NR-667 · Week 2

NR-667 Week 2 The Reproducible Evidence Search: How to Write It

The short answer

NR-667 Week 2 usually turns the capstone question into a hunt with rules: which databases, which terms, which limits, and which screening decisions, all recorded so completely that a stranger could rerun the search and land on the same pile of studies. Reproducibility is the graded property. Your section may print this as NR 667 or NR667; 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.

Search weeks run alongside clinic days, and the capstone's precepted hours remain your program's territory end to end. What a desk can carry is the documentation burden of the search itself, the part that happens at a keyboard, cited and checkable.

NR-667 Week 2 grading scale at Chamberlain, the criterion levels this assessment is scored on, from Chamberlain Tutors
How Chamberlain grades NR-667 Week 2, visualized by Chamberlain Tutors.

What NR-667 Week 2 asks for

Expect the territory to cover the machinery of systematic searching scaled to a capstone: translating the question's concepts into controlled vocabulary and free-text terms, combining them with Boolean logic, choosing databases whose coverage fits the question, setting date and design limits before results are seen, and writing inclusion and exclusion rules that screening can apply without improvisation. The week's deeper lesson is that a search is an experiment: its methods section is what makes its findings mean anything.

Deliverable shapes commonly include a search strategy document with the count trail recorded, a methods-style narrative describing the search for the eventual paper, or a table of the retained studies with the path that found them. If your section runs a discussion this week, it often asks for your strategy plus one problem you hit, terms too broad, a flooded result set, a database that returned nothing, and how you adjusted.

Your week's rubric will likely weigh the trail hardest: databases named with reasons, the exact combinations logged, counts at each stage, and rules that were written before they were needed.

The NR-667 Week 2 method, step by step

Six moves that make a search something a reader can rerun.

  1. Break the question into its concepts

    The bounded question from the opening week decomposes into concept blocks, typically the population, the change being considered, and the outcome. Each block gets its synonyms, its controlled vocabulary where the database offers it, and its spelling variants, written down before any searching starts.

  2. Choose databases for coverage, not habit

    Name each database you will search and the reason it earns its place for this question: nursing and allied health coverage, biomedical breadth, or a specialty focus. Two well-chosen databases with reasons beat five listed to look thorough.

  3. Log the combinations exactly as run

    Record every search string with its Boolean structure and the count it returned, in the order you ran them. This log is the paper's raw data; reconstructing it from memory afterward produces the vagueness graders can spot in one read.

  4. Set limits before you see results

    Date windows, language, design filters, and the reasons for each, decided in advance and applied uniformly. A limit invented after seeing the results is not a limit, it is a preference wearing one's clothes.

  5. Screen with rules, counting as you go

    Apply your inclusion and exclusion rules first to titles and abstracts, then to full texts, recording how many entered and left at each stage and the reason category for exclusions. The counts become the flow your reader follows.

  6. Archive everything, then format

    Export citations, save the strings, date the searches, since databases update and a dated search is part of reproducibility. Then the mechanics: current APA for every retained source, headings in your rubric's wording.

A search strategy document, section by section

Targets assume roughly 1,100 words plus the log itself. If your section wants the search folded into a larger methods section, keep these proportions inside it.

SectionWhat belongs thereWord target
The question restatedThe bounded question, unchanged from its approved form, opening the document60 to 90
Concept blocks and termsEach concept with its controlled vocabulary, free-text synonyms and variants170 to 210
Databases with rationaleEvery database searched and the coverage reason it was chosen120 to 160
Limits and their reasonsDate, language and design limits, each with its justification, set before results140 to 180
Screening rules and countsInclusion and exclusion criteria, and the stage-by-stage count trail220 to 270
The retained setThe final studies, one line each: design, population, and why it survived screening200 to 250

The count trail is the section that fails silently: totals that do not reconcile between stages are the most common flaw faculty find, and a five-minute arithmetic check before submission removes it.

Limits, grey sources, and the currency problem

Date limits need clinical, not cosmetic, justification. A five-year window suits fast-moving treatment questions; a foundational instrument or a landmark trial may warrant a named exception, and the honest structure is a general rule plus listed exceptions with reasons, stated in the methods rather than discovered in the reference list.

Grey literature, practice guidelines, government reports, quality collaboratives, can strengthen a capstone search when handled openly: say where you looked, what you took, and how you weighed material that never passed peer review. What sinks papers is laundering, citing a guideline's summary of a study as if you had appraised the study itself.

Currency is a verb here, not a state. Guidelines central to your question should be checked against their issuing body's site for newer editions on the day you finalize, and the check itself, with its date, belongs in your documentation. A capstone anchored to guidance that was replaced mid-session, when one check would have caught it, converts a strong search into an outdated one retroactively.

Five mistakes that cost points in a search week

  • Searching until the answer looks right. Stopping when the results support your hunch, rather than when the strategy is exhausted, is bias with a timestamp.
  • Limits applied after the fact. A date window that happens to exclude the inconvenient trial reads exactly as it is, and methods sections make it visible.
  • The missing count trail. "The search yielded relevant articles" tells the reader nothing survived counting; numbers at each stage are the section's substance.
  • One database standing for the literature. Single-source searches miss systematically, and naming a second database costs a paragraph.
  • Criteria written after reading. Inclusion rules drafted to fit the studies already chosen invert the entire method, and the inversion usually shows.

Before you submit

  • Every concept block lists its controlled and free-text terms
  • Each database carries a coverage reason
  • Search strings are logged exactly as run, with dates
  • All limits were set before results and carry justifications
  • The count trail reconciles at every stage
  • Guideline currency was checked and the check is documented

Search log turning into a second job?

Send the question and whatever trail you have. A clean strategy document with the counts reconciled returns inside 24 to 48 hours, revised free until it lands.

Three questions students send about this week

My search returned almost nothing. Is my project dead?
Rarely, and the fix is diagnostic before it is fatal. Check the usual chokepoints in order: terms too narrow or missing synonyms, concept blocks over-combined with AND, limits set tighter than the question needs, or a database whose coverage never fit. Widen one variable at a time and log each widening. If the literature is genuinely thin after honest widening, that thinness is a finding, document it, search the adjacent population or setting, and present the extrapolation openly. Faculty accept a sparse evidence base handled honestly; they do not accept one discovered in week six.
My search returned thousands of hits. What do I tighten first?
Tighten specificity before scope. Convert free-text terms to controlled vocabulary where the database supports it, require your population and outcome concepts rather than letting either float, and apply the design filter your question deserves. Only then consider narrowing the date window, and never narrow it just to shrink the pile. Record each tightening and its new count, because the sequence of adjustments is itself graded methods content, and a flood managed visibly reads far better than a suspiciously convenient few dozen results.
Can I include the article that gave me the project idea?
Only if the search finds it or your rules admit it through a named route. Run the strategy first; a genuinely relevant article usually surfaces on its own, which settles the question. If it does not, you may add it through a declared supplementary route, hand-searching or citation chasing, provided you say so and apply the same screening rules to it as to everything else. What you cannot do is quietly seed the retained set with a favorite the method never touched, because that single untraceable study undermines the reproducibility claim the whole week is scoring.

Keep going

Online now