A host of studies during the past two decades have explored the relationship between how much citizens know about science and how supportive they are of science and scientists, as well as how optimistic they are for positive social outcomes arising from scientific and technological innovation (Miller 2002). The broad conclusion from this research is that a more informed citizenry tends to be one that is more sympathetic to science (Allum et al 2008). The reasons why this should be so are much less well-understood. One plausible hypothesis is that the observed correlations between knowledge and attitudes in surveys about science and technology are due to some other unmeasured characteristic of people who are highly knowledgeable about science. For instance, those who learn more about science at school or college are probably also exposed to more positive images of science and learn to value science as an institution and a profession. So knowledge may not itself be so important, but rather, the other things that tend to come with knowledge acquisition.
In this paper, we take advantage of an unique opportunity to test this hypothesis provided by GSS data collected in 2006. To estimate the
causal effect of scientific knowledge on attitudes towards science, for which there are measures in the GSS, we capitalise on the cognitive
ability measure (‘wordsum’) that was also fielded in that year. Using instrumental variables estimation we use cognitive ability as an
instrument for science knowledge in order to make more secure causal inferences about the real effect of science literacy on attitudes. We
compare the IV estimator with a number of other estimators of the same effect and summarise the findings. This is very much work in progress
and we are keen to use the seminar to improve the paper.
Presented by:
Nick Allum, Department of Sociology (joint with Paul Stoneman, Department of Sociology)
Date & time:
May 6, 2009 12:00 pm - May 6, 2009 1:00 pm
Internal seminars home