The paper indicates that a meta-analysis of primary effects in published research shows a significant effect, but for secondary effects in published research (t

The paper indicates that a meta-analysis of primary effects in published research shows a significant effect, but for secondary effects in published research (those not related to study aims), no effect. Ok. Both sorts of reports (of primary and secondary effects) have their own sorts of biases. In general, while there is tendency to only publish and report significant effects (and this applies for both primary and secondary outcomes), study design matters, and a study may not be well designed to detect, or account for other factors, for secondary outcomes- in N, choice of population, control procedures and variables used, or even how accurately and systematically that data was recorded. Does make it hard to evaluate.