Multilevel international data have been released from the Third International Mathematics and Science Study (TIMSS), providing an opportunity to apply multilevel modeling techniques in educational research. In this paper, TIMSS factors are classified in fixed and random categories according to the project design. Classifying fixed and random factors may help clarify the relevance of hierarchical linear modeling for TIMSS investigations. Relevant methods for the multilevel data analyses are discussed in light of the fundamental relationship between traditional dummy variable regression and hierarchical linear modeling. (SLD)