GOHEAVY.COM Forums

We have made several changes to our policy here at GOHEAVY. Here they are:
  • POST WITH YOUR REAL NAME (FIRST & LAST!)
  • POST WITH YOUR REAL EMAIL ADDRESS
  • NO FLAMES
  • NO ADS
  • NO TROLLING
  • PLEASE KEEP BAD LANGUAGE TO A MINIMUM
The moderators of this forum reserve the right to delete any post that is deemed inappropriate for this forum.

GOHEAVY.COM Debates Forum

Re: if the data showed that then

Posted By: James Chantler (66-193-79-4.static.twtelecom.net)
Date: Tuesday, March 11, 2008, @ 9:59 a.m.

In Response To: if the data showed that then (Scott Slaterly)

Absolutely right. I am well versed in statistical analysis and methods and do not criticize them. What I criticize is the integrity of data collection and poor assumptions made in deciding how and where and when to conduct statistical analyses. An example is here where you made an incorrect assumption that I was talking about powerlifters - the "lifting belt" data was collected among warehouse workers and as you correctly point out, is not necessarily transferable to powerlifters....

See what I mean!?

Regarding medical testing - again, no problem with the math or statistics, just the deception. Quite often, a new drug will "reduce risk of a certain type of cancer by 100%" which sounds fantastic. What is not said is that the chance of getting the disease goes from 1 in a million to 1 in 2 million. This is indeed a 100% reduction, but the chance of NOT getting the disease goes from 0.999999 to 0.9999995 which is not a statistically significant change......

What I'm saying is that statistics are very useful, but extra thinking still needs to be done. Simply having a large number of people with a certain opinion doesn't make it any more right or wrong.......

Messages In This Thread

GOHEAVY.COM Debates Forum is maintained by Administrator with WebBBS 5.12.