Author: Roger D Davis
Date: 17:20:48 09/03/02
Go up one level in this thread
>I don't know if you faked the results to look better or not. Maybe I don't >want to know. But whatever be of it, there is little scientific ground >to keep them standing, IMHO. > >-- >GCP Wow, this comment is in exceptionally bad taste. You don't question the scientific integrity of a researcher lightly, particularly in a public forum. I can tell you my own experience. My disseration was on methods of creating shorts forms of personality tests. The test I choose to work with was one on which we had a large amount of archival data, about 1000 subjects. The whole point of my dissertation was to show that short forms of psychological instruments are psychometrically unstable, and that while the scores on 10-item abbreviated versions of a 30-item scale might indeed have a high correlation in group data, at the level of the individual, the variability of scores was such that a great many people who were classified as having a personality disorder on the longer scale were not thus classified with the shorter scale, and vice versa. In other words, the shorts forms were unusable in a clinical situation. Unfortunately, my committee missed this point of my dissertation. And since I wasn't particularly interested in arguing with them, and since dissertations typically just sit on the shelves gathering dust anyway, I let it go. My experience since then has confirmed my experience with my committee: The the researcher usually knows his or her research better than anyone else. I have seen many journal editors insist on including details which have little or no true relevance to the article, or insist that this be altered or that be altered, even though, from the perspective of the author, this created some form of distortion. Typically, the power dynamics at work are such that junior authors make changes asked of them by senior professors simply out of respect, and senior professors make junior authors jump through such hoops to pay their dues. That's life. In psychological journals, there are failures of replication all the time, and the reasons are well known. They typically have to do with two researchers who use a slightly different way of doing things, and end up with divergent results. Sometimes the operational definitions of their constructs are slightly different, sometimes their methodologies are different, or a thousand other things. I'm convinced that faking of results is, thankfully, quite rare in scholarly journals. Roger Davis, PhD
This page took 0.01 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.