Discussions

Hello,

I'm working with the Advan data and I'm a little worried about the truncation in VISITOR_HOME_AGGREGATION. This excludes visits of just one person from the tract or cbg and also rounds up to 4 when 2 or 3 visitors were recorded. For my particular project, it's really more interesting to know which CBSA the visitors come from. Of course I can calculate this from VISITOR_HOME_AGGREGATION, but the truncation could cause those numbers to be far off, since there are many tracts in a CBSA. Would it be possible to have a variable from the same sample as that used for VISITOR_HOME_AGGREGATION but aggregated to the CBSA (or county or state) level? That way we can look at broader trends while minimizing truncation bias and still respecting privacy.