A review was made of 27 studies describing early intervention for biologically impaired infants and young children. Although every study provided some type of documentation of outcome, most studies failed to meet common criteria for scientific research, such as specification of inclusion criteria, documentation of reliability, random assignment, and/or the use of control/contrast groups. Studies were grouped into four classifications on the basis of experimental design as follows: retrospective, prospective-no control group, prospective-control/contrast group, and prospective-random group assignment. A comparative analysis was made of the effectiveness of findings on the basis of statistical evidence and clinical support. Statistical procedures were used in 59% of the studies and statistical support for the effectiveness of early intervention was reported in 48% of the studies. Effectiveness on the basis of subjective, clinical conclusions was reported in 93% of the studies. Implications of the discrepancy between effectiveness based on statistical evidence and clinical conclusions are discussed.