DNP 679 Module 2 Independent t-Test Write-Up Example

Course	DNP 679 Biostatistics Principles of Statistical Inference
Module	Module 2
Paper type	Independent t-test write-up
Length	About 734 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 2

Do Shift Workers Sleep Worse? An Independent t-Test on Pooled Insomnia Severity Scores and Why a Null Result Is Not a No

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 asks the clinical question in everyday words and warns that the answer is a nonsignificant result that must be read with care.

Do Shift Workers Sleep Worse? An Independent t-Test on Pooled Insomnia Severity Scores and Why a Null Result Is Not a No

Research Question and Hypotheses

Shift work disrupts circadian timing, and clinicians often assume that people who work nights or rotating shifts have more insomnia symptoms. Using the pooled cohort data described after Immersion 1 (75 adults, hypothetical teaching values), this analysis asks whether mean Insomnia Severity Index (ISI) totals differ between respondents who work rotating or night shifts (n = 20) and those who do not (n = 55).

H0: The population mean ISI score of shift workers equals that of non-shift workers.

H1: The population means differ (two-tailed, alpha = .05).

The independent variable, shift work, is nominal with two categories, and the outcome is the ISI total treated as interval, so an independent samples t-test is the planned test, with the Mann-Whitney U test as a nonparametric check.

What this page is doingThe section frames a clinical belief as a testable question, writes both hypotheses in full and justifies the test from the level of each variable.

The Instrument

A test result is only as good as the measure behind it. Reliability refers to consistency and validity to whether a tool measures what it claims (Polit & Beck, 2021). The ISI has shown acceptable internal consistency and correlates with sleep diaries and clinical interviews (Bastien et al., 2001), and later work confirmed its reliability and its ability to detect insomnia cases in community samples (Morin et al., 2011). In our data, however, 15 different students administered the questionnaire informally, which could reduce reliability by adding variation unrelated to sleep. Lower reliability inflates error variance and makes real group differences harder to detect.

What this page is doingTying the week's validity and reliability lecture to the test shows why measurement quality affects the result, not just the methods section.

Choosing Between the Two Tests

The independent t-test is a parametric test: it compares means and assumes that the outcome is roughly normal within each group and that the groups have similar variances. The Mann-Whitney U test is its nonparametric counterpart. It pools both groups, ranks every score from lowest to highest and asks whether one group's ranks tend to be higher, so it does not depend on normality and is less affected by extreme values (Polit & Beck, 2021). It answers a slightly different question, about the distribution of ranks rather than means, and with normal data it is somewhat less powerful. Running both, and reporting whether they agree, protects the conclusion when assumptions are borderline, as they are here.

What this page is doingThis section connects the Week 5 lecture to the decision in this analysis, explaining what each test compares and why both are reported.

Assumptions and Visualization

Group sizes were unequal (20 and 55), so the equal-variance check matters more than usual. Levene's test was not significant, F = 3.36, p = .071, so equal variances were assumed, though the p value close to .05 was noted. Shapiro-Wilk tests within groups gave p = .110 for shift workers and p = .054 for non-shift workers, borderline for the larger group, which supports running a Mann-Whitney U test as a check. A bar plot of group means with 95% confidence interval error bars showed heavily overlapping intervals, and a box plot showed the shift worker median slightly higher with similar spread.

What this page is doingEach assumption is reported with its statistic and a decision, and the borderline values are acknowledged rather than glossed over, which justifies the nonparametric check.

Results

Shift workers scored 1.30 points higher on average than non-shift workers (M = 11.10, SD = 3.73 versus M = 9.80, SD = 4.79), but the difference was not statistically significant, t(73) = 1.10, p = .276, 95% CI [-1.06, 3.66]. The effect size was small, d = 0.29. The Mann-Whitney U test agreed, U = 665.5, p = .167, so non-normal scores would not change the verdict. We fail to reject the null hypothesis.

Group	n	Mean	SD	Median
Shift workers	20	11.10	3.73	11
Non-shift workers	55	9.80	4.79	9
Test	Statistic	df	p	Additional values
Levene's test	F = 3.36	1, 73	.071	equal variances assumed
Independent t-test	t = 1.10	73	.276	mean difference 1.30, 95% CI [-1.06, 3.66]
Cohen's d				0.29
Mann-Whitney U	U = 665.5		.167

What this page is doingThe results give both tests in one table and one paragraph, with the interval and effect size beside the p value, and end with the formal decision in the course's language.

Interpretation

Failing to reject H0 is not evidence that shift work has no effect on insomnia. The confidence interval runs from about 1 point lower to almost 4 points higher in shift workers, so the data are compatible with no difference and with a difference large enough to matter. The study simply lacked power: with only 20 shift workers, a small effect like d = 0.29 would usually be missed. Detecting it with 80% power at alpha = .05 would take roughly 190 people per group. The sample is also a convenience sample, so even a significant result would not generalize. A clinician reading this result should not conclude that shift workers can be spared insomnia screening.

What this page is doingThe interpretation turns a null result into useful information by reading the interval, estimating the needed sample and drawing a careful practice conclusion.

References

Bastien, C. H., Vallieres, A., & Morin, C. M. (2001). Validation of the Insomnia Severity Index as an outcome measure for insomnia research. Sleep Medicine, 2(4), 297-307. https://doi.org/10.1016/S1389-9457(00)00065-4

Morin, C. M., Belleville, G., Belanger, L., & Ivers, H. (2011). The Insomnia Severity Index: Psychometric indicators to detect insomnia cases and evaluate treatment response. Sleep, 34(5), 601-608. https://doi.org/10.1093/sleep/34.5.601

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

DNP 679 Module 2 instructions, in plain terms

Week 6 brings the second immersion and, after it, the Post Immersion 2 Written Assignment. The schedule asks for an independent t-test run in Intellectus and a write-up filed by the Week 6 deadline, for 5 points on a pass or fail basis. Week 5 introduced parametric versus nonparametric testing, including the independent samples t-test and its rank-based Mann-Whitney counterpart, and Week 6 adds bar and line plots and the validity and reliability of instruments, so a complete write-up touches all of them: hypotheses, assumptions, a visualization, the test result, a nonparametric check where warranted and a word on the measure. The written assignments in this course use hypothetical datasets, and the detailed prompt and dataset are in Canvas.

How the DNP 679 Module 2 example is put together

Every part of a short results report appears in order. The research question comes from a common clinical assumption and is stated with both hypotheses. A paragraph on the instrument links reliability and validity to the result, noting that informal administration by many students could weaken a real difference. Assumption checks report Levene's test and Shapiro-Wilk values, including the borderline ones, and describe the bar plot with confidence interval error bars. The results table holds descriptive statistics for both groups and every test statistic, and one paragraph reports the t-test with its interval and effect size, then the Mann-Whitney U test. The interpretation explains why a null result leaves the question open and estimates the sample a proper test would need.

DNP 679 Module 2 rubric: what earns full marks

The Post Immersion 2 assignment is pass or fail, so the rubric in Canvas is best read as a list of elements that must all be present and correct. For an independent t-test write-up those elements are usually correct hypotheses, a reported check of equal variances that determines which result is read, accurate test statistics with degrees of freedom and p values, and a correct conclusion in words. The course's emphasis on writing up tests rewards reporting the interval and d alongside p, and a plain statement that a null result leaves the question open. Each late day costs 3%, and after the fifth day the item scores zero. Generative AI is not permitted at any stage, including drafts and brainstorming.

DNP 679 Module 2 help from the desk

The usual error is writing "there is no difference between groups" after a nonsignificant test. Write that the difference was not statistically significant and say what the confidence interval allows. Another is reading the wrong line of the output; check Levene's test first. With unequal group sizes, report both n values. If a group's distribution is skewed or the sample is small, add the Mann-Whitney U test and say whether it agrees. Avoid copying descriptive statistics without labels; a table with group names, n, mean, SD and median is clearer. Describe your plot in a sentence. A second reader helps here: share your t-test output and the prompt with the desk and the reporting can be checked line by line.

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 2 questions, answered

Where can I find a free DNP 679 Module 2 sample paper?

A complete DNP 679 Module 2 sample is on this page: the Post Immersion 2 independent t-test write-up comparing shift and non-shift workers' insomnia scores, with a Mann-Whitney check.

When should I use the Mann-Whitney U test instead of an independent t-test?

Use it when the outcome is ordinal, when groups are small and clearly non-normal, or as a check when normality is doubtful; it compares ranks rather than means.

What does a nonsignificant t-test mean?

It means the data did not give enough evidence to reject the null hypothesis; it does not prove the groups are equal, especially when the sample is small.

What is the difference between reliability and validity?

A reliable tool returns consistent scores under the same conditions; a valid one captures the concept it claims to capture. Consistency alone does not guarantee validity.

What happens at the DNP 679 immersions?

Students meet on campus three times in the semester to work through analyses together, and post-immersion written assignments follow the first two.

DNP 679 Module 2 Post Immersion 2 Written Assignment: Independent t-Test and Mann-Whitney U Test Example