Excess significance and power miscalculations in neurofeedback research
Bibliographic record
Abstract
We use the term "repeated measures (mixed) ANOVA" to describe a study design with one independent measure and one repeated measure (e.g., two independent groups measured at two time points).GPower uses the term "ANOVA: Repeated measures" for this type of design.4 Kohl and coauthors (including me-RTT) were unaware of this default option in GPower.I was only led to become aware of this issue when reading the fMRI-nf review and noticing a surprising amount of statistical power.5 Kieslich (2020) provides a detailed explanation of how GPower calculates f for the different options for repeated measures (mixed) ANOVAs.The GPower team informed us that Cohen proposed Cohen's f for between-subjects designs and that this is how GPower defines f when using the default option, regardless of the ANOVA design the user selects.6 We performed our calculations in R.They are all available in our open code.We reproduce Table 3 from Tursic et al. and Table 5 from Kohl et al. using the sample sizes and statistical tests which each review provides in their supplementary material (i.e., we did not re-extract this information from the original studies or check if the statistical tests in the original studies were appropriate).The WebPower package we used assumes a correlation of 0.7 for repeated measures for ANOVAs, which is slightly more conservative than the 0.8 correlation used in the reviews.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.520 | 0.866 |
| Meta-epidemiology (narrow) | 0.002 | 0.003 |
| Meta-epidemiology (broad) | 0.005 | 0.007 |
| Bibliometrics | 0.024 | 0.024 |
| Science and technology studies | 0.002 | 0.015 |
| Scholarly communication | 0.009 | 0.013 |
| Open science | 0.008 | 0.011 |
| Research integrity | 0.006 | 0.007 |
| Insufficient payload (model declined to judge) | 0.012 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".