| Course | PU525 Biostatistics for Public Health |
|---|---|
| Unit | Unit 4 |
| Paper type | Hypothesis test worksheet |
| Length | About 910 words, 6 pages double-spaced |
| Format | APA 7 student paper |
| School | Purdue University Global |
| Program | Master of Public Health (MPH) |
| Updated | October 2026 |
Free sample paper for PU525 Unit 4
Four Questions, Four Different Tests: A Hypothesis Test Worksheet Drawn From a Composite County's Immunization, Blood Pressure and Hospital Data, Where Each Item States the Hypothesis, the Measurement Level and the Group Structure Before It Names the Test That Follows
[Student Name]
Purdue University Global
PU525: Biostatistics for Public Health
Unit 4 Hypothesis Test Worksheet
[Instructor Name]
[Date]
The county data and every figure are composite and bracketed for the writer's own data; the methods cited are real.
Item 1: Is Adult Flu Vaccination Below the Target?
Question: does adult influenza vaccination coverage fall short of the county's [70] percent target? Hypotheses: null, the population proportion vaccinated equals [0.70]; alternative, it is less than [0.70]. Measurement: vaccinated yes or no, a binary outcome. Groups: one sample compared with a fixed value. Test: one-sample test of a proportion. Assumptions: random sample; expected counts large enough for the normal approximation, which they are with [600] respondents. Result: [372] of [600] vaccinated, [62.0] percent. The test statistic z is about [negative 4.3], with a P value well below [0.001]. Interpretation: coverage is very unlikely to be at the target; the shortfall is about eight percentage points.
Item 2: Does Coverage Differ by Insurance Status?
Question: is vaccination associated with insurance type? Hypotheses: null, vaccination is independent of insurance type; alternative, it is not. Measurement: vaccinated yes or no by private insurance, Medicaid or uninsured: two categorical variables. Groups: independent. Test: a chi-square test for association, with insurance in three columns and vaccination in two rows. Data: private [210] of [300] vaccinated, [70] percent; Medicaid [122] of [200], [61] percent; uninsured [40] of [100], [40] percent. Result: chi-square about [28.8] with two degrees of freedom, P below [0.001]. Interpretation: coverage differs by insurance type, lowest among the uninsured. The test says the groups differ, not which pairs differ; follow-up comparisons would be needed. The chi-square test answers whether coverage depends on insurance at all; it does not say where the difference lies.
Item 3: Did Blood Pressure Fall Among Program Participants?
Question: did systolic blood pressure change among [40] participants from enrollment to six months? Hypotheses: null, mean change equals zero; alternative, it does not. Measurement: systolic pressure in mmHg, continuous. Groups: paired, the same people measured twice. Test: paired t-test on the differences. Assumptions: differences approximately normal; checked with a histogram of the [40] differences, which was roughly symmetric. Result: mean change [negative 6.2] mmHg, standard deviation of change [11.5]; t about [negative 3.4] with [39] degrees of freedom; P about [0.002]. Interpretation: average systolic pressure fell about six points, but without a comparison group, the change could reflect regression toward the mean or other factors.
Item 4: Do Lengths of Stay Differ Between Two Hospitals?
Question: do lengths of stay for pneumonia differ between the county's two hospitals? Hypotheses: null, the distributions of length of stay are the same; alternative, they differ. Measurement: days, continuous but strongly right-skewed. Groups: independent, [180] and [210] patients. Test: Wilcoxon rank-sum test, because the skew makes a t-test's comparison of means less meaningful for typical stays. Result: medians [3] and [4] days; P about [0.004] from software. Interpretation: stays tend to be longer at the second hospital. With large samples, a t-test on the means would also be valid, but it would answer a different question, about averages rather than typical stays (Fagerland, 2012).
Where the Data Came From
Item 1 and Item 2 use the county's adult health survey, a random sample of households. Item 3 uses the hypertension program's own records, measured with the same automated cuffs at enrollment and six months. Item 4 uses hospital discharge data for adults admitted with pneumonia in one year. Each source shapes the question it can answer: a survey sample supports inference about the county's adults, a program record describes only people who enrolled, and discharge data cover only patients who were admitted. Stating the source before the test keeps the interpretation within what the data can support.
Choosing One-Sided or Two-Sided Tests
Item 1 uses a one-sided alternative because the county asked only whether coverage fell short of its target, and a result above the target would not change any decision. The other three items test in both directions, since a gap either way would change what the county does. The choice is made before seeing the data and recorded, so it cannot be adjusted to reach significance.
How the Tests Were Chosen
Measurement level, number of groups and whether groups are paired decide the test (Rosner, 2016). A binary outcome against a fixed value calls for a test of one proportion; two categorical variables call for a chi-square test; the same people measured twice call for a paired test; and a skewed outcome in two independent groups may call for a rank-based test.
What a P Value Does and Does Not Mean
Each P value gives the share of repeated samples that would produce a result at least this far from the null, if the null held. It is not the chance the null hypothesis is correct, and it is silent on how large or how meaningful a difference is, so every item also reports the difference (Whitley & Ball, 2002).
Common Errors to Avoid
Analyzing before-and-after readings on the same people as if they were two separate groups. Treating a chi-square result as identifying which groups differ. Reporting a P value without the effect size.
References
Fagerland, M. W. (2012). t-tests, non-parametric tests, and large studies: A paradox of statistical practice? BMC Medical Research Methodology, 12, 78. https://doi.org/10.1186/1471-2288-12-78
Rosner, B. (2016). Fundamentals of biostatistics (8th ed.). Cengage Learning.
Whitley, E., & Ball, J. (2002). Statistics review 3: Hypothesis testing and P values. Critical Care, 6(3), 222-225. https://doi.org/10.1186/cc1493
What the PU525 Unit 4 assignment asks for
Unit 4 in PU 525 typically asks students to select and carry out appropriate hypothesis tests for several research questions. Instructions usually require writing hypotheses, justifying the test choice, checking assumptions, reporting results and interpreting them. Put selection before calculation: state the question, write hypotheses about the population, identify the outcome's measurement level and whether groups are independent or paired, then name the test. Check assumptions under their own heading. Report the effect size, not only the P value. Note what each test cannot show, such as cause without a comparison group. Describe the data source for each item so conclusions stay within its population. Cite a statistics text in APA style, and explain one-sided or two-sided choices in advance.
How the PU 525 Unit 4 example is put together
The sample presents four boxed items with identical subheadings, with a margin note on putting selection first. Item 1 tests flu coverage against a target. Item 2 uses a chi-square test on coverage by insurance, and a highlighted line notes that the test shows a difference exists but not where. Item 3 uses a paired t-test on blood pressure and carries a margin note on what it cannot establish. Item 4 compares skewed lengths of stay with a rank-based test, citing research on when t-tests remain valid. A section on data sources carries a third note, followed by one-sided choices, how tests were chosen, the meaning of a P value and common errors.
Where the points sit in the PU525 Unit 4 rubric
Hypothesis test worksheets in PU 525 are graded on correct test selection, properly stated hypotheses, assumption checks, accurate results, interpretation with effect sizes and awareness of each test's limits. Selection earns the most weight, so reasoning about measurement level and pairing should come before any formula. Hypotheses must be written about the population. Assumptions should be checked, not assumed. Results should include the size of the effect alongside the P value. Interpretation should stay within what the data source and design allow. Worksheets that state what a significant result cannot prove, such as cause in a before-and-after design, tend to receive the highest marks. Naming each item's data source earns credit too, since it keeps conclusions inside the population sampled.
PU525 Unit 4 help from the desk
Test worksheets often lose points by choosing a test before identifying the outcome type and group structure, by using an independent test on paired data, or by reporting only P values. Start each item with the question and hypotheses. Identify measurement level and whether groups are paired, then choose the test. Check assumptions. Report results with effect sizes and intervals where possible. Explain what the P value means and does not mean. Point out what the design leaves open, for instance a before-and-after result with nobody untreated to compare. State each data source so conclusions stay within its population. Decide one-sided or two-sided tests before seeing the data. Keep each item to about a page so a grader can follow the reasoning quickly.
Get a PU525 Unit 4 example written to your instructions
Paste your PU525 Unit 4 worksheet items, with any data supplied and the rubric that goes with them. The desk matches each question to its test through measurement level and group structure. That first custom sample costs nothing, and within 24-48 hours it arrives with hypotheses stated about the population and every assumption checked in writing. 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 PU525 papers and MPH samples
- PU525 Unit 1: Skewed Average Drinking Survey Post
- PU525 Unit 2: Birth Data Description Exercise
- PU525 Unit 3: Public Health Probability Problem Set
- PU525 Unit 5: Confidence Interval Exercise
- PU525 Unit 6: Output Interpretation Memo
- PU525 Unit 7: Seminar Reflection
- PU525 Unit 8: Regression Analysis
- PU525 Unit 9: Statistical Analysis Plan
- PU525 Unit 10: Biostatistics Report
- PU520 Unit 4: Case-Control Study Design Critique
- PU508 Unit 4: Postpartum Activity Construct Mapping
- PU501 Unit 4: County Health Department Profile
PU525 Unit 4 questions, answered
Where can I find a free PU525 Unit 4 sample paper?
The complete PU525 Unit 4 hypothesis test worksheet is here, choosing four different tests for a composite county's data.
How do you choose a hypothesis test?
By the outcome's measurement level, the number of groups and whether groups are independent or paired.
Why use a rank-based test for length of stay?
Lengths of stay are strongly skewed, so comparing typical values may be more meaningful than comparing means.
What does a P value not tell you?
Neither the chance that no effect exists nor how large any effect is.
How long is the PU525 Unit 4 worksheet?
Four boxed items of about a page each.