Tropentag, September 16 - 18, 2015 in Berlin, Germany "Management of land use systems for enhanced food security – conflicts, controversies and resolutions"

The Effects of Household Definitions on Survey Results and their Possible Implications for Policy Design: Evidence from Tanzania's Maasai

University of Hohenheim, Institute of Agricultural Economics and Social Sciences in the Tropics and Subtropics, Germany

Abstract

Household definition plays a crucial role in the nature and outcome of any research based on household level data. While developing and evaluating policy interventions, survey designers and analysts often assume a 'standard' definition that may not fully reflect the local understanding and/or the actual structure of the households under consideration. This may lead to misusing survey data and consequently to misleading conclusions and doubtful policy recommendations. The need for a clear definition is particularly apparent in pastoral societies such as the Maasai of East Africa due to them living in areas with dispersed population, their partial mobility, and the complex family structure involving polygamy.

Considering this background, this study addresses the underrated but highly relevant household definition debate. Based on an intensive literature review, it first discusses different concepts and understandings of a household, particularly in pastoral settings. Second, it uses available data from two Maasai household surveys which were carried out in Tanzania in 2009 and 2012 to econometrically test different household definitions in order to identify possible variations in regression results, results' interpretation or subsequently recommended policy interventions. The effects of modifying this key household characteristic on outcome variables, i.e. a set of welfare indicators (income, TLU ownership, asset value), is assessed by running different regression models and testing the results' sensitivity using likelihood ratio tests, F-test and Hausman test.

Results of the study's analyses are expected to show considerable changes in the socio-economic indicators associated to each household definition, and hence significant variations among outcomes and conclusions obtained from identical household data. Identifying the potentially large variation will highlight the importance of carefully designing household questionnaires with a properly defined variable representing the structure of the families under consideration.

Keywords: Household definitions, Maasai, policy implications, survey data, Tanzania

Contact Address: Vinicio Villalba, University of Hohenheim, Institute of Agricultural Economics and Social Sciences in the Tropics and Subtropics, Ecksteinstr. 9, 70599 Stuttgart, Germany, e-mail: viniciovuni-hohenheim.de