Statistical copy is the most technically demanding element of APA 7th formatting, and the errors that accumulate in it are among the most resistant to correction by authors working alone. The rules are specific, the manual's treatment of them is distributed across multiple sections, and the conventions that feel intuitive — about when to italicize, when to use abbreviations, how to handle spaces around operators — frequently turn out to be wrong.
The following covers the areas where errors are most common and most consequential.
Italicization of statistical symbols. APA 7th requires that most statistical symbols be italicized when they appear in text. This includes the common inferential statistics — t, F, r, p, M, SD, N, n, df — as well as Greek letters used as statistical symbols, though the convention for Greek letters is more nuanced and depends on whether the letter represents a population parameter or a sample statistic. Authors frequently apply italics inconsistently, italicizing in some instances and not others within the same document, or applying italics to symbols that do not require them.
Spacing around operators and values. APA 7th requires a space on both sides of mathematical operators and comparison symbols when they appear in running text. The correct form is p = .043, not p=.043. The same rule applies to inequality symbols: t(38) = 2.14, not t(38)=2.14. The omission of spaces around operators is among the most common statistical formatting errors in dissertations and one of the easiest to correct with a targeted find-and-replace pass.
Reporting of exact p values. APA 7th no longer recommends reporting p values as inequalities (p < .05) except when the value falls below .001, in which case it is reported as p < .001. For all other values, the exact p value should be reported to three decimal places: p = .043, p = .127. Authors trained in APA 6th conventions frequently continue to report inequality forms for all values and are not aware that this convention changed in the 7th edition.
Leading zeros. APA 7th does not use a leading zero before the decimal point for values that cannot exceed 1.0 — including p values, correlation coefficients, and effect sizes expressed as proportions. The correct form is p = .043, not p = 0.043. For values that can exceed 1.0, the leading zero is used. This distinction is simple once stated and reliably missed in practice.
Effect size reporting. APA 7th requires that effect sizes be reported alongside inferential statistics for all primary analyses. The specific effect size measure should be appropriate to the statistical test — Cohen's d for mean comparisons, r or r² for correlations, η² or partial η² for ANOVA designs — and should be reported with its confidence interval where feasible. The omission of effect sizes is both a formatting error under APA 7th and a methodological gap that reviewers will flag.
Sample size notation. Italicized N refers to the total sample. Italicized n refers to a subsample. These are not interchangeable, and their incorrect use — particularly N where n is required — appears regularly in the results sections of dissertations.
The standard for statistical copy in APA 7th is exacting, and achieving it requires a format-specific review pass that is separate from the substantive editing of the document. Authors who have invested significant work in the analysis itself deserve a final manuscript that presents that work with the precision it requires.
608-535-6324
editors@offthegridediting.com