NR-518 Week 4 asks where your numbers came from before it asks what they mean. Every rate quoted in the earlier stages of this course was produced by a system that decided who counted as a case, who was obliged to report it, and how long the report took to arrive, and the middle of an epidemiology session is where those decisions become the subject rather than the background. Your section may print this as NR 518 or NR518; it is the same course. Chamberlain publishes no syllabi outside Canvas. The placement here is our teaching judgment from the course's catalog arc; your section's rubric decides what your week actually asks.
What NR-518 Week 4 asks for
The territory is the machinery of population data: notifiable condition reporting, vital records, disease registries, national health surveys, syndromic and event-based systems, and the administrative and clinical records that get reused for surveillance long after they were created for billing or for care. Each of those has a different route from event to database, and the route decides what the data can honestly answer. A rubric row that asks you to appraise a data source is asking about the route, not about the figure sitting at the end of it.
Two ideas do most of the work at this stage. The first is the case definition, which is the written rule the system uses to decide whether an event enters the count at all, usually built from clinical features, laboratory confirmation and a time and place restriction. Change the rule and the count changes with no change whatsoever in the health of the population. The second is the difference between passive collection, where the system waits for reports to arrive, and active collection, where someone goes looking. Passive systems are cheap, wide and incomplete. Active systems are expensive, narrow and much closer to the truth, which is why they are usually reserved for short periods or for conditions that matter enough to fund.
The deliverable at this point in an eight-week session is often a source appraisal or a short data-driven description of a problem in a defined group. Whatever shape it takes, the graded move is the same: name the system, quote its rule, and say who never reaches it. If your section runs a discussion this week, prepare the source details before you open the text box, since posts do not reopen once submitted in Canvas and an unattributed figure is difficult to repair in public.
The NR-518 Week 4 method, step by step
Six moves that turn a downloaded statistic into an appraised piece of evidence your grader can follow back to its origin.
-
Read your week's rubric for what the data has to do
Mark whether each row wants a description of a system, an appraisal of its quality, or a health problem described using data from it. Those are three different papers. A careful appraisal submitted where a description was wanted scores in the middle band for the same reason a description scores badly where appraisal was wanted, and the rows say which one you are in.
-
Write the question before choosing the source
One sentence: what do I need to know, about whom, for what period. Sources chosen first tend to drag the question toward whatever they happen to publish, and the paper ends up describing a national picture when the group under discussion was a county.
-
Find and reproduce the case definition
Go to the system's own documentation and locate the rule for a confirmed case, then paraphrase it in your draft with the year attached. Most students have never read one. A single sentence showing that you did separates an appraisal from a summary of a dashboard.
-
Trace the route from event to record
Say who notices the event, who is required to report it, to whom, and by when. Then say how long the whole path takes. Timeliness is a property of the route rather than of the disease, and it explains why last month's figure is provisional and why it will move.
-
Name who never enters the system
Every source has a shadow: people who never sought care, who were tested where results are not reported, who lack the insurance that generates a claim, or who fall outside the sampling frame of a survey. Write who is missing and which direction their absence pushes the estimate. This is the paragraph that separates the top band from the middle.
-
State what the source can and cannot support
Close the appraisal with two short lists in prose: the claims this data can carry, and the claims that would need a different source. A reader should finish knowing exactly how far your later argument is allowed to travel.
A layout and word budget for a data source appraisal
Sized for a piece of roughly 1,000 to 1,250 words. This is our drafting frame rather than a university-issued form, and your week's rubric outranks it wherever the two disagree. Scale each target proportionally if your assigned length differs.
| Section | What belongs in it | Word target |
|---|---|---|
| The question and the group | What you need to know, about whom, over what period, in one paragraph with no preamble ahead of it. | 70 to 90 |
| The system itself | Who runs it, why it exists, what it collects and at what geographic level it reports. | 150 to 180 |
| The case definition | The rule for inclusion, with its year, and one line on what a change to that rule would do to the count. | 150 to 180 |
| Collection route and timeliness | Passive or active, who reports and by when, the usual lag, and whether recent figures are provisional. | 170 to 210 |
| Coverage and undercount | Who reaches the system, who does not, and the direction each gap bends the estimate. | 200 to 240 |
| Fit for your question | The claims this source supports, the claims it cannot, and the source you would add to close the gap. | 170 to 200 |
| Close | The one sentence a reader should carry away about trusting this number. | 50 to 70 |
Evidence craft for surveillance writing
Cite the agency that produced the data, not the article that mentioned it. Surveillance systems publish their own tables, technical notes and definitions. Going to the producer gives you the base, the collection period and the caveats in one place, and those three details are what turn a quoted figure into evidence a grader can score.
Separate the year the data describes from the year it was published. These are almost never the same, and papers that report only the publication year make recent data look stale and old data look current. Write both into the sentence and the ambiguity disappears.
Say when a count is provisional. Recent periods in most reporting systems are incomplete by design, because reports are still arriving. Treating a provisional figure as final produces a fake downturn at the right hand edge of every trend, and it is one of the easiest errors for a grader to spot.
Never build a trend line out of two different systems. If the first half of your series comes from one source and the second half from another, the change you are describing may be a change in counting. Where you have no alternative, mark the join in the text and argue only the parts that sit on one side of it.
Five mistakes that cost points in this week's territory
- A figure with no case definition behind it. The number is meaningless until the reader knows what was counted as a case, and this is the week that scores exactly that.
- Comparing years across a definition change. When the rule was revised, the two counts are not the same measurement, and a comparison that ignores the revision reports an artifact as a finding.
- More reports read as more disease. Better testing, a new requirement or an outbreak in the news all raise counts on their own. Rule out detection before claiming incidence moved.
- Billing records treated as clinical records. Administrative data records what was coded for payment, which is close to but not the same as what happened, and the gap deserves a sentence.
- Silence about who is missing. An appraisal that lists strengths and calls the limitations minor has skipped the analytic work the stage is built on.
Before you submit
- The system is named along with the body that runs it and the level at which it reports
- The case definition appears in the draft with the year it applied
- Collection is identified as passive or active, with the usual reporting lag stated
- At least one named group is described as missing from the data, with the direction of the bias
- Every figure carries its base, its period and its producing agency
- Every reference appears in the text and every in-text citation appears in the list
Appraising a data source this week?
Send the prompt and its scoring guide from Canvas. A premium original draft returns in 24 to 48 hours with the case definition, the route and the undercount written out, and revision runs free until the grade lands.