Profile picture
Andrew Althouse @ADAlthousePhD
, 8 tweets, 2 min read Read on Twitter
Question time (some of the specifics that follow have been changed slightly, but hopefully not enough that it affects the answqers): suppose that you've reported results of an RCT with HR=0.82 and p-value=0.08 for one of the trial's secondary outcomes.
You used the phrases "tended to ____ less" / "modest reduction in ____" / and "there was a small ____ benefit" to describe these results.
Not just one, but several of the manuscript reviewers say things along the lines of "The ____ results are not statistically different" and "The reported benefit of ____ has been overstated. There is no statistical significance"
I do not believe the phrasing used was unreasonable. The incidence of ____ was lower in one of the treatment arms. It was not "significantly lower" at the 0.05 threshold but it was, you know, lower. It's a descriptive word, folks.
Authors should be permitted to state something like "The incidence of ____ was lower in XXX versus YYY" - they are reporting a fact about the results that occurred in the trial. Whether this constitutes strong evidence that the apparent difference is reflective of a true benefit admittedly another question. But saying that all findings with p>0.05 must be called "no difference" is clearly absurd. If the incidence of an endpoint was 15% in one arm and 20% in the other, there's NOTHING wrong with saying that it was lower in one arm than the other.. long as your phrasing is correct. You're reporting the results that occurred in the trial. 15 percent is lower than 20 percent. The statistical hypothesis test that accompanies this is meant to address whether the observed data are sufficient to conclude that...
...these data would be unlikely to occur under the null / are incompatible with the null hypothesis of no difference.
Missing some Tweet in this thread?
You can try to force a refresh.

Like this thread? Get email updates or save it to PDF!

Subscribe to Andrew Althouse
Profile picture

Get real-time email alerts when new unrolls are available from this author!

This content may be removed anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member and get exclusive features!

Premium member ($30.00/year)

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!