psychbull-writing-style

Writing Style (psychbull-writing-style)

Psychological Bulletin prose must do something a primary paper's does not: **integrate and evaluate an

entire literature while meeting APA 7th-edition** style and the matching reporting standard —

MARS for meta-analyses, PRISMA for systematic reviews, JARS for quantitative/mixed work.

This skill guides structure and prose; the numbers come from the analysis skills.

When to trigger

• Drafting or restructuring the manuscript
• Writing the ≤ 250-word abstract
• Making sure the methods section satisfies MARS/PRISMA item-by-item
• Tightening an over-long or descriptive review into an evaluative argument

Structure for a synthesis

1. Introduction — the open question, why the field disagrees, what a synthesis can resolve; the

contribution, not a study-by-study recap.

1. Method (MARS/PRISMA) — eligibility criteria, search (databases, strings, dates, counts),

screening, coding and reliability, effect-size metric, model, heterogeneity, moderators, and

bias plans. Written so an item-by-item MARS/PRISMA check passes.

1. Results — pooled effect with CI and prediction interval, heterogeneity, moderators, bias

diagnostics, sensitivity; lean on the exhibits (forest/funnel/PRISMA).

1. Discussion — the theoretical contribution, scope conditions, limitations of the evidence base,

and a research agenda (see psychbull-theory-integration).

The abstract (≤ 250 words)

• State the question, the synthesis type, the size of the literature (k), the pooled effect,

key moderators, and the bottom line on robustness. APA structured-abstract conventions help.

Evaluative, not descriptive

• Synthesize, don't annotate. Group studies by argument and finding, not one paragraph per paper.
• Write for the whole field: define constructs and methods so non-specialists follow.
• Keep terminology and effect-size language consistent with the exhibits and tables.
• Masked: no author clues in the text; neutralize self-citations (see psychbull-submission).

Anti-patterns

• An "annotated bibliography" that lists studies instead of integrating them
• A method section that omits search strings/dates/counts (fails PRISMA/MARS)
• An abstract over 250 words, or one with no k / pooled effect / bottom line
• Burying the contribution under descriptive summary
• Inconsistent effect-size terms between text, tables, and figures

The prose bar that separates a Bulletin synthesis from a methods report

As the APA's flagship integrative-review outlet, Psychological Bulletin rewards prose that argues

about a literature, not prose that merely catalogs it. Referees read for an evaluative voice:

| Writing dimension | Reads as a Bulletin synthesis | Reads as a methods-only report (reject risk) |

|-------------------|-------------------------------|----------------------------------------------|

| Introduction | Frames the field's unresolved tension | Recaps each study in turn |

| Organization | Grouped by argument and mechanism | One paragraph per paper, chronologically |

| Results prose | Ties the pooled estimate to the open question | Dumps coefficients with no interpretation |

| Discussion | Advances or refines a theoretical account | Restates the forest plot in words |

| Audience | Written for the whole discipline | Assumes the reader is a meta-analysis specialist |

Worked vignette — writing up the intervention synthesis

Illustrative numbers only. For the self-affirmation synthesis (k = 42, g = 0.34 [0.24, 0.44],

I² = 68%, bias-adjusted toward ≈ 0.25), the write-up under this skill's rules:

• Abstract (≤ 250 words) names the synthesis type, k = 42, the pooled g and CI, the delivery-format

moderator, and the bottom line that the effect is positive but likely inflated by selective reporting.

• Method is written so a MARS reviewer can tick each item: databases and dated strings, screening

and reliability (κ = 0.81), Hedges' g with small-sample correction, a random-effects/RVE model, and

the pre-registered moderator and bias plan.

• Results lead with the prediction interval, not just the CI, so the 68% heterogeneity is honest in

the prose, then walk through the converging bias diagnostics.

• Discussion explains why earlier studies disagreed (delivery format) rather than re-listing them.

Referee pushback → venue-specific fix

• "This reads as an annotated bibliography." → Re-outline around arguments and mechanisms; merge the

per-paper paragraphs into thematic syntheses.

• "The abstract omits the headline numbers." → Insert k, the pooled effect with CI, and the robustness

bottom line within the 250-word cap.

• "The method won't pass a MARS item check." → Add the missing search strings, dates, counts, and the

reliability statistic, item by item.

Output format

【Structure】intro / method(MARS-PRISMA) / results / discussion present? [Y/N]
【Method completeness】search + coding + model + bias all described? [Y/N]
【Abstract】≤ 250 words; states k + pooled effect + bottom line? [Y/N]
【Evaluative】synthesizes vs annotates? [Y/N]
【APA 7th + masked】consistent style, no author clues? [Y/N]
【Next】psychbull-open-science-and-transparency

Supplementary resources

• [../../resources/external_tools.md](../../resources/external_tools.md) — APA-style reference managers; MARS/PRISMA/JARS standards
• [../../resources/official-source-map.md](../../resources/official-source-map.md) — APA 7th, 250-word abstract, MARS/PRISMA/JARS requirements