The booming of higher education has witnessed an expansion of universities and colleges both in number and in scale, and a hike of tuition fees. The acceleration of marketization in higher education has led to a keen competition for education resources among universities and colleges. Meanwhile, higher education, as an institution for talents\' training and scientific research, has been of great concern for every walk of life, e.g. the government, the institutions, students and parents,—they desire for all kinds of information representing the quality of universities and colleges. This stirs up the enthusiasm for college and university rankings (CUR) and a research on ranking results and methods as well. This dissertation will review the development of the college and university rankings (CUR) system both home and abroad, its research areas and research outcomes, and as such ascertain the orientation for the CUR research.First, the rationality of methods and results in rating universities and colleges has been of top concern, particularly the rationality of indicators\' weight concerned and whether universities and colleges should be evaluated by types and by class. Most of the current research has been grounded on qualitative analysis and as a result lack of theoretical and empirical evidence as they seldom provide criteria for evaluating the rationality of rating methods and results. Secondly, current research shows that the outcome of the CUR has exerted a great impact on institutions, students and relevant spheres, particularly on the competition for education resources among institutions. Therefore, it is of realistic importance for universities and colleges to get advantage in ranking. But little research has been carried out in this area. Thus the dissertation is determined to substantiate the above two research areas in the CUR.Within the research framework outlined above, the dissertation first draws a comparison of methodology used in the CUR home and abroad in terms of elements of statistical comprehensive evaluation methodology, elaborates on the problems of the CUR methodology and forwards suggestions for improvement based on the idea that elements of statistical comprehensive evaluation methodology lay a theoretical foundation to test the rationality of the rating results and provide an impetus for standardization in the methodology of the CUR. In the following, based on the internal quantitative relationship between evaluation indicators and evaluation outcomes of the CUR and by making use of the econometric models, this paper presents a quantitative analytical method to test the rationality of the indicators\' weight and the rating outcomes, and to show how institutions can gain competitive edge in the CUR. By using the China\'s university ranking data published annually on the website(www.netbig.com ) to carry out an empirical analysis, and thus provides criteria for testing rating methodology, particularly the rationality of the indicator\'s weight, and offers ways to help institutions choose the most competitive strategies to better their position in the rankings. The empirical analysis also provides empirical evidence for the necessity that universities and colleges should be evaluated by types and by class.The main contribution of this paper lies in the research providing a test for the rationality of the evaluation method and result in the CUR and how institutions can seek more effective competitive strategies to better their position in the CUR. These two research areas are seldom covered by scholars home and abroad and this paper is tempted to shed some highlights in these two areas by a quantitative analysis based on the Models with Discrete Dependent Variables.