NR-575 · Week 3 of 8 · The evidence synthesis behind the capstone

NR-575 Week 3 Evidence Synthesis: How to Write It

The short answer

By the third stage of a capstone practicum the written work usually turns to evidence: not a literature review that tours what has been published, but a synthesis that establishes what the body of work supports, where it disagrees, and what it never tested. The test of a synthesis is whether a reader could predict your product from it. If your product could be replaced by any other intervention without changing a word of the synthesis, the synthesis has not done its job. Your section may print this as NR 575 or NR575; 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. Clinical hours, logs and preceptor evaluations remain your own record and are never drafted or reconstructed with help.

NR-575 Week 3 grading scale at Chamberlain, the criterion levels this assessment is scored on, from Chamberlain Tutors
How Chamberlain grades NR-575 Week 3, visualized by Chamberlain Tutors.

What a capstone synthesis has to deliver

Audit a folder of student literature sections and the pattern is easy to see from the headings alone. Smith found. Jones concluded. A third study reported. Each paragraph belongs to one article, the articles appear roughly in the order they were downloaded, and nothing in the section requires the reader to have read the paragraph before it. That structure is the reliable signature of a summary. A synthesis is organized by claim, and every paragraph in it holds several studies at once.

The graded shift is from what did the authors say to what does the evidence support. A synthesis paragraph opens with the finding, not the researcher: interventions that place the escalation criteria at the point of documentation reduce time to reassessment across several settings, though the effect is smaller where the criteria were added without changing the workflow. Three or four sources sit inside that sentence and the sentence is a claim a grader can score.

A capstone synthesis carries an extra obligation that a research course synthesis does not. It must arrive somewhere. By the end the reader should know which intervention component is best supported, in what population, at what strength of evidence, and what the literature has not settled. That closing paragraph is the bridge to your product, and the product you build in the following stages should be visibly assembled from the components the synthesis identified.

Acute care raises the transferability question sharply. Much of the usable evidence on rapid deterioration, transitions of care and structured escalation comes from settings that differ in staffing model, patient acuity or country. That is not a reason to exclude it; it is a reason to say what changes when it moves. A synthesis that notes a finding was produced in a closed-model intensive care unit and argues what would and would not carry into a step-down environment is doing precisely the analytic work the rows reward.

Deliverables at this stage are commonly a written synthesis, often with an appraisal table, and sometimes a posted summary of the strongest evidence found. Claims about studies are easy to check, so treat any post as final copy; posts do not reopen after submission in Canvas.

The NR-575 Week 3 method, step by step

Six moves that produce a synthesis instead of a reading list.

  1. Turn your question into search terms and record the search

    Take the population, intervention and outcome from your week one question, list the alternative vocabulary each one uses in the literature, and keep a log of database, terms, limits and yield. Capstone rubrics frequently score a described search, and reconstructing one later is far harder than recording it now.

  2. Set inclusion rules before you read anything

    Population, design, setting, recency, language, outcome measured. Written down first, these rules are a method. Written down afterward, they are a description of what you happened to find, and graders can tell the difference from the way the numbers fail to add up.

  3. Build an extraction table, one row per study

    Design, setting, sample and attrition, what was actually delivered, outcome and how it was measured, direction and size of effect, and your one-line judgment of quality. The table is a working document; the columns are what makes the next step possible.

  4. Sort the rows by finding until themes appear

    Rearrange the table by outcome, then by intervention component, and the claims emerge on their own. Three studies clustering around the same result is a paragraph. One study standing alone is either a limitation or a lead.

  5. Write claim-first paragraphs and put the citations inside them

    State the finding, name the studies that support it in the same sentence, then handle the study that disagrees and say why it might. A paragraph that begins with a researcher's surname has already committed to summary shape.

  6. Finish with the gap that your product fills

    Name what the evidence does not settle, in one paragraph, and make it the specific gap your capstone addresses. This is the sentence the implementation and evaluation stages are built on.

A layout and word budget for a capstone synthesis

Our frame for a synthesis section, sized for roughly 1,400 to 1,800 words. It is our own outline rather than anything the university issues, and your week's rubric outranks it wherever the two disagree.

SectionWhat belongs in itWord target
Search and selectionDatabases, terms, limits, yield at each stage, and the inclusion rules stated as rules rather than as description.180 to 230
Strength of the body of workWhat kinds of studies you have, how many of each, and what that mix can and cannot support before any finding is reported.150 to 200
Claim one, best supportedThe finding with the most convergent evidence, several sources inside the paragraph, effect direction and size in real units.280 to 350
Claim two and its qualifierThe second theme, including the condition under which it holds and the setting where it did not replicate.280 to 350
Disagreement handledWhere studies conflict, the methodological reason for the conflict, and which side your setting resembles.200 to 260
Transferability to your settingWhat changes when these findings move into your unit type, acuity and staffing model.180 to 230
The gap your product fillsWhat is unsettled, stated narrowly enough that the product in the next stage is the obvious response.130 to 180

Evidence craft for synthesis at capstone level

Name the design in the sentence where you use the study. A randomized trial and a single-site improvement report carry different weight, and a reader should never have to consult your table to find out which one you are leaning on. In a stepped-wedge trial across four units and In a before-and-after project at one hospital both take four words and both change how the claim lands.

Report effects in clinical units before statistical ones. Time to reassessment fell by roughly forty minutes tells a clinician something. The difference was statistically significant does not. Where the source gives an interval around the estimate, carry it, because a wide interval is information about sample size that a point estimate conceals.

Match verbs to design. Observational work supports was associated with. Reduced, improved and prevented belong only where an intervention was assigned and compared. This is one of the most consistently marked errors in graduate synthesis writing and it is entirely avoidable with one pass through your verbs.

Use a named appraisal framework and say which one. Structured critique tools exist for different designs. Naming the one you applied lets the grader check your quality judgments against a standard rather than taking them on trust, and it makes your table columns defensible when somebody asks about them at the intensive.

Keep the local layer separate and labelled. Your own observations from practice belong in the capstone and are valuable, but they are not literature. Where you use them, mark them as local context, de-identified, and do not let them share a sentence with published findings in a way that blurs which is which.

Five mistakes that cost points at this stage

  • One paragraph per article. The single most reliable indicator of a summary. If your section headings could be replaced by surnames, restructure before you edit anything else.
  • Convenient silence about disagreement. A body of evidence in which every study agrees usually means the ones that disagreed were left out, and a grader who knows the literature will notice.
  • An unrecorded search. Rubrics that ask for a described strategy cannot award the row from a reference list, and reconstructing the search four weeks later never produces the same numbers.
  • Evidence that does not point at your product. If the synthesis would fit any intervention equally well, it has not narrowed anything and the following stages inherit the problem.
  • Age of sources unexamined. Old work is sometimes the best work, but a synthesis leaning on evidence from a very different practice era should say so and argue why it still holds.

Before you submit

  • Every paragraph opens with a finding rather than an author
  • The search is described with databases, terms, limits and yield
  • Inclusion rules appear before the results of applying them
  • Each claim carries more than one source inside the paragraph
  • Effects are reported in clinical units and verbs match the designs
  • Conflicting evidence is named with a methodological explanation
  • The closing gap is narrow enough that your product follows from it

Building the NR-575 evidence section?

Send the rubric and your article set out of Canvas. A premium original draft comes back in 24 to 48 hours organized by claim rather than by author, with disagreement handled and the gap written to carry your product, and revisions run until the grade lands.

Questions students ask about this stage

How many sources does a capstone synthesis need?
Enough to support each claim more than once, which in practice tends to mean somewhere between eight and fifteen studies doing real work, plus whatever guidance documents and standards the argument leans on. Counting is the wrong instinct, though; density is the right one. A synthesis with ten sources where every paragraph holds three of them reads as far stronger than one with twenty-five where each appears alone. If your rubric names a minimum, meet it, then check the distribution: if half your citations appear exactly once and never again, the set is wider than the argument and some of those articles are decorating rather than supporting.
There is almost no research on my specific problem. What do I do?
Widen along one axis at a time and be explicit about it. If nothing exists on your exact intervention in your exact population, look for the same intervention in an adjacent population, or a different intervention aimed at the same mechanism, and say in the text which axis you widened and why the transfer is reasonable. That reasoning is itself gradeable content and it usually reads better than a thin set of perfectly matched but weak papers. A genuine absence of direct evidence is also the strongest possible gap statement for your capstone, so write it plainly rather than disguising it: the mechanism is supported, the application to this setting has not been tested, and that is what the product addresses.
Can I reuse the synthesis I wrote for an earlier course?
Check your section's policy on reused work first, because self-plagiarism rules vary and the consequences are not worth guessing at. Where reuse is permitted with citation, the honest version is to treat the earlier work as a starting point rather than a section to paste. A capstone synthesis has a job an earlier assignment did not: it has to converge on a buildable product for a specific setting. That means new searching for recent work, a different organizing spine, and a closing gap paragraph written for this problem. What genuinely carries over is your extraction table and your familiarity with the sources, which is a real head start.
What belongs in the appraisal table and what belongs in the prose?
The table carries the per-study facts: citation, design, setting, sample and attrition, intervention delivered, outcome and measurement, result, and a quality judgment. The prose carries what the studies mean together. If a sentence in your synthesis reports one study's sample size, it is probably duplicating the table and could be spending those words on a claim instead. Use the table as the evidence a reader can audit and the prose as the argument you are asking them to accept. Where the rubric asks for both, that division also solves the length problem, since the per-study detail that inflates most drafts moves into cells where it belongs.

Keep going

Online now