Date of Degree


Document Type


Degree Name



Educational Psychology


Barry J. Zimmerman

Committee Members

Linnea Ehri

David Rindskopf

Subject Categories

Educational Psychology


Struggling adolescent readers are distinct from others in two important ways: (1) They are adolescents; and (2) they have a history of struggle with reading.

Good pedagogy prescribes that effective programs "meet students where they are." For middle-school students, this means meeting them in adolescence. Adolescents are more concerned with social norms and more susceptible to peer influence than younger children. Additionally, the fact that these youth are still struggling after years of reading instruction suggests that their motivation to persist at reading is likely to have suffered. To fully support and engage such adolescents, reading programs must leverage social processes and include explicit support for motivation and strategy use.

This dissertation investigated the effects of a peer modeling instructional intervention on early adolescents' question asking, reading motivation and comprehension. Videotaped peer models demonstrated the use of question asking for comprehension and motivated participants to use the strategy.

Participants were 48 sixth graders who attended public schools in New York City. Eighty-five percent were classified as reading below grade level. After completing an interactive tutorial on question asking, all students read a moderately challenging, computer-based science text. While doing so, participants in the Peer Modeling condition observed same-age, similar-ability peer models asking authentic questions about the text. After reading, all participants generated their own questions, completed a short survey, and were assessed for reading comprehension.

The research found that peer modeling had a positive effect on the quality of questions that students asked and their text comprehension. Participants exposed to peer modeling asked more questions that were not answered in the text and more deep-level "I'm Confused" questions. They better understood the solution component of the text and recalled more critical idea units. Peer modeling did not affect participants' motivation or accuracy of comprehension judgments.

In contrast to previous research, the study also found that students asked numerous deep-level questions, but that these questions were not necessarily linked to greater understanding. Additionally, higher motivation was not associated with greater comprehension. These findings have implications for the design of systems to support struggling readers and for theory-building about reading comprehension.


Digital reproduction from the UMI microform.