Study: Don't Use Wi-Fi Near Your Junk

[dyre]

I assume the control group was a bunch of sperm next to non-wifi laptops, not just a bunch of sperm sitting on a table?

Also, what's up with the tiny sample sizes? All the studies I read about on the internet seem to be content to study under the standard minimal sample size of 30 people :\ (and imo 30 is really small)

29 subjects is actually sufficiently large for statistical purposes if you're not dividing them into groups. If you take a Stats class (and you should, because Stats is awesome), you'll notice pretty quickly that a 30-subject sample provides you with a pretty scary level of confidence even for huge populations. Late-Phase Clinical Trials for drugs use much higher numbers because of the seriousness of potential errors and the need to run the trial on many different populations, but for the purposes of this kind of experiment 29 is honestly kinda on the large side.

And yes, the control group were kept at the same temperature, albeit without the laptop. This should control for temperature, though they certainly could have been more specific about it.

hey, I DID take a stat class! But all I did was memorize the methods the prof taught and when to use them >_>

thanks for the info though

edit: lol, woah, I noticed that after your response to my post, you responded to everyone else in the thread. You seem to be a bit miffed at people not bothering to read the paper >:O. But to be fair, the paper wasn't even linked, and well, to laymen like myself, scientific papers are boring as fuck to read. I usually don't get past the abstract. And when I get to the method section...ugh. We're usually supposed to just read the summary from the news writer :\

 

[noxymoron19]

I only need one to make it anyways.

 

[jonnosferatu]

I assume the control group was a bunch of sperm next to non-wifi laptops, not just a bunch of sperm sitting on a table?

Also, what's up with the tiny sample sizes? All the studies I read about on the internet seem to be content to study under the standard minimal sample size of 30 people :\ (and imo 30 is really small)

29 subjects is actually sufficiently large for statistical purposes if you're not dividing them into groups. If you take a Stats class (and you should, because Stats is awesome), you'll notice pretty quickly that a 30-subject sample provides you with a pretty scary level of confidence even for huge populations. Late-Phase Clinical Trials for drugs use much higher numbers because of the seriousness of potential errors and the need to run the trial on many different populations, but for the purposes of this kind of experiment 29 is honestly kinda on the large side.

And yes, the control group were kept at the same temperature, albeit without the laptop. This should control for temperature, though they certainly could have been more specific about it.

hey, I DID take a stat class! But all I did was memorize the methods the prof taught and when to use them >_>

thanks for the info though

edit: lol, woah, I noticed that after your response to my post, you responded to everyone else in the thread. You seem to be a bit miffed at people not bothering to read the paper >:O. But to be fair, the paper wasn't even linked, and well, to laymen like myself, scientific papers are boring as fuck to read. I usually don't get past the abstract. And when I get to the method section...ugh. We're usually supposed to just read the summary from the news writer :\

I actually wrote the other post first and then decided that my response to you didn't belong in with the rest of them.

I recognize that the paper isn't directly linked, but a vital part of critical thinking (and reading ANY news article) is to pay attention to the source of whatever's being said. Greg didn't do a horrible job, but AFAIK he isn't a science writer by any stretch of the imagination (and probably just summarized the Reuters article), and the guy who wrote the Reuters article either is either not a science writer, or a pretty bad one. Following the source is actually pretty fast for most articles like this one, because sites like Reuters generally link directly to the paper.

I also probably should have mentioned (as I will in the thread I'm writing on this, pretty much in response to this one) that nobody actually READS the methods section unless they're trying to design a follow-up experiment. Usually the sequence goes:

Abstract -> Conclusions -> Results (look for tables and graphs to verify Conclusions)
After that we might skim the Methods, but only if we're particularly interested in how the research was done or want to make an argument about its validity.

THAT SAID, I don't mind people not reading the original paper or even a Reuters piece on it. What I DO mind is people reading about research and then making very stupid statements about it, which is what happened here. Reading this thread was actually a very good thing for me, because it reminded me of something I'd forgotten: The main reason that so many discoveries are overlooked or rejected in the popular consciousness is that people who don't know what the fuck they're talking about won't let that stop them from chiming in. This is a problem.

(I'd write something about Global Warming here but I'm pretty sure you get the gist of what I'd say just by my mentioning it)

 

[dyre]

I actually wrote the other post first and then decided that my response to you didn't belong in with the rest of them.

I recognize that the paper isn't directly linked, but a vital part of critical thinking (and reading ANY news article) is to pay attention the source. Greg didn't do a horrible job, but AFAIK he isn't a science writer by any stretch of the imagination, and the guy who wrote the Reuters article isn't a very good one. Following the source is actually pretty fast for most articles like this one, because sites like Reuters generally link directly to the paper.

I also probably should have mentioned (as I will in the thread I'm writing on this, pretty much in response to this one) that nobody actually READS the methods section unless they're trying to design a follow-up experiment. Usually the sequence goes:

Abstract -> Conclusions -> Results (look for tables and graphs to verify Conclusions)
After that we might skim the Methods, but only if we're particularly interested in how the research was done or want to make an argument about its validity.

THAT SAID, I don't mind people not reading the original paper or even a Reuters piece on it. What I DO mind is people reading about research and then making very stupid statements about it, which is what happened here. Reading this thread was actually a very good thing for me, because it reminded me of something I'd forgotten: The main reason that so many discoveries are overlooked or rejected in the popular consciousness is that people who don't know what the fuck they're talking about won't let that stop them from chiming in. This is a problem.

(I'd write something about Global Warming here but I'm pretty sure you get the gist of what I'd say just by my mentioning it)

Hmm, ok, fair enough. About being upset about people dismissing research without actually reading the research, that is.

Don't think it's just you science types that get dismissed by the popular consciousness though! People are just as eager to ignore history. For example, Iran's situation today, which is the culmination of about sixty years of US-Iran relations, mostly involving the US bullying Iran by supporting "anti-communist" dictators and squashing nationalist, democratic leadership, gets simplified into "those Iranians are a bunch of extremists who just hate our way of life." I'm sure the same applies to any subject.

Oh, and I appreciate the knowledge that I can just read the abstract, the conclusion, and the results (though, results before conclusion? not what I would've guessed)

 

[jonnosferatu]

Abstract and Conclusions are the easy bits with lots of information. The Results are usually at least mildly interesting, but that's mostly because they give you something you can use to frame the Conclusions. Research in peer-reviewed journals will almost always be pretty honest.*

*I know someone's going to come in and get on my case for this, so here's the spiel:
The Lancet article on MMR Vaccines and Autism is an example of a pretty rare occurrence - generally speaking, when people are publishing bad science, they do it by only publishing the results of a small number of experiments. This is usually only viable in industry, where you don't have to justify your expenses to an outside grant agency, and usually have ~2 decades before anyone else can do their own experiments to verify the results (patents). Academic research is generally less susceptible to this.

 

[Roxor]

How is this a bad thing? The planet is already overpopulated.