Data Analysis in Health-Related Research

Types of Study Used in the Articles

All of the articles under consideration present the findings of the original research. Walker, Farzan, Gaydos, DeCastro, and Jonassaint (2016) apply qualitative analysis to discover the perceptions of Hydroxyurea among patients suffering from sickle-cell disease (SCD). This study applies the context analysis framework to develop deductive codes that allow defining the changes in the understanding of suggested treatment strategies. Research by Lê et al. (2015) is a quantitative one. Its purpose is to reveal the effects of traditional treatment with the help of the cohort study. Finally, the study by Lebensburger, Grosse, Altice, Thierry, and Ivankova (2015) utilizes both qualitative and quantitative types of study to investigate different factors that influence patients’ ability to acquire new knowledge.

Types of Statistical Tests and Rationale for Their Selection

Statistical tests are integral to quantitative research due to their ability to support research by numbers. Thus, research by Lê et al. (2015) utilizes such statistical tools as Graphpad Prism1 software 5.0, Kruskal–Wallis test, Dunn’s multiple comparison test, Fisher’s tests, Kaplan–Meier, the log-rank test, and P-value. This set of tests was used because they allow analyzing data in a way to answer research questions. Thus, Kruskal–Wallis test compares continuous data of multiple groups, and Dunn’s multiple comparison test was applied to compare the results obtained from pairs of groups. Fisher’s tests were included in the study to deal with proportions. Kaplan–Meier is related to the confidence interval, which was estimated as 95%. Also, survival rates were compared through a log-rank test. Moreover, the P-value was adjusted to multiple testing. Statistical significance was established attained when a P-value was less than 0.0167 for multiple testing by the Bonferroni method.

The study by Lebensburger et al. (2015) used descriptive statistics to interpret the research findings. Descriptive statistics proved to be a suitable method that empowered this cohort study to compare patient outcomes among people treated traditionally and those treated with the help of Hydroxyurea. The use of descriptive statistics as a component of the mixed research design contributed to the determination of the major trends in patients’ health status. Finally, the qualitative research by Walker et al. (2016) used the codes that derived descriptive statistics based on NVivo 11 software. This type of statistical analysis can be applied in qualitative research supporting the findings with numbers and thus adding validity.

Applicability of Statistical Tests

The majority of studies could have been improved in case some other types of analyses were applied. For example, in the qualitative research by Walker et al. (2016) the utilization of a mixed-method would have allowed evaluating the extent of the change in patients’ perceptions. Moreover, it could have provided additional information for future research. In addition, application of quantitative tools would have made the results of the analysis less subjective, thus eliminating this limitation of the initial design. In case of quantitative study by Lê et al. (2015), research could have been improved by the involvement of the mixed research design. This type of design could have helped to identify the major trends in patients’ health status using descriptive statistics. As for the study by Lebensburger et al. (2015), it could have benefited from the use of purely qualitative design because it could have allowed focusing on research content and lead to more coherent results.

Differences between Parametric and Nonparametric Tests

Statistical tests are applied to generalize data bout the population included in the research sample. They empower decisions about the reasonableness of research hypothecs and are divided into parametric and nonparametric tests. Parametric tests are those that contain information about a certain parameter of the population from the sample. In turn, nonparametric tests do not involve population parameters. These types of tests have different measurement levels. Thus, while the results of parametric tests are presented with the help of interval or ration, nominal or ordinal measurement level is typical of nonparametric tests. Another core distinction is that in parametric tests, population data are known due to its parameters. For example, parametric tests include t-test, f-test, z-test, and ANOVA. Nonparametric tests do not operate knowledge about the study population or its parameters. These tests comprise Mann-Whitney, rank-sum test, Kruskal-Wallis test, etc. In this analysis, only the quantitative study uses parametric tests while qualitative and mixed ones utilize nonparametric tests to manage data.

Assessment of Reliability and Validity Factors

Reliability and validity are crucial for every research. The study by Walker et al. (2016) can be considered valid and reliable due to the methods used by researchers. The framework created for this research is flexible, which makes it applicable to other tasks related to the research problem. Research by Lê et al. (2015) is characterized by a strong validity due to its concentration on the quantitative method, which allows controlling variables. Moreover, the findings can be considered valid due to the opportunity for their generalization. The study by Lebensburger et al. (2015) has a relatively high validity due to the focus on a certain issue of nursing care, which provides an opportunity to discover this issue in detail. Nevertheless, its reliability and validity could have been improved by reducing the number of influences under consideration thus providing better control over them.

Summary of the Studies’ Application to Practice

All of the studies considered in this research have implications for nursing practice. Thus, the study by Lê et al. (2015) presents the experience of treatment for sickle cell disease contributing to evidence-based practice development. The researchers compare diverse treatment options and provide statistical data that reflect the effectiveness of these treatment methods. Research by Lebensburger et al. (2015) contributes to the nursing understanding of the significance of health education for parents of children with sickle cell anemia. The awareness of education interventions necessity can have a positive impact on patient outcomes due to nurses’ attention to this issue. Finally, Walker et al. (2016) attract attention to an opportunity to evaluate the perceptions of hydroxyurea among sickle cell disease stakeholders through the use of social media. Social media can become a powerful tool that allows focusing on the importance of sickle cell disease problem among the interested parties.

References

Lê, P. Q., Gulbis, B., Dedeken, L., Dupont, S., Vanderfaeillie, A., Heijmans, C.,… Rozen, L. (2015). Survival among children and adults with sickle cell disease in Belgium: Benefit from hydroxyurea treatment. Pediatric Blood & Cancer, 62(11), 1956-1961.

Lebensburger, J. D., Grosse, S. D., Altice, J. L., Thierry, J. M., & Ivankova, N. V. (2015). Understanding and improving health education among first-time parents of infants with sickle cell anemia in Alabama: A mixed methods approach. Journal of Pediatric Hematology/Oncology, 37(1), 35-42.

Walker, A. L., Farzan, R., Gaydos, L. M., DeCastro, L., & Jonassaint, C. (2016). Assessing perceptions of hydroxyurea among sickle cell disease stakeholders using social media. Blood, 128(22), 318.

Appendix A: Methodologies Beneficial for the Studies