The problems of conducting research in schools are discussed. Solutions to social psychological problems are suggested, and an improved design for research is recommended. Clearer definition of terminology is recommended to avoid the confusion that can result from ambiguous terms. A research design should be adaptable to the administrative and instructional styles of individual schools. The researcher should observe the proper hierarchy in gaining permission to conduct research and should report results immediately. Deadline dates and specifications of event sequences with names of responsible persons are requisites for successful interpersonal relations. The following steps are recommended to avoid the frequent failure of educational research to improve educational practice: formulation, instructional specifications, component preparation and tryout, product preparation, quality verification, and operations analysis. (BS)