Over a third of American adults, 35 percent, currently own smartphones, with that number expected to keep growing. That means that over one third of Americans keep a user-friendly networked computing device with them at all times. The unprecedented access to computing created by smartphones is good for more than just Angry Birds, with companies like Pollbob using smartphone apps to poll users on any subject then share questions and results via social media.
Pollbob is an application available for free on Apple‘s App Store that allows users to quickly answer simple, one question opinion polls or build polls of their own. To do so, you come up with a short question, select how many options respondents will have to choose from, between two and five, and a category, such as Entertainment, Health, or Travel. Users can also share their poll on Facebook or Twitter, see if friends answered, and follow other users. In theory, technology like Pollbob has tremendous potential, giving smartphone users the ability to make their own public opinion polls to see what their peers think about issues relevant or interesting to them, both on a national scale on, by following their friends, a personal level.
I recently downloaded the Pollbob App to try it out. I set up an account and made my first poll, which was easy, intuitive, and took about a minute. Inspired by one of those Facebook privacy articles that pops up every week, I made my question “Do you trust Facebook to keep your information private?” and, without sharing my question on Facebook or Twitter, I had 12 responses in 24 hours and counting (If you’re curious, 17% trusted Facebook, 58% did not, and 25% said they didn’t use Facebook).
I also answered some questions, which was fairly entertaining and allowed me to see poll results. There was a huge range of questions, some insightful and professionally written and others misspelled, unclear or unhelpful (for example, “Do you enjoy vampires?” What? How?) or introduced bias through question and answer phrasing or format. Since all questions are user-submitted, there is no quality control and currently the quality is fairly low.
I’m also not convinced that we can gain meaningful data from these polls due to selection bias. Pollbob users are not representative of the nation or the world as a whole, and hence do not provide a random sample. Even questions with more responses often deviated from national statistics and gave varying results when asked multiple times, such as questions on party affiliation, which provided very different distributions in several instances. Some of this selection bias could be mitigated if the app had more users and an even broader range of the population had smartphones. Other questions become inaccurate due to impersonal, informal self-reporting. For example, in one sizable poll, 31% say they are overweight, but health statistics show that 68% of American adults are overweight or obese.
Yet Pollbob’s slogan isn’t “Discovering accurate population statistics,” it’s “Quickly discovering opinions” and there, it delivers. In my own example, I had assumed that most people were fairly comfortable with Facebook’s privacy protection, and while my limited results don’t prove anything, they do give me enough opinions to question my assumptions in a matter of hours. If this app or others like it gain in popularity, the results will get more accurate and arrive even faster. Pollbob may not yet be a useful tool, but already, it’s a pretty cool toy, and as this genre of apps evolve, it may better exploit the groundbreaking potential of smartphones.