The plot in the linked article gives so little information that it is absolutely useless at concluding any sort of bias. It does establish that there is a tren

The plot in the linked article gives so little information that it is absolutely useless at concluding any sort of bias. It does establish that there is a trend, with higher referrer traffic to sites identified as left/liberal. In order to establish bias, this would need to be identified as something intentional and inaccurate on the part of google. What differences exist in user patterns? Do users of conservative news typically browse certain websites, while users of left-leaning media search for articles? No attempt was made to look at the relationship between website size and google referral traffic. Glancing at the websites listed, most of the conservative sites are much smaller, and so I would expect them to have lower search rankings and referral traffic as a result.