NR-516 · Week 2 of 8 · Searching the literature

NR-516 Week 2 Searching the Literature: How to Write It

The short answer

NR-516 Week 2 turns the question into a search, and a search is a documented procedure rather than an act of luck. The graded skill at this stage is reproducibility: another nurse should be able to take your databases, your terms, your limits and your keep or drop rules, run them, and land on roughly the articles you landed on. Your section may print this as NR 516 or NR516; 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-516 Week 2 grading scale at Chamberlain, the criterion levels this assessment is scored on, from Chamberlain Tutors
How Chamberlain grades NR-516 Week 2, visualized by Chamberlain Tutors.

What NR-516 Week 2 asks for

The territory of a search stage is retrieval and its documentation. Which databases hold nursing evidence and how they differ, how controlled vocabulary differs from free text, how Boolean operators and truncation widen or narrow a set, what filters do to a result count, and how you decide which of the returned articles actually belong in your paper.

Most students already search well enough to find something. What they cannot yet do is describe what they did in a way that survives a reader. That gap is the whole point of this stage. A sentence saying a database search was performed earns nothing. A short paragraph naming three databases, the terms combined with their operators, the year and language limits applied, the counts at each step and the reason articles were excluded is a search description, and it is a component many rubric rows in this course will ask for again in later weeks.

Deliverables at this point in a session are often a short written piece with a table attached, or a posted contribution about how a search went, sometimes both. Whatever the shape, write down the search while you are doing it rather than reconstructing it afterwards. Reconstruction is where the numbers stop adding up, and a search paragraph whose counts do not reconcile is the easiest thing in the course for a grader to mark. If your section runs a discussion this week, draft outside the box first, because a Canvas post cannot be reopened once submitted.

The NR-516 Week 2 method, step by step

Five moves that produce a search someone else could repeat, plus the record that proves it.

  1. Break the question into concepts before you type anything

    Take your question apart into two or three concept blocks, usually the population and the action, occasionally the outcome. Each block gets its own list of synonyms. Older adult, elderly, geriatric, aged belong in one block. Combining synonyms with OR inside a block and joining blocks with AND is the structure that makes a search behave predictably.

  2. Run controlled vocabulary and free text side by side

    Subject headings catch articles that use words you never thought of; free text catches articles indexed badly or too recently to be indexed at all. Running one without the other leaves a predictable slice of the literature invisible, and saying that you used both is a mark of a graduate level search.

  3. Apply limits one at a time and record what each one costs

    Add the year limit, note the count. Add peer reviewed, note the count. Add language, note the count. Filters are decisions with consequences, and a paper that reports a single final number cannot show its reader whether an important limit removed half the evidence.

  4. Screen in two passes, titles then full text

    Read titles and abstracts against written inclusion rules first, then read the survivors in full. Two passes with stated rules is faster than one pass with vague ones, and it produces the exclusion reasons your write up needs: wrong population, wrong design, no comparison, not retrievable.

  5. Write the search paragraph the same day

    While the numbers are still in front of you, write the databases, the strings, the limits, the counts and the exclusion reasons into a paragraph and a small table. Doing this at the end of the week means guessing, and guessed counts are visible because they are round.

A layout and word budget for a search write up

Here is the frame our tutors use for a search documentation piece of roughly 1,000 to 1,200 words, with a table sitting outside the count in most cases. It is our own outline, not a university issued form, and your week's rubric wins any disagreement.

SectionWhat belongs in itWord target
The question restatedOne sentence, in its final form, so the reader can judge whether the terms match it.50 to 70
Databases and why thoseThe sources searched and one clause each on what that source covers that the others do not.150 to 180
Terms and combinationsConcept blocks, synonyms, subject headings, operators and truncation, written out rather than described.230 to 270
Limits and their effectEach filter applied, in order, with the result count before and after and a reason for the filter.180 to 210
Screening and exclusionThe inclusion rules, the two screening passes and the reasons articles were dropped, by category.200 to 240
What was retainedThe final set, described by design and recency, with a sentence on what the search failed to find.120 to 150

Evidence craft while you are still retrieving

Report counts, not impressions. A search that yielded a manageable number is not a finding. A search that returned 412 records, reduced to 68 by a five year limit and to 9 after full text screening, is a finding, and it lets a reader see where the losses happened.

Keep the grey literature question open and answer it. Dissertations, agency reports and registry entries are outside most database defaults. You do not have to include them, but saying whether you looked and what you decided is what separates a search from a browse.

Publication bias belongs in this paragraph. Trials that found nothing are published less often and later. One sentence acknowledging that your retained set is skewed toward positive findings costs you nothing and earns credit under any row about the limits of the evidence.

Record retrieval details as you go. Digital object identifiers, volume and issue numbers and page ranges are trivial to capture at the moment you download and painful to reassemble at midnight. A reference list assembled from memory is where citation errors enter a course that grades citation accuracy.

Five mistakes that cost points in this week's territory

  • Typing the whole question into one box. A full sentence entered as a phrase asks a database for an exact match it will never find, and the near empty result set gets misread as an absence of evidence.
  • Naming a general web search engine as a database. It is a discovery tool, it is not indexed by subject heading, and citing it as your search source signals that the library instruction did not land.
  • Using only one database. Coverage differs by discipline and by journal, and a single source guarantees a blind spot you cannot describe because you cannot see it.
  • Silent exclusions. Dropping articles because they were awkward, without recording a reason, breaks the reproducibility the whole stage is about.
  • Confusing peer reviewed with primary research. A peer reviewed journal publishes editorials, commentary and reviews as well as studies. The filter narrows the venue, not the article type.

Before you submit

  • Every database searched is named, with a reason for including it
  • The actual search strings appear in the draft, operators and all
  • Result counts are reported before and after each limit
  • Inclusion and exclusion rules are written as rules, not as impressions
  • The retained set is described by design and by year, not just by number
  • Every reference appears in the text and every in-text citation appears in the list

Stuck on the search this week?

Send your question and the scoring guide from Canvas. We return a premium original draft in 24 to 48 hours with the strings, counts and screening rules written out, and revisions stay free until it lands.

Questions students ask about this stage

My search returns almost nothing. Do I change the question?
Usually you change the search first and the question second. Check the obvious failure points in order: a full sentence in the box, a missing synonym block, a subject heading that does not exist in that database, a year limit set too tight, or a spelling that differs between countries. If a properly built search still returns very little, that is a real result and you write it down, then widen one concept deliberately, most often by loosening the population or dropping the comparison term, and report the change. Documented widening reads as method. Silent widening reads as drift.
How many articles am I supposed to end up with?
Your week's rubric decides the number and it varies by assignment, so read the row rather than copying a classmate. What is constant is the standard for how you got there. A defensible set of four beats an undefended set of twelve, because every article you keep is one you will have to describe by design, sample and measurement later in the session. If the count feels arbitrary, work backwards from the appraisal work each article will demand and keep only the ones you can afford to treat properly.
Can I use an article I found through a reference list rather than a database?
Yes, and following citations backwards is a recognised technique rather than a shortcut. Name it in your write up as citation chaining or hand searching so the reader knows this article did not come from the strings you reported. The only thing that causes a problem is presenting a hand found article as though the documented search retrieved it, because then the counts no longer reconcile. Add a line to your search paragraph saying how many additional records came in this way and where they came from.

Keep going

Online now