[Wherein our hero looks at our general online credulousness, and why

So we recorded our second trial of the Psychobabble Podcast the other night, I'm yet to edit down into a nice ipod-size programme, but I thought I might make a post regarding an interesting paper, and series of points that came up.

---

But first, please help me help a PhD! She has half the data set she needs, but another 100 participants would be awesome. http://bit.ly/i9GLPq There's not much research in this particular area of decision making, and the results look to be really interesting, and really novel.

---

Back in the day, when I was studying business, a lecturer described the internet (pre-bubble) as a wild-west-like frontier, where you could pretty much get away with anything. People were still grappling with the technology, and bunch of really smart people took it by the reigns and made a killing. I think a lot of it had to do with novelty, but a lot of it had to do with the credulousness of people, too. Yet I'm not sure it was entirely their fault. We were all unfamiliar with the what constitutes credentials online - our best cues to legitimacy were often poor, and included things like how professional the site looked, and other simple things like spelling errors. The problem with these kind of cues is that credible institutions had equal access to good webdesigners to the same extent as Nigerian Princes.

So what are the consequences of conquering such a frontier? Chesney & Su (2010) looked impact of credibility of anonymous blogs. They had three conditions - completely anonymous (known by a pseudonym only), semi-anonymous (where a psuedonym was accompanied by some demographic details, like age, sex ) and non-anonymous (actual name, photo and demographic details). Now for those of you into the wider science-blogging scene you can probably make a pretty good guess at how it turns out. Chesney & Su (2010) found no difference in the degree of credibility between any of these three conditions. They even replicated this study cross-culturally (in the UK and in Malaysia) (suggesting that net-norms might really transcend borders?).

This lead to a second experiment based on conclusion of the first - perhaps readers were simply responding to 'the feel' of the blog. And so they introduced some grammatical/spelling errors into one post, and some factual errors into another (content held the same). Again, Chesney & Su (2010) found that although the author was perceived as being slightly less credible in the poorly-constructed condition, both blogs were perceived as being equally credible. They suggest that the blog format is fairly liberal and we've all come to expect some spelling and grammatical mistakes, due to the lack of editing and redrafting.

And so this raises the argument that science-bloggers are not doing good. Without such editing they can write any ol' crap and people will lap it up, as they surely do. To be fair, this is not limited to science-bloggers, but arguable they do the most harm with their misinformation (I mean, what's peer-review for homeopathy? It's all woo).

The natural counter-argument is that science-bloggers speed the 'peer-review' process up. Where it would normally take months to deconstruct bad science, the science-bloggosphere leaps to and crushes bad science when and where it appears.

So is this good or bad? Well, let's take the arsenic-life story. Everyone is blaming NASA for saying something stupid. Now I was following this story at Universe Today, on twitter, and generally across the blogosphere. The first I heard of it was (from Universe Today) generally exclaiming a bit of excitement, but not making any bold claims. Then I saw the tweets - NASA to reveal alien life... blah blah blah. You know what, I'd put money on it that for every science blogger that said something stupid, 10 kooks said something even more stupid. And so rumour and speculation spread like wildfire. People lapped it up, due to this apparent problem of free credability on the internet.

Here's what the original announcement said:

"...to discuss an astrobiology finding that will impact the search for evidence of extraterrestrial life. Astrobiology is the study of the origin, evolution, distribution and future of life in the universe."

To discuss a finding that will have an impact. There's a bucketload of unaccountable Methane on Venus, but people didn't lose their shit about gassy Venutians. That finding has an impact on astrobiology, too. A big one. We're just not sure how to interpret it.

My point is that no-one cuts deeper than a science-blogger. If you're going to believe what you read, there are worse things to read than the writing of a science-blogger. For every scathing review of a paper, a science blogger is slicing into another science blogger. Science Bloggers self regulate, not to mention they respond - NASA made their announcement and every microbiologist worth their phosphate had something to say. To say they're doing science an ill-service is to be completely niave of what science bloggers actually do, and it's something you don't see the kooks and the laymen doing. Kooks aren't going to self-regulate, science-bloggers do. And so long as it's well written, anyone can say anything, and people will swallow it.

Given this, to complain about science-bloggers doing an ill-service to science is to complain about all bloggers, everywhere. It's an every day affair to find some academic routing another on some point of contention, and they don't mince their words.

In my opinion (for whatever that's worth) science-bloggers are the good guys. If people are going to google 'arsenic life' it's better they stumble onto some microbiologist, than on to some kooks. Sure, they might get some technical points wrong, maybe even grossly, but the correct sentiment will usually emerge, in their post, in their comments, or in some other science-bloggers rebuttal.

And given that we that pretty words dazzle us this is a good thing.

---

Chesney, T., & Su, D. (2010). The impact of anonymity on weblog credibility International Journal of Human-Computer Studies, 68 (10), 710-718 DOI: 10.1016/j.ijhcs.2010.06.001