| Course | PHLT 8331 Fundamentals of Biostatistics |
|---|---|
| Week | Week 10 |
| Paper type | Effect size brief |
| Length | About 720 words, 5 pages |
| Format | APA 7 student paper |
| School | Walden University |
| Program | PhD in Public Health |
| Updated | October 2026 |
Free sample paper for PHLT 8331 Week 10
Significance or Size? An Effect Size Brief on Three Results From a Composite Anticoagulation Registry and Which Number Should Drive Each Clinic Decision
Student Name
Walden University
PHLT 8331: Fundamentals of Biostatistics
Instructor Name
Month Day, Year
Significance or Size? An Effect Size Brief on Three Results From a Composite Anticoagulation Registry and Which Number Should Drive Each Clinic Decision
Why This Brief
A large registry can make almost any difference statistically significant, because the standard error shrinks as the sample grows. A p-value signals to the clinic network that a difference would be surprising if the null were true; whether that difference is large enough to act on is a separate question (Wasserstein & Lazar, 2016). This brief gives three results from the composite registry an effect size, both standardized and in original units, and argues for each which number should drive the decision. All figures are composite.
Result 1: Improvement After Moving to Pharmacist Management
The mean within-patient improvement in time in therapeutic range was 7.7 percentage points, p < .001. For paired data the standardized effect scales the average change by the spread of individual changes: 7.7 over 18.6, or 0.41. By the widely used benchmarks of Cohen (1988), that is a small-to-medium effect. In original units, it is about 14 more days in range over six months.
Which Number Matters
Here both point the same way, but the original units carry the decision. A clinician deciding whether to refer patients to the pharmacist service cares about days in range, which relate to the risk of bleeding and clots, not about standard deviations. The significance result adds confidence that the change is not noise; the size, in days, says it is worth having, subject to the caution about regression to the mean.
Result 2: Pharmacist Versus Physician Clinics
Mean time in range was 5.3 points higher at the pharmacist-managed clinics, p < .001. Divided by the pooled standard deviation of 16.5, the standardized difference was 0.32, a small effect on the same benchmarks. In original units, it is about 10 more days in range over six months.
Which Number Matters
The size matters more than the p-value, but a third consideration matters most: the comparison is between clinics that differ in more than management. A large, significant, clinically meaningful difference can still reflect confounding. The effect size tells the network the difference is worth investigating; it does not tell them the pharmacist service caused it.
Result 3: Number of INR Tests
Patients at the pharmacist clinics had a mean of 11.8 INR tests in six months, compared with 10.9 at the physician clinic, a difference of 0.9 tests, p = .03. With a standard deviation near 5.8 tests, the standardized difference was about 0.16, below even the conventional 0.2 mark for a small effect.
Which Number Matters
Here significance and size diverge, and size should win. A difference of less than one test over six months has no practical consequence for staffing, cost or patient burden. The result is statistically significant only because the groups are large. Reporting it as a significant difference in testing would mislead the network into thinking the services operate differently when, for practical purposes, they do not.
Summary of the Three Results
Table 1
Significance and Magnitude for Three Registry Results (Composite Figures)
| Result | Difference, original units | Standardized effect | p | Which drives the decision | |
|---|---|---|---|---|
| Improvement after move | 7.7 points (about 14 days) | 0.41 | < .001 | Size, in days in range | |
| Pharmacist vs physician clinics | 5.3 points (about 10 days) | 0.32 | < .001 | Size, with confounding caveat | |
| Number of INR tests | 0.9 tests | 0.16 | .03 | Size: too small to matter | |
Note. Composite figures. Standardized effects are mean differences divided by the relevant standard deviation.
Should Standardized Effects Always Be Reported?
Standardized effects help compare results measured on different scales and are needed for meta-analysis and power calculations. Sullivan and Feinn (2012) recommend reporting effect sizes alongside p-values in all quantitative research. But for a clinic audience, original units, days in range and number of tests, communicate more directly. The brief reports both and leads with original units.
What Counts as Meaningful
The registry's clinical leads consider a change of 5 or more percentage points in time in range, about 9 days over six months, worth acting on. Results 1 and 2 meet that standard; Result 3 has no comparable threshold but is plainly trivial. Setting such thresholds in advance, with clinicians, keeps the decision from being driven by whichever results happen to be significant.
References
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum Associates.
Sullivan, G. M., & Feinn, R. (2012). Using effect size, or why the P value is not enough. Journal of Graduate Medical Education, 4(3), 279-282. https://doi.org/10.4300/JGME-D-12-00156.1
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 PHLT 8331 Week 10 instructions
Week 10 commonly asks students to calculate effect sizes for results from their dataset and discuss whether statistical significance or magnitude should guide decisions. The expectation is accurate calculation of standardized effects and clear presentation in original units, with an argument for each result about which number matters. Students should explain why large samples can make trivial differences significant. Thresholds for meaningful effects should be discussed, ideally in terms the decision makers would use. Briefs usually run two to three pages with a summary table, citing sources on effect sizes and p-values. A summary table pairing each result's p-value with its effect size and original-unit difference helps readers compare.
How this PHLT 8331 Week 10 example is built
The brief opens by showing how a large dataset can make a trivial gap significant, drawing on the ASA's p-value statement. Three sections each present a result with its standardized effect and original-unit difference and argue which should drive the decision, citing conventional benchmarks for standardized effects. The first and second results both matter in size, the second with a caveat about clinic differences; the third is significant but trivial. Table 1 summarizes the three. A section discusses whether standardized effects should always be reported, citing guidance on reporting effect sizes. A closing section describes thresholds set with clinicians.
Where the marks sit in the PHLT 8331 Week 10 rubric
Credit goes first to correct calculation and clear presentation of effect sizes in both standardized and original units. The sample's case-by-case argument, showing size outweighing significance in the third result, demonstrates the judgment the assignment targets. Explaining that large samples make small differences significant shows conceptual understanding. Keeping the confounding caveat attached to a meaningful effect shows that size alone does not settle causation. Discussing thresholds with decision makers shows practical maturity. A tidy table and sound referencing earn the final points. Presenting all three results in one table, with the deciding number named for each, makes the argument easy to follow and verify. Instructors also reward recognizing that a standardized effect cannot settle whether a difference was caused by the factor of interest.
PHLT 8331 Week 10 help: mistakes that cost marks
Effect size briefs often report standardized effects and conventional labels without asking what the numbers mean for the decision. Lead with original units the audience uses. Recognize that a significant result in a large dataset may be too small to matter, and say so. Conversely, a meaningful effect may still reflect confounding. Use conventional benchmarks with caution; what counts as meaningful depends on context. Set thresholds with the people who will act on the results, before seeing the data if possible. We will write a free effect size brief for results from your own study. Present your results side by side so readers can see where significance and size agree and where they diverge. Name, for each, the number that should drive the decision.
Get a PHLT 8331 Week 10 example written to your instructions
Attach the Week 10 prompt and rubric along with the results your section wants weighed. A brief pairing each result with its effect in natural units and giving a case-by-case verdict comes back in 24-48h, with the first free of charge. Every effect size it reports is invented to show the reasoning, not to be quoted.
More PHLT 8331 papers and PhD in Public Health samples
- PHLT 8331 Week 1: Population and Sample Concept Post
- PHLT 8331 Week 2: Variable Distribution Brief
- PHLT 8331 Week 3: Hypothesis Statement Paper
- PHLT 8331 Week 4: Paired t-Test Write-Up
- PHLT 8331 Week 5: Two-Group Mean Comparison
- PHLT 8331 Week 6: Test Condition Peer Replies
- PHLT 8331 Week 7: Chi-Square and McNemar Tests
- PHLT 8331 Week 8: ANOVA With Trend Contrast
- PHLT 8331 Week 9: Confidence Interval Interpretation
- PHLT 8331 Week 11: Plain-Language Findings Report
- PHLT 8270 Week 10: Outbreak Response Action Brief
- PHLT 8078 Week 10: Hotel Ergonomics Rule Memo
- PHLT 8076 Week 10: Misleading Health Map Analysis
- PHLT 8074 Week 10: Exposure Assessment Plan Paper
PHLT 8331 Week 10 questions, answered
Where can I find a free PHLT 8331 Week 10 sample paper?
The complete Week 10 effect size brief appears on this page, weighing significance against magnitude for three results from a composite anticoagulation registry.
Why can a significant result in a large registry be unimportant?
Because the standard error shrinks as the sample grows, so even tiny differences can reach significance. The sample's 0.9-test difference in INR testing is significant at p = .03 but too small to matter for staffing or patients.
Should standardized effect sizes always be reported?
They help compare results across scales and are needed for meta-analysis and power calculations, so reporting them is good practice. For a clinic audience, original units such as days in range communicate more directly and should lead.
What counts as a meaningful change in time in therapeutic range?
That depends on the clinical context. The sample's clinical leads consider 5 or more percentage points, about 9 days over six months, worth acting on, a threshold set with clinicians rather than borrowed from general benchmarks.
What are Cohen's benchmarks for effect sizes?
Cohen suggested standardized mean differences of about 0.2, 0.5 and 0.8 as small, medium and large. They are rough guides and should be interpreted in light of the field and the decision at hand.