This week we’ve been exploring the world of Text Analysis, using tools such as Wordle, Voyant and Many Eyes. To do so, we needed text to examine, and for that purpose we reused the data sets created from our previous TAGS and Altmetrics exercises. I’ll be writing a post this weekend in which I’ll tell you how I got on and what I’ve found out, but today, it seems apposite to reevaluate the benefits and limitations of TAGS.
TAGS is an app developed by Martin Hawksey @mhawksey which mashes up the Google and Twitter APIs in order to collect tweets and their metadata, and display them on Google spreadsheets for analysis. It’s quite an ingenious solution to a serious issue confronting information professionals and other social scientists – how can we make sense of the masses of data created on Twitter? Or as Richard Rogers puts it in Twitter and Society (2014), “debanalising Twitter”. The following quote from Rogers reinforces the pertinence of TAGS, “Twitter is particularly attractive for research, owing to the relative ease with which tweets are gathered and collections are made” (preface, p. xxi)
Hashtags are an important component of Twitter, facilitating the creation of communities and conversations. I particularly like the fact that this was a development which came from the user base of Twitter rather than the designers of the app, since for me, it’s a perfect illustration of the web’s democratic nature and the potential therein to create previously unthought of positive outcomes. The TAGS app allows us to search the tweets under any hashtag, and thereby draw some conclusions about the communities the hashtags represent.
I made two searches, one for #citylis and the other for #libchat. I searched for #citylis because it’s the hashtag for my LIS course at City University and so I was curious to see the patterns of the hashtag’s usage. I searched for #libchat because it’s also a hashtag I follow on Twitter, particularly when they’re having a prearranged live discussion, similar to the one just announced by#uklibchat
#uklibchat topic for December 2 6.30pm - 8.30pm will be on Negotiating and Influencing.—
#uklibchat (@uklibchat) November 18, 2014
The app was pretty straightforward to use although it does require a bit of patience at times; when my classmates and I were all using TAGS simultaneously it was noticeably slower than when I repeated a search today at home.
I wont share my findings of both searches because it will be repetitive, but will report back on my findings regarding #libchat. If you would like to read an analysis of #citylis then please refer to my blogroll which list the blogs of my classmates, although this post by Dominic deals with it very effectively.
TAGS created an archive, displaying the number of tweets and top tweeters:
A list of the most retweeted tweets in the past 24 hours:
An hourly display of activity under the hashtag:
And a line graph displaying tweet volume over time.
The last graph clearly shows spikes of activity whenever a prearranged topic is occuring, but we can also see that the hashtag is used at other times as well.
To conclude, TAGS is an imperfect but very useful app which can be used in many different ways, and is ideally suited to making sense of the data which is generated on Twitter. I plan to use it much more in the future, and in my next post you will be able to read how I used the data derived from TAGS to conduct a text analysis.