An average conceals a distribution, and this stage opens the average up. Disparity analysis stratifies your population by the characteristics along which outcomes differ, measures the gaps in both absolute and relative terms, and explains them by pointing at conditions rather than at the groups themselves. The technical constraints are small numbers, suppressed cells and the risk of reading a difference into noise; the writing constraint is language that describes structural causes without slipping into characterizing populations. Your section may print this as NR 538 or NR538; 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-538 Week 7 asks for
Split a county's older adult population by where people live and two very different pictures appear. Residents of the incorporated town sit within two miles of a pharmacy, a clinic and a senior centre. Residents in the outlying townships average eleven miles from the nearest of those, and the share of households without an available vehicle in the oldest age band is meaningfully higher there. The county-wide averages you calculated earlier describe neither group. Stratification is what makes those two populations visible, and this stage asks you to do it deliberately: to pick the axes that plausibly matter, to measure the difference properly, and to say what is producing it.
Measurement precision is graded here. A difference can be expressed absolutely, as the arithmetic gap between two rates, or relatively, as a ratio. The two answer different questions and they can point in opposite directions in importance: a relative difference of double sounds dramatic and may represent a small absolute number of events, while a modest ratio applied to a common outcome can represent a large number of affected people. Reporting both, and saying which one bears on the question you are asking, is the mark of a competent analysis.
The vocabulary needs care too. A difference between groups is not automatically a disparity. The term as used in population health carries a sense of a difference that is systematic, avoidable and connected to social or structural disadvantage, which is why the explanation section has to do real work rather than simply presenting the gap. Equity and equality are also distinct: equal distribution of a service across a population and an equitable distribution proportionate to need are different targets, and the difference matters when the resource inventory from the previous stage is read against the gaps found here.
Deliverables here usually involve stratified tables, calculated differences and a written interpretation, often with a section on data limitations because subgroup analysis runs into suppression constantly. Where a discussion runs, it commonly asks which disparity is most significant, and the graded answer distinguishes statistical, practical and equity significance rather than treating them as one thing.
The NR-538 Week 7 method, step by step
Six moves for a disparity analysis that holds up.
-
Declare the stratification axes before running any numbers
Return to the subgroups you named in the first stage, revise them with what you now know, and record the revision. Choosing axes after seeing results invites the appearance of fishing.
-
Check subgroup denominators before interpreting subgroup rates
A rate built on a handful of people in the denominator moves wildly. Report the denominator for every stratum, and decline to interpret the ones too small to support a claim.
-
Report absolute and relative differences together
Give the arithmetic gap in events per unit of population alongside the ratio, then say in one sentence which one answers your question and why.
-
Handle suppressed and missing cells explicitly
Say which strata were unavailable and why, whether privacy suppression or unreported categories. Silence about a missing group reads as an oversight rather than as a data constraint.
-
Explain each gap by pointing at conditions
Connect the difference to the access findings, the environmental observations and the determinant pathways already established. The explanation must be about circumstances, not about the characteristics of the people in the group.
-
Separate the three senses of significance in your conclusion
Whether the difference is unlikely to be chance, whether it is large enough to matter practically, and whether it is unjust in the sense the field means by disparity. Address each explicitly.
Layout and word budget for a disparity analysis
Our frame for a stratified analysis with tables attached, sized for roughly 1,300 to 1,600 words of prose. It is our own outline rather than anything the university issues, and your week's rubric outranks it wherever they disagree.
| Section | What belongs in it | Word target |
|---|---|---|
| Axes and rationale | The characteristics you stratified by, the reason each was expected to matter, and when you chose them. | 170 to 210 |
| Stratified findings | Rates by stratum with denominators, presented so the reader can see the size of each group. | 270 to 330 |
| Difference measures | Absolute and relative gaps for the comparisons that matter, with the question each addresses. | 230 to 290 |
| Data constraints | Suppressed cells, unreported categories, small denominators, and what each prevents you from concluding. | 200 to 250 |
| Explanation | Each gap connected to access, environmental or structural findings from the earlier stages. | 280 to 340 |
| Significance, three ways | Statistical stability, practical magnitude and equity, held apart and answered separately. | 190 to 240 |
Evidence craft for disparity writing
Publish the denominator for every stratum. A rate for a subgroup of 60 and a rate for a subgroup of 6,000 look identical on a chart and mean entirely different things. The denominator is what lets a reader calibrate your claim.
Attribute causes to conditions, never to groups. Write that the outlying townships lack any pharmacy open past six, not that a population is less engaged with follow-up care. The first is a fact about circumstances; the second attributes an outcome to people and is both analytically wrong and heavily penalized.
Use the categories your source used and say so. Race, ethnicity, rurality and income are operationalized differently across datasets. Report the category labels as the source defined them, note when two sources categorize differently, and avoid collapsing groups to make a table tidier.
Say when a difference is too small or too unstable to interpret. Restraint reads as competence in this stage. Naming the comparisons you will not draw a conclusion from strengthens the ones you do draw.
Five mistakes that cost points in this week's territory
- Ratios without absolute differences. A doubled rate on a rare outcome can represent very few people, and reporting only the ratio distorts the picture.
- Explanations that attribute outcomes to culture or behaviour. This is the fastest way to lose the analysis row and the professional-language row simultaneously.
- Subgroups reported without denominators. Unstable rates presented as findings mislead the reader and cannot survive a grader who checks the source.
- Missing groups unmentioned. A table with a category silently absent looks like an omission rather than the suppression it usually is.
- Significance treated as one idea. A gap can be stable, small and still inequitable, and collapsing the three senses loses the distinction the stage is teaching.
Before you submit
- Stratification axes are declared with their rationale and their timing
- Every stratum reports its denominator alongside its rate
- Absolute and relative differences both appear for key comparisons
- Suppressed or missing categories are named with the reason
- Every explanation points at a condition rather than at a group
- Statistical, practical and equity significance are answered separately
Analysing disparities for NR-538?
Send the stratified data and the scoring guide out of Canvas. A premium original draft comes back in 24 to 48 hours with denominators published, both difference measures reported and every explanation aimed at conditions, and revisions run until the grade lands.