Reliability and Validity in Qualitative Research Methods

Reliability and Dependability

Dependability refers to the need for researchers to take into consideration the ever-evolving context within research as it unfolds (Creswell & Miller, 2000). The responsibility of qualitative research is to describe the changes naturally occurring in the setting as well as how such changes affect the researcher's approach to the study. To address dependability within a qualitative research study, a researcher can perform member checking of data interpretation, a pilot test, and a peer review. These techniques allow for accurate assessment of dependability regarding the information collected for the study.

When beginning qualitative research, researchers may perform member checks to determine the reliability and dependability of the information collected. "Member checking, also known as participant or respondent validation, is a technique for exploring the credibility of results. Data or results are returned to participants to check for accuracy and resonance with their experiences" (Birt, Scott, Cavers, Campbell, & Walter, 2016, p. 1802). Such reporting may be criticized due to its interpretive stance concerning qualitative research; however, it is a commonly employed technique. The primary reason it is used is that it provides an opportunity to understand whether the author's intended meaning has been captured accurately, and to assess whether interpretations contain errors. It functions as a playing-back process that serves as respondent validation of informant feedback.

The next technique often employed is peer review. Peer review has many names — some refer to it as auditing, expert review, or independent scientific review (Wicherts, 2016). Regardless of the label, it functions similarly to the member check in that someone else reviews the information for reliability, except that the work is subjected to the scrutiny of experts within the same field before it can be published as a book, in a journal, or in conference proceedings (Wicherts, 2016).

The final technique discussed here is the pilot study. Pilot studies allow researchers to identify potential barriers and issues concerning the recruitment of participants and to observe phenomena from a phenomenological perspective. Additional benefits that increase dependability include an opportunity for reflection on the significance of the epoché process, the challenges of performing phenomenological inquiry, and the potential modification of interview questions (Kallio, Pietilä, Johnson, & Kangasniemi, 2016). Pilot studies serve as an extra layer of reflection and assessment that can lead to more thoughtful interpretation of qualitative data while improving the dependability and accuracy of that data.

Validity and Credibility

Much like dependability, credibility often involves employing certain techniques to determine what may or may not work regarding data collection and interpretation. To establish credibility, member checking can again be employed. Member checking allows researchers to establish the level of expertise and education of those who have written or contributed to the work (Varpio, Ajjawi, Monrouxe, O'Brien, & Rees, 2016). It is an important tool for establishing a valid interpretation of information through verification of the credentials of the person who collected and recorded the data.

Another important technique used to establish credibility is triangulation. Triangulation can be used both qualitatively and quantitatively, and it can improve the credibility of scientific information by refining both internal consistency and generalizability through the combination of quantitative and qualitative methods within the same study (Northrup & Shumway, 2014). Because of this dual nature, data can be evaluated in various ways, providing a richer understanding of the researcher's interpretive capabilities.

Participant transcript review is another tool that helps establish credibility. It allows researchers to determine whether participants fully and correctly responded to the interview questions, and it gives researchers an opportunity to refine the interview guide if needed. "Analysis begins after reviewing the first interview to examine whether participants are responding to the research question related to your area of interest in diabetes, or whether your interview guide needs refining" (Stuckey, 2014, p. 6).

Transferability

Transferability is another important aspect of qualitative research. To establish transferability, researchers must provide readers with evidence concerning the applicability of findings to other times, situations, populations, and contexts — evidence that must come from sources other than the researcher themselves (Sacks, 2015). For example, thick description is an approach that allows the qualitative researcher to offer a detailed and robust account of their experiences while collecting data (Sacks, 2015). By making explicit connections to social and cultural contexts, this approach provides a strong means of checking for transferability.

Ethnography is another avenue for establishing transferability. Ethnography involves the study of people in their natural environment through methods such as face-to-face interviewing and participant observation (Gentry, Milden, & Kelly, 2017). Such methods allow researchers to present information through descriptive processes that others can then examine for aspects related to transferability. It is a common means of providing information without imposing direct interpretation.

Case studies also support transferability by enabling empirical generalization. "Critical realism recognizes the role of case study research in empirical generalization, theoretical generalization, and theory testing" (Tsang, 2014, p. 174). In addition, case studies allow for theoretical generalization, which can facilitate easier transfer of findings across different contexts.