The 7th edition of the Publication Manual for the American Psychological Association was published in October, and I finally finished reading it now. (Hey, give me a break! I had a book to finish, another book to revise, and a new baby to take care of.)
Yes, I read the manual cover-to-cover, all 395 pages (a total which does not include the references, sources for tables and figures, or the index). The 6th edition of the manual was published while I was in graduate school, and my dissertation co-chair made me read it–twice! APA manuals are not engrossing page turners, but I was glad I read the manual because it helped me master APA style quickly. So, I decided to read the new edition.
One thing that has struck me is how much open science is already impacting publication guidelines. Some of the most profound changes include:
- Writing up exploratory hypotheses as if they were pre-planned is now “. . . a violation of basic ethical principles” (APA, 2020, p. 12).
- A journal editor may deny publication if an author refuses to share materials or data during the peer review process (APA, 2020, p. 15).
- If authors do not share data within the data retention period, then the journal editor can retract an article or issue an Expression of Concern (APA, 2020, p. 16). There are exceptions for when data sharing is not mandatory, such as when it would violate participants’ privacy or when the requester is unqualified to analyze the data.
- The Journal Article Reporting Standards (JARS) now distinguish between “primary and secondary hypotheses” from exploratory hypotheses (APA, 2020, pp. 76, 80, 81).
- JARS mandates reporting intended sample size and achieved sample size (APA, 2020, p. 79).
- JARS also mandates that researchers report all variables, “. . . including measures collected but not included in the report” (APA, 2020, p. 79).
- Open data/citing data. This is a change from the 6th edition of the APA manual, which made no mention of making data freely available (whether online or in archives). Instead, the last APA manual mandated data retention for at least five years and had a weak statement that “APA encourages the open sharing of data among qualified investigators” (APA, 2010, p. 12). The current version says that authors in APA journals “are invited” (APA, 2020, p. 14) to make their data open to the scientific community, and there are provisions for citing data sets that are available online (APA 2020, p. 338). Journal editors also can reject a manuscript if the author does not share data with peer reviewers, and refusal to share data after publication is now grounds for retracting an article (APA 2020, pp. 14-15). While I wish that open data were a mandate, this is an improvement from the prior (incredibly weak) policy regarding data sharing in the previous APA manual.
- The most telling change is the JARS language against selective reporting of outcomes or analyses:
Mention all relevant results, regardless of whether your hypotheses were supported, including results that run counter to expectation; include small effect sizes (or statistically nonsignificant findings) when theory predicts large (or statistically significant) ones. Do not hide uncomfortable results by omission.American Psychological Association (2020, p. 86; see also p. 87).
What makes all this amazing is that the replication crisis started in 2011, and the reforms in psychology were already mature enough to be incorporated into this year’s APA manual as ethical mandates with teeth.
American Psychological Association. (2020). Publication manual of the American Psychological Association (7th ed.).