Similar strategies can be applied for the analysis of outcomes other than case enumerations.
Case enumerations at the Census county, tract, and block group level are considered as the outcome of interest when applying and evaluating the imputation methods.
We consider total case enumerations at the Census county, tract, and block group level as the outcome of interest when applying and evaluating the methods.
The procedure allows for the potentially biased and likely under reported outcome, case enumerations based on only the geocoded records, to be presented with imputed statistically adjusted results.
These imputed cases were then combined with the experimental geocodes providing a complete data set (geocoded plus imputed nongeocodes) for which case enumerations can be calculated and compared to that obtained using the complete geocoded data set.
The assumption of independence, however, is clearly no longer justified when an outcome of the geocoding process that is a function of more than one case is considered, such as with case enumerations.
The procedure allows for the potentially biased and likely under reported outcome, case enumerations based on only the geocoded records, to be presented with a statistically adjusted count (imputed count) with a measure of uncertainty that are based on all the case data, the geocodes and imputed nongeocodes.
Evident from these results is that imputation for case enumerations at the county, tract, and block group level improves as we move towards Strategy 3 both in terms of the percentage of times the imputation interval contained the true total as well as average width of the intervals.
For instance, in the case of enumeration using 3 processors and division depth d = 2 for C2O2H2, vertices 0, 1, 2, 3 are assigned to processors 0, 1, 2, 0, respectively by BfsEnumP1 (see Figure 1).
In the case of enumeration, the subjects had to indicate the number of perceived items by pressing the corresponding number on a keyboard immediately after target or mask offset.
