Reporting of instrument validity and reliability in selected clinical nursing journals, 1989

Before research findings are applied to practice, the quality of the research must be assessed so that flawed research does not lead inadvertently to flawed practice. Two critical indicators of research quality are the validity and reliability of the data collection instruments. This article summarizes the principles of instrument validity and reliability and identifies deviations from these principles in a random sample of 55 research studies published in 1989 in five refereed nursing journals targeted toward practicing clinicians. Using a valid and reliable instrument, the investigators found that even with a policy of giving authors "the benefit of the doubt," 47% of the research studies contained no evidence of validity for any data collection instruments and 36% had no evidence of reliability; 29% had no evidence of either validity or reliability. Content validity, a basic requirement for all research instruments, was addressed in only 27% of the studies. This article provides documentation, justification, and suggestions for nursing educators, journal editors, and researchers to take action to improve the reporting of instrument validity and reliability to help ensure the quality of the research on which nursing practice is based.