On the first day of the recent federal election campaign, April 11th 2019, I constructed a search constituted of election-related hashtags, terms related to the election, and the names of parties participating in the election.
The intention was to map twitter talk about the election, but without attempting to gather every tweet referencing the election, which would have been a mammoth data-set, very time-consuming to process over a six week period: not feasible in my spare time.
Instead, the sample of tweets collected are from one day each week, from 8:30am to 11pm EST. The election was called on a Thursday, so I went on collecting tweets during the same time period, with the same search terms, each Thursday. Then, when particular discussions of interest came up – on climate crisis policy, on Labor’s pledge to continue turning back asylum-seeker boats, etc. – I collected tweets on those separately. As it turned out the regular maps from the weekly sample provide a good ‘baseline’ picture of Australian political twitter for comparison with maps on particular topics and events.
This post is not intended as analysis of the maps – I’ll get to that another time – only to put them up so that they’re publicly available somewhere. There are two maps from each dataset. The first is a map of accounts tweeting about the election. The second is a map of the most prominent 300 terms used in the sample of tweets.
One point worth making here is that the largest number of tweets captured from the six week time-frame was on day one. After this, tweet volume halved in week two, and then crept up again, on the 36th day not quite matching the first.