Validity, Reliability and Qualitative Research

In research, a reliable measure is the one that gives the same result over and over again (assuming there is no change in what is being measured) (Trochim, 2006a); while a valid measure is the one that gives the correct value (assuming there is such a thing as a “correct” value for the given measurement) (Trochim, 2006b). However, according to Cronbach (1975), all phenomena, even those that can be quantitatively measured, will, sooner or later, change. This is especially true in social and behavioral sciences, where (1) the studied phenomena change very rapidly, and (2) it is often impossible to tell whether two different measurements of the same phenomenon were made under identical conditions; because isolating the variables of interest, from all external influence, may be impossible (Cronbach, 1975, pp. 122-123; as cited in Lincoln & Guba, 1985, p. 115). Those social and behavioral phenomena which are primarily studied through qualitative methods (often because it is hard to study them through quantitative methods) are especially prone to such unpredictability. Hence, whatever is being “measured,” through qualitative research, is usually constantly changing. And, as a result, there is no way to known what its “correct” value is at any one time; because this value is also constantly changing.  Therefore, when it comes to qualitative research design, the concepts of validity and reliability seem to be inapplicable.

References

Cronbach, L. J. (1975). Beyond the two disciplines of scientific psychology. American Psychologist, 30, 116-127.

Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills, CA: Sage.

Trochim, W. M. K. (2006a). Theory of reliability. Retrieved from http://www.socialresearchmethods.net/kb/reliablt.php

Trochim, W. M. K. (2006b). Reliability & Validity. Retrieved from http://www.socialresearchmethods.net/kb/relandval.php