A rejoinder to King 2 March

King (1999) has replied to our review of his book. After summarizing the issues, we will respond to the main points and a few of the minor ones. The book proposes a method for ecological inference and makes sweeping claims about its validity. According to King, his model provides realistic estimates of uncertainty, with diagnostics capable of detecting failures in assumptions. He also claims the model is robust even when assumptions are wrong. We showed, by example, that these claims are seriously exaggerated. King’s method works if its assumptions hold. If assumptions fail, estimates are unreliable: so are internallygenerated estimates of uncertainty. Diagnostics cannot distinguish between cases where his model works and where it fails.