World Reputation Rankings 2011 methodology

Behind the numbers

The Times Higher Education World Reputation Rankings are a subsidiary of the annual World University Rankings, and they are based entirely on the results of a worldwide survey of academics

They are a measure of a university's reputation for excellence, in both teaching and research, among experienced university academics around the world.

The reputation rankings are drawn from an Academic Reputation Survey carried out by polling company Ipsos for our rankings data provider, Thomson Reuters, as part of the Thomson Reuters Global Institutional Profiles Project.

The same survey results formed two of the 13 performance indicators used to create the World University Rankings 2010-2011, published on 16 September 2010. The reputation data are revealed here in isolation for the first time.

The invitation-only survey was sent to tens of thousands of experienced academics, based on the United Nations' estimates of global academic researchers by geographical area. The survey was offered in eight languages: Japanese, Simplified Chinese, Spanish, French, German, Brazilian Portuguese, European Portuguese and English.

A key feature of the survey was the opportunity for narrow disciplinary focus: respondents could highlight what they believed to be the strongest universities, regionally and globally, in their specific fields, selecting from hundreds of disciplines and from more than 6,000 academic institutions. "Action-based" questions - such as "where would you recommend a top undergraduate should study for the best postgraduate supervision?" - were used to encourage more thoughtful responses and more meaningful results.

The survey was distributed between March and May 2010 and 13,388 people from 131 countries provided usable responses. The average respondent had been working at a higher education institution for more than 16 years and had published more than 50 research papers.

The key to understanding: reading the table

Our table ranks institutions according to an overall measure of their esteem that combines data on their reputations for research and teaching.

The two scores are combined at a ratio of 2:1, giving more weight to research, because feedback from the global higher education community suggests that academics have a greater confidence in their ability to make accurate judgements on research quality.

The reputation scores are based on the number of times an institution was cited by survey respondents as being "the best" in their narrow fields of expertise. Each respondent was able to nominate a maximum of 10 institutions.

The number one ranked institution, Harvard University, was selected most often. The scores of all the other institutions in the table are expressed as a percentage of Harvard's score, set at 100. For example, the Massachusetts Institute of Technology received 88.4 per cent of the number of nominations for research that Harvard received, giving it a score of 88.4 compared with Harvard's 100.

This scoring system is different from the one used in the World University Rankings, and is intended to provide a clearer and more meaningful perspective of the reputation data.

Phil Baty is editor of Times Higher Education Rankings