NR-586AT · Week 3 of 8 · Comparison, benchmarking and making a number mean something

NR-586AT Week 3 Comparison and Benchmarking: How to Write It

The short answer

Comparison is the stage where an indicator set becomes an argument. A local figure alone is a fact; a local figure beside a benchmark measured the same way is a finding, and the size and direction of the gap is what a leadership audience will remember. The work is partly arithmetic and mostly discipline: matching definitions across both sides of a comparison, refusing crude comparisons where age structure distorts them, and stating a gap in people rather than percentage points wherever you can. Your section may print this as NR 586AT or NR586AT; 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.

NR 586AT Week 3 grading scale at Chamberlain, the criterion levels this assessment is scored on, from Chamberlain Tutors
How Chamberlain grades NR 586AT Week 3, visualized by Chamberlain Tutors.

What NR-586AT Week 3 asks for

Why does a benchmark sometimes make a good service look bad? A behavioral health service line inside a system of four hospitals reports a thirty-day readmission figure that sits well above the peer group average, and the finance committee wants to know what went wrong. Nothing did. Two of the peer facilities do not operate an acute psychiatric unit at all, one transfers its most complex presentations out of the region, and the fourth counts readmissions only where the second admission was to the same facility. Four denominators wearing one label. The leadership skill this stage teaches is the ability to interrogate a comparison before accepting what it appears to say, and to write that interrogation in a paragraph a committee can follow without a statistics background.

A defensible comparison rests on four checks. Is the case definition the same on both sides? Is the denominator built from the same population at risk? Do the two figures cover the same period, or is a 2023 local value being set against a 2019 benchmark? And is the comparison population similar enough in age and risk profile that a crude difference means something? Where any of those fails, you have two honest options: adjust if you can, or state the mismatch and its direction and let the reader discount accordingly. What you cannot do is present the raw gap as though the checks had passed.

Choosing the benchmark is itself an argument. A state figure tells you whether your population is unusual for where it sits. A national figure tells you whether it is unusual at all. A peer figure from similar organizations tells you what is achievable with similar resources, which is usually the most persuasive comparison for a leadership audience because it removes the excuse that circumstances are unique. A trend against your own past performance answers a different question entirely, which is whether you are improving. Strong papers use two of these deliberately rather than one by default.

Deliverables at this depth in an accelerated session run roughly 1,000 to 1,400 words, often with a comparison table permitted outside the count. The interpretation is what is scored. A gap reported without a sentence on what it means in patients, appointments or admissions has done the arithmetic and skipped the analysis, and in a leadership course the analysis is the deliverable.

The NR-586AT Week 3 method, step by step

Six moves for a comparison that will survive a committee.

  1. Choose the benchmark type on purpose and say why

    State, national, peer or your own trend. Name the question each one answers, then pick the one that matches the claim you intend to make.

  2. Verify the definition on both sides before comparing

    Read how the benchmark source defines its numerator and denominator. Definitional mismatch is the single most common cause of a comparison that misleads.

  3. Align the periods or state the offset

    Same year where possible. Where not possible, name both years and say what changed in between that might account for part of the difference.

  4. Decide whether the crude comparison is fair

    If your population is markedly older, younger or higher acuity, a crude gap is partly demographics. Say so and state which way the distortion runs even where you cannot adjust.

  5. Express the gap in people, not only in points

    Four percentage points over a population of 3,200 is roughly 128 people. That number is what a director will act on, and the percentage alone is what they will forget.

  6. Write one sentence of consequence per gap

    What the difference costs in admissions, appointments, staffing or outcomes. A gap without a consequence attached does not survive to the prioritization stage.

A layout and word budget for a benchmarking analysis

Our frame for this stage, sized for roughly 1,000 to 1,400 words with the comparison table outside the count where that is permitted. It is our own outline rather than anything the university issues, and your week's rubric outranks it wherever they disagree.

SectionWhat belongs in itWord target
Benchmark chosen and justifiedWhich reference you selected, the question it answers, and why it fits the claim you intend to make.140 to 180
Definitional checkHow each side defines the measure, where the definitions differ, and what the difference does to the gap.200 to 250
The comparison walked throughEach indicator with both values, both years, and the gap stated in points and in people.280 to 340
Fairness of the comparisonAge, acuity and payer differences between the populations, with the direction of any distortion named.200 to 250
What the gaps costEach meaningful difference translated into service consequences a decision-maker recognizes.180 to 220
The two gaps that will carry forwardWhich differences are large, real and actionable enough to enter the prioritization stage.110 to 140

Evidence craft for benchmarking

Quote the benchmark's own definition, briefly. One clause naming how the reference source counts is worth more than a paragraph of interpretation built on an assumption about how it counts.

Never compare a crude local rate with an adjusted published one. This is the arithmetic error most likely to survive into a final document because both numbers look like the same kind of object. Check which the source published before you set them side by side.

Report the gap with both values visible. Writing that the local rate is nearly double the state figure without giving both numbers forces the reader to trust you. Give both, and the sentence becomes checkable.

Be careful with peer group composition. Peer sets are constructed by somebody using criteria, and those criteria decide the answer. Say which peer definition you used and what kind of organization it includes.

Flag small numerators on either side. A benchmark built on a handful of events is as unstable as a local figure built the same way, and instability is not visible once a rate has been calculated.

Keep internal comparisons de-identified and aggregate. Comparing units, clinics or teams within your organization can be useful and can also identify individuals in a small setting. Aggregate to a level where no person is recognizable, and say that you have done so.

Five mistakes that cost points in this week's territory

  • A benchmark adopted because it was the first one found. The reference is an argument and should be chosen as one.
  • Definitions never checked. Two measures with the same name and different denominators produce a gap that is arithmetic rather than health.
  • Years silently mismatched. A recent local figure against an older benchmark flatters or damns the service for no defensible reason.
  • Gaps left as percentages. Leadership audiences act on counts of people, and the translation is one line of arithmetic.
  • Every gap treated as equally important. A comparison section that does not rank ends up handing the prioritization stage a flat list.

Before you submit

  • The benchmark type is named and justified rather than assumed
  • Both sides of every comparison carry a definition, a year and a denominator
  • No crude figure is set against an adjusted one without a stated caution
  • Each gap appears in percentage points and in people
  • Population differences that distort the comparison are named with their direction
  • The section closes by identifying which gaps carry forward and why

Working the comparison section?

Send the rubric and your indicator set out of Canvas. A premium original draft comes back in 24 to 48 hours with benchmarks selected on purpose, definitions checked on both sides, and every gap translated into people and consequences, with revisions until the rows read clean.

Questions students ask about this stage

What if my population looks better than the benchmark on everything?
Then either you have chosen the wrong comparison or you have chosen the wrong indicators, and both are worth writing about honestly. The commonest cause is a benchmark that is too broad: a national figure will make a well-resourced suburban catchment look excellent on almost any measure, which tells a decision-maker nothing they can use. Switch to a peer comparison, to a higher-performing reference, or to internal variation between sites or subgroups within your own population, where differences almost always exist. Internal variation is frequently the most useful analysis available to a leader, because the resources and policies are held constant and the difference must therefore come from something you can change. A paper that reports uniformly good performance against a soft benchmark reads as though it avoided the question; one that finds the variation inside a good average reads as leadership.
How do I compare when my organization defines a measure differently from the benchmark?
Say so, quantify the direction if you can, and then decide whether the comparison is worth making at all. Sometimes the difference is small and you can note it in a clause. Sometimes it is decisive, as when a local measure counts only returns to the same facility while the benchmark counts returns anywhere in a region, which will make local performance look systematically better. In that case the responsible move is to state that the two are not comparable, and either rebuild the local figure using the benchmark's definition if the data allows it, or switch to a different indicator where definitions align. Writing that a comparison cannot be made and explaining why is a legitimate finding, and it is far stronger than presenting a gap you privately know to be an artifact of counting rules.
Should I compare against my own past performance instead?
Use it as a second comparison rather than the only one, because a trend answers a different question from a benchmark. Your own history tells you whether you are improving, which is what a quality committee wants to know, and it controls for almost everything about your setting since the population and the measurement system are largely the same. What it cannot tell you is whether the current level is acceptable, since a service can improve steadily and still sit well below what comparable organizations achieve. The strongest structure at this stage pairs the two: here is where we stand against peers, and here is the direction we have been moving. Where the two disagree, that tension is the interesting part of the paper, and it is usually where a prioritization argument finds its footing.

Keep going

Online now