I don't know if anyone wants to read book reviews, but I like to know what people think of books, so I am going to do them from time to time. This is of the book I just finished, Intuition by Allegra Goodman. I would give it a rating of fair to good.
A small plot summary (no spoilers): This book was set in a cancer lab examining ways to cure cancer using mice. The main problem in the book was the question of research integrity.
The good: The characters were exceedingly well developed. I really cared about what would happen. For that reason, I found it hard to put the book down once I started reading. The fundamental questions the book dealt with were important and interesting to me. In addition to the main issue (research integrity), there were many interesting side issues regarding supervision of students and postdocs, mentoring from researchers at later levels of their careers, the research process (when to publish, how much to extrapolate from data).
The bad: The storyline was obvious to me. I could see where she would end up. However, she tried to hide it by obscuring details (or leaving them out). It was hard to tell which she was trying to do. I do not like it when all of the relevant facts are not presented or when the way something ends up is not clear from the premise laid out in a book. I think she would have been better served by a less obvious ending.
SPOILERS FOLLOW
To better explain my point about the plot: Two of the postdocs were dating, then broke up. One of them made a big discovery, became famous. The other was jealous and tried to find evidence that the first had falsified data. In the end, (about 20 pages from the end of the book), the postdoc admitted to leaving some of his animals out of the data he published. But this is an animal lab. I can only imagine all of the animals are accounted for. Data for 13 of them could not simply vanish without a trace. And there was no discussion of the fact that the data or whereabouts of 13 mice were unaccounted for. I do not believe that would be tolerated. I know in my lab, if 13 subjects' data (humans in this case) were to disappear someone would be answering hard questions.
But the kernel of a good story was there. The author brought up the way "intuition" in research can lead you to false conclusions, that you must follow the data. It does happen where one researcher will not believe another's data, for whatever reason. Most often, the data is fine. It is the researcher's belief, "intuition," that the other did something wrong which is false. And that is an important point. I kept hoping that would be the point of the book. It is rare where data are actually falsified, at least in my experience.
Subscribe to:
Post Comments (Atom)
5 comments:
o.O Even in animal lab studies, all data must be tracked. There's a reason why lab notebooks have to be carefully numbered, and notes written in pen. And forget it if the book claimed that data was deleted from a computer system. *forehead smack*
Sounds like an interesting book.
But, I worked in an animal lab for 18 months in college (one of about 20 undergrad RAs), and I wouldn't find it too surprising to hear that some data went "missing" from such an operation (intentionally -- it might be harder to do it by accident, at least on a major scale). We had I think two sets of 3 or 4 rooms of 100+ rats in each. It was a diet/cancer study, and each animal had to be fed a specific amount of food of a specific type each day, with the precise amounts varying from day to day. If I remember correctly, an animal that was given the wrong amount of food or the wrong diet on a given day was culled, so eliminating problem animals would simply have been a matter of claiming that some bonehead undergrad RA (e.g. yours truly) had over/under/mis-fed it.
What, you don't think the postdocs were down there measuring out food (to the nearest 1/10 gram) and putting it, by hand, into the cages of hundreds of sometimes ravenous, eventually cancerous rats, do you?
Yes, there was a *lot* of data tracking, but there was also a *lot* of data to track.
That said, the two cases of alleged academic fraud that I've read anything in detail about were a lot more prosaic, and easily identified. If I remember correctly, in one case the researcher simply re-used a figure from an earlier study that had come out "right", while in the other case the claim (disputed by the accused postdoc) was that the postdoc had simply written down false numbers, in ink, in the numbered-page notebooks (suggests a lot of forethought, to figure out what numbers would make the eventual analysis come out right...)
They were using mice as a cure for cancer?
I enjoyed The Family Markowitz by the same author.
Not sure if I would read this, though. The subject matter doesn't do much for me.
They weren't using the rats as a cure, they were looking at a link between diet and cancer -- all the rats were injected with a carcinogen to increase the probability that they would eventually get cancer. In the meantime they were fed a diet that was either high or low in animal or vegetable fat, and high or low in animal or vegetable protein. Some got all they could eat, others were calorie-restricted to some degree.
I think the hypothesis was that the unrestricted, all animal diet would result in the quickest appearance of tumors, while the calorie-restricted, all vegetable diet would postpone the inevitable somewhat longer.
Here are some links to some other studies by the same group (can't find the one I worked on - we definitely fed *rats*, not mice -- nasty, bitey, big, sharp-toothed rats).
Post a Comment