David M Rothschild on Posted on

Our state-of-the-art method uses machine learning to convert fast mobile-based polls into results that are at least as accurate as results from traditional landline polls. With smartphones reaching saturation in the US, our polls reach a broader audience in a shorter time and our high-quality panel is supplemented with a rich array of passive data exclusive to mobile data collection. We ensure that our results are representative of the country using algorithms we have fine-tuned over years of tracking public opinion in the political sphere.

First, we model the raw responses to each question, given each respondent’s age, gender, location, education level, race, marital status, party identification and income. This information divides the population into thousands of demographic categories. For each sub-group and poll question, we predict the percent of people that would provide each answer if the entire country showed up to the poll.

Each of these predictions is informed by all responses, including responses received in previous weeks. To achieve this, we have developed a complex dynamic model that allows us to parse out variance in sample composition from true swings over time. This is crucial for assessing trends in polling data, as illustrated by this cautionary tale from the world of political polling.

Second, we project our estimates for each sub-group onto our best estimate of the likely adult population, a process known as post-stratification. Specifically, we weight our predictions by the fraction of that sub-group in the overall target population. We derive the target population from a Big Data combination of population-level census data, and proprietary financial and political data sets, including background information on all registered Americans.

The transformed data provides meaningful information about many segments of the population. Our approach has been validated in peer-reviewed articles published in leading academic journals (see here and here) and real-world event predictions (see here and here). Our polling has lead to major publications in the US’ leading news platforms, including but not exclusive to The New York Times (see here and here), the Washington Post (see here and here), or Slate (see here).