In recent years, anthropologists and sociologists have been concerned with the problem of the definition of ethnic groups within various societies. This paperl suggests that the purpose of the researcher's definition largely determines the composition of the group and that research on ethnicity may be seen as analogous, in that respect, to approaches to research on language.

For some types of research, it is sufficient to equate ethnic identity with tribal affiliation. This is true in areas of the world where it makes sense to record answers from informants to the question “What is your tribe?”. However, it is necessary in such areas for the researcher to ask the question directly, rather than ask one person to label another.

In a recent article, A.N. Tucker (1973) discusses this same problem with regard to placenames. He cites, for Africa, cases where it was discovered that the official name of a particular place was not the place's name at all but had been supplied to researchers (linguists, anthropologists, geographers, or whatever) by a guide from another culture or linguistic group. Tucker (1973: 163) states that:

As research in social anthropology has turned from studying tribal units in more or less isolated areas toward urban-centered research, it has become apparent that one feature of urban social organization seems to be emergent ethnic or tribal groupings (Arens, 1973: 441). The criteria underlying such groupings, as opposed to those underlying traditional tribal affiliation, are often of a complex nature. Where, in rural areas, it has been possible to ask informants “What is your tribe?” and receive an unambiguous answer (usually related to parentage and/or land inheritance), in urban poly-ethnic communities today as well as in rural yet urban-linked areas, criteria for determining ethnicity are changing. In fact, as Arens has pointed out, in today's Africa common tribal identity in the sense of ethnic consciousness seems to be of particular importance in the cities, where ethnicity—in the absence of kinship—underlies social organization. The common assumption that rural areas are static and urban areas dynamic is being challenged.