%0 Conference Paper %1 googlearchy-hindman %A Hindman, Matthew %A Tsioutsiouliklis, Kostas %A Johnson, Judy A. %B Proceedings of the Annual Meeting of the Midwest Political Science Association %D 2003 %K qual search-bias %T Googlearchy: How a Few Heavily-Linked Sites Dominate Politics on the Web %X Claims about the Web and politics have commonly confounded two different things: retrievability and visibility, the large universe of pages that could theoretically be ac- cessed versus those that citizens are most likely to encounter. While the governing assumption of much previous work has been that retrievability would translate in- exorably into visibility, we cast doubt on that claim. Drawing on a large literature in computer science that ties a site's visibility to the number of inbound hyperlinks it receives, this paper proposes a new methodology for measuring the link structure surrounding political Web sites. Our technique involves iterative, extremely large- scale crawls away from political sites easily accessible through popular online search tools, and it uses sophisticated automated methods to categorize site content. In every community we examine, we find that a small handful of Web sites dominate. Online political communities on the Web thus seem to function as ``winners take all'' networks, a fact that would seem to have widespread implications for politics in the digital age.

@inproceedings{googlearchy-hindman, abstract = {{Claims about the Web and politics have commonly confounded two different things: retrievability and visibility, the large universe of pages that could theoretically be ac- cessed versus those that citizens are most likely to encounter. While the governing assumption of much previous work has been that retrievability would translate in- exorably into visibility, we cast doubt on that claim. Drawing on a large literature in computer science that ties a site's visibility to the number of inbound hyperlinks it receives, this paper proposes a new methodology for measuring the link structure surrounding political Web sites. Our technique involves iterative, extremely large- scale crawls away from political sites easily accessible through popular online search tools, and it uses sophisticated automated methods to categorize site content. In every community we examine, we find that a small handful of Web sites dominate. Online political communities on the Web thus seem to function as ``winners take all'' networks, a fact that would seem to have widespread implications for politics in the digital age.}}, added-at = {2011-09-28T23:59:32.000+0200}, author = {Hindman, Matthew and Tsioutsiouliklis, Kostas and Johnson, Judy A.}, biburl = {https://www.bibsonomy.org/bibtex/2510fdc39df1ca58b5f54e08aacffa8f7/dimitargn}, booktitle = {Proceedings of the Annual Meeting of the Midwest Political Science Association}, citeulike-article-id = {3051534}, interhash = {cb886ffe5bc562e3ffdc3601f258b841}, intrahash = {510fdc39df1ca58b5f54e08aacffa8f7}, keywords = {qual search-bias}, posted-at = {2011-09-09 19:50:44}, priority = {3}, timestamp = {2011-10-13T17:47:05.000+0200}, title = {{Googlearchy: How a Few Heavily-Linked Sites Dominate Politics on the Web}}, year = 2003 }