Rigor in qualitative and quantitative health science research

¶ … Oxford Dictionaries (2014) define 'rigorous' as "extremely thorough and careful," "strictly applied or adhered to," and "harsh and demanding." Florczak (2011) defines 'rigor' in scientific research as "… being scrupulously accurate & #8230;" (p. 202). Rigor (or rigour) is essential, according to Florczak, for creating a logical line of reasoning that slices through the chaos that surrounds us. In other words, research findings and interpretations would not accurately represent the phenomenon being investigated without rigor.

Houghton et al. (2013) defined the four criteria needed for rigor in qualitative research. Credibility, the first criterion, is defined by how believable and demonstrably credible an investigative strategy appears to other investigators. Some of the research strategies that can increase credibility are prolonged engagement and observation, peer debriefing, and the use of multiple methods to study the same phenomenon. The second criterion is dependability, which is defined by how reproducible the findings will be over time. A close cousin of dependability is confirmability, which is how unbiased and accurate the data is. The rigor of dependability and confirmability depend on an accurate recording of the decision-making process, in addition to the rationale behind each decision. The final criterion, transferability, is the generalizability of the findings across individuals and settings. The primary method for increasing transferability is for researchers to provide a detailed description of the context in which the study was conducted, thereby empowering independent investigators with enough information to replicate the findings.

By comparison, the rigor of quantitative health science research has been defined by some in terms of the study design, such that meta-analysis, systematic review, randomized-controlled trial (RCT), cohort, case control, case series, and animal research represent the most to least rigorous in that order (Florczak, 2013). A meta-analysis applies statistical tests to the findings from multiple independent studies, thereby increasing the validity, reliability, and generalizability of the findings. By comparison, animal studies are amenable to tighter controls, but the relevance of the findings to human health is often questionable. Such considerations are framed by the three criteria used to evaluate the rigor of quantitative studies: (1) validity, (2) importance, and (3) relevance. Florczak (2013) took issue with the state of health science research because of the lack of sampling rigor and the widespread and often inappropriate use of the term RCT.