The language of citizen science: short strings and 'we' as a group marker

Citizen Science is a form of science crowdsourcing where members of the public offer their time to help scientists tag, classify and collect data in the field. The concept arguably began when the US Audubon society decided to count birds rather than shoot them as a Christmas day activity in 1900 (Silvertown 2009). Over the last eight years Zooniverse.org has grown to be the world's largest online platform offering volunteers the opportunity to classify galaxies based on NASA images, work with camera traps in the Serengeti, and monitor penguins in the wild.

In this chapter, I introduce the discourse of citizen science discussion forums and ask what role short strings of words such as that we are and we are in play in terms of group identity. I compare the linguistic culture of Zooniverse forums to that of a popular blog - ScienceBlogs. The corpus analysis software WordSmith Tools (Scott 2012) is used to focus on strings containing the word we and the data is tagged according to the referent of we in each line such as we meaning human, the scientific community or a particular team of researchers.

Analysis shows the tendency for one thread of Zooniverse to focus on its shared membership of the group human rather than split up into smaller groups whereas the results from ScienceBlogs are more consistent with an atmosphere of debate. The posters, for example, use we to identify with sub-groups groups such as USA or Christians. Certain strings also seem to be associated with particular contexts in unexpected ways such as a high frequency of that we are when we refers to all humans; results are discussed in terms of Hoey's Lexical Priming (Hoey 2005) and my previous work on Lexical Selection (Hadikin 2015). Overall, the chapter highlights the need for careful consideration of the context when trying to understand communication in online groups.