| Course | HD520 Health Education Program Assessment |
|---|---|
| Unit | Unit 3 |
| Paper type | Sampling plan |
| Length | About 860 words, 6 pages double-spaced |
| Format | APA 7 student paper |
| School | Purdue University Global |
| Program | MS in Health Education |
| Updated | October 2026 |
Free sample paper for HD520 Unit 3
Everyone on the Roster, and the People the Roster Misses: A Unit 3 Sampling Plan for a Refugee Health Orientation Assessment
[Student Name]
Purdue University Global
HD520: Health Education Program Assessment
Unit 3 Sampling Plan
[Instructor Name]
[Date]
The agency, the roster and all counts are composite; the texts on nonresponse, sample size and clustering cited are real.
Who the Funder Means and Who the Roster Holds
The funder's interest covers every adult, aged 18 or older, resettled in Calder City during the assessment year. The sampling frame is narrower: the agency's arrival roster, which lists adults placed by the agency itself. Adults who reach Calder City through another agency, who move here from another state after their first placement, or who arrive under statuses the agency does not serve are absent from the roster. In the last year, caseworkers estimate that about [40] such adults sought help from the agency after arriving on their own. Any gap between the population of interest and the list used to reach it is a source of coverage error (Lohr, 2021). Their absence matters because secondary migrants often move to join relatives, who may already know how to use local clinics.
Selection
All adults on the roster are invited at their first caseworker meeting. Sampling within the roster was considered and rejected. The roster is expected to hold about [268] adults, and drawing a sample from it would save little assessor time while leaving the smallest language groups with too few people to describe at all. Results will be reported by the three main languages, Dari, Arabic and Swahili, and by a combined group of smaller languages served through telephone interpreters.
A separate purposive subsample of [18] adults will be drawn for interviews in Unit 5, chosen from task and appointment results rather than by educators. Adults who join the roster late in the year are included until the final class of the assessment year begins.
Sample Size
The sample size is set by the roster, so the plan asks what that roster can detect rather than how many people would be ideal. The agency's director said a gain of less than about 20 percentage points in task completion would not justify the class's staffing over caseworker coaching. Assuming baseline completion near 30 percent, about [146] early starters and [122] adults waiting for a seat, a loss of 25 percent by day 42, and a design effect of about 1.3 because early starters are taught in classes of up to twelve (Donner & Klar, 2004), the census should detect a true 20-point difference about four times in five.
A smaller true effect, such as 10 points, could go undetected, and the report will say so plainly rather than describe a nonsignificant result as evidence that the class does nothing.
Exclusions and Expected Losses
Each group below is listed with its expected size, how it may differ from those who remain and the likely direction of bias.
Early movers. Families who leave Calder City within six weeks, often to join relatives, are expected to number about [25] adults. Those joining established relatives may already have help using clinics, so their loss could make remaining adults look less capable at baseline. Effect on the comparison is uncertain, because movers may leave either group.
Night-shift workers. Adults who start overnight work in their first month, mainly at two meat-processing plants, often cannot attend the day-42 task. Expected loss is about [15] adults. They are likely younger and more confident on the phone, so their loss may lower task scores in both groups.
Phone-interpreted languages. About [30] adults speak languages, such as Tigrinya or Kinyarwanda, for which the agency has no in-person interpreter. They can take the task through the telephone line, but class sessions reach them less well. Their results will be reported separately, and their small number means no group-level conclusion will be drawn for them.
Adults who decline. Some adults will decline any assessment, especially those wary of forms connected to their legal status. Expected refusal is about [10] percent. Refusers may be more distrustful of institutions, which could relate to how they use clinics.
Lost to follow-up for other reasons. Illness, childcare and changed phone numbers will account for the rest of the expected [25] percent loss.
Nonresponse Procedures
Assessors, not caseworkers, will contact adults who miss the day-42 task. Caseworkers control housing and benefits, and a request from them would be hard for a newly arrived adult to refuse freely. Each missed appointment will receive up to three contact attempts on different days and times, including one evening attempt for shift workers. Reasons for loss will be recorded in a short log, because knowing why people were lost is as important as knowing how many (Groves, 2006).
Reporting the Sample
The final report will include a flow diagram from roster to analysis showing the numbers invited, enrolled, assessed at baseline and assessed on day 42 in each group, with reasons for each loss. Losses that differ in size between the two cohorts will be flagged, since unequal losses threaten the comparison more than equal ones.
References
Donner, A., & Klar, N. (2004). Pitfalls of and controversies in cluster randomization trials. American Journal of Public Health, 94(3), 416-422. https://doi.org/10.2105/AJPH.94.3.416
Groves, R. M. (2006). Nonresponse rates and nonresponse bias in household surveys. Public Opinion Quarterly, 70(5), 646-675. https://doi.org/10.1093/poq/nfl033
Lohr, S. L. (2021). Sampling: Design and analysis (3rd ed.). CRC Press.
HD520 Unit 3 instructions, in plain terms
Unit 3 usually asks for the sampling section of an assessment: who the target population is, what list or source will be used to reach them, how participants will be selected and how many are needed. Many prompts also expect a discussion of who will be missed and how nonresponse will be handled. Define the target and the frame separately and describe the gap between them. Justify sample size from the effect that matters for the decision, or, if the population is fixed, state what that population can detect. Consider clustering when participants are grouped. Give each excluded or lost group an estimated size and a sentence on likely bias. Three to four pages is typical.
How the HD 520 Unit 3 example is put together
The plan opens by setting the funder's population, every adult resettled in the city that year, beside the roster the agency can actually reach, and names those missing from it. Selection is a census, with results reported by Dari, Arabic, Swahili and a combined phone-interpreted group. The sample size section works backward from the fixed roster and an allowance for clustering in classes of twelve. The highlighted sentence commits the report to stating plainly that a smaller effect could be missed. Five groups of expected losses follow, each with size, difference and bias direction. Nonresponse is assigned to assessors rather than caseworkers, with three contact attempts including one in the evening.
Where the points sit in the HD520 Unit 3 rubric
Sampling plans tend to be graded on clear definitions of target and frame, an appropriate selection method, a justified sample size, a thorough treatment of exclusions and nonresponse, and accurate use of sources. Definitions score well when the gap between target and frame is named. Selection earns credit when the method fits the frame's size and the reporting needs. Sample size is stronger when tied to a decision-relevant effect and when its limits are admitted. Treatment of exclusions is often where plans gain or lose the most, since vague lines such as "some people may not respond" offer nothing to check. Clear organization and APA format carry the remaining weight.
HD520 Unit 3 help from the desk
The usual weakness is a sample size taken from a rule of thumb with no link to the decision. Another is listing exclusions without estimating their size or saying which way they push the results. Some students forget clustering when participants are taught or treated in groups, which makes later intervals too narrow. Others assign follow-up to staff whose role makes refusal difficult. Treating the frame as the population is a quieter but common error. If your assessment uses a fixed list, such as a clinic panel or a class roster, share its size and makeup, and our team can estimate the smallest effect it could reveal.
Get an HD520 Unit 3 example written to your instructions
Describe who your Unit 3 plan is meant to cover, the sampling frame realistically available and any size requirement, along with the rubric. Selection, size, exclusions and nonresponse procedures are set out, with each missing group named alongside its likely effect on results. First custom sample: free, and back within 24-48h. 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 HD520 papers and MS in Health Education samples
- HD520 Unit 1: Discussion Board Post
- HD520 Unit 2: Design Rationale Paper
- HD520 Unit 4: Instrument Appraisal
- HD520 Unit 5: Mixed Methods Protocol
- HD520 Unit 6: Seminar Reflection
- HD520 Unit 7: Participation and Governance Plan
- HD520 Unit 8: Data Analysis Section
- HD520 Unit 9: Limitations Analysis
- HD520 Unit 10: Assessment Protocol
HD520 Unit 3 questions, answered
If everyone is invited, is it still a sampling plan?
Yes. A census of a frame still requires defining the target, examining what the frame misses and planning for nonresponse, which is most of what a sampling plan does. The example explains why inviting everyone on the roster suits a population of this size, and states that coverage and nonresponse, not sampling error, are the main threats.
Why report results separately by language?
Because the groups may experience the class and the health system differently, and a combined figure could hide a group the class serves poorly. Separate reporting shows each group's result with its own uncertainty. The plan warns that small groups will have wide intervals and commits to reporting them anyway rather than folding them into a total.
Where can I find a free HD520 Unit 3 sample paper?
You will find the complete Unit 3 sampling plan here, built on a resettlement agency's arrival roster and the groups it leaves out.
Why should caseworkers not follow up nonresponders?
Caseworkers control housing and benefits, so a request from them is hard to refuse freely. Using assessment staff keeps participation voluntary.
What is a design effect?
It is the factor by which clustering inflates the variance of an estimate. When people are taught in classes, their results are more alike, and the sample behaves as if it were smaller.