This paper examined the history and current issues in the evaluation of the education of Native students. Although public schools are the principal providers of services to Native students, most programs receive the greater part of their funding from the Federal Government and are subject to federal evaluation. Previous federal evaluation efforts have focused on efficiency and cost effectiveness rather than on overall effects and program responsiveness to parent perceptions of their children's needs. Also, despite various data collection and analysis efforts at federal, state, and local levels, a national database on Native education does not exist. Coordination of data collection efforts will depend, in part, on agreement about tests of academic achievement. The public school system's reliance on standardized testing may hurt Native students, as several factors can bias these tests against them. Some educators suggest other indicators of learning, such as assessment of skill mastery, student portfolios, and attitudinal measures. Providers, consumers, and monitors all "have a stake" in the continuous evaluation of Native education, but their evaluation needs differ. A naturalistic inquiry model of evaluation offers strategies that address the complex and culturally diverse realities and conflicting goals of Native education, and ensure that all stakeholders receive the information they need. These strategies rely heavily on information gathering and dissemination and include examination of each stakeholder group's views and interests, developing an understanding of different perspectives, consensus building, and negotiations. This paper contains 53 references. (SV)