My research, at the moment, is thoroughly low risk. The methods I’m considering do involve human subjects, but only by way of their tweets and Tumblr posts – none of which are put online under any promise of privacy, and as I understand it, using public online postings is tantamount to using newspaper letters to the editor for research purposes. The people involved are (as far as I know) not particularly vulnerable, nor will my research put them in any danger that their posting to Tumblr or Twitter didn’t place them in the first place. But since humans are, however indirectly, involved, I’d expect to be eligible for a Delegated Ethics Review.
The one complication that I’ve been thinking about is the difficulty of determining when someone is technically a child or not based solely on their social media posts. I wonder what the impact of potentially including ie a bunch of 17 year olds in my research would be; and what tools I might have to find them out, if their inclusion does pose ethical concerns.