DuckDuckGo study says Google search personalization perpetuating ‘filter bubble’

Over the years, there has been considerable discussion of Google's "filter bubble" problem. Put simply, it's the manipulation of your search results based on your personal data. In practice this means links are moved up or down or added to your Google search results, necessitating the filtering of other search results altogether. These editorialized results are informed by the personal information Google has on you (like your search, browsing, and purchase history), and puts you in a bubble based on what Google's algorithms think you're most likely to click on.

The filter bubble is particularly pernicious when searching for political topics. That's because undecided and inquisitive voters turn to search engines to conduct basic research on candidates and issues in the critical time when they are forming their opinions on them. If they’re getting information that is swayed to one side because of their personal filter bubbles, then this can have a significant effect on political outcomes in aggregate.

Back in 2012 we ran a study showing Google's filter bubble may have significantly influenced the 2012 U.S. Presidential election by inserting tens of millions of more links for Obama than for Romney in the run-up to that election. Our research inspired an independent study by the Wall Street Journal (paywall):

A Wall Street Journal examination found that the search engine often customizes the results of people who have recently searched for "Obama"—but not those who have recently searched for "Romney."

Now, after the 2016 U.S. Presidential election and other recent elections, there is justified new interest in examining the ways people can be influenced politically online. In that context, we conducted another study to examine the state of Google's filter bubble problem in 2018.

Findings

Google has claimed to have taken steps to reduce its filter bubble problem, but our latest research reveals a very different story. Based on a study of individuals entering identical search terms at the same time, we found that:

Most participants saw results unique to them. These discrepancies could not be explained by changes in location, time, by being logged in to Google, or by Google testing algorithm changes to a small subset of users.
On the first page of search results, Google included links for some participants that it did not include for others, even when logged out and in private browsing mode.
Results within the news and videos info-boxes also varied significantly. Even though people searched at the same time, people were shown different sources, even after accounting for location.
Private browsing mode and being logged out of Google offered very little filter bubble protection. These tactics simply do not provide the anonymity most people expect. In fact, it's simply not possible to use Google search and avoid its filter bubble.

d58afe6116389a865111848d5312aad36bbce4ce20a3ab4fca