Reply to comment

How do you rank universities from the developing world, whose graduates end up in the most selective universities for graduate work, or who end up passing the competitive professional exams in the most industrialized countries ? . To me, if universities in resource poor settings can produce high quality graduates who are competitive against the graduates of the most selective and monied schools, then the universitities of the developing world are doing a superb job, and may in fact be better at teaching students !
There needs to be a separation of teaching quality (but how do we judge this?) from research quality in the overall assessment. The rankings are good to the extent that they encourage healthy competition to improve global academic standards. They must not be used to decide university choice, student recruitment, as the parameters measured are not always universally applicable or objective. Citations is subjective and sometimes 'nationalistic". Acceptance of papers to top journals is not totally based on novelty or impact, geographical and nominal factors play a role.
All that is required is a list of league tables - top universities in different fields- based on teaching of their students, and the objective outcome of such teaching (such as scores achieved on USMLE 1 and 2) , GRE, etc. A separate league table for research universities that is field oriented is also useful, but numerical ranking of universities is misleading and confusing to those not so familiar with academic analytics and its limitations.