Research validity is an important factor that affects the outcomes and usefulness of scientific studies. Researchers have to create a setting that will support different types of validity to produce results that are real, logical, coherent, and appropriate for being used in real life. One can consider four types of validity: internal, external, statistical conclusion, and construct (Polit & Beck, 2017). All four of them are important as they support different spheres of the study’s usability and effectiveness. However, the intention of scholars to enhance a certain type of validity also has an impact on other ones. This effect can be both positive and negative. Using the study “Effectiveness of an Aspiration Risk-Reduction Protocol” by Metheny, Davis-Jackson, and Stewart (2010), the possible flaws in internal validity are described below along with strategies for improvement and outcomes to nurses’ neglect of validity.
The Study’s Validity Concerns
Internal validity can be compromised in a variety of ways. Polit and Beck (2017) state that internal validity in quantitative research is the extent to which external influences affect the causality of the study’s events. Metheny et al. (2010), for example, discuss the possible impact of the ARRP (Aspiration Risk Reduction Protocol) on critically-ill patients with mechanical ventilation who receive tube feedings. To assess the usefulness of this program, which is based on educating nurses about particular monitoring techniques, scholars collected data from one hospital at different points in time. The first group of participants was studied from 2002 to 2004, while the second was examined in the period from 2007 to 2008. During the years between data acquisitions, the hospitals’ ICU nurses were trained to document information about the patient’s head-of-bed elevation, insert feeding tubes, and assess high volumes of gastric residuals.
One possible concern linked to the study’s internal validity is the lack of data about nurses’ skills working with the first group. For example, while in the intervention group nurses were asked to keep the patients’ beds at a certain angle, those professionals’ level of knowledge in the first group was never defined. Thus, they could follow the same procedures without any specific training. Moreover, the difference between the two sample groups also threatens internal validity because one group was significantly larger than the other and was stationed in the hospital multiple years earlier. Thus, this study may be subject to some concerns regarding its validity.
In order to improve the research’s validity, one can create specific environments for both groups, where nurses’ actions are easily documented in both samples. This change may enhance the construct validity as well because it will show that the new intervention is being implemented accordingly. Statistical conclusion integrity may also be positively impacted because the changes would be more transparent than before. External validity, however, might be threatened because the study’s environment would be highly controlled, preventing its outcomes from being easily generalized.
Conclusion
Different types of validity are linked to the coherence of studies. The research by Metheny et al. (2010) reveals such concerns as the lack of available data and the ambiguity of compared variables. Thus, it is possible that some other variables such as the nurses’ prior experience might have affected the study’s outcomes. Moreover, as the two groups were studied at different times, the results also become questionable. The scholars could create a more controlled environment which would benefit all types of validity except the external one. Such attention to validity is important for nurses who seek to utilize studies’ findings. If they neglect to assess the validity of research, they may fail to recognize possible external influences and inconsistencies between theory and practice.
References
Metheny, N. A., Davis-Jackson, J., & Stewart, B. J. (2010). Effectiveness of an aspiration risk-reduction protocol. Nursing Research, 59(1), 18-25.
Polit, D. F., & Beck, C. T. (2017). Nursing research: Generating and assessing evidence for nursing practice (10th ed.). Philadelphia, PA: Wolters Kluwer.