A collection of essays by Bill (website@ccjj.info) accompanied by feedback from his friends.

Wednesday, July 27, 2011

Ideology and Science

Ideology and Science Don't Mix

In some Scientific American articles I've read over the years, I felt there was considerable liberal bias, but they published a particularly offensive article this month: In Defense of Wishful Thinking, where the writer admits that he lets his liberal ideological bias influence his scientific thinking, and he makes no apologies for it.

It's disgraceful that anyone saying such a thing should be allowed to publish in such a prestigious science magazine. I see any ideological bias in science as a very bad thing.

Ideological Bias in Global Warming Science

I have observed that many Libertarians are global warming skeptics, and I have heard liberals speculate about why this is. Some speculate that they have been bought off by the oil companies, and in fact I have heard my Republican brother say that he feels that most of the resistance from the American political right is due to the influence of the fossil fuels lobby.

I think the fossil fuels lobby is obviously doing its level best to oppose any action to fight global warming, but an important point is that Libertarians don't have to be bought. Their ideology, and their political faith, leads them to believe that anthropogenic global warming theory cannot be scientifically true.

Libertarian ideology holds that most of what our government does is unnecessary, that the level of taxation we suffer from is orders of magnitude more than necessary or desirable, but further, taxation and regulation are not only harmful and unnecessary, but morally wrong.

When posed with the question "What if severe taxation or government regulation really were, in some instance, necessary?", a Libertarian will answer "That is never the case.". And, when the questioner starts airing scenarios, the Libertarian will come up with explanations, in each case, how either A: "That would never happen." or B: "The situation could be dealt with without regulation.".

As I am not an extreme Libertarian, I see a lot of these explanations as rationalizations, and a lot of the Libertarian world is an industry for manufacturing such rationalizations.

But behind the rationalizations is a faith, a desire to believe certain things, a desire to reach certain conclusions. If someone gets swept up in the Libertarian movement, they eventually convince themselves that the principle that massive regulation is never a good thing is one of the central characteristics of the world, like the conservation of energy. They get an almost theistic belief that it is a fundamental quality that a creator endowed the universe with.

If global warming is really a threat, it is virtually impossible to think of a realistic solution to it that can be achieved without massive government intervention. Given the fervent faith of a hard core Libertarian, it just seems impossible to them that the creator would put us in such a situation. It's as unthinkable as agreeing we should eat our young. Given that the science is hard to follow and most of the scientists involved are liberals who tend to feel that regulation is a good thing in and of itself, it is easy for Libertarians to mistrust the scientists and reject the science.

It is a very bad thing when faith gets in the way of science.

Now, my liberal friends (which means nearly all of my friends) will think this is an illustration of why Libertarians are bad, and the moral of this story is we should not listen to Libertarians, but that is not my point. My point is that it is bad when people let their ideology blind them to the facts.

A Scenario Where Liberal Bias Could Be Harmful

Let's create a hypothetical situation where liberal faith might get in the way.

Scientists could do tests on people, like putting them in a MRI scanner and watching their brains while they see videos of suffering, and see how much they are repulsed or turned on, to determine whether they were empathetic, sadistic, or indifferent. And suppose, with such tests on people of all ages from babies to adults, scientists determined that a dominant gene, HYNC3, caused people to be severely sadistic. Only a small proportion of the American population carried it, but it was very common in the prison population.
While very few Americans carried HYNC3, a very poor country on our border, Pralaxia, had a population twice as large as that of the US where 75% of the population carried it. Pralaxia was in a shambles, a horrid, brutal society, and many people there were illegally immigrating to the US.
Conservatives would look at this state of affairs and say that we should go to great lengths to stop illegal immigration from Pralaxia. How would you expect liberals to respond?

I think liberals would deny the science, claiming that the MRI scans were not as meaningful as the scientists claimed them to be, that correlation does not establish causality, they would go on and on. Why would they manufacture all these objections? Because the whole state of affairs would violate several articles of liberal faith:
  • Genes are not a very important factor in determining human behavior.
  • Ethnic discrimination is always morally wrong. Nothing good ever comes of it. It is never justified.
  • We are a "Nation of Immigrants". It is always wrong for anyone descended from immigrants to say we should bar any other immigrant.
  • There is no such thing as a bad social / ethnic group, other than perhaps white American males.
Given that Pralaxia was much larger than the US, a failure to stop the immigration would quickly put us into a situation where 40% of our population was severely sadistic and probably reduce us to the same sorry state as Pralaxia. We are talking about the total destruction of our country. I think liberals would still deny the science, and resist measures to stem the tide every inch of the way.

Idealistic people are often likable and inspiring. But ideology should be extremely suspect among scientists.

What About a Real Situation????

A liberal might say "Well, that's just a hypothetical, fictitious situation. There are no real situations where liberal bias is causing harmful policy.".

I chose to give a hypothetical situation because I feel the liberal world, like the Libertarian world, is an industry of self-justifying rationalizations, and if I were to talk about any real liberal policy that I see as harmful, it would be ardently defended by liberals who have long ago generated rationalizations for it.

Sunday, July 24, 2011

Standardized Tests

College Tuition Costs and Standardized Tests

One of the biggest problems American society faces is that college tuition has become so high that many people just can't afford college, or if they go, they graduate massively in debt.

There are standardized tests, the SAT and ACT, that are widely used for entrance to college. Colleges put a lot of weight on these, because grades are non-standard and not very meaningful. A good grade could just mean that one had an easy teacher or dumb classmates. The presence of standardized testing for college admissions has been very beneficial, it allows outstanding students at unexceptional schools to get recognition.

We have a standardized tests for graduate school admissions, such as the GRE. However, the service that administers the GRE refuses to make their test scores available to private companies hiring college graduates. Given that grades are an unreliable indicator, this means that all the companies have to go by when considering applicants is the reputation of the school and the student's GPA.

What determines the reputation of a college? The quality of the undergraduate program has little to do with it. The quantity of research being done has a lot more, and it is common for professors at elite schools to neglect their teaching so they can focus on their research. The university administration doesn't get very upset about it, because it is the research and not the quality of instruction that determines the reputation of the school.

There is also a big problem of grade inflation. Many teachers give nearly all of their students good grades, because that discourages students from complaining about poor instruction quality. Rarely do administrators do anything about it.

There is only so much fame to go around. The average hiring manager can only have heard of a certain number of nationally famous schools. The vast majority of people are going to have to go to relatively unknown schools. And it would be nice if someone could go to a cheaper, less famous school, and still be recognized as an outstanding performer. Widely available, standardized tests taken at the end of one's college education would be a great way to achieve this.

Another benefit of widely available, standardized tests taken at the end of one's college education is that one could compare the test scores of students graduating from schools with the SAT scores they got while applying, and see which schools had the most beneficial impact on test scores.

One criticism of testing is that schools will "teach to the test". This certainly happens, but the solution to it is simple: construct an intelligently designed test such that the best strategy for achieving a high score is a mastery of the subject matter. I have heard, mostly from people who are against testing, that the "No Child Left Behind" tests are particularly bad. The solution is to improve the tests, not do away with them.

A big problem liberals have with standardized tests is that different social groups perform differently on them.
If these tests were mismeasuring talent due to unfair cultural bias, I would expect there to be groups who are high performing in society who perform poorly on the tests and groups that are low performing in society who are performing well on the tests. This is not the case -- generally, performance of groups on standardized test correlates very highly with the per capita intellectual performance of individuals in those groups in society, even before those tests existed. This is evidence that the tests are not unfairly biased. Given that society stands to benefit so much from the institution of more widespread standardized testing, I think the burden of proof should be on those who maintain that the tests are unfairly biased.