A new analysis suggests that measuring political polarization in the United States is more complex than previously thought, with findings indicating that the results depend heavily on how questions are framed and who is being considered. Researchers from the University of Michigan argue that when assessing polarization, it's important to consider whether Americans are thinking about politicians or everyday voters. The political context surrounding the questions also plays a role in shaping responses. The research highlights that seemingly small choices in survey design can produce dramatically different pictures of how much Americans dislike the opposing party. For example, people appear more polarized when asked about politicians in the other party, but less so when the question is clarified to refer to ordinary voters from the opposing side. This shows that the way questions are worded can significantly affect the results of polarization studies. The study also notes that polarization statistics often exclude people who identify as independent and have no leaning toward either major party. As more individuals identify as independent, this exclusion means that the understanding of public polarization becomes less comprehensive, as it excludes larger groups who do not align with either party. This can lead to an incomplete picture of the overall political climate. Additionally, the research suggests that imbalances in political power in government could influence polarization, with people potentially developing more negative feelings toward the opposing party when that party holds more power. However, data on which party is more polarized remain inconclusive and inconsistent across different studies and data sources. The study, published in the journal Political Psychology, was co-authored by James Druckman of the University of Rochester, Samara Klar of the University of Arizona, and Matthew Levendusky of the University of Pennsylvania.