When does a data stream become a tsunami?

November 1, 2012

Could society be placing too much reliance on the quantity of information at its disposal when it should be equally focused on its accuracy?

That is one of the key questions that will be asked by Harvey Goldstein, professor of social statistics at the University of Bristol, in a debate that will explore the opportunities and challenges posed to social scientists, policymakers and the public by ever-expanding data streams.

The Big Data event at the British Academy, due to take place on 5 November, will assemble academic researchers and statisticians to discuss how best to use the increasingly accessible sets of public data on anything from education to road safety.

Other than Professor Goldstein, panel members for the event - for which Times Higher Education is media partner - are Paul Boyle, chief executive of the Economic and Social Research Council, Paul Woobey, director of the strategy and standards directorate at the Office for National Statistics, and Farida Vis, research fellow in the social sciences at the University of Sheffield.

The British Academy showed its commitment to the importance of open data when, in 2011, it launched "a four-year programme to support quantitative skills in the humanities and social sciences", stating that "people need to have the tools to interpret and critique the increasing amount of data now available".

Professor Goldstein will argue that it is often assumed that more measurement is always a good thing for policymaking, as if an increasing volume of information always translates into better decisions.

Although the availability of data has obvious benefits, Professor Goldstein will raise a number of questions about "the data tsunami threatening to engulf us".

School league tables are widely cited, for example, yet there has been little public debate on the issues of "added value" and "uncertainty intervals" that are crucial to interpret them, he will argue.

Neither policymakers nor advocates for "data freedom" feel any responsibility to explain why these statistical details actually matter.

Although he applauds initiatives such as the Royal Statistical Society's "getstats" campaign "to improve how we handle numbers", Professor Goldstein will suggest that far more needs to be done to divert the data "tsunami" into safer channels.


Register to continue

Why register?

  • Registration is free and only takes a moment
  • Once registered, you can read 3 articles a month
  • Sign up for our newsletter
Please Login or Register to read this article.