I have a few questions about patterns data. Visitor_home_cbgs is for each CBG, and visitor_home_aggregation is for each Census Tract, but if I add all cbgs' visitor's number in one Census Tract, it's not the same, I don't know why?

Hi, I’m Ryan from University at Buffalo. I have one question about pattern data, could you help me? visitor_home_cbgs is for each CBG, and visitor_home_aggregation is for each Census Tract, but if I add all cbgs’ visitor’s number in one Census Tract, it’s not the same, I don’t know why? Also, sometimes I find out that CBG has access data for this area, but Tract does not. Tract data contains CBG data, right?


This topic was automatically generated from Slack. You can find the original thread here.

Hi @Ryan_Zhenqi_Zhou_SUNY_Buffalo ! If there is a census block group with only one visitor, that would not be reported in visitor_home_cbg for privacy reasons. However, it’s possible that the one visitor could be counted in visitor_home_aggregation if there’s another census block group in the same tract with 1 visitor (as 2 is above our privacy threshold). However, please note that in this situation it would appear as “4” due to our differential privacy policy. That’s why you might be seeing a disrepancy between the sum of visitor_home_cbgs and visitor_home_aggregation.

Note that we also add a jitter (Laplacian noise) to the two columns separately as an additional privacy measure. So it’s random noise first, and then the transformation that Niki describes above.

Thanks @Niki_Kaz @Eugene_Chong_SafeGraph , a lot of help~

No problem! To prevent any further questions from being overlooked, I’ll go ahead and close this thread out. If you have any more questions or follow-up questions, we’re always here to help! Just be sure to make a new post to help, as we aren’t monitoring old threads at this time. Thanks!