Is it possible to confirm this information? And what would it mean?
The questions I ask are as follows.
Is the graph real? Can it be reproduced by anyone?
What data is used?
If those are satisfied, so that it actually represents a data analysis, what does it mean?
2 comments:
Fell into your blog by accident ... and just had to comment on this because it's a particular big beef with me. In short... the devil is in the details. Reproduceability (chrome says this isn't a word, and I say it is) means little compared to understanding all the variables that influence a set of data. This is one of the major problems with the way EBM (evidence based medicine) is managed ... they focus more on the statistics than one the crap that is all mixed in confounding the statistics. There are other even bigger problems with the way they (those who get stuff by the FDA) do EBM (in particular with regards to cost and exclusivity turning it into profit based medicine), but this (drawing conclusions about a set of data points about which we know next to nothing) is one of the whoppers. In fact the atrocities committed in the name of EBM have completely tarnished the phrase "Evidence Based Medicine" for me, and it should for everyone, so perhaps "evidence based science" should also be suspect. What's better than evidence? Awareness? I don't know ... I'm looking for a word that includes both evidence and an awareness of all the factors that effected the evidence. I'm looking for a word that looks at evidence with respect to the bigger picture of all previous observations supporting and refuting a perceived phenomena. Still working on that one. Not "omniscience" because that insinuates perfection... but what term is as close to perfect awareness of all mitigating factors as possible? Anyway ... that's what I think is missing in science today.
Just came across this (while deleting spam comments). You make a good point. Science is not easy, but corrupting it to make a profit is.
Post a Comment