A hypothesis developed based on the data is often more likely to be true, than if you had not used the data. The problem is not whether the hypothesis is true. The problem is the hypothesis was not severely tested. You can't *claim* it is true until you test it on new data.

Often a single Registered Report is more informative than a meta-analysis. The meta-analysis will show a non-zero estimate and we will not know if it is due to bias. Heterogeneity is huge, so the main recommendation of a meta-analysis is that future research is needed anyway.

There are 3 PhD's hired on these projects. I will also hire a post-doc, and there is money for a fifth team member, if I think we need one. So, lots of space to follow up on other ideas as well.

Happy to announce @SciMed_Football’s new 𝗣𝗿𝗲𝗹𝗶𝗺𝗶𝗻𝗮𝗿𝘆 𝗥𝗲𝗽𝗼𝗿𝘁 category for small-sample studies, a pragmatic step to address the spread and misreporting of underpowered football research. Grateful to all co-authors for months shaping the submission requirements.

It is because you want this to be true. Whenever you see something you want to be true, time to turn on your internal critic.

My paper with @lakens and Anna van 't Veer - “Replication value as a function of citation impact and sample size” - is published in @Meta_Psy! open.lnu.se/index.php/metaps…

"the very core of science – an institution that promotes mutual criticism, while accepting our fallibility" x.com/lakens/status/19834384…

New study in @JSportsSci exploring a potential strategy for increasing affective responses during exercise while emphasizing the need to rigorously test related auxiliary assumptions and determine a SESOI for confirmatory research: tandfonline.com/doi/full/10.…

Of course, here it is sportrxiv.org/index.php/serv…

It is ironic how confidently this incorrect explanation of confidence intervals in presented. Anyone looking to brush up on stats @lakens has some fantastic free courses available on these exact concepts.

The effects of lengthened-partial range of motion resistance training of the limbs on arm and thigh muscle area: A multi-site randomised trial doi.org/10.1080/02640414.202…

Not having to deviate from your preregistration is how you impress peers. You demonstrate you know something, because knowledge allows you to make good predictions. The more deviations, the less impressive the tests in a study are.

Too often, I see people talk about a replication as if the first study has established something, and the replication study is a double-check. What people often fail to understand is that we do not do replication studies to *check* a finding, but to *establish* a finding. 1/x

Thanks for trying! It is still amazing how slow the uptake of equivalence testing is, given that it is just a 90%CI of the same test you ordinarily do, but against a smallest effect if interest. Yes, setting it is a challenge, but important enough to spend some time on!