Web Survey Bibliography
The paper analyses a split-ballot repeated experiment on knowledge questions on basic EU facts in a panel web survey carried out in Italy before and after the 2014 European Elections (N > 2500). The control group received a neutral prompt to the knowledge questions, while the treatment group received a normative prompt inviting the respondents to answer without using internet. The first results show a consistent lower proportion of correct answers for the treatment group, lack of relation between “cheating” and socio-demographic characteristics, and a misleading effect of increasing consistency of the knowledge scales when “cheating” is more diffuse.
European survey research associaton conference 2015, ESRA, Reykjavik >>