The user population of academic databases is heterogeneous consisting of users of disparate languages background, computer skills, different needs and different ways in expressing those needs. Academic databases provide scholars with the information to build their knowledge. English language is considered the dominant language in these databases. However, a wide range of academic databases are now available in other languages including Arabic language and more databases offer multilingual retrieval facilities. Searching in different languages might affect users' behaviour, which in turn is reflected in the various search strategies such as: problem formulation, query constructing, information seeking, evaluation of documents and relevance judgment. The variation with which users interpret their needs and follow different ways in searching the tasks in both languages is what motivated this research which will try to explain the information seeking behavior (ISB) of users when searching similar tasks in the Arabic and the English database. This project aims to identify, compare and model the ISB of academic staff when searching Arabic academic databases and English academic databases, by examining the behavior of the academic staff of social sciences faculties in Jordanian universities when searching two academic databases. One in Arabic (E-marefa) and one English language database(Science direct). The research explores whether, and how, the language choice may affect their needs, aiming at the end to propose an information behavior model to account for Cross-Language Information Seeking behavior (CLISB).