Remember that "Horns" study that got all the attention, substantial corrections have now been posted to the results, discussion, figure captions and author conflict of interest statements. But here's why it's still misleading and the journal is complicit:
@GidMK @scireports
-
Show this thread
-
The graph of "distribution" of cases in age groups as a percentage is wildly misleading. It looks like there's a higher percentage in the youngest group and oldest group, right? Very misleading.pic.twitter.com/6gh9u8BmVU
1 reply 0 retweets 1 likeShow this thread -
The problem is this graph is "what percentage of total cases fell into each age decade", it's problematic because they didn't recruit equal numbers in each age group! The largest groups were unsurprisingly the first and last. This is a plot of numerators with no denominators 3/
1 reply 0 retweets 2 likesShow this thread -
As an analogy, if an author took a sample of 500
@SpringerNature papers,100 Elsevier papers,100 Wiley papers, and graphed the "percentage of retractions occurring in each journal" showing a much higher "percentage" of the retractions in SpringerNature they would probably sue! 4/1 reply 1 retweet 1 likeShow this thread -
It seems bizarre to me that the editors accepted a materially misleading graph, and now have accepted this clearly invalid and misleading graph that exaggerates the main finding (central to the author's conclusions) with a slightly rejigged caption. 5/
1 reply 0 retweets 1 likeShow this thread -
I pointed this out to the editors of
@SciReports and even provided them with what the graph should look like (much flatter) when plotted in a sensible way as prevalence in the authors' preferred age groups from original data. But that wouldn't support the authors' conclusions 6/1 reply 0 retweets 1 likeShow this thread -
I'd write a letter to the editor but
@SciReports doesn't accept them, or any other commentary! I think we have to question the bona fides of a journal that does not allow criticism of their papers in any form. Science is a conversation, not a series of proclamations. 7/71 reply 2 retweets 5 likesShow this thread -
Kyle Sheldrick Retweeted Kyle Sheldrick
Kyle Sheldrick added,
Kyle Sheldrick @K_SheldrickIf anybody is interested this is what the graph should look like. I provided this to the editors or@SciReports/@SpringerNature along with the original data set some months ago. This has not "slipped through", this is a deliberate decision. https://twitter.com/K_Sheldrick/status/1176277907825516544 … pic.twitter.com/PAesM7DqY3Show this thread1 reply 1 retweet 1 likeShow this thread
Something fun for you @jamesheathers
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.