified by a high level panel including experts in statistics from the Inter-American Development Bank, the International Monetary Fund, the World Bank, the UN, the National Statistical Office of Spain and the US Census Bureau) who analysed the information provided and selected those programmes that received the highest points.

To see full details of the competition and the awards at the World Bank website click: http://www.worldbank.org/lacstatistics

Young Lives submission
Click here to download the GRADE/Young Lives 30-page technical brief [208 KB pdf] for this competition which emphasises the uniqueness of the Young Lives data. No longitudinal research of this size, scope and complexity has ever been undertaken in the developing world and the methodology developed by GRADE and Young Lives is designed to address the multiple challenges faced in ensuring the accuracy and validity of a very large, and growing, dataset.

A number of methods have been combined to help ensure its integrity:

• Sampling: the sampling frame allows the linkage the results of the project to national representative samples like ENAHO and DHS Surveys, so as to improve the effectiveness of the project when engaging in policy debates. Young Lives data comes from a multi-stage, cluster stratified random sample.

• Mechanisms to minimise attrition: The introduction of a detailed account of family and friend networks for each family to minimise attrition rates throughout the 15-year period.

• Field worker training: Through intensive role-playing, a good level of standardisation was achieved. A paper entitled ‘Selection and Training of Supervisors for Fieldwork: Experiences from the Young Lives Study in Peru’ documents the process. We know of no other publication that specifically addresses the selection and training of leaders or supervisors for community field research or similar situations involving autonomous groups working as a team adhering to rigorous quality control while confronting varied and challenging circumstances.

• Back translation: Two levels of back translation were developed. First, back translation from English to Spanish and then back to English ensure that core questions are comparable across countries that are also part of the study. In addition back translation was also used to ensure that questions are consistent between Spanish and local languages (in particular Quechua).

• Evaluation of biases: The inclusion of information about field workers and supervisors (education, gender, quality of their work as assessed by their supervisors, etc.) for ex-post evaluation of potential biases introduced during the data gathering phase.

• Double-data entry: The construction of a data entry system that includes double entry data automatic levels of consistency checking aimed at minimising data entry errors.

• Electronic storage of survey forms: The design and construction of a software system that administrates the scanned forms collected, giving researchers access to the actual survey forms without jeopardising the anonymity of respondents.

• Ethical guidelines: Strict adherence to ethical guidelines approved by both local and international ethics committees ensures that the project achieves the highest ethical standards.

Click here to download the GRADE/Young Lives summary competition programme brief [67KB, pdf]