Validity in Research Design
Validity in Quantitative Research Designs
Validity refers to the degree to which physical evidence supports an experimental concept or conclusion. In statistics, it refers to the level of confidence that researchers hold on a particular causal effect of an experiment. The most common challenges to internal validity in the research design occur from the selection process of target population as well as the sample size (Yuan et al., 2009).In particular, the selection process in the article reflects only persons who have volunteered for the research. Ideally, this represents a threat to internal validity since the individuals used in the experiment provides skewed results that do not significantly represent the whole target population.
Improving the internal validity of the article involves ensuring sufficient levels of randomness in the selection process. Firstly, the sample selected for analysis should well represent the entire target population to guarantee any feasibility in the outcomes found after the experiment. In addition, the sample selected for the experiment should be significant enough to guarantee the efficacy of the research. Like the selection process, sample size needs to have a significant representation of the target population (Polit & Beck, 2008). However, the article considered a small sample size that is insufficient to represent the entire population. Furthermore, the selection process considered only volunteers for the research. In particular, this reflects the high bias in the responses given by the interviewees. Therefore, a recommendation of for improvement in sample size as well as randomization of the selection process will improve the internal validity of the article.
Threats to Internal Validity
On the other hand, failure to consider validity in research designs leads to many risks in providing feasible results of the experiment. For example, maturation is a validity risk where the researcher lacks the capacity to control the behavior of the interviewee before the date of the interview (Cantrell, 2011). Furthermore, failure to consider randomness in the selection of the sample hinders the reliability of the experiment because validity is depended on the sufficiency of the results.
References
Cantrell, M. A. (2011). Demystifying the research process: Understanding a descriptive comparative research design. Pediatric nursing, 37(4), 188.
Polit, D. F., & Beck, C. T. (2008). Nursing research: Generating and assessing evidence for nursing practice. Lippincott Williams & Wilkins.
Yuan, S. C., Chou, M. C., Hwu, L. J., Chang, Y. O., Hsu, W. H., & Kuo, H. W. (2009). An intervention program to promote health‐related physical fitness in nurses. Journal of clinical nursing, 18(10), 1404-1411.
