The criticisms describing the effect of missing data on the analyses of county-level UCR data indicated that the effect of population-weighted regression was ignored; all datasets have measurement error; the wrong States were used in analyses; state-level analyses were similarly affected by missing data; and city- and State-level analyses show the same pattern as the county-level analysis. The team of Maltz and Targonski counters that the re-analysis of the data did not adjust for the errors in those States and called this step inadequate. They also claim that the errors in the county-level dataset are not insignificant and there are a number of different types of errors that affect analyses. These types of errors are measurement error due to missing data, instrumentation error, and stochastic variation. In regard to measurement error due to missing data, one needs to understand and model the process that generates the “missingness” to compensate for it. The instrument performing the measurement may also be in error. Regarding stochastic variation, the uncertainty associated with each datum constitutes a high hurdle that a finding of deterrence must overcome. Categorizing States is far from exact, and depends on interpreting State statutes. State-level analyses are not affected to the same extent as county-level analyses by missing data. The changes in concealed carry laws were probably prompted by the dramatic increase in homicide that occurred during the late 1980's and early 1990's. The subsequent drop should have been anticipated, due to regression to the mean. There is a natural “regression to the mean” in that when the times become more dangerous more efforts are put into play to bring the crime rate down. 3 footnotes, 20 references