Nonparametric and Parametric Tests Response

RESPONSE R (TWO)

Response R (Two)

K. Kim

In the first post, K. Kim opted to focus on a descriptive analysis to gather socio-demographic data and medical history of the sample to assess the prevalence of lifetime and current prescription-opioid use disorders and opioid misuse. According to Macfarlane et al. (2020) opioids are routinely used in the treatment of cancer patients. However, misuse of the said opioids has been rather rampant (Macfarlane et al., 2020). Apart from the descriptive analysis highlighted by my colleague, the person-level sample weight can be used to analyze data (Hudgins et al, 2019). In this method, the weights of the sample are adjusted by use of generalized exponential model. The said model adjusts in case of extreme weights, post-stratification, and survey non-response so as to yield characteristics and occurrences that are unbiased (Hudgins et al, 2019). STATA Version 15 is used to analyze the data. The strength of the study is that the response rate of the analyzed years was high. However, the study did not include people in prison, military personnel, and homeless people. In addition, patterns of opioid use were not evaluated owing to the fact that a longitudinal design was not employed. The findings from the study indicate a high prevalence of p-OUD. Therefore, the relevant interventions to prevent and treat the said disorder should be taken into account.

R. Cantu

In the second post, R. Cantu highlights the fact that systematic methods would be effective in examining the effects of advanced nursing practices on patient satisfaction in critical care, patient cost, care quality, and clinical outcomes. There are two tests that can be used to analyze datasets. According to Orcan (2020), nonparametric tests analyze datasets that do not require normal distribution assumption while parametric tests require equal variance and normality. However, the said tests were not used in the article by Woo et al (2017). In essence, nonparametric tests, as Orcan (2020) points out, are not founded upon assumptions. The author further indicates that in those instances whereby data follows a specific distribution, parametric tests (such as t-test and ANOVA) come in handy. One strength of the study has got to do with the methodology deployed. This would largely have a positive impact on evidence-based practice. However, the study failed to conduct a meta-analysis which would have been appropriate in aggregating estimates of the roles of advanced practice nurses in emergency and critical care settings.

References

Hudgins, J. D., Porter, J. J., Monuteaux, M. C. & Bourgeois, F. T. (2019). Prescription opioid use and misuse among adolescents and young adults in the United States: A national survey study. PLOS Medicine, 16(11), e 1002922.