The effects of three factors on environmental perception were investigated: (1) medium of presentation; (2) evaluative context, and (3) observer sample. A quasi-experimental design and field testing were used to present an automobile tour of a geographical locale to systematically drawn sub-samples of the population under varying media conditions and evaluative contexts (response formats and instructional sets). Four media of presentation were employed: (1) direct; (2) film of direct; (3) model film; and (4) model video. Evaluative contexts consisted of: (1) structured self-report inventories administered post-tour only; (2) enroute structured ratings plus post-tour procedures; and (3) unstructured ongoing descriptions enroute plus post-tour procedures. Observer samples were drawn to represent variation in (1) professional standing in environmental design, planning and management, and (2) familiarity with the research site. A least squares analysis of variance was performed for each of three factorially derived cross-procedure environmental perception variables. Results indicate that each of the facets of research design investigated has a small but clearly discernible effect on environmental perception, and that media effects are manifested differentially for professional versus nonprofessional observers.