Measurement limitations bias elite and interest-group influence estimates downward.
Evidence
The rather low explanatory power of all three independent variables taken together (with an R-squared of just . in Model 4) may partly result from the limitations of our proxy measures, particularly with respect to economic elites (since our “affluent” proxy is admittedly imperfect) and perhaps with respect to interest groups (since only a small fraction of politically-active groups are included in our measure). Again, the implication of these limitations in our data is that interest groups and economic elites actually wield more policy influence than our estimates indicate. — Martin Gilens and Benjamin I. Page, Testing Theories of American Politics, 94
Categorization
- QUESTION: The imperfect ‘affluent’ proxy — and the correction direction (true influence higher) — matters for assessing whether the measured public-opinion null is real or artifact.
- XREF: connects to: No prior study jointly assessed interest-group, public, and elite influence on policy. — Martin Gilens and Benjamin I. Page, Testing Theories of American Politics:…, 46; Measurement error in the affluent proxy biases elite influence estimates downward, not upward. — Martin Gilens and Benjamin I. Page, Testing Theories of American Politics:…, 58; Opposing measurement biases strengthen confidence in the financialization trend — Greta R. Krippner, The Financialization of the American…, 85 (+3 more)
Broader Connections
- Survey research is vulnerable to systematic measurement errors that skew results
#corroborates(0.82) - The method of limits is inherently prone to response bias
#exemplifies(0.78) — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 918 - Opposing measurement biases strengthen confidence in the financialization trend
#expands(0.76) — Greta R. Krippner, The Financialization of the American…, 85 - Critics charged that sampling only high and low scorers biased the results
#exemplifies(0.74) — Martin Jay, The Dialectical Imagination, 744