As the presidential election campaign in the United States reaches the home stretch, one thing has become abundantly clear—barring any truly egregious mistakes by either campaign, this election is going to be particularly close. Thanks to the quirks of the Electoral College, the results in what are popularly known as ‘swing states’ are acutely important. However, the voters that are still in play in most of these states, ‘undecided voters,’ are, by many accounts, generally under-informed about the campaign. Typically, they consider themselves too busy to actually keep up with the issues, but still vote out of a sense of civic duty.

The first question this prompts is: how are these voters deciding? According to a trove of political science research, voters with ‘low information’ on the candidates, policy proposals, or the campaign itself, often use heuristics—mental shortcuts—to narrow down their choices. Much of this academic research has focused on what these heuristics are. In general, they include partisan party identification, visible physical characteristics, such as gender and race, candidate job occupations, and most disturbingly, ballot positioning.  Past research has found that the order of candidates’ names on the ballot can influence voters’ decisions.

Political campaigns have come to believe that ‘low information’ voters can be swayed by even more trivial cues. In the tightly contested 2008 Obama-Clinton Democratic campaign, strategists put a concerted effort into getting candidates onto ‘soft’ entertainment and lifestyle television programs, based on the belief that a decisive group of voters would be swayed by their haircut and clothing preferences, among other things.

So what can be done? Many of the historical ‘checks’ on American democracy were crafted with the express purpose of preventing the rule by an uninformed mass. Some have argued that a renewed fear of uninformed voters is simply a new manifestation of this old elitist loathing for the common man. However, the point of democracy, especially as opposed to a more totalitarian form of government, is not simply that citizens will vote in elections. Rather, the point is that a representative mass of the population, armed with proper information on policies, would reach generally better decisions than an autocratic dictatorship.

So what is to be done? The inevitable first response is to improve civic education at the primary level, but education can wear off as people age. The more immediate solution, at least in the case of the United States, would actually be to eliminate the Electoral College and replace it with a raw popular vote-based system. This would bring the entire country into play, instead of a select few states with particularly fickle populaces, and solve or at least ameliorate the problem of ‘low information’ voters in two ways. First, candidates would have to talk to a larger cross section of the country, rather than ignoring states with clear party leanings. This would revive voter engagement by making their votes matter more substantively. Second, the area in play during the election would be extended from a group of about seven states, thereby reducing the influence of easily persuadable ‘low information’ voters in ‘swing states’ to begin with.

The optimal solution to this whole problem of the ‘low information’ voter is not to keep them from participating in elections, as the harm of such efforts would far outweigh any benefit gained. Rather, we should give them the information to participate more gainfully, or to reduce their influence to the point that they aren’t the decisive demographic in our elections.