Reading comprehension emerges as an important skill set in the early elementary grades. It is supported by component skills including decoding, linguistic knowledge including vocabulary and syntactic knowledge, as well as more complex, higher-level components such as inference making and comprehension monitoring. Theoretical understanding of reading comprehension has historically included reader-focused models, as well as models that include reader-text interaction and text-task interaction. Together, the dimensions of reader, text, and task represent the conceptual space in which comprehension can occur. Using a crossed random-effects model, the probability of a correct item response can be modeled as a function of reader, text, and item characteristics. This approach has been used in several studies of reading comprehension, with informative results. However, to date this work has focused on older readers, or has used relatively small samples of readers. In this study, a crossed random-effects modeling approach was used to analyze a large data set consisting of item response data from a sample of 2,723 Grade 3 students. Student-level predictors of vocabulary knowledge, syntactic knowledge, and word recognition, as well as several categorizations of item type, and passage-level predictors of lexile and several indices of passage complexity were found to be significant predictors of reading comprehension. Cross-level interactions were investigated, and significant interactions were found between student and item predictors, and between student and passage predictors. Approximately 50% of variance in reading comprehension across students was explained by the student-level predictors, but only 18-22% of variance across items was explained by the passage-level and item-level predictors. Results from this study suggest that for Grade 3 readers, the strong predictive relations between student predictors of syntactic and vocabulary knowledge to reading comprehension may be moderated by some aspects of item and task demands. However, for this large-scale, multiple choice assessment of reading comprehension, variability in items and passages was largely unexplained. Results are discussed in the context of theoretical accounts of reading comprehension, from which the item and passage predictors are derived. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]