This quantitative, quasi-experimental study conducted by Yaun et al. (2009) aimed to determine the effects of an exercise intervention on nurses’ health-related physical fitness. The researchers also expressed an explicit interest in the relationship between physical fitness and the incidence of musculoskeletal disorders. Taiwanese nurses from five different units volunteered to be part of the study. The participants were divided into two groups with 45 nurses in the experimental group and 45 nurses in the control group. There was no randomization, but all the participants gave written informed consent (Yaun et al., 2009).
According to Polit and Beck (2017), internal validity pertains to the empirical relationship between the independent variable and the final results. Researchers must establish that the intended cause created the effect, and that it was not influenced by other variables (Polit & Beck, 2017). After all, correlation does not equal causation, and an astute researcher will adeptly identify and control convoluting variables. Further, Andrade (2018) asserts that internal validity assesses whether the design of the study, the conduct of the researchers, and the analysis of the results answer the research question without bias (Andrade, 2018).
Consequently, the research conducted by Yaun et al. did have some issues that negatively impacted the internal validity of their research. Firstly, convoluting variables were not adequately controlled. The exclusion criteria consisted of cardiovascular disease, diabetes, hypertension, renal disease, pulmonary disease, severe musculoskeletal aches, and pregnancy. However, other significant variables such as age, gender, marital status, educational level, or other medical issues. It is worth noting that the diet and exercise habits of the participants were not limited by the researchers.
Moreover, the nurses in the experimental group worked a fixed schedule whereas nurses in the control group worked alternating shifts. Secondly, the lack of randomization coupled with the fact that the participants worked for the same organization could have contaminated the results. Thirdly, while the results of the research showed the exercise intervention improved the physical fitness of the participants in the experimental group, participants were not evaluated for musculoskeletal improvements.
Recommendations to Strengthen Internal Validity
A different research design would have strengthened internal validity. Randomization is the most effective way to control individual characteristics of participants. Randomization also eliminates for the Hawthorne Effect, which occurs when participants behave differently because they know they are being studied. Moreover, a cross-over design is highly effective when groups are being compared to one another. Although, this design is subject to carryover bias, in which an effect carries over from one experimental condition to another (Polit & Beck, 2017).
I contend that a randomized control trial with a cross over design would have increased the strength of the internal validity in this study. In a cross-over design participants serve as their own control group, which would negate the convoluting variables that influenced the results of this study, and would more accurately gauge changes resulting from the exercise intervention. I would also add a metric to assess the musculoskeletal status of the participants. To limit the effects of carryover bias, the health metrics of the participants would be obtained before the exercise intervention to establish a baseline, then after the exercise intervention, and finally, after a wash-out period, the metrics should be re-recorded.
The Impact of Changes on Other types of Validity
In contrast to internal validity, statistical validity is not concerned with the causal relationship between variables, but rather measures the mathematical correlation of all relationships that occur between the variables (Polit & Beck, 2017). The randomized control, crossover design would improve statistical validity because the participants would serve as their own control group making statistical analysis more powerful. Construct validity determines if the outcome measured corresponds to the theoretical construct of the study (Polit & Beck, 2017). In this research, the theoretical construct was Pender’s health promotion model. Construct validity also would have been improved by changing the design of the study. The same health promotion strategy yields different outcomes for different participants based on individual differences. The modification of the study’s design would have negated these individual differences. External validity indicates if the results of the research will remain the same when applied to other people or settings (Polit & Beck, 2017). Again, a change in the design of this research would optimize external validity which would increase the likelihood of the results influencing evidence-based practice.
Failure to Consider Validity in Research
Failing to properly account for and control variables threatens the validity of the results yielded from the research. The rigor of the research design may be the most important factor in strengthening or weakening validity, as evidenced by the hierarchy of research studies in the evidentiary pyramid. Other elements such as biased statistical analysis, unreliable implementation of an intervention, carryover bias, and the Hawthorne Effect are just a few variables that can threaten the validity of a research study (Polit & Beck, 2017). Since research guides evidence-based practice, failure to ensure the validity of results directly affects patient outcomes; unfortunately, the effects of poorly executed research impacts all research. People are inherently inclined to remember negative consequences over positive outcomes. Improper research regarding vaccines has created an anti-vaccination movement that is highly problematic. Big tobacco companies produced improper research that may have resulted in people continuing to smoke longer than they otherwise would have. The failure to appropriately consider validity in research is a grave mistake that should be avoided at all costs.
Andrade, C. (2018). Internal, external, and ecological validity in research design, conduct, and evaluation. Indian Journal of Psychological Medicine,40(5), 498. doi:10.4103/ijpsym.ijpsym_334_18
Polit, D. F., & Beck, C. T. (2017). Nursing research generating and assessing evidence for nursing practice. Philadelphia: Wolters Kluwer.
Yuan, S., Chou, M., Hwu, L., Chang, Y., Hsu, W., & Kuo, H. (2009). An intervention program to promote health-related physical fitness in nurses. Journal of Clinical Nursing,18(10), 1404-1411. doi:10.1111/j.1365-2702.2008.02699.x