NU704 Unit 4 Statistical Test Selection Worksheet Example

Course	NU704 Biostatistics, Epidemiology, and Social Determinants of Population Health
Unit	Unit 4
Paper type	Statistical test selection worksheet
Length	About 590 words, 5 pages double-spaced
Format	APA 7 student paper
School	Purdue University Global
Program	Doctor of Nursing Practice (DNP)
Updated	October 2026

Free sample paper for NU704 Unit 4

Six Questions, Six Tests: A Unit 4 Statistical Test Selection Worksheet for an Insulin Discharge Project

[Student Name]

Purdue University Global

NU704: Biostatistics, Epidemiology, and Social Determinants of Population Health

Unit 4 Statistical Test Selection Worksheet

[Instructor Name]

[Date]

The project, questions and data are composite and illustrative; the sources cited are real.

What this part is doingPromising a justification for every choice in the lead shows the worksheet values reasoning over recall.

How the Choices Are Made

Choosing a test starts with three questions: what type of variable is the outcome, how many groups are compared, and are the groups independent or paired? Assumptions such as normal distribution and equal variance then decide between parametric and nonparametric options (Mishra et al., 2019).

Question 1

Did the proportion of patients returning within 30 days for glucose problems change between the twelve months before the bundle and the six months after? The outcome is categorical, returned or not. There are two independent groups, different patients before and after. A chi-square test of independence fits, switching to Fisher's exact test when the table is too sparse for the chi-square approximation. Alongside the p-value the worksheet calls for the relative risk with its interval, since the size of the change matters more than significance alone.

Question 2

Did mean knowledge scores on a ten-item insulin quiz differ between patients taught with return demonstration and patients taught before the change? The outcome is a score, treated as continuous, in two independent groups. If scores are roughly normal, an independent-samples t-test applies; if they are skewed, which is likely with a ceiling near 10, the Mann-Whitney U test is the safer choice.

With a ten-point quiz most patients will score near the top, and that ceiling decides the test.

Question 3

Did the same patients' confidence scores, rated 0 to 10 before and after the supervised injections, change? The outcome is ordinal, and the measurements are paired, the same patient twice. The Wilcoxon signed-rank test fits; a paired t-test would assume interval data and a normal distribution of differences, which a 0-to-10 rating may not meet.

Question 4

Did return demonstration rates differ among three groups defined by preferred language: English, Spanish and Marshallese? The outcome is categorical and the groups are independent, so a chi-square test of independence applies across the three groups. With small Marshallese numbers, Fisher's exact test, or combining categories with care, may be required.

Question 5

Is the number of supervised self-injections before discharge associated with 30-day return, after accounting for age and literacy screen result? The outcome is binary and there are several predictors, some continuous. Logistic regression fits, reporting adjusted odds ratios with confidence intervals. The project's sample may limit how many predictors can be included reliably.

What this part is doingNoting the sample-size limit on regression shows the worksheet considers whether a method is feasible, not only correct.

Question 6

Is the weekly demonstration rate improving over the pilot? The data are a proportion measured repeatedly over time on one process. Rather than a significance test, a run chart with a median line and standard rules for shifts and trends is appropriate for a quality improvement project (Perla et al., 2011).

Summary of Choices

Across the six questions, the choice of method followed from the outcome type, the number and independence of groups and the data's distribution. Several choices also anticipate practical problems, such as ceiling effects, small subgroups and limited sample size, that could make a standard test misleading. Each analysis will report an effect measure with an interval, not only a p-value, in line with the American Statistical Association's warning that a p-value by itself says nothing about how large or how important a difference is (Wasserstein & Lazar, 2016).

What this part is doingEnding with the reporting rule ties the worksheet to how results will be presented in the project.

References

Mishra, P., Pandey, C. M., Singh, U., Keshri, A., & Sabaretnam, M. (2019). Selection of appropriate statistical methods for data analysis. Annals of Cardiac Anaesthesia, 22(3), 297-301. https://doi.org/10.4103/aca.ACA_248_18

Perla, R. J., Provost, L. P., & Murray, S. K. (2011). The run chart: A simple analytical tool for learning from variation in healthcare processes. BMJ Quality & Safety, 20(1), 46-51. https://doi.org/10.1136/bmjqs.2009.037895

Wasserstein, R. L., & Lazar, N. A. (2016). The ASA statement on p-values: Context, process, and purpose. The American Statistician, 70(2), 129-133. https://doi.org/10.1080/00031305.2016.1154108

Reading the NU 704 Unit 4 instructions

The fourth unit commonly gives students a set of research or project questions and asks which statistical method fits each, with justification. Prompts expect students to identify the type of outcome variable, the number of groups compared, whether observations are independent or paired and whether parametric assumptions are likely to hold. Instructors reward worksheets that anticipate real problems with the data, such as ceiling effects, small subgroups or limited sample sizes, and that choose nonparametric or descriptive approaches when appropriate. For quality improvement questions about change over time, many sections expect run charts rather than significance tests. A brief explanation per question is usually more valuable than long theory. Sources commonly include a guide to selecting statistical methods, a reference on run charts and, often, guidance on reporting effect sizes and p-values.

How this NU704 Unit 4 example is built

The worksheet opens with the three questions that drive every choice, citing Mishra and colleagues. Question 1 compares return rates before and after the bundle and chooses a chi-square test, with an exact test for sparse tables, plus the relative risk and interval. Question 2, on quiz scores, explains why a ceiling near ten favors the Mann-Whitney test, under a highlighted line. Question 3 uses the Wilcoxon signed-rank test for paired ratings. Question 4 extends the chi-square test to three language groups, with a caution about small cells. Question 5 chooses logistic regression and notes a sample-size limit, marked by a margin note. Question 6 uses a run chart for weekly process data, citing Perla and colleagues. A summary ties every choice to data type and design and commits to effect measures with intervals, citing Wasserstein and Lazar.

Where the points sit in the NU704 Unit 4 rubric

Test selection worksheets are marked on correct identification of variable types and design, appropriate choice of method, justification, attention to assumptions and practical limits, and reporting. Identification credit requires naming the outcome as categorical, ordinal or continuous and the groups as independent or paired. Choice of method must follow from those features. Justification is weighted heavily, since a correct answer without reasoning earns little. Graders reward recognition of ceiling effects, sparse cells and sample-size limits, and of when a run chart suits better than a test. Reporting plans that pair each test with an effect measure and interval show statistical maturity. Concise entries that a project team could actually use are preferred to theory. Accurate citations for the selection guide and run chart reference complete the criteria in most sections of this DNP course.

NU704 Unit 4 help from the desk

Worksheets lose marks most often when they choose a familiar test, usually a t-test, for every question regardless of the outcome type. Another frequent error is treating paired measurements as independent, or the reverse, which changes the test entirely. Some students apply parametric tests to ordinal ratings or skewed scores without comment. Others choose a significance test for a quality improvement question that a run chart would answer better. Ignoring small subgroups can lead to tests that are technically invalid. Begin each question by writing the outcome's type and the groups in one line; the test usually follows from that line. Then ask what could go wrong with the real data. Your project questions and the worksheet instructions are enough for a custom worksheet to be built.

Get an NU704 Unit 4 example written to your instructions

Send the questions or study designs and the worksheet instructions with the rubric. A custom worksheet identifies variable types and design for each question and justifies the test chosen. It arrives within 24-48h; first requests are free. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.

More NU704 papers and DNP samples

NU704 Unit 4 questions, answered

How do you choose a statistical test?

By the type of outcome variable, the number of groups, whether groups are independent or paired and whether data meet the test's assumptions.

When is a nonparametric test used?

When data are skewed, ordinal or from small samples that do not meet the assumptions of parametric tests.

Should the worksheet explain each choice?

Yes. A brief justification for each test shows understanding rather than memorization.

Where can I find a free NU704 Unit 4 sample paper?

You can read the complete NU704 Unit 4 statistical test selection worksheet on this page. It matches six questions from an insulin discharge project to statistical methods and justifies each choice by outcome type, design and data limits.

Why does the NU704 Unit 4 worksheet use a run chart instead of a test for weekly data?

During a pilot the team wants to know week by week if the demonstration rate is moving. A run chart with a median line shows change as it happens and uses defined rules to separate signal from noise.