An article from
MuslimMatters:
As Muslims defensively chant “not all terrorists are Muslims” and “there are terrorists of every faith,” the question that the Islamophobes invariably ask is: “why is it that 99% of terrorists are Muslims?” Alternatively, they state: “twenty thousand acts of Islamic terrorism have occurred in the last year compared to three acts from every other religion combined.” (Admittedly, I am unsure of the exact numbers they use, but it doesn’t really matter since these sorts of figures are usually concocted anyways.) Nonetheless, the point appears valid: there seem to be a lot more Muslim terrorists nowadays; so does Islam breed terrorism? It’s a fair question.
In order to arrive at a scientifically valid answer, we’d have to account for confounding variables. Otherwise, incorrect conclusions could be drawn. One study showed for example that people with more ashtrays in their houses were more prone to lung cancer. A faulty conclusion would be that ashtrays
cause lung cancer. The confounding variable in this case is of course smoking. In other words, a linear correlation between ashtrays and lung cancer does not necessarily establish a causal relationship. To give a slightly more complex example: a study found that those who drank more coffee were more likely to develop lung cancer; could researchers then claim that coffee consumption causes lung cancer? No. It turns out that smokers tend to drink more coffee than non-smokers; it was the smoking, not the coffee, that caused the cancer.
When clinical drug trials are conducted, researchers give the drug to be tested to one population (called the variable population) and the existing drug therapy to another (the control population). The effect of the drug is then ascertained by comparing the morbidity and mortality in both populations. However, in order for the comparison to be valid, the two populations have to be similar. If they are not, the study becomes compromised. For example, if the city of Berkeley, California is chosen for the variable population, and Miami, Florida is chosen as the control population, there could be heavily skewed results. Berkeley hosts a college campus, and thus a disproportionately high number of young adults; Miami on the other hand is home to many old people who go there to retire (“It’s the law” according to Seinfeld). It would not be surprising then if there was less morbidity and mortality in Berkeley than in Miami. Perhaps such a tactic would be useful for pharmaceutical companies to market their drugs, but it would certainly be bad medicine. In other words, the raw data must be stratified or normalized before final conclusions can be drawn.
Such empirical studies ought to be done with academic rigor and scholarly honesty, something which we can hardly expect of Islamophobes. Many of us are familiar with
the indefatigable work of John Esposito who has led the charge in using the scientific method to verify (or in this case, reject) the hypotheses of hate-mongers; Gallup poll obtained some much needed data in this regard. But I’d like to draw the reader’s attention to another less familiar study–one which is closer to our null hypothesis. In 2002, Professor Daniel Price of Kent State University
published a paper in the
Journal for the Scientific Study of Religion. He asked: does Islam repress human rights?
Islamophobes will often compare the develop
ed “Christian world” with the develop
ing Islamic one, and then draw faulty conclusions based on that. But surely such a comparison is unfair–it’s comparing Berkeley to Miami. Confounding variables include gross domestic product, literacy rates, poverty level, and a host of other factors (not to speak of foreign occupation). If, however, one compares a majority Christian country in Africa, for example, with an Islamic country in the same continent, the results would be far less dramatic–and much more accurate. Professor Price normalized the data and concluded that there was no causal relationship between Islam and repression of human rights; he
wrote:
I test[ed] the relationship between Islam and human rights across a sample of 23 predominantly Muslim countries and a control group of non-Muslim developing nations, while controlling for other factors that have been shown to affect human rights practices. I found that the influence of Islamic political culture on government has a statistically insignificant relationship with the protection of human rights.
Similarly, we can ask: does Islam cause terrorism? Although I do not claim to have conducted a scientific study on the matter, I would like to present a case series to hopefully shed some light on this matter.
Read the rest of the article
here.