From raw counts to a stratified comparison, the HS345 Unit 10 report example tests whether text reminders cut missed clinic visits and settles on about four per 100. Searches like "hs 345 unit 10 assignment example", "hs345 unit 10 sample" and "hs345 unit 10 example" land here.
What a finished HS345 Unit 10 statistical analysis report looks like
The report runs about eight pages under six headings. A half-page summary opens it in plain language: reminders appear to prevent about four missed visits in every 100 appointments, not the seven a first look suggests. The data section describes 2,400 scheduled visits over six months, 1,560 with a reminder sent and 840 without, and states how reminders were assigned, by whether a mobile number and consent were on file. A two-way table gives no-show percentages of 14.0 and 21.0, a chi-square of 19.4 and a risk difference of 7.0 points with a 95 percent interval of 3.7 to 10.2. A second table splits new and established patients. The adjusted difference, 3.8 points with an interval of 0.5 to 7.0, carries the recommendation. Limitations and a proposed staggered rollout close the report.
How a HS345 Unit 10 example is structured
The report is built around one question and never adds a second. Description comes first, and it is candid about the design: reminders went to patients whose numbers were on file, which already made the two groups different. Testing follows, and the crude result is reported in full before being questioned, so the reader sees what a quick analysis would conclude. Stratifying is where the argument turns. New patients miss more visits and were far less likely to have reminders set up, making up 17 percent of reminded appointments but 40 percent of the rest, so part of the crude gap reflects who received reminders. Within each patient type the difference shrinks, and the pooled adjusted estimate is about half the crude one. From the adjusted figure comes the recommendation, and the limitations section names what the data cannot rule out.
Summary for the clinic director
Four sentences, no test statistics. The estimate is stated as missed visits per 100 appointments, with a note that the true benefit could be as small as half a visit.
How reminders were assigned
Not at random. A reminder required a mobile number and texting consent in the registration record, and the report treats that as the study's central weakness.
The crude comparison
Fourteen against 21 missed visits per 100, chi-square 19.4, p below 0.001, and a risk difference with its interval. Reported first, questioned second.
New and established patients apart
New patients: 23.8 against 30.0 percent. Established: 12.0 against 14.8. Neither stratum alone reaches significance, and both point the same way.
An adjusted estimate
A Mantel-Haenszel risk difference of 3.8 points, interval 0.5 to 7.0. The report explains the pooling in two sentences and shows the weights.
Recommendation and a fair test
Collect mobile numbers at scheduling so new patients receive reminders, and start clinics in staggered months so the next report can compare like with like.
Where marks go in HS345 Unit 10
Final analysis reports in HS345 generally carry rubric rows for the research question, data description, appropriate methods, correct results, interpretation, limitations and professional presentation. Methods credit depends on matching a test to two categorical variables and on recognizing that the groups formed themselves. This report stands apart on interpretation, declining to present the crude seven-point difference as the effect once the patient-type split shows it inflated. Results credit needs every figure to agree across the summary, the tables and the text. Limitations credit goes to naming the specific bias, that patients with numbers on file differ from those without, and its likely direction. Common deductions include reporting only p values, recommending policy from a crude comparison, generic limitations lists, and summaries full of terms a clinic director would skip.
Get a HS345 Unit 10 example written to your instructions
Final HS345 Unit 10 projects usually rest on a dataset chosen in an earlier unit, and yours is where a custom sample begins. Provide the data or output, the research question, the project directions and the rubric. The first sample is free, turned around in 24-48 hours, and follows that question from the first summary table to a recommendation the data can bear.
HS345 Unit 10 questions, answered
Why does the adjusted estimate differ so much from the crude one?
Because patient type is linked to both variables. New patients miss more appointments and were less likely to receive reminders, so comparing all reminded visits with all unreminded ones mixes the reminder's effect with the difference between patient types. Comparing within each type and then pooling removes that particular distortion, though other differences between the groups may remain.
Is a Mantel-Haenszel estimate expected in an introductory course?
Not always. Many sections would accept the stratified table with the two within-group differences described in words. The example includes the pooled estimate because it gives the recommendation a single number with an interval. A custom sample can leave it out, or replace it with whatever adjustment method your course has covered, if the rubric does not call for it.
Does the report prove that reminders work?
No, and it says so. The data are observational, and patients who give a mobile number and consent to texts may differ in other ways, such as stable housing or phone access, that also affect attendance. The recommendation therefore pairs expanding reminders with a staggered rollout, which would allow a fairer comparison in the next report.