The Internet is making it harder to measure how “informed” voters are

Quick: what’s your best friend’s phone number? If you’re like me, you have no idea. But that doesn’t mean I can’t give them a call; I can whip out my phone and access the necessary information as if not more quickly than if I typed in their digits from memory.

As one could imagine, this principle lies behind changing dynamics of knowledge for more than just getting in touch with friends. The Internet, and particularly our ability to access it anywhere and at any time, is shifting much of what we would traditionally call things we “know” to things we have access to. The average American—especially the average American between the ages of 18 and 34—is probably less likely to know the year of the Spanish Armada, the national unemployment rate or the name of the Chief Justice of the Supreme Court off the top of their head than they were twenty years ago. However, since the proverbial man on the street is infinitely more likely to be carrying an iPhone now than they were twenty years ago, they have no problem producing that information when asked.

This is one of the reasons why the United Nations has declared Internet access a basic human right. Living in the 21st Century without an Internet connection is, in a practical sense, a mental handicap. The 64% of American adults with a smartphone arent just using it for texting, Candy Crushing and swiping right; theyre using it as a cloud-based repository of information that no longer has to reside in their brains.

This widening gap between what we know and what we have access to is being observed in particularly stark terms in political science, where the Internet is making it harder take one of the most basic measurements in the field: political knowledge. Steadily declining contact rates in telephone surveys have pushed the bulk of survey research online. As one might predict, observed levels of political knowledge are higher online than they are on the phone. Taking a survey online gives respondents the ability to act on our innate desire to “self-enhance,” allowing them look up the correct answers on an online political knowledge survey that they wouldnt be able to if their phone was being used to, well, take the survey.

This has the potential to create the impression that political knowledge is increasing when there may be reason to believe that the opposite is true. Over time, expanding media options have given Americans a choice between the news and entertainment, and they have chosen entertainment at the expense of exposure to factual knowledge about current affairs.

From a measurement perspective, knowing how much voters know is important insofar as that information correlates with other things that we want to keep track of. Political knowledge has traditionally been positively correlated with voter turnout, polarization and ideological consistency, while being negatively correlated with persuadability, or the likelihood of attitude change in response to discordant information. More informed citizens have also been shown to be more accepting of basic democratic principles. But if the Internet is muddying the waters by inflating observed levels of knowledge, those predictions become harder to make. As political scientists Hillary Shulman and Franklin Boster note:

 A measure is valid to the extent that it measures what it purports to measure and nothing else, and the validity of this political knowledge measure could decrease substantially if participants look up answers on the Internet.

The changing ways in which we consume political information may lead us to reconsider both how to measure and understand political knowledge. Just because students may only answer about 25 percent of political knowledge questions correctly when taking a survey in pencil and paper format—much lower than online scores—doesn’t mean the rest of their political behavior follows suit as would be predicted by past correlations between information and action. Young citizens today, while less “informed” in the academic sense of the word, are more politically active than their predecessors. They may not know who the Chief Justice of the Supreme Court is, but they do have a coherent slate of opinions as to how they feel the Court should rule on marriage equality later this month and are willing to express those opinions in the public sphere. At the end of the day, which is more important?

All of this suggests that we need to rethink our understanding of “political knowledge.” As the Internet mediates and supplements our political minds, the blurring of the line between knowledge and access to information matters much less in practical terms than it does in academic measurements. Political deliberation has moved from the town square to Facebook, Twitter and Reddit—media on which there is no functional difference between recalling and querying information. When political knowledge is tested in the public sphere, we arent measuring what we know; were measuring how adept we are at finding things out.

That isnt necessarily a bad thing. The tradeoff—and it does feel like a tradeoff—between remembering less and having access to more information has the potential to expand and improve discourse, or at least make discourse more verifiable. As Cunningham’s Law stipulates: “The best way to get the right answer on the Internet is not to ask a question, but to post the wrong answer.” Wikipedia, the accuracy of which has been shown to be comparable to more traditional encyclopedias, operates on this principle.

Theres a reason why people lie less online than they do in person: the lie comes with a paper trail, the veracity of which is only one tab away. This has helped drive a growing market for companies like SnapStream to make offline statements—in their case, TV shows—searchable online by importing closed-captioning transcripts or otherwise automatically codifying language that would otherwise be difficult if not impossible to track down. Buzzfeed recently launched an in-house opposition research team based on a similar, if slightly less systematic, idea.

However, this doesnt mean that changing dynamics in political knowledge arent cause for concern. An increasing reliance on the Internet as a substitute for factual political knowledge leaves us vulnerable to manipulation—both internal and external. On the one hand, the essentially limitless expanse of information sources on the web allows netizens to create hermetic bubbles of information, which can lead to increasing polarization that poses problems for political discourse. Todays consumers of political information, armed with nothing more than a Twitter account, can expose themselves to mountains of political information that all conforms to their existing worldview. And given the choice, were far more likely to seek out belief-affirming information than we are to seek out information that challenges what we hold to be true.

However, while Twitter has customized the news, Google has in many ways homogenized it. As Matthew Hindman argues in The Myth of Digital Democracy, despite the expanding number of sources of information, citizens are no more likely to receive their news through alternative or non-corporate sources than they were before the Internet Age. In what Hindman terms a “Googlearchy,” users who seek information on the Internet are still filtered through a particular set of information pathways that guide them to online versions of the same information they used to receive in print: large, corporate media empires. Prominent placements on newsstands have been replaced, not disrupted, by high SEO rankings.

As was the case in the past, you can find information that isnt dominating the market, but you have to dig for it. That shouldnt come as a surprise; an algorithm designed to produce results the user is most likely to click on is going to be based on what other users are clicking on, which makes Hindmans findings predictable. In other words, it wasnt necessarily Googles fault that searches for “Bernie Sanders” produced above-the-fold results for a “rape fantasy” essay he wrote in 1972 for weeks instead of his proposals for tuition-free public college and arguments on how and why to reduce economic inequality. Users just find 43 year-old essays about rape fantasies more clickable. 

At least thats our operating assumption: Value-neutral search engines that respond only to what the market for clicks desires. And since we have neither the will nor the capacity to regulate search engines so as to verify that assumption, we had better hope that assumption holds true. If Google ever did decide to be evil, or at least something other than value neutral, the results could be disastrous for democracy—and we wouldnt have any idea.

ipad internet

Internet, via Shutterstock

In a 2013 study, psychologists Robert Epstein and Ronald Robertson created a mock search engine, dubbed Kadoodle,” and asked American respondents to use it to gather information about two candidates in an Australian election. The control group got a value-neutral search algorithm; the treatment groups were exposed to search results that favored one or the other candidate. They found that not only did the Search Engine Manipulation Effects (SEME) influence participants’ candidate preference, but that the bias could also be masked such that participants were not aware that they were being nudged in one direction or another.

Epstein and Robertson further found that even a small bias in search results could, over time, produce large information effects, as there is likely a synergistic relationship between search engine bias and voter preference. As noted above, page rankings are affected by what users are already clicking on, so even a slight bias—if its enough to shift preferences—can compound on itself over time.

As Epstein has noted, it would be absurd to rule out the possibility that Google has thought about the immense power they wield as the gatekeepers of political information:

The formula is simple: Find a tight race, identify the candidate who best suits your needs, and contribute to his or her campaign. Next, identify undecided voters and repeatedly send them customized search rankings that make the favored candidate look better than the other ones — easy as pie with Google’s massive database of email content and search histories.

Your candidate wont know about your fiddling, and he or she will be beholden because of the contribution. Sweeter still, regulators will have no way of detecting your manipulation because it was directed at a small number of people who received customized rankings, as many of us do these days.

This is especially problematic given how many of the voters who are most persuadable—the ones most susceptible to manipulation—arent necessarily those who are least informed. On the contrary, theyre often what political scientists D. Sunshine Hillygus and Todd Shields call “cross-pressured partisans” whose political opinions do not conform to a coherent party platform: think of the Ohio steelworker who strongly supports collective bargaining rights but has mixed feelings about same-sex marriage and abortion. These are voters who are both likely to be actively searching for information about political candidates, and who are likely to be influenced to a great degree by the results their searches return (notably, Epsteins research has found that people from the noted swing state of Ohio are more susceptible to SEME than people from other states).

All this is to say that when we take information that would normally be stored in our heads and instead keep it on the Internet, we arent really the ones doing the thinking anymore. This has always been true to a certain extent—Socrates had the same concerns about writings effects on memory that others are now having about the Internet—but its more true now.

On balance, the Internet is absolutely a net force for social good, as anyone with a connection has access to essentially limitless information provided that they know what to look for. However, the history of progress is marked by giant leaps forward that come with (often unseen) steps to the side, creating new problems that need to be taken into account. We can’t put these genies back in their respective bottles, nor would we want to, but we must instead recognize that genies are useful insofar as they are powerful beyond our control or even our comprehension.

Jon Green graduated from Kenyon College with a B.A. in Political Science and high honors in Political Cognition. He worked as a field organizer for Congressman Tom Perriello in 2010 and a Regional Field Director for President Obama's re-election campaign in 2012. Jon writes on a number of topics, but pays especially close attention to elections, religion and political cognition. Follow him on Twitter at @_Jon_Green, and on Google+. .

Share This Post

© 2021 AMERICAblog Media, LLC. All rights reserved. · Entries RSS