ROBERT BATES California Institute of Technology Approaches to the Study of Ethnicity* INTRODUCTION This paper has two objectives The first is to isolate several of the dominant approaches to the study of ethnic behavior Africa The second is to test these approaches against data obtained from the Nkana townships in Zambia We must register several caveats The data were drawn from sample of the residents of mine township.1 As result it is only with great caution that our findings can be generalized to other urban settings Secondly there is the definitional problem of the term .2 In this paper have adopted natural definition when focusing on the Bemba for example group together persons who have come to identify themselves as Bemba on the copperbelt Ushi Kazembe Lunda Bisa Ngumbo etc. are thus counted as Bemba for they regard themselves and are regarded by others as
While collecting the data for this study was sponsored by an SSRC and ACLS Foreign Area Fellowship and by grant from the National Science Foundation Grant Number G.S 1210 While writing this article was supported by the Center for International Studies of the Massachusetts Institute of Technology which published much earlier version of the paper presented the paper in its present form at panel of the annual meeting of the African Studies Association in October 1969 The data were drawn from the personnel records of 20 systematic random sample of the labor force of Rhokana Corporation one of the largest mines in Zambia comparison of my data with census data published by the mines indicates low level of sampling error The sample is sufficiently unbiased to enable generalizations to the population with errors of less than with of not less than 30 One of the most enlightening discussions of the semantics of this term is that of Pierre VAN DEN BERGHE ed. Africa Social Problems of Change and Conflict San Francisco 1965 pp 1-9 Introduction
