This thesis studies the assessment of academic journal impact via the analysis of citations. In the research community, to most common approaches have been widely adopted to evaluate the journal impact are (i) the conduction of expert survey, and (ii) the calculation of cross citation quantity, such as the well-known SCI impact factors. The 2ndapproach is generally accepted for its objectiveness. However, using impact factor only is also considered to have bias to a certain degree, due to a widely-held notion that citation quantity does not represent citation quality....[ Read more ]

This thesis studies the assessment of academic journal impact via the analysis of citations. In the research community, to most common approaches have been widely adopted to evaluate the journal impact are (i) the conduction of expert survey, and (ii) the calculation of cross citation quantity, such as the well-known SCI impact factors. The 2ndapproach is generally accepted for its objectiveness. However, using impact factor only is also considered to have bias to a certain degree, due to a widely-held notion that citation quantity does not represent citation quality.

This thesis has addressed the citation quality in journal from two perspectives. One perspective is to reflect the citation importance. We studied an eigenvector analysis (called Page Rank) to reflect the importance so that citations from prestigious journals are weighted higher. We have shown extensively that the PageRank results have matched the experts' perceptions much better than impact factor, from the following three perspectives. The other perspective is to reflect citation relevance. We propose a revised PageRank method, which first clusters journals according to their citation patterns so that a cluster of "pure" journals can be identified for a particular discipline, and then evaluate the impact for each journal, according to its citations that are received from "pure" journals and weighted by citation importance.

Effectiveness and the models and the methods proposed in this thesis have been examined by their applications on ORMS journals and MIS journals. By extending results obtained, we have developed an online journal ranking system, which is called Journal-Ranking.com and is perhaps the first online journal ranking system in the world.