I have the following question: are there ethical considerations to be aware of when using Safegraph data in academic research? For instance, what statement should I include in any papers I publish? Also related to this, do I need to obtain research ethics board approval from my university separately? I haven’t downloaded the data yet so am not familiar with the level of aggregation.
I have the following question: are there ethical considerations to be aware of when using Safegraph data in academic research?
@Razvan_Romanescu_University_of_Manitoba thanks for your question. two forms of data ethics that most researchers focus on are the following:
info regarding privacy can be found here in the places manual : Places | SafeGraph Docs
safegraph uses differential privacy techniques and aggregates at the CGB level, thus protecting individual user data and stoping this data from being derived
- sample bias
SafeGraph has done some work to assess how representative its sample of devices is of the entire population. Specifically, check out the Measure and Correct Sampling Bias section of the Data Science Resources. A recent external audit was done that might also be of value. The audit finds SafeGraph’s panel underrepresents older people and minorities. We hope that normalization techniques correct for some of that bias, but it is still an important consideration.
regarding getting approval from your university’s ethics board, since the data is not on the individual level, there generally shouldn’t be any need for an IRB approval for using safegraph data for research. You should however double-check with your university to make sure.
let me know if you have more questions
just a little more information on this:
the citation and publishing section of the FAQ doc should also help with your question
Let me know if there are any more questions
Thanks, Pranav. This is very helpful.