Week 3 skill · pulling an empirical article apart into its working pieces

The Article-Dissection Worksheet

One article, one page, eleven boxes. You will do this to every source in your literature review — so learn it on your two strongest.

What this is for

A literature review is not a stack of summaries. It is an argument built from the parts of studies — who was measured, what was measured, how, and what came out. You cannot compare two studies until you have taken both apart the same way. This worksheet is that same way. Fill one in for each article and the synthesis matrix in Week 4 builds itself: every row of the matrix is a box from this sheet.

Fill it in in your own words, in your shared journal Doc, in a new tab called Dissections. Copy the template below once per article — or print the PDF version (two pages, tagged) and work on paper first, then type it up. There is also a Google Doc copy of the template in the seminar folder to duplicate into your journal. Quote the authors only in the two places that ask for it (the hypothesis as stated; the statistic as reported), and give a page number for every finding. If you paste anything else, label it as pasted — the syllabus rule.

How to work it

  1. Read first, the three-pass way. Abstract, then figures and tables, then the whole thing (How to Read a Scientific Article). Do not open the worksheet until pass two is done.
  2. Fill boxes 1–9 from the PDF alone. No notebook, no AI. If a box is hard, that is information: it usually means the article did not report it clearly, which is itself a limitation for box 9.
  3. Then the notebook check (box 11). Ask your capstone notebook the same questions — "What was the sample?", "How was X measured?", "What did the authors list as limitations?" — and compare. Anything it surfaces that you missed, go back to the PDF and find it. Write what you missed and the page you found it on. Log the notebook use in your AI log.
  4. Box 10 last. Only after the article is fully apart can you say what it gives you.

Budget about forty-five minutes per article the first time. By Week 6 you will do one in fifteen.

The worksheet

0 · CitationFull APA 7 reference, with DOI. Then the filename in your references folder, so I can find the PDF.
1 · The questionWhat did the authors want to know? One sentence, your words. Usually the last paragraph of the Introduction.
2 · HypothesesAs the authors stated them (quote, with page). Then: directional or not? Confirmed or not?
3 · SampleN; who they were (age, population, country); how recruited (students for credit, MTurk/Prolific, clinic, community); exclusions and final N after exclusions.Method → Participants.
4 · DesignCorrelational, experimental, quasi-experimental, longitudinal, or qualitative. If experimental: what was manipulated, how many conditions, random assignment yes or no. If longitudinal: how many waves, how far apart.
5 · VariablesOne row per construct. Role is IV / predictor, DV / outcome, moderator, mediator, or covariate. Operational definition is the instrument or manipulation: name, number of items, response scale, reliability if reported (α).
ConstructRoleOperational definition (instrument / manipulation, items, scale, α)
6 · ProcedureWhat a participant actually did, start to finish, in three sentences.
7 · Key findingsThe main result as reported — test, statistic, p, effect size if given (e.g., r = .34, p < .001; t(198) = 5.39, d = 0.76), with page. Then the same result in one plain-English sentence a friend outside psychology would understand.
8 · InterpretationWhat the authors say it means, and one alternative explanation they did not rule out.
9 · LimitationsThe ones the authors name (Discussion, usually near the end). Then one they did not name — sample, measure, design, or generalizability.
10 · So what, for my projectWhich theme of my review it belongs to. What it gives me: a measure I can use, a finding I build on, a gap it leaves, a design I can borrow. One or two sentences.
11 · Notebook checkThe question(s) I asked my capstone notebook. What it surfaced that I had missed. Where I then found it in the PDF (page). Logged in the AI log, dated.

What a complete worksheet looks like

Every box has something in it, in your words, and boxes 2 and 7 carry page numbers. Box 5 has at least one row for every construct in your research question that the article touches — if your question names four skills and an outcome, and the article measures two of them, two rows say how, and a note says the other three are absent. Box 7 has a real statistic, not "they found a significant relationship." Box 9 has one limitation the authors did not write down. Box 11 shows you checked your own work against the notebook and went back to the source.

Incomplete looks like this: box 3 says "college students," box 5 says "self-compassion — survey," box 7 says "self-compassion predicted well-being." Those are labels, not dissections. A dissection tells me the instrument, the N, and the number.

Week 3 assignment. Two completed worksheets, on your two strongest sources — the ones that measure the constructs in your own question — in your journal Doc before your meeting the week of September 28. Pick empirical articles: a systematic review or meta-analysis cannot be dissected this way (there is no single sample, procedure, or statistic). Save those for Week 4, where they earn their keep in the synthesis matrix.

Where this comes from

Bem, D. J. (2004). Writing the empirical journal article. In J. M. Darley, M. P. Zanna, & H. L. Roediger III (Eds.), The compleat academic: A career guide (2nd ed., pp. 185–219). American Psychological Association.

Landrum, R. E. (2012). Undergraduate writing in psychology: Learning to tell the scientific story (Rev. ed.). American Psychological Association.