Bias is pretty much unavoidable during any experiment conducted by psychologists, and the range of bias is extensive and can be caused by a variety of things. However an important and, unfortunately, a forgotten bias, (one, to be honest, I was completely unaware of till I began researching for this blog topic) is the bias where research is overlooked and not published in journals and not made known to the scientific world. This bias is known as Publication bias, but also goes by the title “the file drawer problem”. This has been defined as “ the tendency on the part of investigators to submit, or the reviewers and editors, to accept [research] based on direction or strength of the studies findings” (Dickersin, 1990). This suggests that only the studies that show a positive effect, or provide strong proof against a null hypothesis get published. Also in many cases it is those that do not support the findings of other’s in the field that fail to be published this therefore can cause a biased pool of research.
When others are researching a topic area that may have been affected by publication bias, it is likely that they will only find published work of those that found positive results. If they were to take this research and review it or create meta-analyses, it could result in an overall bias, when in fact there may be hundreds of unpublished work that reports to opposite. For example Simes (1986) created a review of research on the effect of combination chemotherapy versus a single agent in the treatment of ovarian cancer. He looked at published trial research and stated that combination therapy was the best way to treat it. However, when another analysis was done on all research, including unpublished research it was seen that, in fact, there was no benefit at all in the use of combination therapy.
There is a range of factors that can be the reasons for why publication may occur. They are:
- Poor research design quality
- Small sample size
- External funding
- Negative findings
- Rejection from journal editors
- Failure of authors to submit their research to be published
The final point is where the term “file drawer problem” comes from. Many psychologist and researchers just simply fail to send their work for publication, this may be down to incomplete research or fear of unimportance of results, or simply because researchers become too busy or lose interest. (Dickersin, Min & Meinert, 1992)
To conclude I think, looking into this subject matter, that this is potentially a huge problem, especially concerning the pollution of evidence pools. If research is stopped from being published or is researchers are fearful of publishing research that contradicts other findings then we are going to find ourselves in a place where we are believing things that simple are not true. A variety of research has to be available to us all.
I agree with you that the file drawer effect is a big problem in psychology although many people forget about it. I, like you, did not know about this before I wrote my blog on this topic. Then, I came to realise how important it is to come up with solutions for reducing this problem. There has been a number of studies over the past years that looked at the file drawer effect.Most of them looked at the research papers and tried to identify the number of them that published only significant results. The outcome of such studies showed that “Positive results in research studies overall, became 22% more likely to appear in scientific journals from 1990 to 2007,” Also, Overall, 70.2% of papers were positive in 1990–1991 and 85.9% were positive in 2007. ” This means that on average, the odds of reporting a positive result increased by about 6% every year. This is a big number if you think about it. All these results together indicate that the publishing process became very competitive and the journals gain a lot of power over researchers. However, even though competition is a good thing, I don’t think it is like that in this case. So, yeah, I think we need to come up with some good solutions for the problem. For example, some of the websites where you can publish the results of your study no matter what they are are pretty good.
Reference:
http://content.usatoday.com/communities/sciencefair/post/2011/09/science-studies-neglecting-negative-results/1#.T188Kodw-So
[…] 3. https://stach22.wordpress.com/2012/03/09/publication-bias-and-the-file-drawer-problem/#comment-50 […]
Ill be honest before this blog I had never heard of this problem. It makes you question how many times this actually occurs and the implication it has on psychology as a science.
Take for example if a individual doesn’t publish some of their work that possibly didn’t show much of an effect so was deemed a failure. There is nothing stopping someone reading that with fresh eyes in the future and having an idea on how to modify the design so an effect may be found. This furthers psychology as a science, all from one study that someone thought not good enough to publish.
A very well written and interesting blog.