Forms of evaluation that involve evaluators working collaboratively with practitioners on applied social research projects are becoming increasingly common, but a body of empirical literature that warrants firm conclusions about collaborative evaluation has not yet been accumulated. This survey studies the views of evaluators and program practitioners using a conceptual framework developed to guide the study that considers pragmatic, philosophical, and political interests and three dimensions of the process: control of evaluation decision making, stakeholder selection, and depth of participation by program practitioners. An 8-page questionnaire was completed by 564 North American evaluators from professional association mailing lists. A subsample of 348 also selected and provided data on a recently completed collaborative evaluation. Sixty-seven practitioners who had collaborated returned parallel questionnaires. Findings show that evaluators tend to support pragmatic as opposed to political or philosophical justifications for collaborative evaluation and subscribe to a stakeholder-service orientation to the role. Attitudes toward such evaluation were generally positive, but they were found to depend on evaluators' experience with and involvement in such activities. Evaluators reported that practitioner involvement was generally more extensive than might be considered typical for traditional stakeholder-based evaluation, but was limited to mostly nontechnical research tasks. A wide range of stakeholder groups participated. Evaluators also tended to lean toward evaluator control rather than a balanced approach to evaluation decision making. In general, these self-reports found the impacts of the evaluation projects to be very favorable. Appendixes present the survey and cover letter. (Contains 86 references.) (SLD)