Explanatory variableand effect type Regressioncoefficient Standarderror   Full random-effects logit model Fixed Effects:       Intercept -1.486*** 0.033   Time (centered) 0.292*** 0.016   Time × time (centered) -0.047*** 0.002   Gender (centered) -0.476*** 0.052   Age (centered) 0.132*** 0.006   Education (centered) -0.032*** 0.006 Random Effects:       Intercept 0.381*** 0.017   Reduced random-effects logit model Fixed Effects:       Intercept -1.472*** 0.031   Time (centered) 0.271*** 0.018   Time × time (centered) -0.047*** 0.002   Gender (centered) -0.446*** 0.051   Age (centered) 0.129*** 0.006 Random Effects:       Intercept 0.380*** 0.017   Reduced fixed-effects logit model Fixed Effects:       Intercept -1.448*** 0.027   Time (centered) 0.229*** 0.007   Time × time (centered) -0.045*** 0.002   Gender (centered) -0.390*** 0.040   Age (centered) 0.116*** 0.003

***p<0.01.

Note: Randomness of the intercept is parameterized by the variance of the random effects.

Table 1: Analytic results of three binary logistic regression models on the interwave death rate in older Americans (N=7,480).