Although Oxford retained its first-place position in The Times Good University Guide 2012, the institution lost out to Cambridge in The Complete University Guide 2012 and in the rankings published by The Guardian.
Other big winners included the University of Exeter, which improved its standing in all three tables, including a substantial nine-place jump from 24 to 15 in The Complete University Guide's table.
For the fourth year, Times Higher Education has compiled a "Table of tables" to provide an overview of how the sector is seen by the national newspapers.
The Times newspaper is no longer formally linked with Times Higher Education, which was sold by News International, the newspaper's owner, in 2005.
Exeter and the University of Bath both re-entered the top 10 of the "Table of tables" this year. Their ascent was bad news for the University of York, which has fallen out of the top 10 after recording drops in all three league tables.
City University London and the University of Kent have entered the top 30 of the "Table of tables" for the first time after improving their positions in The Guardian's rankings.
This success comes at the cost of Aston University, the University of Dundee and the University of Leeds, all of which have dropped out of the top 30.
Aston and Leeds had previously featured in the top 30 for all previous "Table of tables" rankings, but slipped out this year after falling out of the top 30 in The Times and The Complete University Guide rankings respectively.
How the table is set
Each year since 2008, Times Higher Education has compiled a "Table of tables" to offer a snapshot of how the UK higher education sector is viewed by the national newspapers.
This year, the "Table of tables" was calculated by giving the top 30 institutions points corresponding to their position in The Times, The Guardian and the Complete University Guide tables (30 for the top place, 29 for second and so on).
The total scores determined the overall positions of the institutions.
We make no claims for statistical rigour and acknowledge the methodological limitations.