The closing stage of NR-716 asks how you will know whether the translated practice worked, and it asks in writing before anything has been implemented. An evaluation plan names the measures, defines each one operationally, identifies where the data will come from, specifies the comparison, states how the data will be displayed and analyzed, and says in advance what result would count as improvement. It also states plainly what a site-level evaluation cannot establish. That last sentence is not a weakness; it is the difference between an evaluation and an overclaim. Your section may print this as NR 716 or NR716; 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-716 Week 8 asks for
A pediatric practice implementing a follow-up standard for children newly started on medication for attention-deficit disorder wants to know whether the change worked. The question sounds simple and it hides four separate ones. Did the process change, meaning are follow-up contacts now scheduled and completed within the intended window. Did anything reach the family, meaning did the contact actually happen for the children it was meant for. Did something get worse, meaning are visit slots now displacing other children or is staff time being pulled from somewhere. And is any observed movement plausibly attributable to the change rather than to the season, the staffing, or the fact that everyone knew they were being watched.
The territory here is measurement built to answer those questions honestly. Process measures tell you whether the change is happening. Outcome measures tell you whether the thing you care about moved. Balancing measures tell you what the change cost somewhere else, and they are the measures students omit most often and the ones experienced reviewers ask about first. Each measure needs an operational definition precise enough that two people counting independently would produce the same number, a named data source, a collection frequency and an owner.
Comparison is the second half. A single before and after pair invites every rival explanation there is. Data plotted over time, with several points before the change and several after, lets you see whether the level was already moving, and it is the standard display for improvement work for exactly that reason. Where possible, adding a comparison group or a comparison site strengthens the inference without turning the project into a trial.
Expect a written evaluation plan, often with a measures table and sometimes a data collection schedule, plus a final posted reflection in many sections. Boards do not reopen after submission in Canvas.
The NR-716 Week 8 method, step by step
Six moves for writing an evaluation plan that could actually be run.
-
Write one measure per question you actually have
One process measure for whether the change is happening, one outcome measure for the thing you care about, one balancing measure for what it might cost. Three well-defined measures beat nine vague ones and are far more likely to be collected to the end.
-
Define each measure operationally, numerator and denominator
Write who counts in the denominator, what event puts someone in the numerator, and what window applies. If a colleague could count differently from your definition, the definition is not finished.
-
Establish a baseline over several points, not one
Pull the same measure for several periods before the change. A single pre-period cannot show whether the level was already drifting, and a project that improves against an unusually poor month has learned nothing.
-
Choose a display before you choose a statistic
Plot the measure over time with the change marked on the chart. For most site-level improvement work the picture carries the argument, and a chart with enough points is more persuasive and more honest than a significance test on two numbers.
-
Name the rival explanations in advance
Seasonality, staffing changes, a concurrent initiative, altered documentation behavior, and the attention created by the project itself. Listing them before data arrive is what lets you address them credibly afterward.
-
State what would count as success and what you will not claim
Write the threshold you would accept as meaningful improvement and the reasons behind it, then write the sentence saying this design shows change over time in one setting and does not establish causation.
A layout and word budget for an evaluation plan
Our frame for a closing evaluation plan, sized for roughly 1,400 to 1,900 words plus a measures table. It is our own outline rather than anything the university issues, and your week's rubric outranks it wherever the two disagree.
| Section | What belongs in it | Word target |
|---|---|---|
| Evaluation question | What you need to know, written as a question the plan can answer, before any measure is named. | 100 to 140 |
| Measures table | Process, outcome and balancing measures with numerator, denominator, source, frequency and owner. | Table |
| Operational definitions | Each measure written out in prose precisely enough that two people would count identically. | 250 to 300 |
| Baseline and comparison | How many pre-periods, from what source, and what the comparison will be after the change. | 220 to 270 |
| Display and analysis | The chart you will use, how many points before and after, and any simple analysis, with the reasoning. | 250 to 300 |
| Rival explanations | The alternatives named in advance and how each will be checked or acknowledged. | 240 to 290 |
| Data handling and limits | Who collects, where data are stored, how they are de-identified, and what the design cannot establish. | 240 to 290 |
Evidence craft for evaluation planning
Report every figure as a count over a denominator with its period. Nineteen of 46 children contacted within the intended window in one month is interpretable. Forty-one percent is a number whose volatility a reader cannot judge, and in small monthly denominators that volatility is the whole story.
Use language proportional to the design. Increased following implementation is defensible. Caused, produced and led to are not available to a single-site pre and post comparison, and misusing them is the most reliably marked error in this stage.
Follow an improvement reporting standard. Standards for reporting quality improvement specify what a reader needs, including context, intervention description, measures, analysis and limitations. Writing the plan to that shape now means the final report writes itself later.
Say who will collect each measure and what it costs them. A plan requiring twenty minutes of a rooming nurse's day will not survive a busy fortnight. Naming the burden and the owner is what makes a plan real.
State the de-identification and storage approach. Say that data will be aggregated, that identifiers will be removed at extraction, and where the working file will live. Also say that your organization's and your program's review steps will be completed before implementation, without implying any outcome of those steps.
Five mistakes that cost points in this week's territory
- No balancing measure. A plan that only measures what it hopes to improve cannot detect the harm it may be doing elsewhere, and reviewers ask about this first.
- A single before and after pair. Two points cannot distinguish a real shift from ordinary variation, and every rival explanation stays live.
- Measures nobody defined. Follow-up completed means at least three different things until you write the numerator, the denominator and the window.
- Causal language on a site-level design. The words you choose in the analysis section are graded, and overclaiming there undoes an otherwise sound plan.
- Rival explanations added afterward. Named before the data arrive, they demonstrate rigor. Named after a disappointing result, they read as excuses.
Before you submit
- Process, outcome and balancing measures are all present
- Every measure has a numerator, denominator, window, source, frequency and owner
- Baseline covers several periods rather than one
- The display is specified with the number of points before and after
- Rival explanations are listed in advance with how each will be handled
- No causal verb appears anywhere in the analysis section
- Data handling, de-identification and review steps are stated without implying any outcome
Writing the NR-716 evaluation plan?
Send the rubric and your measures out of Canvas. A premium original draft comes back in 24 to 48 hours with operational definitions that two people could count identically and language proportional to the design, and revisions run until the grade lands.