Kleck's complaint is that when the data was originally collected two decades ago, CDC researchers could have calculated an estimated number of DGUs from the raw

Kleck's complaint is that when the data was originally collected two decades ago, CDC researchers could have calculated an estimated number of DGUs from the raw data and reported summary statistics relating to it, but didn't do so. They did do public estimates for other categories, just not this one. Kleck thinks they didn't release any analysis of this data because they didn't like the conclusion. (Which seems plausible but would be a hard thing to prove) Kleck wrote this report once he realized the raw unanalyzed response data (which is what you found) had at some point become available at the CDC site so he could do his own analysis. >I was able to find this page through Google search You didn't need to - Kleck's data source is the very first item in the "References" section of OP. It was: > Centers for Disease Control (CDC). 2018. Behavioral Risk Factor Surveillance System website at https://www.cdc.gov/brfss/annual_data/annual_data.htm https://www.cdc.gov/brfss/annual_data/annual_data.htm. Accessed 2-14-18.