HLT-362V Article Analysis 1: three quantitative articles

The whole assignment is decided at the search stage, and most of the pain in it comes from choosing badly and discovering it three sections into the template. The brief asks for three *different* health care articles that use *quantitative* research, none of them from the Topic Materials or the textbook. Quantitative is the binding word: the template will ask you for the hypothesis, the independent and dependent variables and their data types, the sampling method, the statistical tests used, and the assumptions and limitations. A qualitative study — interviews, focus groups, thematic analysis — has none of those, and a mixed-methods paper will give you half of them. Before committing to an article, open its methods section and check that it names a statistical test. If it does not, it will not fill the template, however relevant the topic looks.

Three articles is three times the work only if you approach it badly. The efficient route is to search once for a topic area you can stand reading about, then filter within it, because three papers on adjacent questions let you reuse your understanding of the domain across all three analyses. The GCU Library restriction is real — search there rather than on the open web, because you will need a full-text methods and results section, and abstracts do not contain sampling method or statistical test. Randomised controlled trials fill the template most easily and cohort or cross-sectional studies work well too. Filter by article type and by date, and read the methods section before the abstract, which reverses the usual habit and saves the most time. Keep a one-line note of why each article passed screening, because if one turns out to be unusable at the analysis stage you will want to know what you already rejected.

The template's hardest row is the one asking you to identify the dependent and independent variables *and the type of data for the variables*, and the second half is what separates a completed template from a correct one. Data type means the level of measurement: nominal categories with no order, ordinal categories with order but no fixed spacing, interval or ratio for continuous measures. This matters because it determines which statistical test is legitimate, so the variables row and the data-analysis row have to agree with each other. If you have written that the outcome is nominal and then reported a t-test, one of the two entries is wrong. Checking that internal consistency across rows is the fastest quality check available on your own work. The same check catches the reverse error, where a continuous outcome has been dichotomised by the authors and the analysis row should reflect what they did rather than what they measured.

Two rows are routinely answered from the wrong part of the paper. *Define hypothesis* — the brief adds "or state the correct hypothesis based upon variables used", which is an instruction for the common case where the authors never state one explicitly. Many published papers give an aim or a research question instead, and you are expected to construct the testable hypothesis their variables imply. And *summary of assumptions and limitations* should come from the authors' own limitations paragraph in the discussion, not from your impressions: sample size, single-site recruitment, self-reported measures, short follow-up, loss to follow-up. Quote what they conceded and then add anything they omitted, which is where independent judgement can legitimately appear. Both rows reward reading the discussion carefully, which is the section students skim hardest and the one where authors are most candid about what their study cannot show.

Practically, the brief says APA style is not required for the body but that documentation of sources must follow APA formatting — so the three articles still need correct APA references even though the template is not an essay. Use the "Article Analysis 1" template as supplied rather than rebuilding it, keep each cell to a sentence or two of your own words rather than quoting the abstract, and answer every row for every article, since a blank cell is the easiest mark to lose. The brief points to a worked example — the hospital-at-home COPD trial alongside the "Article Analysis Example 1" — and reading that pairing before starting is the single cheapest thing you can do, because it shows the expected depth per cell. Working across the three articles row by row rather than paper by paper also helps, because the comparison keeps your standard for each cell constant.