It is charged that remedial and experimental programs in reading ignore the results of relevant research and that research is often poorly developed. Faults of programs and research include the following: (1) There is no firm definition of the term "learning disabilities" and so they are not measured consistently. (2) Good sampling techniques often are not used in selection of subjects for experiments. (3) Too much emphasis is placed on methods of teaching and not enough on content taught. (4) Results are often slanted by the researcher or by the teacher administering tests. (5) Criteria for measuring improvement often do not reflect actual progress. (6) Editors tend not to publish studies which do not show significant results and researchers do not publicize findings which might not reflect well on them. (7) There is a tendency to extrapolate beyond the point where results from one study could apply to another group of subjects. It is suggested that innovative programs be funded. (BD)