Reputation and Reality in the Study of "Community Power"

The currently popular research design of ranking a community's leaders according to their reputations for power is found to be seriously deficient as a technique for the study of a local political system. The frequent assumption that power is equally distributed for all issues results in inaccurate descriptions and in respondents tacitly basing "general power" rankings on a specific salient issue. It is difficult to avoid confusion between status and power without using questions so complicated and qualified as to be impractical in an interview. Even politically experienced respondents are unreliable observers. In any case influence rankings are not useful because: (1) there is no way to assess the relative power of top-ranked individuals compared to presumably less powerful persons without making unwarranted assumptions; (2) identification of leaders is not an adequate description of a political system; (3) the reputational method assumes a static distribution of power.