DNP 679 Module 5 One-Way ANOVA Write-Up Example

Q: Where can I find a free DNP 679 Module 5 sample paper?

The full DNP 679 Module 5 sample is on this page: a one-way ANOVA write-up comparing blood pressure change across three follow-up models, with Tukey comparisons and a Kruskal-Wallis check.

Q: What is the Kruskal-Wallis test?

It is the nonparametric alternative to one-way ANOVA; it compares the rank distributions of three or more independent groups and does not assume normality.

Q: What does eta squared tell me in an ANOVA?

Eta squared tells what fraction of all the variability in the outcome lines up with the groups; compute it as SS between over SS total.

Q: Why use Tukey's HSD after ANOVA?

Tukey's HSD tests every pair of means while keeping the familywise error rate at the chosen alpha, a protection separate t-tests lack.

Q: When is the DNP 679 one-way ANOVA assignment due?

The schedule sets it for the end of Week 10, after the lecture on one-way ANOVA and the Kruskal-Wallis test, and it is worth 10 points.

Course	DNP 679 Biostatistics Principles of Statistical Inference
Module	Module 5
Paper type	One-way ANOVA write-up
Length	About 736 words, 5 pages
Format	APA 7 student paper
School	Arizona State University
Program	Doctor of Nursing Practice
Updated	October 2026

Free sample paper for DNP 679 Module 5

Three Ways to Follow Up High Blood Pressure: A One-Way ANOVA of Twelve-Week Systolic Change With Tukey and Kruskal-Wallis Results

Student Name

Edson College of Nursing and Health Innovation, Arizona State University

DNP 679: Biostatistics: Principles of Statistical Inference

Instructor Name

Month Day, Year

What this page is doingThe title frames the analysis as a comparison of care models, which is how a clinic leader would read it, then names the three tests reported.

Three Ways to Follow Up High Blood Pressure: A One-Way ANOVA of Twelve-Week Systolic Change With Tukey and Kruskal-Wallis Results

Question and Hypotheses

Many clinics are redesigning how they follow patients whose blood pressure is not controlled. Team-based approaches that add pharmacists have produced larger reductions than usual care in trials and reviews (Carter et al., 2009; Margolis et al., 2013). In this hypothetical dataset, 75 adults with a systolic blood pressure of 140 mmHg or higher were randomly assigned to one of three 12-week follow-up models (n = 25 each): usual in-person visits with the primary care provider, video visits with the same provider or pharmacist-led medication management with home blood pressure monitoring. The outcome was change in office systolic blood pressure from baseline to 12 weeks, where negative values mean a reduction.

H0: Mean systolic change is equal in all three populations.

H1: At least one population mean differs (alpha = .05).

What this page is doingThe question is grounded in published trials, and the outcome is defined with the meaning of its sign, which prevents confusion when the means are negative.

Why a One-Way ANOVA

With three groups, the tempting shortcut is three separate t-tests. With alpha at .05 for each, three comparisons push the familywise error rate to roughly 14%. One-way ANOVA tests all three means in a single step at alpha = .05 by comparing the variance between group means with the variance within groups, and a post hoc procedure such as Tukey's HSD then compares pairs while holding the overall error rate at the chosen level (Polit & Beck, 2021). The dataset was set up with one row per patient, a group code from 1 to 3 and the change score, calculated as 12-week minus baseline systolic pressure, and every patient had both readings.

What this page is doingExplaining why ANOVA replaces multiple t-tests, and how the outcome was calculated, gives the reader the reasoning behind the test before the assumptions are checked.

Assumptions

One-way ANOVA assumes independent observations, approximately normal outcomes within groups and equal variances. Random assignment placed each patient in one group. Shapiro-Wilk tests were not significant in any group (p = .367, .150 and .360), and Levene's test showed no evidence of unequal variances, F(2, 72) = 1.43, p = .246. The Kruskal-Wallis test, which compares rank distributions, was run as the nonparametric counterpart.

Results

Systolic blood pressure fell in all three groups, by a mean of 4.92 mmHg with in-person visits, 4.04 mmHg with video visits and 15.52 mmHg with pharmacist-led care. The one-way ANOVA showed a statistically significant difference among the models, F(2, 72) = 9.08, p < .001, and follow-up model accounted for 20% of the variance in systolic change, eta squared = .20, a large effect. Tukey's HSD showed that pharmacist-led care produced a greater reduction than in-person visits (difference 10.60 mmHg, 95% CI [3.42, 17.78], p = .002) and than video visits (difference 11.48 mmHg, 95% CI [4.30, 18.66], p = .001), while in-person and video visits did not differ (p = .954). The Kruskal-Wallis test agreed, H(2) = 16.23, p < .001.

Follow-up model	n	Mean change (mmHg)	SD
Usual in-person visits	25	-4.92	9.38
Video visits	25	-4.04	12.04
Pharmacist-led with home monitoring	25	-15.52	10.20
Total	75	-8.16	11.70
Source	SS	df	MS	F	p
Between groups	2041.04	2	1020.52	9.08	< .001
Within groups	8091.04	72	112.38
Total	10132.08	74
Tukey HSD comparison	Mean difference	p	95% CI
In-person minus video	-0.88	.954	-8.06 to 6.30
In-person minus pharmacist-led	10.60	.002	3.42 to 17.78
Video minus pharmacist-led	11.48	.001	4.30 to 18.66

What this page is doingThe writer reports the omnibus test, the effect size and only then the pairwise results, in that order, and states each difference with its interval so the reader sees how large it may be.

Interpretation

The results suggest that what changed outcomes was not where the visit happened but what happened during follow-up. Video and in-person visits produced almost identical, modest reductions, which supports offering video visits for convenience without expecting better control from them alone. The pharmacist-led model, which combined frequent home readings with active medication adjustment, lowered systolic pressure by about 11 mmHg more than either visit-based model, a difference large enough to matter for cardiovascular risk. With patients randomized, the follow-up model itself is the likeliest source of the gap, within the limits of a 12-week, 75-patient study. A clinic would still need to weigh cost and pharmacist availability, and a longer study would show whether the gap persists. The effect size helps here. An eta squared of .20 means that one-fifth of the variation in blood pressure change was tied to the follow-up model, which is large for a health services comparison, yet four-fifths of the variation came from other sources, such as baseline pressure, adherence and diet. Within the pharmacist-led group, individual changes still ranged widely around the mean of 15.5 mmHg, so some patients would need further intensification regardless of the model.

What this page is doingThe interpretation names the practical meaning of each comparison, links the design to the strength of the causal claim and lists what a decision maker would still need to know.

References

Carter, B. L., Rogers, M., Daly, J., Zheng, S., & James, P. A. (2009). The potency of team-based care interventions for hypertension: A meta-analysis. Archives of Internal Medicine, 169(19), 1748-1755. https://doi.org/10.1001/archinternmed.2009.316

Margolis, K. L., Asche, S. E., Bergdall, A. R., Dehmer, S. P., Groen, S. E., Kadrmas, H. M., Kerby, T. J., Klotzle, K. J., Maciosek, M. V., Michels, R. D., O'Connor, P. J., Pritchard, R. A., Sekenski, J. L., Sperl-Hillen, J. M., & Trower, N. K. (2013). Effect of home blood pressure telemonitoring and pharmacist management on blood pressure control: A cluster randomized clinical trial. JAMA, 310(1), 46-56. https://doi.org/10.1001/jama.2013.6549

Polit, D. F., & Beck, C. T. (2021). Nursing research: Generating and assessing evidence for nursing practice (11th ed.). Wolters Kluwer.

What the DNP 679 Module 5 instructions ask for

Week 10 of DNP 679 is Statistical Analysis V, covering one-way ANOVA and the Kruskal-Wallis test, with a one-way ANOVA to run in Intellectus and a write-up due when the week closes. It is worth 10 points, the third of the course's 10-point written assignments, and it uses a hypothetical dataset supplied with the Canvas prompt. A strong write-up states the question with its hypotheses, checks normality within groups and equality of variances, reports the ANOVA table and the F statistic with both degrees of freedom, gives an effect size, uses a post hoc procedure to show which groups differ and mentions the Kruskal-Wallis result when it is relevant. The week before, the course covers correlation, so this assignment returns to group comparison.

How the DNP 679 Module 5 example is put together

In the order a results section follows, the sample opens with published evidence for team-based hypertension care, describes the hypothetical randomized design and defines the outcome, including what a negative change means. Assumptions are checked with Shapiro-Wilk tests for each group and Levene's test, and the Kruskal-Wallis test is introduced as the nonparametric counterpart. Three tables follow: descriptive statistics by group, the full ANOVA summary and the Tukey comparisons with confidence intervals. The results paragraph reports the means, the F test, eta squared and each pairwise result, then the Kruskal-Wallis statistic. The interpretation explains what the pattern means for a clinic's follow-up model and states what the design does and does not support.

DNP 679 Module 5 rubric: what earns full marks

The ANOVA assignment is graded against a Canvas rubric for 10 points. Typical criteria are a correct choice of test, assumptions checked and reported, the F statistic given with both degrees of freedom and a p value, an effect size, an appropriate post hoc analysis after a significant F and an interpretation tied to the question. Submissions lose points for running post hoc tests without reporting the omnibus test, for reporting only p values without mean differences, for omitting the effect size and for treating a nonsignificant pairwise result as a finding that two groups are equivalent. Write-ups that turn the results into a practice implication, within the limits of the design, show the doctoral judgment the course aims for. Late submissions lose 3% per day.

DNP 679 Module 5 help: mistakes that cost marks

The most frequent problem is a results section that lists every number Intellectus produces without saying what they mean. Lead with the group means in the outcome's units, then the F test and effect size, then the post hoc results. Report the F statistic as F(2, 72), with between and within degrees of freedom. Remember that a significant F only says that some means differ; the post hoc test says which. When means are negative, explain the sign so a reader knows that a larger negative number is a bigger reduction. If assumptions are doubtful, give the Kruskal-Wallis result and say whether it agrees. Bring the ANOVA and Tukey tables plus the prompt to the desk for a check on how each number is reported.

Write yours, or have the desk draft it

This paper is an original model document written by our desk, not a submitted student paper and not an official Arizona State University document. Read it for the moves, then write your own to the instructions in your classroom. If you want one built to your exact prompt and rubric, the first custom sample is free and arrives in 24 to 48 hours.

More DNP 679 and Doctor of Nursing Practice sample papers

DNP 679 Module 5 questions, answered

Where can I find a free DNP 679 Module 5 sample paper?