ICT-mediated learning appears to hold great promise for achieving the goal of education for all such as, reducing the long-existing disparity between North and South, reducing poverty and promoting social inclusion. However, the integration of ICTs in education requires considerable investment in time and resources. Consequently, when planning to integrate ICT in education and training policy makers should be able to use evidence-based information for making sound decisions. In spite of the critical importance of sound research to guide policy and practice, it appears that there is a lack of valid and reliable evidence-based information in the field of learning technology. Many studies conducted during the past 70 years have failed to establish a significant difference in effectiveness between learning technology and traditional methods. While these findings tend to suggest that learning technology does not considerably improve learning, the fundamental question that remains unanswered is: "were the researchers assessing the effectiveness of ICTs or were they simply assessing the effectiveness of instructional treatments that were initially less than perfect? If the instructional treatment is weak or flawed it may lead the researcher to reach false conclusions. This purpose of this paper is to propose a framework for establishing, a priori, the effectiveness of ICT-mediated instructional treatments used in educational research.