Tracking the Trackers
Our analysis of privacy policies revealed that a majority of the top 50 websites allow third party-tracking. Because this practice is essentially invisible to end users, we conducted research to discover how many of these trackers these companies allow and how many of the top 100 websites each tracker covers.
Our investigation of web bugs was based on data from Ghostery, which is an add-on for the Firefox web browser.
The data was generously provided by David Cancel, creator of the software. Ghostery identifies and informs the user of hidden web bugs by looking for hidden signatures inside the HTML of the pages a user visits . The software has an optional ‘GhostRank’ feature that allows the software to report the web bugs found on each site to a central database operated by Ghostery. Mr. Cancel provided us with data from GhostRank so that we could determine how many web bugs have been identified on each of the top 100 websites as well as how many of those sites each tracking company is present on. The data provided to us was for the entire month of March, 2009. During the month of our analysis there were approximately 300,000 users. Of those who have downloaded the software, approximately 10-15% (30,000-45,000 users) participate in the GhostRank reporting feature. During the month of March these users reported on 393,829 unique domains.
While data from this source cannot comprehensively cover the entire Internet, the potential for a self-selection bias is mitigated by the large sample size and large set of unique domains reported. Furthermore, we primarily sought to examine the use of web beacons on the top 100 sites, and since this data set did cover each of those sites, it was well-suited for our purposes.