Self-Study: Comparison of Means, Comparison of Means, Part II: ANOVA and Kruskal Wallis
Throughout the course, there will be a self-study Discussion pertaining to an important concept or topic covered within the assigned week. These Discussions are designed to give you the opportunity to collaborate with your peers and faculty, test your knowledge, ask questions, practice research analysis, and assist your peers.
You are not required to post to this forum; however, you are encouraged to post, review the posts of others, as well as answer questions and/or provide clarity and collaboration with your peers. Discussions will be graded as either Complete or Incomplete.
Resources
Be sure to review the Learning Resources before completing this activity.
Click the weekly resources link to access the resources.
· Laerd Statistics: Sign up for a one-month plan using this link
to an external site.
·
One Way ANOVA in SPSSLinks to an external site.
·
Kruskal-Wallis Test in SPSSLinks to an external site.
· Salkind, N., & Frey, B. (2019).
Statistics for people who (think they) hate statistics (7th ed.). SAGE Publications.
· Chapter 14, “Analysis of Variance: Two Groups Too Many?” (pp. 269–273, 279)
·
Document:
Statistical Output: Kruskal-Wallis (Word document)
Download Statistical Output: Kruskal-Wallis (Word document)
·
Document:
Statistical Output: ANOVA (Word document)
Download Statistical Output: ANOVA (Word document)
Niedz, B. (2024).
Comparison of means, part II: ANOVA and Krustkal Wallis [Video]. Walden University Canvas.
Document:
Comparison of Means, ANOVA and Kruskal Wallis (PowerPoint presentation)
Required Resources for Topic: ANOVA
· Turgut, M., & Yıldız, H. (2023).
Investigation of grief and posttraumatic growth related to patient loss in pediatric intensive care nurses: A cross-sectional studyLinks to an external site..
BMC Palliative Care, 22(1), 195.
To prepare:
· Read and view the Learning Resources in Doherty and Skalsky, and Dang et al. (2021) in Required Readings.
· View the video on comparison of means.
Use this Discussion to collaborate with your peers and faculty as an open office hours/ Q&A forum.
Post answers to the following:
· Summarize the ANOVA or the Kruskal-Wallis output after completing the self-learning module and completing the Required Readings.
· Turgut and Yıldız (2023) used both a
t test and an ANOVA and presented these findings in Table 2 on page 5.
· Compare and contrast the comparisons for the impact of
duration of work on the unit, on the Texas Revised Inventory of Grief (ANOVA), and the
education on terminal period and grief (
t test).
· Were these the correct tests to be used in these analyses? Explain why.
For this Self-Study Discussion, you may post throughout Week 5. You are not required to post to this forum; however, you are encouraged to post, review the posts of others, as well as answer questions and/or provide clarity and collaboration with your peers. Discussions will be graded as either Complete or Incomplete.
Use this Discussion to collaborate with your peers and faculty as an open office hours/ Q&A forum.
Post answers to any or all of the following:
· Summarize the ANOVA or the Kruskal-Wallis output after completing the self-learning module and completing the Required Readings.
· Turgut and Yıldız (2023) used both a
t-test and an ANOVA and presented these findings in Table 2 on page 5.
· Compare and contrast the comparisons for the impact of
duration of work on the unit, on the Texas Revised Inventory of Grief (ANOVA), and the
education on terminal period and grief (
t-test).
· Were these the correct tests to be used in these analyses? Explain why.
Our interactive discussion addresses the following learning objectives:
· Differentiate between ANOVA and Kruskal-Wallis tests of significance
· Summarize statistical findings for ANOVA and for Kruskal-Wallis
RESPOND TO THIS DISCUSSION POST
Michelle
· ANOVA (Analysis of Variance) is a statistical test used to compare the means of three or more groups to see if at least one group is different from the others. It works well when the data is normally distributed and the variance (spread of data) is the same across all groups. For example, if you’re comparing the grief scores of three different groups based on their work duration, ANOVA would test whether the average scores are different between these groups. However, if these conditions aren’t met, an alternative test like the Kruskal-Wallis test is used. This test compares the ranks of the data instead of the actual values, making it more suitable when the data isn’t normally distributed or has unequal variances.
· Turgut and Yıldız (2023) used ANOVA to study the impact of work duration on grief scores using the Texas Revised Inventory of Grief. If they are comparing multiple groups (like people with different years of work experience), ANOVA would be the right choice, as long as the data fits the assumptions of normality and equal variance. If the data didn’t meet these assumptions, they would have needed to use the Kruskal-Wallis test instead. In their study, ANOVA would also be suitable for comparing grief scores across multiple groups if other factors, such as work role or age, are involved, as long as the data is normally distributed.
· For the comparison of grief scores based on education about terminal periods, Turgut and Yıldız used a t-test, which is correct when comparing just two groups. For example, they might compare grief scores between those who received education about terminal periods and those who didn’t. This test works well if the data follows a normal distribution and the variance is similar
· between the two groups. However, if there are more than two groups or if the normality assumption is violated, they would need to use ANOVA (for multiple groups) or a non-parametric test like the Mann-Whitney U test for two groups. Maintaining statistical rigor ensures the appropriate use of statistical analyses that uphold the scientific method and contribute credible results to the scientific literature (Hodges et al., 2023). Overall, the tests they used seem appropriate, but only if the data meets the necessary assumptions.
·
References
· Hodges, C. B., Stone, B. M., Johnson, P. K., Carter III, J. H., Sawyers, C. K., Roby, P. R., & Lindsey, H. M. (2023). Researcher degrees of freedom in statistical software contribute to unreliable results: A comparison of nonparametric analyses conducted in SPSS, SAS, Stata, and R.
Behavior Research Methods,
55(6), 2813-2837.
· Salkind, N., & Frey, B. (2019).
Statistics for people who (think they) hate statistics (7th ed.). SAGE Publications.
· Turgut, M., & Yıldız, H. (2023). Investigation of grief and posttraumatic growth related to patient loss in pediatric intensive care nurses: A cross-sectional study.
BMC Palliative Care, 22(1), 195.
·
