Friday, April 11, 2008
Don't belive everything experts tell you about the Internet
Apparently a study by researchers shows that if you search for "suicide" you are more likely to find material pro than anti. Mental health campaigners are now calling for "something to be done", says a report of the research at the BBC web site.
The original research was conducted by psychiatrists - three of whom, no less, were professors. So they must be experts in their subject. So, let's check; just go to Google, type in "suicide" and you get over 66m results - wow! But take a look at the first five - all of them are either objective, such as the Wikipedia entry, or trying to present a balanced or positive view.
People rarely look beyond the first five results on Google. They may look at the adverts - and you can see that the sponsored results are from people like The Samaritans, hardly "pro" suicide. So how come the "expert" researchers came up with their "results"?
Well they were doing something that Internet users don't do. They were looking for the most frequently occurring sites across four different search engines. The research shows, it seems, that the three most indexed sites across different search engines were all pro suicide. So, it sounds as though the search engines are presenting a biased view of suicide sites.
What the researchers didn't do, it seems, is what Internet users do. Search on Google, look at the first handful of results, then search again on Google with a different phrase if they can't find what they want, repeating the search on Google with a further phrase if they still don't discover anything useful. Only then do they resort to alternative search engines.
So whilst I'm not doubting the "frequency" result which these scientists discovered about suicide web sites, the conclusions people are drawing from their study only makes sense if you don't know how people behave online. Once you take into account online search behaviour, the results are pretty meaningless. Besides which, the first page of Google results are almost entirely positive and supportive - hardly pro-suicide. An individual would have to be really serious about their own demise to search for something negative. The casual searcher won't face the negative information the health campaigners are worried about.
Labels: internet, internet psychology
Add this story to:
| BlinkList | BlogMarks | del.icio.us | Digg | Furl | Google | LinkRoll | Lycos |
| ma.gnolia | Netscape | Newsvine | Ning | reddit | Simpy | Spurl | Squidoo | Wink |
Email this story to your friends: 
Readers' Comments:
At
April 12, 2008 6:18 AM 
Graham Jones said…
At
April 12, 2008 10:01 AM 
Hjörtur Smárason said…
You only see what you want to see. If you are really suicidal, will you be typing in "suicide" or "how can I kill myself"? Those who type in suicide are maybe people looking for info, studies or assistance as family or friends to someone suicidal. My guess is that those who really are suicidal, are going to be looking for methods, a "DIY suicide guide". And they won't click any anti-suicide links because they are not looking for that and don't want to see that.
At
April 13, 2008 11:40 AM Paul Canning said…
Hallelujah! Good to see some people criticising this 'research'.
The research is flawed, as I explain at length in this blog post http://paulcanning.blogspot.com/2008/04/why-suicide-prevention-charities-are.html
Anyone with any knowledge of search and search behaviour could point out numerous, extremely basic, problems with the methodology.
What is worse is that charities and government have it within their power to counteract these websites. Blaming people like ISPs is an abdication of responsibility.
At
April 14, 2008 2:33 PM said…
This reminds me of that ad that ran in America:
Kids in trench coats shoot up school.
Washington's response? Ban trench coats.
It beggars belief that these so called experts aren't grasping the opportunity afforded by the Internet to reach out to suicidal people.
Perhaps they should have asked what are people really looking for - a way to kill themselves or a glimmer of hope and some help to end their suffering rather than their life?
Further to my report of this BMJ study, I notice another psychologist has criticised the study for being poor. See: PsychCentral