So if you haven't already figured it out, I'm a fan of CBT. I'm actually a huge fan of CBT. Which is odd, considering the fact that I have never actually READ or SEEN a CBT manual. I know what CBT is, I know that it is based on SCIENCE, when a bunch of other forms of therapies are not, so I decided I'm a huge fan of it. This week made me happy because I got to learn more about CBT and I got lots of evidence, in the form of meta-analyses, to support the efficacy/effectiveness/efficiency of CBT.
Now that I've made it clear that I am honestly a fan of CBT, I want to take a little bit of time to complain about meta-analyses. I have decided that they are a very sneaky way to report information. I don't think that the authors of meta-analyses are always purposefully being sneaky...but I think it is much easier to fudge a few facts or hide a few errors in a meta-analysis than it is in a report on a single experiment.
Here are a few instances in which I believe meta-analytic authors are being "tricky" in the Butler (2006) meta-meta-analytic article we read for class:
-It is mentioned that Parker et. al. did not report the criteria they used to select studies for their review paper and as a result, "it is difficult to interpret their conclusions" (p. 20). In my mind, this means the authors were probably doing something sneaky when selecting which articles to include. It is very possible that they only chose articles that supported their own opinions. This seems especially likely when you take into account that "researcher allegiance accounted for half the difference between CBT and other treatments" in a different meta-analytic study done by Dobson in 1989 (p.20).
-In a 1998 meta-analysis, Gloaguen et. al. found that CBT had significantly better outcomes than medication for depression. However, Gloaguen et. al. "included some early studies comparing CT with medications, which had methodological features that favored CT" (p. 23). Was this done on purpose? If so, it is a very sneaky way to prove your point; purposely include studies that have methodological advantages for whatever you favor.
-Ten studies (well, ten meta-analytic articles) were excluded from Butler's (2006) meta-meta-analysis because they were written in a foreign language. What if all ten of these articles pointed to results that were extremely different from other results? What if they would have greatly impacted the meta-meta-analysis? We will never know. We will also never know if Butler honestly did not include these studies because of a language gap or if he did not include them because there was something in these articles that he wanted to hide. I doubt this is the case, but it is a possibility. Sure it would be expensive to translate the articles to English, but I think it would have been a good idea if Butler et. al. had the funds.
Perhaps I am just paranoid and all of these things I find "sneaky" are actually very normal. However, I really do think meta-analyses are a great way to try to prove your point without really having to explain every little aspect of your procedure.
In other news, I still love CBT.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment