An article comparing single-sex vs. coed on a domain called singlesexschools.org? Without even looking into the background of this: Can we really expect the article not to be biased?
Without even looking into the background of this
The fact that someone made up their mind on an issue doesn't by itself mean they're wrong though.
Having a stated belief upfront puts your work in obvious danger of confirmation bias[0]. It can and should be assumed that you will promote results which support your belief and ignore results that don't.
1. I wonder about the effectiveness of single-sex vs. coed schooling.
2. I do a bunch of research.
3. I conclude that single-sex education is significantly more effective, and that students are being harmed by the dominant paradigm of coed schooling.
4. I establish a website to promote my views, and I call it "singlesexschools.org".
5. I display, on my website promoting single-sex schools, the arguments and papers that convinced me single-sex schools were a good idea.
Pretty much anyone who possesses knowledge will be able to give you a "stated belief up front" if you ask them about it. That doesn't mean they started with the belief. Knowledge causes beliefs. In my five-step scenario, what should I have named my website?
The idealist but misguided situation you describe is not the case here, anyway. This reaseach was published by a group formerly known as the NASSPE - the National Association for Single Sex Public Education.
No, it wouldn't. Someone more interested in dismissing my ideas than in learning what they were could just as easily say "Effective Schooling is a known advocate of single-sex education, and can therefore not be trusted to report on the issue" (indeed, people say this all the time). The fact that someone is willing to tell you what his ideas are doesn't automatically make them bad ones. And everyone choosing to talk about an issue is going to have an opinion on it.
Not only that, but it's entirely proper, even desired, to state your hypothesis before testing it. Scientists are fighting for it right now, some are proposing a registry of hypotheses to be tested to fight another problem -- publication bias.
The article here touts the word "random", which is essential to the science of the claim - but despite being claimed repeatedly, it remains totally unexamined.
There is bias and there is bias - this article makes no attempt to clear any source of bias, and as such is about as biased as you can get.
http://link.springer.com/article/10.1007/s13524-012-0157-1
Also, posted by another commenter in this thread:
http://newsroom.ucla.edu/portal/ucla/graduates-of-all-girls-...
It's enough for me not to dismiss the idea.
While this particular article presents facts in favor of single sex schools, I would think that our duty is to educate our kids and peers to critically question anything they read and at least do a little research (scanning the site to get a basic understanding of the agenda the author may be pushing; to follow the supposedly supporting links, ...).