The I/EM refers to the process in which a person compares his or her ability in a subject with that in another subject to construct a subject -specific self-concept when evaluating his or her ability in the subject. It investigated theoretically into the effects on the construction of academic self-concept produced by both the intra-individual and inter-individual frames of reference, proving that the construction of students’ academic self-concept is influenced by external (or social) comparison as well as internal comparison of individual’s different subject-specific ability. Since the I/EM has been proposed, many researchers have used different methods to test its cross-culture generalizability widely, such as examining it in only one country or culture, examining the moderating role of country after a meta-analysis of the model was made, making a comparative studies in different countries or cultures using the same instrument and method, all of these studies supported the cross-culture generalizability of the I/EM. However, lots of studies suggested that the construction of academic self-concept of Chinese mainland adolescents had its specificity due to the distinct difference in self-concept and its neural mechanism between western and Chinese mainland adolescents. Therefore, to more critically examine the cross-culture generalizability of the I/EM, we should also focus on the analysis of its regional and ethnical difference when examining it using national big samples. M?ller and Marsh (2013) proposed the dimensional comparison theory (DCT) by means of summarizing the studies of the model, and defined dimensional comparison a process that a single individual compared his or her ability in a domain with his or her ability in another domain. The DCT became another important comparative theory following the social comparison and temporal comparison. Dimensional comparison may produce contrast effects and assimilation effects, dimensional comparison predicts contrast effects when the nature of contrasting domains has a big difference (far comparisons: e.g., the negative effect of verbal achievement on math self-concept), while much weaker negative contrast or even positive assimilation effects may be produced when the nature of contrasting domains have a small difference (near domains: e.g., positive effects of verbal achievement on English self-concept). Although this explanation to certain degree seems to be reasonable, some studies also found that contrast effects could be produced despite of the small difference of the nature of the contrasting domains. More studies are needed to identify the context where the contrast or assimilation effects of dimensional comparison are more likely to occur. As the studies getting further, the researchers gradually found that the I/EM was insufficient construction of academic self-concept, and tried to integrate it with other theories or models in order to explain the construction of academic self-concept more comprehensively. Recently the integration with the big fish little pond effect (BFLPE) was one of the most important study. the I/EM and its integrated models have just investigated the role of dimensional comparison and social comparison in construction of academic self-concept, temporal comparison was ignored. Future studies should strengthen further its integration with temporal comparison, and sort out the role of the three comparison in processes of construction of academic self-concept. There are at least two aspects about the enlightenment for academic self concept research and educational practice: Firstly, strengthening the subject specificity research of academic self-concept and other academic variables; Secondly, teachers and parents should help students to select proper comparison style and frame of reference.