Bad Science and Bad Analytics That Misleads–TED Video Takes on Advertising on Steroids and Danger of Mis-Matched Analytics
Posted Sep 29 2011 2:58pm
This is great and goes along with some of my recent posts with analytics on steroids for marketing only and also makes a point about what gets put in journals today too. Dr. Ben Goldacre, talks fast so listen up carefully, but he gets it. Fish Oil pills is one he takes on as well as coffee and cancer, yes it does and no it doesn’t. On the pills they calculate and project how the children would have done without the pills..bunk.
I am so glad this is finally getting attention so consumers may start asking questions if posed with something you know is weird and yet it persists in the papers every day, day in and day out! You have to laugh at some of this for goodness sakes. Now if you want to go the next step and see how you can be deceived, read the post below and mathematical deception is the next step when steroid marketing kicks in. Even the real scientists hate this stuff and it adds confusing for advocacy groups to untangle.
How all this really gets a stronghold in society today is the use of math which is what it considered to be the most accurate way to prove truth, well it was until modern day algorithms kicked in, where writing the unreadable starts to be an issue. So what is WellPoint going to load up in Watson…that needs scrutiny to work. Again worth a mention I had to laugh when one of our lawmakers asked Google about their “secret sauce”..come on folks digital literate people call those algorithms:)
We know our beliefs do this, look at what politicians do with this and he talks about pharma distorting reports as well with stacking up the bad stuff so the good stuff looks better or the negative information goes missing in action. He says he’s a nerdy doctor and does a lot of reading. One drug he uses an example that 76% of the data was withheld from the doctors. Anti-depressant data drug information has been withheld.
Do you wonder why doctors get upset when they are graded on the drugs they prescribe? It’s gets hard when you have Quants and health insurance companies breathing down their backs all the time doesn’t it?
One more comment, add on this insanity with a credit agency trying to tell doctors that they have the “secret sauce” to determine if patients will take their meds and they blatantly sell this to the pharma companies doing clinical trials and insurance companies.
Have you been mathematically deceived of late? It’s a good question to think about and that’s why I write such articles like this, to open eyes and hopefully get people to think before the algorithms for profits completely take over, don’t believe all the stats you see out there today and your social networks are running a gambit trying to also convince you as such with flawed profiles, etc. they collect with data scraping and it just aint so all the time. Algorithms are only as good at the data it sits on top of. It’s interesting to crunch numbers and get stats and nothing wrong with that but when flawed data as such is brought down to the individual level to risk asses and judge, the numbers are not right. Be aware as this is part of the insane world of algorithms for profit distorting the good science we have out there. BD