Purpose: The purpose of this article is to identify and describe existing students' statistical types by their learning strategies and to show the connection with factual computer literacy. Methodology: The empirical-experimental part of the present study is based on a series of diagnostic studies of 1004 surveyed Lithuanian students. The article shows respondents' typologies according to their learning strategy and the relation between the distinguished statistical types and factual computer literacy. Results: The highest computer literacy level can be related to the interactive learning in a group. It seems much easier to acquire new computer operations with the help of an experienced specialist, who immediately can show and explain what to do and how to do it. From this point of view, listing a "thick" computer manual, studying all illustrations and trying to learn from "trials and errors" is neither really useful, nor attractive. The technology of "statistical types" gives an opportunity to apply a classical, didactic approach concerning differentiation and individualization of learning as a more subtle and effective way. Conclusions: The research data has shown that the usage of a cluster analysis method, searching for statistical types of students' population by the variation of their learning strategies, has served its purpose. During the research the factually existing statistical types of students by their learning strategies were revealed; students oriented towards individual learning, dysfunctional learning and interactive learning were identified and described, and the relation with their factual computer literacy was also determined. The highest level of computer literacy can reasonably be related to interactive learning in a group. Qualitative description of "pure" statistical groups, depending on the expression of features and the determination of these groups by percentage in the general population, is essential information with which the process of formation of computer literacy should be optimized and promoted. While selecting the already existing and/or developing new educational computer literacy strategies and methods, and preparing textbooks and media, we should focus not on abstract "faceless" learners, but on very definite and truly existing types of learners. In other words, the search technology for statistical types provides an opportunity to apply such classical and highly relevant didactical approaches as learning differentiation and learning individualization more effectively and efficiently. (Contains 2 figures and 3 tables.) [This paper was published in: Bruillard, Eric, Aamotsbakken, Bente, Knudsen, Susanne V., and Horsley, Mike (eds) (2006). Caught in the Web or Lost in the Textbook? STEF, IARTEM, IUFM de Basse-Normandie, Paris: Jouve, October 2006, p235-242.]