Answers are grounded in this paper and cite their sources.
This research note compares survey pooling and multilevel regression for estimating death penalty support, highlighting disparities in results.
This study compares two approaches to estimating state-level death penalty opinion. One of these, survey pooling, is relatively simple, and the other, multilevel regression and post-stratification (MRP) is considerably more involved. Using General Social Survey (GSS) geocoded data, we estimate state-level support for the death penalty using survey pooling and MRP. The estimates are then compared with available probability-based state-level polls. We find that the survey pooling estimates are generally closer to the state-level polls than are the MRP estimates. MRP consistently underestimate the level of death penalty support when compared to the state-level polls, but this is not the case with the survey pooling method. Researchers should carefully consider the data context and benchmark when choosing between these methods.
No takes yet. Share an insight, caveat, or question.
Wu et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: