“the survey asked every blogger to list three other regularly read science blogs?”. What here is the meaning of “blogger”. I presume it means the one person per blog that runs the blog. eg, Judith Curry for Climate Etc, Gavin Schmidt for RealClimate, etc. Now obviously, these people will actually read many science blogs, but they will predominantly read those of their own “side” of the argument. So when asked to nominate only three, the division shown by the chart is inevitable. IOW, it is an artefact of the procedure used.
But it gets worse. By asking only a limited number of “sceptical” bloggers, and a large number of “alarmist” bloggers (and the survey has clearly extended the notion of “scientific blog” quite dramatically on the “alarmist” side only), obviously the link counts will be much higher for the “alarmist” blogs than for the “sceptical” blogs. So the link counts are also an artefact of the procedure used.
And then just for good measure, a dark colour was used for “alarmist” sites, and a pale colour for “sceptical” sites, so that optically the “alarmists” dominated even more.
WUWT lists 50 blogs with “skeptical views”. If the survey had covered all 50, the divide would probably still have been there, but the “sceptical” websites would have dominated.the chart, with the “alarmist” sites looking like the minor side. IOW, the appearance of the chart depends only on the number of bloggers used on each “side” and on the colours used – and nothing else.