CMIST Afternoon Seminars: Combining data-sets to test the normalization of protest thesis'
|Starts:||16:00 21 Jun 2016|
|Ends:||17:30 21 Jun 2016|
|What is it:||Seminar|
|Organiser:||School of Social Sciences|
|Speaker:||Dr Clare Saunders|
Scholars have for a long time claimed that protest is becoming an increasingly common thing to do, and that the people doing it are becoming increasingly normal. But these claims are based on shaky evidence, often using an 'ever' variant of a question about protest participation, which artificially inflates trend lines, and relying on a very small sub-sample of protesters. Measuring the extent to which protesters are normal requires supplementing data from cross-national samples in which protesters are relatively scarce (e.g. around 3.8% of the population 2008-2014) with data from protest surveys.
With Natalie Shlomo, I have been experimenting with ways to combine non-random samples from protest surveys with randomised cross-sectional survey data from the European Social Survey. Three approaches will be compared: 1) weighting the number of protesters and non-protesters according to their frequencies in random samples; 2) sub-classification using propensity scores; and 3) sub-classification propensity score matching with ONS benchmarks. By so-doing, we are able to comment about the differences between non-protesters and those who protest with varying degrees of intensity and frequency. Our preliminary findings show that those who protest the least are most similar to non-protesters, whereas those who protest more are substantively different. Overall, it has not been possible to substantiate the claim that protesters are normal.
No registration needed. All Welcome. Tea & coffee from 15:45.
Dr Clare Saunders
Role: Associate Professor
Organisation: University of Exeter
Travel and Contact Information
Humanities Bridgeford Street