DNP 679 Module 3 Paired t-Test Written Assignment Example

Reviewed by Ingrid Vasterling, MSN, RN Arizona State University Updated October 2026

This DNP 679 Module 3 sample is the Paired t-test Written Assignment in Biostatistics: Principles of Statistical Inference, which ASU's Doctor of Nursing Practice students take as their core statistics course. In Week 7, after a lecture on the paired t-test and the Wilcoxon signed rank test, ASU DNP 679 asks students to run a paired t-test in Intellectus on a hypothetical dataset and write up the results for 10 points. The composite family nurse practitioner student analyzes A1c values from 28 adults measured before and after a 12-week diabetes self-management group at her clinic. She checks the distribution of the differences, gives the mean change with its interval and a paired effect size, runs the Wilcoxon test as a check and separates a statistically significant drop from a clinically meaningful one.

CourseDNP 679 Biostatistics Principles of Statistical Inference
ModuleModule 3
Paper typePaired t-test write-up
LengthAbout 755 words, 5 pages
FormatAPA 7 student paper
SchoolArizona State University
ProgramDoctor of Nursing Practice
UpdatedOctober 2026

Free sample paper for DNP 679 Module 3

1

Half a Point in Twelve Weeks: A Paired t-Test of A1c Before and After a Diabetes Self-Management Group

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 leads with the size of the change, which is the number a clinician cares about, and then names the test and the program.
2

Half a Point in Twelve Weeks: A Paired t-Test of A1c Before and After a Diabetes Self-Management Group

Background and Question

Diabetes self-management education teaches skills such as meal planning, medication use and glucose monitoring, and reviews link it to modest improvements in glycemic control (Chrvala et al., 2016). The hypothetical dataset for this assignment follows 28 adults with type 2 diabetes who attended a 12-week group program at a primary care clinic. A1c, the percentage of glycated hemoglobin, was measured at enrollment and again at 12 weeks.

The question is whether mean A1c changed from enrollment to 12 weeks. Because the same people were measured twice, the observations are paired, and the analysis works on each person's difference score.

H0: The population mean difference in A1c (enrollment minus 12 weeks) is zero.

H1: The population mean difference is not zero (two-tailed, alpha = .05).

What this page is doingThe background explains the program in a sentence and the reason the data are paired, so the choice of test follows directly from the design.
3

Data Preparation

The dataset holds one row per participant with an identification number and two columns, A1c at enrollment and A1c at 12 weeks. Paired analysis requires both values for every person, and all 28 participants had both, so no cases were dropped. A difference column, enrollment minus 12 weeks, was created so that positive values mean improvement; this direction is stated once and kept throughout the write-up. Values were checked against the plausible range for A1c, and none fell outside it.

What this page is doingStating the direction of subtraction before the results prevents the most common confusion in paired write-ups, a sign that reverses between the table and the text.
4

Assumptions

For a paired comparison, it is the difference scores that must be roughly normal, and each person's pair must be independent of everyone else's. The 28 difference scores had a Shapiro-Wilk p value of .284, and their histogram was roughly symmetric with no extreme values, so normality of the differences was reasonable. Each participant contributed one pair, so pairs were independent. The Wilcoxon signed rank test, which compares the ranks of the differences without assuming normality, was run as a check.

What this page is doingThe writer tests normality on the differences rather than on the raw scores, which is the correct assumption for a paired test and a common source of lost points.
5

Results

Mean A1c fell from 8.55% (SD = 1.28) at enrollment to 8.01% (SD = 1.47) at 12 weeks, a mean reduction of 0.54 percentage points, 95% CI [0.33, 0.75]. The change was statistically significant, t(27) = 5.36, p < .001, and the effect size for paired data was large, d = 1.01. The Wilcoxon signed rank test led to the same conclusion, W = 26.5, p < .001. Twenty-two participants improved, five rose and one did not change. We reject the null hypothesis.

MeasurenMeanSD
A1c at enrollment (%)288.551.28
A1c at 12 weeks (%)288.011.47
Difference (enrollment minus 12 weeks)280.540.53
TestValue
Correlation between time pointsr = .93
Paired t-testt(27) = 5.36, p < .001
95% CI of the mean difference0.33 to 0.75
Effect size (d for paired data)1.01
Wilcoxon signed rank testW = 26.5, p < .001
Direction of change22 lower, 5 higher, 1 unchanged
What this page is doingThe results report the change in the outcome's own units first, then the test, interval and effect size, and add how many individuals moved in each direction, which keeps the average from hiding the spread.
6

Interpretation

Statistical significance here reflects the strong correlation between the two time points (r = .93): comparing every participant with herself or himself removed the large differences between people, so even a modest shift stood out clearly. Clinical significance is a separate question. A reduction of about half a percentage point is similar to what reviews of self-management education report (Chrvala et al., 2016), and it is a worthwhile change for an individual. Still, the mean at 12 weeks remained above 8%, and only four participants were below 7% at follow-up, compared with three at enrollment. The large standardized effect size (d = 1.01) comes partly from the small standard deviation of the differences and should not be read as a large clinical effect.

The design also limits what can be concluded. With one group and no comparison, the drop could reflect regression to the mean, since people often enroll when their A1c is unusually high, as well as medication changes or seasonal effects that occurred during the 12 weeks (Polit & Beck, 2021). A comparison group or a longer series of measurements would be needed to attribute the change to the program. Durability is a further question. A meta-analysis of self-management education found that the improvement in glycemic control was largest at the end of the intervention and smaller at later follow-up (Norris et al., 2002), so a 6- or 12-month measurement would show whether this group's half-point gain holds.

What this page is doingThe interpretation separates three ideas students often merge: why the test was significant, whether the change matters clinically and whether the design supports a causal claim.
7

Conclusion

In this hypothetical dataset, A1c fell by a mean of 0.54 points over 12 weeks, a statistically significant change confirmed by a nonparametric test. The change is clinically modest and, without a comparison group, cannot be credited to the program alone.

References

Chrvala, C. A., Sherr, D., & Lipman, R. D. (2016). Diabetes self-management education for adults with type 2 diabetes mellitus: A systematic review of the effect on glycemic control. Patient Education and Counseling, 99(6), 926-943. https://doi.org/10.1016/j.pec.2015.11.003

Norris, S. L., Lau, J., Smith, S. J., Schmid, C. H., & Engelgau, M. M. (2002). Self-management education for adults with type 2 diabetes: A meta-analysis of the effect on glycemic control. Diabetes Care, 25(7), 1159-1171. https://doi.org/10.2337/diacare.25.7.1159

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

DNP 679 Module 3 instructions, in plain terms

The paired t-test write-up falls due as Week 7 closes and opens the set of three 10-point written assignments. Week 7's topic is Statistical Analysis II, the paired t-test and the Wilcoxon signed rank test, and the schedule directs students to conduct a paired t-test in Intellectus and submit a written assignment. The course description of the written assignments explains that each one applies statistical knowledge by conducting, interpreting and writing up a test on a hypothetical dataset. Expect the Canvas prompt to provide the dataset and to ask for hypotheses, an assumption check, the test result in proper form and an interpretation. Including the nonparametric alternative as a check shows you understood both halves of the week's lecture.

Inside the DNP 679 Module 3 example

Opening with two sentences of clinical background, the sample explains why the A1c data are paired and states both hypotheses. The assumptions section checks normality on the difference scores, not the raw values, and introduces the Wilcoxon signed rank test as a check. Results sit in two tables, one for the descriptive statistics at each time point and for the differences, and one for the correlation, test statistic, confidence interval, effect size, Wilcoxon result and the count of people who improved or worsened. The results paragraph reports the change in percentage points before the test. The interpretation explains why a paired design detects small changes, compares the size of the change with published reviews and names the threats a single-group design leaves open.

Reading the DNP 679 Module 3 grading rubric

Canvas holds the Week 7 rubric, and the syllabus lists the assignment at 10 points. In paired t-test write-ups, points generally go to a correct statement of hypotheses, the right assumption checked the right way, accurate reporting of the mean difference with its confidence interval, t, degrees of freedom and p value, and an interpretation in plain language. Write-ups that distinguish statistical from clinical significance, and that say what the design cannot show, tend to reach the top of the scale. Errors that commonly cost points include degrees of freedom of n minus 2, normality tested on the raw scores and a causal claim from a single group. Graded work appears within a week of the due date, and late submissions lose 3% per day.

DNP 679 Module 3 help with common mistakes

Many students report that the test was significant and stop. Give the change in the outcome's units first, here percentage points of A1c, then the test. Do not confuse the direction of subtraction; say whether the difference is before minus after or the reverse, and keep the sign consistent. Check normality on the differences. If the differences are skewed or the sample is very small, report the Wilcoxon signed rank test and say whether it agrees. Avoid calling the program effective when there is no comparison group; describe the change and the alternative explanations. Every word of this assignment must be your own, because the course bars generative AI throughout. With your Intellectus output and the prompt, the desk can help you check the write-up.

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 3 questions, answered

Where can I find a free DNP 679 Module 3 sample paper?

The DNP 679 Module 3 sample is above in full: a paired t-test write-up on hypothetical A1c values before and after a diabetes self-management group, with a Wilcoxon check and interpretation.

What are the assumptions of a paired t-test?

The difference scores should be approximately normally distributed, and each pair should be independent of the other pairs; normality of the raw scores at each time is not required.

What is the Wilcoxon signed rank test?

It is the nonparametric alternative to the paired t-test; it ranks the absolute differences and compares positive and negative ranks, so it does not assume normality.

What is the difference between statistical and clinical significance?

A result earns statistical significance when sampling error alone makes it unlikely; clinical significance asks if patients or clinicians would notice and value the change.

How many points are the DNP 679 written assignments worth?

The two post-immersion assignments are 5 points each and pass or fail; the paired t-test, chi-square and one-way ANOVA assignments are 10 points each.