Monday, August 28, 2006

Academic Dishonesty

Today I had lunch with a fellow academic who has interests in evolutionary biology. I learned that a well-published evolutionary researcher, Anders Møller, was cited a while back for academic dishonesty. "A Danish government committee has ruled that one of the world's leading evolutionary biologists, Anders Pape Møller, is responsible for data fabricated in connection with an article that he co-authored in 1998 and subsequently retracted," the journal Science reported in 2004. According to my colleague, some professors don't have their students read Møller's work. That's a shame because not all of it is bad. In fact, Møller launched an intriguing line of research on fluctuating asymmetry that has generated a ton of papers, not to mention PhDs.

Fluctuating asymmetry (FA) refers to deviations in size of the left and right halves of an animal's body. Animals, including humans, who are very symmetrical (their left and right sides are near mirror images) are more attractive to potential mates. Møller's original line of research showed that barn swallows, who have long forked tails, mate more successfully if the forks of their tails are symmetrical. Similar studies have been performed with humans with the same conclusion: symmetry is beautiful. Some deviations can be visible to the naked eye, as was the case with Møller's research, however some are so vanishingly small that it's a genuine mystery how the signal could ever be received and therefore useful in mate selection.

I was very curious what my colleague could tell me about the reputation of FA research. He told me about a 1999 meta-analysis published in Nature. I tried to follow as best I could. It still intrigued me by the tiome I got home, so I looked up the article. And, I am glad I did. I left lunch feeling like the link between mate quality and symmetry was debunked 7 years ago and I was the last scientist to know about it. It turns out that the only part that was debunked was a piece that I wasn't really aware of - probably because I have been reading newer research based on a refined understanding of the phenomenon. Symmetry signals fitness. Scientists originally thought this occurred because homozygous genes caused asymmetry, however, the 1999 meta-analysis of 42 genetic studies showed this wasn't the case. The authors concluded that heterozygosity explained only a very small amount of variation in symmetry.

What is now thought to cause asymmetry?

Parasites, infection, pollution, poor nutrition... basically growing up in a harsh environment. Possessing an immune system that can't handle the deleterious side effects of testosterone is another one.

Stalk-eyed flies would seem to make a great candidate for investigating the link between symmetry and reproductive success because their eyes are at the end of inch long stalks that jut out from their head like antennae. A lot could go wrong when growing them so long, symmetricla ones seem like they'd be a good fitness signal. This group has found no connection however. Perhaps the female flies are so rapt with the length of the stalks that they can't see past them to the symmetry.

Nevertheless, there are enough studies that have found a link between mate quality, attractiveness and symmetry for me to continue to be intrigued by it AND to have my students read some of the literature.

Recently I read an old paper (1999) in Proceedings that showed a publication bias in research on how parasites change their host's behavior by making the parasite reproduce more prolificly. The author, whom I recognized from his research on cleaner fish grooming behavior, concluded that a paradigm shift occured. Scientists are no longer sure that parasites manuipulate their host's behavior, and papers published more recently are less likely to show it. This doesn't necessarily mean that parasites don't affect their host's behavior. It could be a publication bias caused by journal reviewers and editors who publish papers that don't show the effect. My conclusion is that science IS a biased endeavor in spite of efforts to be objective. I supppose one may even conclude there is no such thing as Truth in science, ever. I disagree. How about gravity?

I actually had a student say that gravity is a social construction during an oral defense!

It took all I had to fight against the force that drew my lower mandible to the floor.

This student said in the same oral that ALL women are mentally ill.

Although I thought her thesis was a pile of bovine waste, we graduated one opinionated woman that day.

Right now I am wondering if anyone is still reading this, and if so, whether you wonder how I plan to transition back to the point of my story about academic dishonesty. It certainly would be a tidy though round about story if that student had failed her oral exam because her committee discovered she had plagiarized.

She may have, but we didn't discover it. In fact, no one got flagged for plagiarism last year.

I bet it happened given these statistics on academic dishonesty based on self-reports from established scientists.

From Nature (2005) by Martinson et al.

1. Falsifying or "cooking" research data (0.3)
2. Ignoring major aspects of human-subject requirements (0.3)
3. Not properly disclosing involvement in firms whose products are based on one’s own research (0.3)
4. Relationships with students, research subjects or clients that may be interpreted as questionable (1.4)
5. Using another’s ideas without obtaining permission or giving due credit (1.4)
6. Unauthorized use of confidential information in connection with one’s own research (1.7)
7. Failing to present data that contradict one’s own previous research (6.0)
8. Circumventing certain minor aspects of human-subject requirements (7.6)
9. Overlooking others’ use of flawed data or questionable interpretation of data (12.5)
10. Changing the design, methodology or results of a study in response to pressure from a funding source (15.5)
11. Publishing the same data or results in two or more publications (4.7)
12. Inappropriately assigning authorship credit (10.0)
13. Withholding details of methodology or results in papers or Proposals (10.8)
14. Using inadequate or inappropriate research designs (13.5)
15. Dropping observations or data points from analyses based on a gut
feeling that they were inaccurate (15.3)
16. Inadequate record keeping related to research projects (27.5)

I love Number 15. I wonder how often that happens in psychology research.

According to the Dutch group that cited Møller in 1999:
"Scientific dishonesty includes actions or omissions in research which give rise to falsification or distortion of the scientific message or gross misrepresentation of
a person’s involvement in the research, and includes:
1. Fabrication and construction of data.
2. Selective and surreptitious discarding of undesirable results.
3. Substitution with fictitious data.
4. Consciously misleading use of statistical methods.
5. Consciously distorted interpretation of results and distortion of conclusions.
6. Plagiarization of others’ results or publications.
7. Consciously distorted reproduction of others’ results.
8. Inappropriate credit as the author or authors.
9. Applications containing incorrect information.
In order to label a conduct as scientific dishonesty, it must be possible to document that the person in question has acted deliberately or exercised gross negligence in connection with the activities under consideration."

It would be interesting to have a list of all of the scientists who have been caught engaging in academic dishonesty, but even more interesting to know what percentage of ACTUAL cheating they represent.

Have you ever discovered a student cheating?

What happened and what did you do?

1 comment:

Alasdair said...

Well, it wasn't a student, but I did catch a reporter plagiarizing news stories, which strikes me as very similar.

I reported the incidents (which I suspect were much more widespread than the three or four examples I found on my quick survey) to a senior editor and the reporter was given a slap on the wrist. To my knowledge, though, the light reprimand ended that reporter's cheating.

Because students (and journalists) are so well versed in what constitutes plagiarism, I say any genuine infraction should result in an automatic zero on the assignment and a mandatory academic honesty hearing.

If that means an A student gets a D-minus (because they cheated on their final paper), so be it. If worse sanctions come from a hearing, that's just the way the cookie crumbles. Real life - not counting the lucky reporter - punishes cheaters. School should, too.