A review has been presented of methods whereby attitudes can be studied in African communities in various stages of transition from tribal to `Western culture. The techniques of the directed interview, group discussion, TAT, incomplete sentences, and attitudes inventory were discussed. In the measurement of African attitudes the problems of rapport and of validity require particular attention. In addition to the usual validity problem whether verbalised attitudes are consistent with actual behaviour, there is the difficulty, in the case of African informants, of ensuring that the verbalised attitudes reflect their actual beliefs, ideas, and feelings. This is as much a matter of the correct choice of method and its skillful manipulation as of the establishment of proper rapport. Important ingredients of successful rapport are the careful timing of investigations with reference to events of cultural or local political significance; the use of the vernacular; the avoidance of pencil and paper except in the case of well-educated groups; the correct contacts and the employment of carefully selected and trained field-workers, preferably graduates in the social sciences. Validity can be determined in a number of ways, depending on the nature of the investigation and the method employed. Checks on facts implicit in the formation of an attitude; group differences in attitudes related to the particular role and status in society of the members of each group; changes in attitudes consistently related to social circumstances, have all been found useful as validity indicators. In work situations, job attitudes can also be related to assessments of work performance by supervisors, though conflicting role expectations in multi-racial societies may render merit rating criteria suspect. The problem of structuring TAT pictures in such a manner that attitudes relating to specific areas of interest will be evoked was discussed in some detail. Research into the development and functioning of perceptual habits in African cultures is necessary, both as a basis for designing TAT series, and for the interpretation of responses. Details were given of a promising inventory, designed to measure attitudes towards Western customs, ethical and legal concepts, and administration of justice. The inventory presents conversations between Africans who discuss the correct mode of conduct in a variety of situations. Each topic is discussed by five speakers whose reasons for acting or not acting in a particular manner are ranked by the subjects. Statistical treatment yields the relative strength of eight different attitudes in individuals and communities. It was shown that the reliability of the scale is high, .92 for the rank order in which the attitudes were placed, and from .85 to .51 for the internal consistency of six of the eight attitudes. Validity data were also presented. [ABSTRACT FROM AUTHOR]