Policy on comparing followers of politicians and journalists

[Hesketh_GD]

Hi,

I’d like to use the API to research the similarity between the followers of different UK politicians and journalists. I would report aggregated follower statistics NOT individual user ids, although I would likely need to specify that some statistics are associated with individual politicians. I may wish to:

Associate patterns in the followers of politicians with their voting record
Associate patterns in the followers of politicians with patterns in the followers of journalists
Write a research article on media bias/diversity
Build an app that recomends journalism to people based on who they follow

Is any of this in violation of the user privacy section of the Developer Agreement or any other agreed terms? I am concerned about VII. A. 4 in particular.

[yoyoel]

Hi! Thanks for your question.

While we can’t comment on specific apps in the forums, the key consideration with section VII.A.4 in the Developer Agreement is not to segment or profile individual users. As we note in the Restricted Uses documentation:

Aggregate analysis of Twitter data that does not process any personal information (e.g., information about identifiable accounts or individuals) is permitted, provided that the analysis also complies with applicable laws and all parts of the Developer Agreement and Policy.

[Hesketh_GD]

Thanks for your reply.

Just to be absolutely clear then, is it the case that VII.A.4 forbids me from writing, e.g. the following sentence in a research article:
“Andrew Neil at the BBC shares 20000 followers with Jeremy Corbyn, Leader of the Labour Party and 30000 followers with Theresa May, Leader of the Tory Party”

Does it make any difference that these are all high profile individuals working in politics and would expect people to analyse their political relationships?

If not this forum, is there a contact where I can discuss my particular case?