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.
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.
I believe strongly that psychological , journals, articles and reports should be written for individuals thus making them accessible by the general public. yes some of the words used would be the psychology specific but the statistics and evaluations of the results should be accessible to all walks of life thus written in a straight forward fashion , explainable to an individual outside the field.
The statistics element is often shown in adverts or in journals use statistic in order to illustrate their point . One example of this is “the Virginia twin study “: this study tested 1412 pairs of caucasian twins on how genetics and the environment can impact on their development during adolescence. The statistics in such studies help the reader to quickly grasp the findings in order to make sense of the possible impact they may have on the world or he selves as statistics shown as percentages are an easy, reliable, straight forward way for individuals to engage with the present findings.
Further more as psychology is somewhat a social science and not just a neural science i believe that i is in psychologists best interests to pass on their knowledge to others thus creating an accessible platform where information can be recieved freely.
among psychology students there is a common misconception that qualitative methods or research is substantially easier than spss or statical programs.
qualitative data is extremely time consuming as it requires te investigator to meticulously analyse the data for themes which represent the research topics and highlight a substantial viewpoint .
janet llieva, steve Baron and Nigel Heculum conducted an online in the pros and cons of marketing research. they found that the main problem with qualitative data is that it is not always representative of the population thus not generalisehttp://www.mendeley.com/research/online-surveys-in-marketing-research-pros-and-cons/.
further studies were conducted by Patrica Johnson and john t. Winstone. this study brought together the pro and cons of data analysis software for qualitative research. they fund that there were many advantages such as improved validity, audabilityand flexibility of the data and the methods used to analyse the data. he concerns were mainly as the software is deterministic and often uses rigid processes to analyse the data it also only focuses on the volume and breath of the data rather than he depth and the meaning of the data which in my eyes would hinder the development of a qualitative research as qualatative is ment to focus on the detail within the data which makes Qualiataive analysis what it is.http://onlinelibrary.wiley.com/doi/10.1111/j.1547-5069.2000.00393.x/abstract
on the whole qualitative research would say is even more complex and demanding than statistics and numerical data due to its subjective nature and th fact that individuals view and present thought can influence the outcome of the analysis.
till next time
As a trainee psychologist i often ponder the question does the end justify the means when conducting research. Is it ethically right to stretch the nerves and break the boundaries of your participants trust through deception in order to obtain data or should this ludicrous practice be stopped.
this question is often at the fore front of my mind when i am thinking about the option of covert observations. it has been noted by individuals such as Roger Homan the possible advantages of sch methods. he described this particular method as “pragmatic expedient, ideally non reactive way to gie access to a secret transaction” as the participant wa free from disturbance and inhibition” . Mr Hammond states a valid and accurate point as covert observations I general do not harm the individuals being observed but rather draw on on methods in order to see natural behaviours which may not be presented within the confines of an experiment. although I still feel disconcerted at the prospect of decieving participants in order to obtain valid and reliable results. http://jstor.org/590062
One individual wo shares my view is Alison Lurie. Lurie states that ” particular ironic version of the means justify ends argument with an excuse that we were seeking truth , we were proposing to lie ourselves blind to the truth seekers ” . bold statements like this highlight the fact that it is never okay to decieve your participants through covert methods no matter the gain you may achieve scientific progress it may hinder. Further more C.D Herrara states that ” deception rampant in society…… methods no more immoral than the behaviour it prevails …. morally indistinguishable from deception.” http://jstor.org/590062 http:/onlinelibrary.wiley.com/doi/10.1111.j.1468-4446..1999.00331.x/abstract
strong poyant arguments such as this have led me to believe that however helpful covert analysis may be in helping individuals observe naturalistic behaviours another thing, it still counts as deception as the participant’s are often often unaware tat they are partaking in n experiments thus breaking ethical guidelines such as the right to with draw and the need for full informed consent.
But in the eyes of some “the goal is justifiable”; Denzin 1968. ccj.sagepub.com/content/http:1111.12/2/97.short
till next time