Using Non-Parametric Statistic Tests Essay
Non-parametric tests are experiments that do not require the underlying population to derive assumptions. These tests are also known as distribution-free since they do not rely on any data referring to parametric probability distribution groups (Carasco et al., 2020). This discussion explores various non-parametric statistical tests, how the tests can be used to analyze data collected in a practice change intervention, and the rationale for using the tests.
The selected practice change problem is poor turn up for diabetes self-management education programs. The intervention that would improve outcomes is the use of group-based DSME. In current practice, patients attend DSME programs individually. Research shows that group DSME programs have shown effectiveness in improving diabetes self-management behavior in diabetes patients since patients encourage and learn from each other’s experiences (Davis et al., 2022).
In an attempt to identify whether the intervention is more effective compared to the current practice, the non-parametric statistic tests that can be used include the Mann-Whitney U test and the sign test. The Mann-Whitney U test is used to compare two independent groups in which the population distribution is the same despite the observations from one group being lower or higher (Schober &Vetter, 2020). The test is appropriate for the intervention because deriving logical comparisons would require similar population distributions from both the intervention and the current practice testing group. The sign test is a statistical non-parametric test used to measure consistent differences between pairs of observations (Schober &Vetter, 2020). The test is suitable in this case since the observations of the intervention and the current practice can easily be compared to determine effectiveness.
In summary, the non-parametric statistical tests that can be used to measure the effectiveness of the intervention described above are the sign test and the Mann-Whitney U test. The tests can be used to derive the differences between observations of the two groups, such as changes in self-management behavior, and derive differences in the two groups, thus the most appropriate tests.
Carrasco, J., García, S., Rueda, M. M., Das, S., & Herrera, F. (2020). Recent trends in the use of statistical tests for comparing swarm and evolutionary computing algorithms: Practical guidelines and a critical review. Swarm and Evolutionary Computation, 54, 100665. https://doi.org/10.1016/j.swevo.2020.100665
Davis J., Fischl A.H., Beck J. & Villalobos, L. (2022) National Standards for Diabetes Self-Management Education and Support. The Science of Diabetes Self-Management and Care, 48(1):44-59. https://doi.org/10.1177/26350106211072203
Schober, P., & Vetter, T. R. (2020). Nonparametric Statistical Methods in Medical Research. Anesthesia and Analgesia, 131(6), 1862–1863. https://doi.org/10.1213/ANE.0000000000005101
The purpose of this discussion is to demonstrate your understanding of the use of non-parametric statistical tests.
Select a practice-change problem and, from the literature, an intervention to impact outcomes. Imagine you are attempting to determine if the intervention is more effective than current practice. Explain the various types of non-parametric statistical tests that might be used to analyze the data collected during the implementation of the intervention. Provide a rationale for the use of non-parametric tests for this data set.
Please click on the following link to review the DNP Discussion Guidelines on the Student Resource Center program page:
Link (webpage): DNP Discussion Guidelines
Links to an external site.
This discussion enables the student to meet the following course outcomes:
Evaluate selected statistical methods for the purposes of critiquing research to complement the critical appraisal of evidence. (POs 3, 5, 9)
Analyze research and non-research data for the purposes of critical appraisal and judgment of evidence for translation into practice. (POs 1, 3, 5, 7, 9)