Why is the “file drawer problem”a problem ????

this is  a question that as a trainee psychologist i have pondered for a while …. what is the file drawer phenomena and why is such a problem when it comes to publishing major research.

the file drawer  problem/ phenomena which was coined by Robert  Rosenthal in 1979, is to do with publication bias ; essentially it  draws on the issue which arises when researches reject their null  due to the statistical insignificance of the results thus leading less  likelihood of their   data  being published. these negative or inclusive results often never reach the  public reasachers , editors, pharmaceutical companies and medical journals are wary of publishing such data which may go against/ contradict common belief.  this in my eyes is a major problem as it shows a direct representation of the data field thus leading to direct problems when it comes to continuation if experiments or extreme bias of data. http://www.ma.utexas.edu/users/mks/statmistakes/filedrawer.html

the file drawer problem has major implications when it comes to question validity as the effect reduces validity: the  literature produces by the file drawer  effect is largely unrepresentative of the population of the studies  for example Merck’s research on Vioxx (arthritis drug) had the results that showed that the drug wasnt as effective. this research is deretimental both  those adminstration and taking the drug as the pseudo result gives an inaccurate image of its true effect as  it largely distorts the data thus giving a bias slant on the results.

 http://www.meta-analysis.com/downloads/PBPreface.pdf

 Iyangar and Greenhouse in 1988 suggested a plausible way of fixing publishing bias; the sensitivity of effect size( 0.05). as this scale can directly show the level of significance an entity posses, thus creating a stricter more efficient presentation of the data without ruling out haf he results ………  the more stringent the alpha level , the better the test  of significance  is. http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.ss/1177013012

to conclude in my opinion the file drawer effect is an out dated practice which should be replaced for a method that best equips the psychological community.

Advertisements

2 thoughts on “Why is the “file drawer problem”a problem ????

  1. sarric says:

    I agree the File- draw problem of only publishing results that show significance and not those that fail to reject the null hypothesis, is a serious problem within research. It could mean unnecessary time and effort is spent researching topics that have already been well-researched, yet just not reported. However I disagree the file draw effect is an outdated practice. I think it is actually becoming more and more common. As the funding government gives to scientific research reduces year upon year, investors are becoming more and more interested in subject area’s sure to guarantee significant findings. Significant results mean an economic return for the providers investment. With scientific progression unable to happen without an economic injection, it is to be expected insignificant findings will be left on the shelf, whilst significant ones are published and researched further.
    As you mentioned Increasing the alpha level makes significant tests more stringent. This is a way of reducing the chance of a type 1 error (accepting a result as significant when actually it is not). Similarly the ‘fail-safe file drawer analysis’ and ‘funnel pots analysis’ are methods used to detect such publication bias’s (*). Such techniques may be the future for reducing errors within results, thus increasing the validity and reliability of science. However, whether or not insignificant findings get published, is in part down to the funding providers and researchers drive for economic success. The whole world revolves around money…!

    * http://www.ma.utexas.edu/users/mks/statmistakes/filedrawer.html

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s