How bad data fed the Ebola epidemic — Rachel Glennerster, Herbert M’cleod and Tavneet Suri

MIT Sloan Associate Prof. Tavneet Suri

MIT Sloan Associate Prof. Tavneet Suri

From The New York Times

The West African Ebola outbreak first hit Sierra Leone in May 2014, followed by an explosion of cases in the capital Freetown in the autumn. The epidemic now counts more than 10,500 cases across Sierra Leone, with signs that the spread is slowing.

The early days of the crisis were characterized by a sense of immense fear, anxiety and alarm, regionally and globally. In Sierra Leone, a three-day, countrywide, military-led lockdown in September fed the fear in West Africa and beyond. Many flights originating in unaffected African countries were restricted. African students were prevented from attending some American schools, and there were countless reports of discrimination against Africans across the globe. Pictures of health workers in full protective suits became a ubiquitous symbol of the panic.

Misleading reports, speculation and poor projections from international agencies, government ministries and the media about the Ebola outbreak exacerbated the problem. The fear that was spread by the dramatic reports that accentuated the negative, undermined confidence, made it harder to encourage people to seek care, and misdirected attention away from Sierra Leone’s urban areas, where data suggest the economic effects of Ebola have been concentrated.

Valid, credible and timely data is essential during a global crisis. Without reliable data, efforts to assist affected people and to rebuild damaged communities can be misdirected and inefficient.

Read More »

How bad data can lead to good decisions (sometimes) — Roger M. Stein

MIT Sloan Senior Lecturer Roger Stein

MIT Sloan Senior Lecturer Roger Stein

From Computerworld

Before companies can profit from big data, they often must deal with bad data. There may indeed be gold in the mountains of information that firms collect today, but there also are stores of contaminated or “noisy” data. In large organizations, especially financial institutions, data often suffer from mislabeling, omissions, and other inaccuracies. In firms that have undergone mergers or acquisitions, the problem is usually worse.

Contaminated data is a fact of life in statistics and econometrics. It is tempting to ignore or throw out bad data, or to assume that it can be “fixed” (or even identified) somehow. In general, this is not the case.

Read More »