Supports· strongMPI-E015
Bösch 2006: micro-PK meta-analysis — tiny effects, huge heterogeneity, publication bias
- what the source states
- A meta-analysis of 380 REG-intention studies found a very small but significant overall effect, extreme heterogeneity, and a strong small-study effect, concluding that publication bias appears to be the easiest and most encompassing explanation for the primary findings.
- what we infer
- The base-rate backdrop for the whole case: the broader literature the matrix experiments grew from is consistent with selection processes acting on null effects. Radin and colleagues' published rebuttal (same issue) contests the smallest-studies weighting — the dispute is noted, not resolved, here.
- limitations
- Targets classical hit-rate micro-PK, not the matrix paradigm itself.
- A published comment (Radin et al. 2006) disputes the meta-analytic weighting choices.