GIS and Academic Portfolio
Like working with any dataset, an analysis of volunteered geographic information (VGI) is an analysis of the dataset and collection methods first, and of the real world second. One point that Crawford and Finn make in their article is that an analysis of Twitter data is incomplete without also an analysis of how the Twitter platform and algorithm affect networks of communication on the app (2014). And this is true. When Wang and Tsou find, for example, that Twitter networks during a wildfire are centered around local news sources, that is not indicative of any aspect of the wildfire’s natural spread. But that doesn’t mean that Twitter data is unhelpful. With so much human communication on Twitter and platforms like it, they have real-world impacts offline that are worthy of study. There are certainly reasons to be cautious of how broad of implications are being drawn from Twitter. As Crawford and Finn note, ‘Twitter use still skews to younger, more urban demographic groups, even in wealthy nations like the US’ (496). Questions of where data are being drawn from are important and need to be taken into account. But this bias and these questions aren’t new to academia or unique to social media analysis.
I find the ethical consent and privacy issues around Twitter data to be really interesting. All of the other VGIs that I’ve used before (OpenStreetMap, CollectEarth) were very obvious in their intent, and anyone contributing would theoretically know the potential uses of the data they provide. With Twitter, however, nobody sends Tweets imagining them to be used as data or seen beyond their limited follow count. And I would guess that many people who post on Twitter with their geolocation enabled don’t even realize it. Especially during high-stress times like a disaster, one sends their tweets spontaneously into a void in rapid succession, hoping to never to hear from them again. This creates ethical uncertainties in so far as how much can and should be done with data that was never meant, intended, or desired to be researched. And while Twitter most likely addresses some of these issues in their terms of service, it can be assumed that most people tweeting agreed to the TOS without actually reading it.
References:
Crawford, K., and M. Finn. 2014. The limits of crisis data: analytical and ethical challenges of using social and mobile data to understand disasters. GeoJournal 80 (4):491–502. DOI:10.1007/s10708-014-9597-z
Wang, Z., X. Ye, and M. H. Tsou. 2016. Spatial, temporal, and content analysis of Twitter for wildfire hazards. Natural Hazards 83 (1):523–540.