Victoria Parker at The Atlantic
Some caveats: Our research, which is available as a preprint, is under review and subject to change. We drew our large samples of respondents from online survey platforms, not from nationally representative polling. We recognize that this sample—and therefore our estimates of the prevalence of liberal and conservative opinions—is not an exact microcosm of the country. Still, other researchers have concluded that these platforms are reasonably comparable to nationally representative polling.
The gap that we identified between what partisans really think and what their opponents think they think shows up again and again—but only on a particular kind of issue. People have a more accurate view of the other side’s position on many standard policy issues, such as taxes or health care. But specifically on culture-war issues, partisans are likely to believe a caricatured version of the opposing side’s attitudes. These misconceptions have hardened into enduring stereotypes: liberal snowflakes and free-speech police, conservative racists and “deplorables.”
In reality, just a third of liberal participants agreed even a little with banning controversial public speakers from college campuses, but conservatives estimated that 63 percent of liberals held that view. Only 22 percent of conservatives expressed hostile and unwelcoming attitudes toward immigrants, but liberals thought that 57 percent of them did. Our data suggest that many people are walking around with an exaggerated mental representation of what other Americans stand for.
