Public source record

Corrections and source updates

Material corrections and clarifications to You Can’t Read Her and its free Part One sample are recorded here with the source and review date.

Current status: The website was corrected on August 8, 2026. The 38-page sample PDF still contains the earlier wording on printed pages 26 and 27. Where the PDF and this record differ, this dated record gives the current account.

By Duncan Reeve · First posted August 8, 2026 · Last reviewed August 8, 2026

1. The five Guéguen retractions

What the sample says: Chapter 3 dates the five retractions to March 2025 and refers broadly to missing ethics approvals across the research corpus.

Current account: Sage’s consolidated notice was first published online on February 25, 2025. A secondary report followed on March 3. The notice covered five articles, two about touch and compliance and three about bust size and hitchhiking, hair colour and hitchhiking, and eye contact and later interest.

The notice described implausible methods, unresolved questions about ethics approval and informed consent, and data-fidelity or analysis problems. That removes the five articles as support. It does not establish that every paper in the wider corpus lacked ethics approval, or that every broader claim about touch is false.

The replication statement

The sample also says that no preregistered replication of the courtship touch-compliance effect exists. That sentence reports the author’s source search, not a finding in Sage’s retraction notice. As of the August 8, 2026 review, no such replication had been identified in the material checked for the book. An unsuccessful search is not proof that no study exists. The narrower conclusion is enough: the two retracted touch papers cannot support the tactic.

2. The red-attractiveness estimate

What the sample says: The “Wear red” section gives d = 0.13 without naming the direction and grades the practical strategy Dead.

Current account: The 2018 meta-analysis separated two directions. The d = 0.13 estimate applies to women rating men. The authors described it as very small and reported evidence consistent with upward bias. The practical grade is the book author’s editorial judgment, not the meta-analysis authors’ label.

DirectionPooled estimateImportant limit
Men rating womend = 0.26 [0.12, 0.40], N = 2,961Substantial heterogeneity; upward-bias evidence was equivocal.
Women rating mend = 0.13 [0.01, 0.25], N = 2,739Evidence consistent with upward bias; eight preregistered contrasts averaged d = -0.10 [-0.24, 0.04].

Source: Lehmann, Elliot and Calin-Jageman (2018), “Meta-Analysis of the Effect of Red on Perceived Attractiveness.”

3. What the flirting percentages mean

No correction is required to Chapter 1, but one interpretive guardrail matters. In Hall, Xing and Brooks’s Study 1, 25 participants reported flirting and their partners agreed in 7 cases. Partners also agreed with 66 of 79 reports of not flirting. The 36% and 18% figures are hit rates conditional on the partner’s yes-or-no self-report, based on 5 of 14 and 2 of 11 cases. They are not general accuracy scores, did not establish a sex difference, and do not measure attraction, sexual interest or consent.

Source: Hall, Xing and Brooks (2015), “Accurately Detecting Flirting.”

Correction policy

Read with the update beside you

The first three chapters are free

Download the 38-page Part One sample, then use this page for the current Chapter 3 source clarifications.