The purpose of this study is to examine factors that influence students' reading of derived words. Recent research suggests that the lexical quality of a derived word depends on the familiarity of the word, its morphemic constituents (i.e., base word and affixes), and the frequency with which the base word appears in other words (i.e., members of the same word family or family frequency). On the premise that better and more experienced readers have higher quality lexical representations, we explore the extent to which accuracy of reading derived words by 4th and 6th graders is related to measures of familiarity, including derived and base word frequencies, family size, average family frequency, and word length. The results of an exploratory factor analysis indicated that these measures formed two factors, one representing morphemic constitution and the second representing exposure to the word family; both factors accounted for significant variance in the students' derived word reading. Comparisons of sets of derived words contrasted on familiarity properties showed that performance on derived words, overall, is better for 6th than 4th graders and for good than poor readers. On the measures of family frequency and family size, there were significant discrepancies between grade level and reading ability and frequency characteristics. These add support to the view that morphemic analysis and wide reading experience contribute to derived word reading.