research methods – Daniel Allington

Public seminar by Daniel Allington
Starts: 16:00 15 Nov 2017
At: Mitchell Centre for Social Network Analysis, University of Manchester

Who follows British politicians on social media? Who stood with Ken Livingstone online? What would it be like to get all your political news from Twitter?

For over a year, I’ve been seeking answers to these questions and more using data scraping and a mixed methods approach centred on social network analysis. Social media have changed British political culture, creating quasi-celebrities out of figures who would otherwise have been condemned to the margins, and giving wide circulation to ideas long believed to be politically defunct – most alarmingly, the belief in an international conspiracy of Jews. In this seminar, I will present theoretical and methodological approaches to the large-scale study of online political culture, as well as sharing preliminary findings.

Open to all. Booking via the University of Manchester events website.

Over the last few months, I’ve often heard that polls can’t be trusted. In particular, I have heard that they can’t be trusted because each one usually involves study of only about a thousand individuals. I have even heard that argument from a retired quantitative linguist.¹ So I’ve put together this essay in order to explain how polls work, why a random sample of a thousand should usually be considered sufficient, and why the results should be treated as informative even though they do not enable us to predict precise numbers of votes (which is a particular problem when the results are going to be close — because then, precise numbers can make all the difference).

Continue reading “Informative, not infallible: why polling mostly works pretty well (even though only about a thousand people usually get sampled)”