We also developed a "pooled" model across these sampling periods to provide an estimate of an annual average.
y is real log hourly pay, OE is the dummy for overeducation, T is a (0/1) time dummy capturing the two periods, and X comprises controls for age, age squared, sex (pooled model only), dummies for ethnicity (Asian, Black, other, missing, and White as reference group), marital status, indicator for dependent children in the household, and a London dummy.
For the analysis we examined a pooled model and a model where we stratified by group membership (i.e., NICU vs. healthy baby sample) to separately examine the contribution of each predictor variable for the two samples.
When the number of output genes is fixed (e.g., top 100), the pooled model found appreciably more truly differentially expressed genes than the individual studies.
Roosting and foraging traits influenced spatial bias, but distance to protected areas was the greatest contributor to spatial sampling bias in a pooled model and 8 out of 10 ecological trait group models.
In the pooled model, we assume that each estimated coefficient is the same for each bank.
We estimate an average policy treatment effect in a pooled model and in a meta-analysis that combines separate estimates for each treatment-control pair.
13I also include a group indicator in the pooled model as an additional variable to avoid having any "residual group differences spilling over into the slope parameters of the pooled model" as recommended by Jann (2008) (pp.6).
These effects appear very clearly in the pooled model, but they appear to be somewhat context dependent.
where lnL P, lnL d, and lnL C represent the log-likelihood function values obtained from the pooled model, the participant, and the non-participant subsamples, respectively.
Note that the average of the coefficients reported in Table A3 - that is the average across the 25 occupational equations - is -0.07, within the range of results obtained from estimating the pooled model that is reported in Table 2.
