A tale of "science" gone bad...
Aug. 14th, 2008 10:14 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
tagrynvia Watts Up With That, a long and involved post titled "Caspar and the Jesus Paper" about how personal politics trumped science in promoting one of the key graphs (the 'hockey stick') used to support the "global warming" hypothesis.
If things are as presented in the post, then its a scandal. For a scientist to refuse to release (a) the data to allow replication of results and (b) a standard measure such as an R-squared which allows others to measure how good a predictor the model is, is simply unacceptable and against the rules that we all play by. This seems to be a case that the researchers felt the issue was too important to permit evidence which would weaken their case to be published.
It reminds me of something I heard in grad school, when one of my classmates in research class said "if I was researching and found something that I felt supported a wrong (i.e. conservative) position, I'd refuse to submit it (suppress it) because of the damage it might do." I was horrified that a scientist would take that stance, am horrified by it when I encounter it still, but am very aware that it happens. Having seen 'how the sausage is made,' I know that science a lot of the time isn't the pure, unbiased process resulting in Truth that many of its idealists assume it to be. It should be, but it isn't; its being done by people, with real flaws and weaknesses, and that's the whole point of requiring replication of results, so that individual's biases are minimized over time.
If things are as presented in the post, then its a scandal. For a scientist to refuse to release (a) the data to allow replication of results and (b) a standard measure such as an R-squared which allows others to measure how good a predictor the model is, is simply unacceptable and against the rules that we all play by. This seems to be a case that the researchers felt the issue was too important to permit evidence which would weaken their case to be published.
It reminds me of something I heard in grad school, when one of my classmates in research class said "if I was researching and found something that I felt supported a wrong (i.e. conservative) position, I'd refuse to submit it (suppress it) because of the damage it might do." I was horrified that a scientist would take that stance, am horrified by it when I encounter it still, but am very aware that it happens. Having seen 'how the sausage is made,' I know that science a lot of the time isn't the pure, unbiased process resulting in Truth that many of its idealists assume it to be. It should be, but it isn't; its being done by people, with real flaws and weaknesses, and that's the whole point of requiring replication of results, so that individual's biases are minimized over time.
