I am looking for the documentation of Advan data and I can’t find it anywhere on the site. Should I use the documentation from SafeGraph as a reference if the Advan documentation is not available? Thanks.
I am writing to inquire about the census block group (cbg) data that is being used in the monthly pattern dataset. Specifically, I am interested in knowing which year of the cbg data is being used for the 2019 and 2022 monthly pattern datasets.
Could you please confirm which year of the census block group data the monthly pattern dataset is used for aggregation? Knowing this information would be very helpful for my analysis.
@lejanek I found this documentation from SafeGraph about the CBGs they use. I believe Advan used the same source when they transferred over this product. Looks like 2016 vintage of the 2010 - 2019 CBG data. Hope that helps
What is the best way to get in touch with someone at Advan regarding the academic data? I need to discuss a licensing question.
When I try to use the Contact Us form on their website, I get an error saying “Please provide a corporate email address.” I have no idea what that means; I’m providing my university email. Can someone help?
I am looking at the sample for the weekly patterns dataset and I have noticed several things which I need to clear up before being able to use this data for my research:
Although the rows are supposed to represent POIs in cities, some boroughs of New York City also appear with their own entries (Brooklyn, Bronx, Staten Island, and Queens). The general New York City entries also cover POIs in those boroughs, so why is this separation made just for some datapoints and why is Manhattan excluded?
When looking at the POI spatial distribution, I noticed that for New York City, Queens is very poorly covered compared to the other boroughs. This drastic heterogeneity makes any analysis of New York City not valid. There is a similar problem with Montréal, which has some regions with no POIs at all. Is there some mistake and this data should be somewhere, or what is the reason for this?
In general, when plotting the coordinates of each POI, I noticed that there are always several datapoints that do not belong to the city (sometimes not even the region) that they are supposed to belong to. What is the reason for this and how can it be fixed?
Can you try checking the SafeGraph Places data to see if you’re seeing the same issues with POI coverage and locations in that dataset?
Advan relies on the SafeGraph’s POI data to compute Patterns. This will help us determine if that trends your seeing are from the underlying POI data, or in the computation of Patterns.
As a note, Patterns is only computed on a subset of total POI since they need to reach a certain threshold of visits from the mobile panel to qualify to be added to the Patterns product.
I am planning to purchase your monthly pattern data through Dewey. However, when expIoring for the free data, I found some missing values for the key variable “raw_visit_counts”, especially for the sub_category of “Gasoline Stations with Convenience Stores.” I would like to understand the reason for this missing value.
Hi @yucheng.wang143 - mind providing a little more about what’s missing in the data vs. expectation? Is it that there are certain POI that have Patterns data, but are missing that one attribute? Or that not all POI of this sub_category have Patterns data? I know that for privacy purposes, most providers will not include visits data if the count from their panel data is only a few visitors per month.
Thanks for your reply. I am looking at the Advan monthly pattern data. Let’s say electricity charging stations, which belong to the sub_category of “Other Gasoline Stations”. All of them are missing the attribute “raw_visit_counts”. I wonder if it is because electricity charging stations do not allow to share of visits data
Where can I get the crosswalk between the safegraph_place_id and the placekey? It seems that safegraph_place_id has been replaced with placekey. I am unable to find the crosswalk file.
Hi @elliotoh, unfortunately there is no way to create a safegraph_place_id ↔ Placekey crosswalk.
The Place ID was retired in July 2021. Since then, the POI’s in the data have gone through a number of additions, drops, changes, etc., so there is no way to create a crosswalk file that would be reasonably accurate or have good coverage.