Saturday, 3 April 2010

Is Google biased in favour of climate change?

Google ranks pages from 0 to 10 using a page rank (named after Lawrence Page). The higher the rank the more likely is Google to send an enquiry to that page. Sites with the maximum rank of 10 include and

About a year ago my colleague and I set up links to the most popular climate change blogs, those with a Google page rank of 5 or above. We’ve just revisited the blogs and found that whilst changes to ‘non-sceptic’ blogs are fairly neutral (as many upgraded as downgraded) for ‘sceptic’ blogs there have been more downgrades than upgrades. The following table summarises the situation.

Site type Change Number
Non-sceptic +1 9
Non-sceptic 0 23
Non-sceptic -1 7
Non-sceptic -2 1
Sceptic +1 2
Sceptic 0 25
Sceptic -1 10
Sceptic -2 3

The non-sceptic blogs are almost in balance with 9 up and 8 down; one has dropped by 2 places but that blog has only posted once since early December 2009. On the other hand only 2 sceptic blogs have increased but 13 of them have fallen, three of them by two points. Unlike the non-sceptic blog which fell by 2 points all three of the sceptic blogs which fell by 2 points had posted something new in the last few days, though one of them had moved its web site to a new address.

It cannot be a function of number of visitors. Alexa, the web information company, gives “Watts up with that” as the most popular climate blog and it has a page rank of 7. On the other hand “Realclimate” according to Alexa has had a quarter of the visitors to ‘Watts’ over the last 3 months but has a page rank of 8. “ClimateAudit”, which also has more visitors than “RealClimate” also has a page rank of 7.

There are also several non-sceptic sites which get less than one percent of the visitors of ‘Watts’ but have the same rank of 7.

The conclusion, unlikely as it may seem, is that Google does have a bias – against climate sceptics.

As several people have pointed out on blogs where we have linked to this, and we ourselves were aware, Google's ranking system is complex, secret and changing. None-the-less it does seem strange that at a time when opinion polls suggests that sceptics are in the ascendant there should be a diminution in the their Google rankings.

1 comment:

charles the moderator said...

Page rank is probably more a function of how may other sites link to a page, not necessarily how many people visit.

Many more sites link to RC than Climate Audit or WUWT even if the traffic is higher on WUWT.

charles the moderator.