A review of literature identified 287 references describing computer applications with deaf children. Of those, 162 investigated computers as a tool in educational settings. The academic applications of computers were distributed among the following categories: academic instruction (120, or 74 percent); psychological investigations (12, or 7 percent); and general education (30, or 19 percent). Further analysis of the academic instruction references found a scant 9 percent (N = 16) reported efficacy data. Although the majority found CAI outcomes to be positive, the degree of CAI success was inversely related to methodological rigor (r = -.69, p < .003). Well-designed studies suggested that CAI is no better than alternative forms of instruction. The available literature suggests computers are neither poison nor panacea, and may be nothing more than expensive placebo. Implications of these findings are discussed with reference to educational decision making and future research activities.