Beginning in 1981, a 3-year study examined implementation of decentralization in rural Alaska education; legislated decentralization had begun in 1975 largely in response to the Native self-determination movement. All rural local school administrators were initially surveyed. Based on their responses and other statewide data, four types of educational control were formulated: localized, regionalized, unified, and mixed. Using a 10% sample of rural schools, 28 schools were selected for more intensified field analysis during the 1981-1982 school year. Significant differences were found among the three relatively pure types of control with respect to the social and political conditions, school governance processes, and their associated outcomes. Although localization is the goal of the decentralization movement, only one-fourth of the schools fell in that category. Localization depended on several conditions: a stable local administrator who shared influence with the school board, had a positive orientation toward the local community, and respected its values; local boards or committees that represented major community interests, including faction, stability (low turnover) of the local board; a district superintendent supportive of local control; district policy which specified substantive, functional areas of school government in which local boards and administrators had deciding influence; and open district-level communication. (BRR)