The Naked Scientists

The Naked Scientists Forum

Author Topic: Are biostatistics useful in a modern world?  (Read 3167 times)

Offline thedoc

  • Forum Admin
  • Administrator
  • Sr. Member
  • *****
  • Posts: 511
  • Thanked: 12 times
    • View Profile
Are biostatistics useful in a modern world?
« on: 13/12/2011 17:21:18 »
Given that people are increasingly using crowd source data in their research, are old style stats still up to the job?  Arnoldo Frigessi explores...
Read a transcript of the interview by clicking here

or  
« Last Edit: 13/12/2011 17:21:18 by _system »


 

Offline CliffordK

  • Neilep Level Member
  • ******
  • Posts: 6321
  • Thanked: 3 times
  • Site Moderator
    • View Profile
Are biostatistics useful in a modern world?
« Reply #1 on: 14/12/2011 05:25:26 »
I don't think statistics is obsolete.

One may still do clinical trials with a new drug, and initially have a well designed study with only a few individuals taking part in the study. 

The examples of computer media and networking opens a few big issues.  How much weight should anecdotal evidence be given?  What about biased reporting?  Those with problems tend to be far more vocal than those without problems.

How does one control for the placebo effect (both positive and negative) without a randomized trial?

Perhaps large cohort studies like the Framingham study could be used to create a model of typical biases that one might encounter with social media.  Then one could use the data from the cohort study to evaluate the strength of trends and biases.
 

Offline Devilmunkey

  • Jr. Member
  • **
  • Posts: 22
    • View Profile
Re: Are biostatistics useful in a modern world?
« Reply #2 on: 12/01/2012 01:17:01 »
Very much so still relevant. Even meta-analysis studies of prior research can yield useful data. This is particulary true in clinical medicine and epidemiology where statistical analysis can highlight unseen flaws in clinical practice or established control measures. This brings to mind a study I read about the treatment of head injuries with steroids, which is/was standard practice, the statistical analysis of which showed that this treatment was associated with a significant increase in risk of subsequent death (I apologise for mentioning studies I do not have a link/reference for, but it should be possible to track down via Pubmed)
 

The Naked Scientists Forum

Re: Are biostatistics useful in a modern world?
« Reply #2 on: 12/01/2012 01:17:01 »

 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums
 
Login
Login with username, password and session length