A distinct group—highly educated people who have experienced downward social or economic mobility—responds to polls at atypical rates and with atypical preferences, biasing samples and standard weighting models. That composition effect can flip perceived public opinion and mislead campaign strategies, media narratives, and institutional planning.
— If true, pollsters, newsrooms, and campaigns must rethink sampling and weighting because a persistent, identifiable respondent class can systematically distort democratic information signals.
Matthew Shor
2026.09.10
100% relevant
Matthew Shor’s City Journal piece claiming 'an overeducated, downwardly mobile segment of the population is scrambling the polls' — the article’s core diagnosis of responder composition exemplifies this idea.
← Back to all ideas