NR-559 Week 2 Search Strategy and Evidence Matrix: How to Write It

What NR-559 Week 2 asks for

The territory here is reproducibility. A reader should be able to take your written method, repeat it, and land in roughly the same set of studies. That is a low bar in principle and a bar most student searches fail, because students search the way they shop: they type a phrase, scroll until something looks useful, and cannot afterwards say how they got there. Graduate rubrics in research courses reward the opposite habit, which is why so many of these assignments ask you to report your search rather than merely its results.

Two written objects usually come out of this stage. The first is a search narrative: a few paragraphs saying where you looked, with what terms, under what limits, and what came back at each step. The second is a matrix, sometimes called a literature table or evidence table, where each study occupies one row and each column asks the same question of every study. The matrix is not the paper. It is the instrument that makes the later paper possible, and its value depends entirely on the columns being consistent.

Chamberlain sessions run eight weeks, so the studies you collect now are the studies you appraise in the middle stages and synthesize at the end. A matrix built carelessly in week 2 produces a week 5 paper that has to be re-researched under time pressure. It is worth an extra evening here. Students who build a clean matrix in this stage routinely report that the rest of the session took half the effort they expected, because every later assignment is a different way of reading the same table.

If your section runs a discussion alongside the written work, expect it to ask about search terms or about what you found and could not find. Be concrete in the post. Naming a database, two search strings and a hit count is more interesting to a grader than a paragraph about how much evidence exists on your topic, and posts do not reopen once submitted in Canvas.

The NR-559 Week 2 method, step by step

Six moves that produce a search someone else could repeat and a table you can still use in week 7.

1. Turn each PICOT element into a term list before opening a database

   Write the concept, then every way the literature might name it, including the clinical term, the lay term and the abbreviation. Delirium screening, cognitive assessment, confusion assessment, mental status screening. This list is the raw material of every search string you will build, and building it away from the search box stops you from anchoring on the first phrasing that returns results.

2. Search at least two databases and say why those two

   Nursing indexes and biomedical indexes cover different journals, and a search in only one is a search with a known blind spot. Name the databases in the paper and give a one-clause reason for each. Where you also used a guideline repository or a professional organization's library, list it separately from the peer-reviewed databases.

3. Record every string exactly as you typed it

   Keep a running log in a plain file with the date, the database, the full string including the Boolean operators, the limits applied and the number of results. This log is what turns a paragraph of vague recollection into a method section. It also protects you when a grader asks how you found a study you cannot otherwise account for.

4. Apply limits deliberately and defend them

   Publication window, peer-reviewed status, language, age group, setting. Each limit is a decision that excluded something, so each deserves a clause. A five-year window is conventional in graduate nursing work and worth stating rather than assuming, and reaching outside it for a seminal source is fine if you say why.

5. Screen in two passes and count what falls out at each

   Title and abstract first, then full text. Report both numbers: how many records the searches returned, how many survived the first pass, how many full texts you read, how many you kept. The arithmetic should reconcile. Nothing signals a careless search faster than a narrative in which 340 hits become 8 studies with no intermediate step.

6. Fix the matrix columns before you fill the first row

   Decide the columns once, then never ask a different question of a later study. Citation, purpose, design, sample and setting, measures, key findings, limitations, and relevance to your question is a serviceable set. Columns invented row by row produce a table that cannot be read down, which is the only direction that matters.

A layout and word budget for a search narrative

Our frame for the written portion, sized for roughly 900 to 1,200 words alongside the table. It is our own outline rather than anything the university issues, and your week's rubric outranks it wherever the two disagree.

Evidence craft for search reporting

An absence is a finding worth reporting. If no randomized trial exists for your intervention in your population, say so and say what you found instead. Students hide thin evidence because it feels like failure; graders read a named gap as evidence that you searched properly, and the gap is often the most useful sentence in the paper for whatever project follows.

Distinguish primary studies from synthesized evidence in every sentence. A systematic review is not a study, and a clinical practice guideline is neither. Your matrix should make the difference visible in the design column, and your narrative should say how many of each you have. Sets built entirely from reviews cannot support statements about a single population, because you have no access to the underlying methods.

Retrieved is not the same as read. Report only what you actually opened. Padding a matrix with abstracts you skimmed produces rows that collapse the first time a grader checks one, and in an eight-week session that check often happens in the appraisal stage when it is too late to substitute.

Cite the studies, not the database. The reference list carries the articles in full citation form. The database belongs in the method narrative as a search location. Students who list the index instead of the journal lose formatting points and, more importantly, make their set impossible to verify.

Keep predatory and non-scholarly sources out of the set entirely. Content farms, term paper repositories, unattributed continuing education summaries and unsigned web pages have no place in a graduate evidence table. Where you use an organizational document such as a guideline, cite the organization as author and give the edition or year.

Five mistakes that cost points in this week's territory

• A narrative with no numbers. Saying you searched several databases and found relevant articles reports nothing a reader could repeat.
• Screening arithmetic that does not reconcile. If the hits, the abstracts and the full texts do not add up, the grader stops trusting the rest of the paper.
• A matrix with ragged columns. Rows answering different questions cannot be compared, and comparability is the only reason the table exists.
• Every source a review. Synthesized evidence is valuable and cannot carry an argument about your specific population on its own.
• Terms invented at the search box. Searching with the first phrase that occurred to you leaves whole bodies of literature invisible and unreported.

Before you submit

• At least two named databases appear with a reason for each
• Full search strings with Boolean operators are reproduced in the text
• Every limit applied carries a one-clause justification
• Screening counts reconcile from hits to abstracts to full texts to included
• Matrix columns are identical across every row
• Each included study is a source you actually read in full
• The narrative names at least one gap the search could not fill

Questions students ask about this stage

Keep going