APA 7 essentials, and the parts people get wrong
The style rules that carry meaning, as distinct from the ones that are merely conventions.
Most of APA style is arbitrary convention and can be looked up. A minority of it encodes substantive commitments about precision, bias and inference, and those are the parts worth understanding rather than checking.
The rules that carry meaning
- Report exact p values, not thresholds
- p = .03, not p < .05. Thresholds discard information and encourage the treatment of .049 and .051 as categorically different, which they are not.
- Always report an effect size and a confidence interval
- Significance tells you whether an effect is distinguishable from zero at your sample size. It does not tell you how big it is or how precisely you know it. APA requires both for good reason.
- Person-first or identity-first, as the group prefers
- APA 7 explicitly accommodates both, and the preference differs: "people with schizophrenia" is generally preferred, while much of the autistic and Deaf communities prefer identity-first. Ask, and follow the group rather than the rule.
- Do not use "subjects" for people
- Participants, or a specific description. The change was substantive rather than cosmetic — it reflects a shift in who the research is understood to be for.
- Singular "they" is correct
- APA 7 endorses it, both as a generic pronoun and for a person who uses it. There is no remaining stylistic argument here.
- Describe samples in enough detail to judge generalisability
- Age, gender, race and ethnicity, socioeconomic indicators, recruitment source, exclusions. Omitting these is what allowed the WEIRD problem to persist unremarked for decades.
Structural conventions
| Element | APA 7 |
|---|---|
| In-text citation | (Rogers, 1957); narrative: Rogers (1957) proposed… |
| Three or more authors | First author et al. from the first citation — no longer spelled out once |
| Direct quotation | Include the page number; block quote at 40+ words |
| DOIs | Formatted as https://doi.org/… — no "Retrieved from" |
| Reference list | Hanging indent, alphabetical by first author surname |
| Headings | Five levels, all sentence case except Level 1 which is centred and bold |
| Numbers | Words below 10, numerals from 10 up, numerals always for statistics and measurements |
| Statistics | Italicise M, SD, t, F, p, d, r, n, N; do not italicise Greek letters |
Writing about people you treated
APA style is not the constraint here — consent is. See the separate guide on writing about patients, which covers what journals now require and why disguise alone is not sufficient.
Sources
American Psychological Association (2020). Publication Manual of the American Psychological Association (7th ed.). APA
manualAppelbaum, M., et al. (2018). Journal article reporting standards for quantitative research in psychology: The APA Publications and Communications Board task force report. American Psychologist, 73(1), 3–25
guideline