This paper examines the core concepts of validity and reliability as they apply to both qualitative and quantitative research designs. Drawing on Golafshani (2003) and Smith (1988), the paper defines each concept and explains why both must be established before research results can be considered credible. It outlines the principal techniques used in quantitative research — content, criterion, and construct validity, along with test-retest, alternative-forms, and internal consistency reliability methods — and contrasts these with the qualitative approach, where credibility of sources and triangulation serve as the primary tools for ensuring data quality.
Validity and reliability are two concepts considered critical in every research study, whether conducted under a qualitative or quantitative research design. These concepts must be disclosed and discussed in detail in every study, since they serve as the foundation of the credibility and strength of the data — that is, how well a study's data answers the research questions and objectives, and how flexible that data is in partially or wholly addressing the questions and objectives of related studies.
Qualitative and quantitative research designs utilize different methodologies and data collection methods; therefore, each design employs different techniques for testing validity and reliability. Conceptually, validity is defined as follows: "validity determines whether the research truly measures that which it was intended to measure or how truthful the research results are" (Joppe, 2000, as cited in Golafshani, 2003, p. 599). Reliability, meanwhile, is the "extent to which results are consistent over time and an accurate representation of the population under study, and if the results of a study can be reproduced under a similar methodology" (Joppe, 2000, as cited in Golafshani, 2003, p. 598).
Under the quantitative research design, validity and reliability are tested using distinct techniques. It is crucial to test for both because a study's data must be proven valid and reliable before end-users can assess the importance of the study's results and recommendations.
In quantitative studies, validity is tested through content, criterion, and construct validity. Content validity determines whether the construct used in the instrument is correctly defined and operationalized. Criterion validity examines how well a construct can accurately predict the construct being measured. Construct validity ensures that the construct measured in the study is accurately defined and that the definition is mutually exclusive to that concept alone (Smith, 1988, p. 48).
Techniques for testing reliability in quantitative research include the test-retest, alternative-forms, and internal consistency methods. The test-retest method administers an instrument to the same population on two different occasions; the difference between the scores from the first and second administrations determines the reliability of the construct under study. The alternative-forms method is similar to test-retest, except that the flow of the instrument in the second administration is reversed, and the difference between the results of the two tests determines the instrument's reliability score. Finally, the internal consistency method examines the construct's consistency when the instrument is tested against itself — that is, if the instrument were deconstructed, would it still generate reliability scores equal to or higher than those produced by other tests or techniques? (Smith, 1988, pp. 46–47).
"Test-retest, alternative-forms, and internal consistency methods"
"Credibility and triangulation as qualitative testing tools"
Smith, M. J. (1988). Contemporary communication research methods. Wadsworth.
You’re 64% through this paper. Sign up to read the remaining 2 sections.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.