Zogby Polls have earned a place as an indispensable and crucial component in the research of public opinion, with continued contributions to knowledge about elections, as well as several timely social issues. Fully acquiring an understanding of how one correctly reads polls and interprets their reliability accurately necessitates examining lessons learned throughout many decades of polling history. This is well investigated in the book titled Beyond the Horse Race: How to Read Polls and Why We Should.
One of the most salient takeaways from the book is how the various technological and social changes have evolved and influenced polling methodologies. The online polling websites, combined with IVR systems, have made data collection much faster and easier. Still, new technologies have introduced so many challenges regarding the ability to effectively communicate and reach out to underrepresented demographic groups, such as older voters and those who live in rural areas. With these changes, pollsters need to carefully reconsider their expectations with regard to the insights and results their polls can return. Further, they must be concerned about the ways technology can affect the overall reliability and validity of their samples. The polls by Zogby have a long history of innovation incorporating the latest technologies.
Whereas nuances of a number of more detailed, interwoven trends may lie in the topline results, moving beyond analysis of the topline will uncover such data. New clusters of voters can be developed by analyzing a range of subgroups, such as those distinct “neo-tribes” found in Zogby’s Tribal Analytics. This provides a much deeper level of understanding about the prevailing public view than any raw polling could. At its very core, polling is a task of capturing the intricacies of human behavior and relying exclusively upon top-level results can often lead to significant misinterpretations or misunderstandings about what the data is truly indicating.
Admittedly, no poll can be infallible; polls, after all, represent no more than a snapshot of public opinion at a given moment in time. Yet, taking the right lessons from whatever missteps do happen when trying to understand polls often tends to produce far greater and more important insights into the behavior and preferences of voters. An important example that reinforces this is the very case from the history of Zogby polls where an obviously incorrect forecast was made for the 2004 presidential election, a lesson on the inherent challenge in polling. Our Zogby Poll indicated that John Kerry would emerge victorious over George W. Bush, which indeed did not happen as forecast. These are valuable lessons that help underline the cautious approach one needs when interpreting the polling data. Pollsters should definitely not be treated as unanimous predictors of election results since late-deciding voters or some unplanned events may always change the tide of the final result in a manner quite unpredictable.
Results from the Zogby polls, conducted during the time of the 2016 election, actually led us to believe that the results from polling were, in fact, quite a bit more accurate and sound than was commonly believed by a number of members of the general public and even commentators. The real problem wasn’t flawed polls but rather proper, methodologically sound ones that the media and various political pundits misinterpreted. What’s more, they didn’t recognize and represent the concerns of major voter demographics, like influential groups such as suburban mothers, whose views were ignored.
The Zogby Polls are useful, but to understand their accuracy, one must read between the lines and be aware of the potential problem of sampling, plus the limitation of polling as a human effort for capturing but not perfectly predicting behavior.
