Coding in the review was non-exclusive, and data from multiple experiments within one paper were separated by semicolons and averaged before computing corpus-level means and medians¶

Claim · All claims
Evidence · 1 study (2 entries) · review r? · q1
Currency · newest study 2020 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q1 i? Coding categories were not mutually exclusive; topics could be marked in multiple categories. → Allison Zengilowski 2020 q1 i? Multi-experiment sample sizes were averaged before calculating overall corpus mean and median sample sizes. → Allison Zengilowski 2020 (2)

Evidence¶

Allison Zengilowski 2020¶

Allison Zengilowski, Brendan Schuetze, Brady Nash, Diane Schallert. (2020). Coding Notes for “A Critical Review of the Refutation Text Literature: Methodological Confounds, Theoretical Problems, and Possible Solutions”. Version 2. https://scholar.google.com/scholar?q=%22A+Critical+Review+of+the+Refutation+Text+Literature%22

q1 · i? · review · r?

The authors describe their coding scheme for the review corpus, stating that "Coding reported here is not necessarily exclusive" and giving scientific and socio-political topics as an example of dual coding.

"Coding reported here is not necessarily exclusive. For example, refutation topics could be both scientific and socio-political and would be marked in both categories."

Allison Zengilowski 2020 (2)¶

Allison Zengilowski, Brendan Schuetze, Brady Nash, Diane Schallert. (2020). Coding Notes for “A Critical Review of the Refutation Text Literature: Methodological Confounds, Theoretical Problems, and Possible Solutions”. Version 2. https://scholar.google.com/scholar?q=%22A+Critical+Review+of+the+Refutation+Text+Literature%22

q1 · i? · review · r?

The authors explain that a semicolon separates data from multiple experiments in one paper, so a sample size of “100; 50” represents two experiments, and these were "averaged before calculating the overall mean and median".

"Studies with multiple experiment sample sizes were averaged before calculating the overall mean and median sample sizes for the full corpus."

Discussion¶

Related Claims¶

• Topic counts in the review's coding could exceed the listed overarching topics because misconceptions separated into individual texts were counted as multiple topics — related
• The review's corpus excluded studies that used refutation texts as stimuli but manipulated other variables so the primary purpose was not elucidating refutation-text effects, mechanisms, or moderators — related
• In the review's coding, equal total and per-condition sample sizes signalled either a correlational study or a repeated-measures experiment, and repeated measures were coded when a pretest entered analysis or a within-subjects design was used
• Posttests were coded as delayed if they occurred at least one day after treatment, with two-delay studies and studies lacking day/week delay metrics excluded from mean and median delay calculations