Academic journal article School Psychology Review

Underlying Changes in Repeated Reading: An Eye Movement Study

Academic journal article School Psychology Review

Underlying Changes in Repeated Reading: An Eye Movement Study

Article excerpt

Repeated reading (RR), an instructional strategy involving the rereading of short passages (LaBerge & S. J. Samuels, 1974), is the oldest and most frequently cited approach for assisting individual learners in becoming fluent readers (Kuhn, 2004; Meyer & Felton, 1999). Despite its simplicity, RR has been associated with significant improvements in fluency since its inception and is now supported by an extensive range of empirical evidence (e.g., Ardoin, Williams, Klubnik, & McCall, 2009; Therrien, Wickstrom, & Jones, 2006). Meta-analyses (Meyer & Felton, 1999; National Institute of Child Health and Human Development, 2000; Therrien, 2004) of RR based procedures suggest that RR improves the reading rate and accuracy of students with and without disabilities on both practiced and unpracticed passages. Meyer and Felton (1999) reviewed studies published between 1981 and 1999 and concluded that, overall, RR resulted in improved reading speed (in words read per minute) for both average and poor readers in elementary school. A number of the studies they reviewed also demonstrated improvements in word recognition accuracy, but findings were mixed regarding transfer of effects and changes in students' comprehension ability. RR in these studies consisted of daily 15-min sessions involving three or four readings per passage. In general, teachers, paraprofessionals, and volunteers implemented these programs, but researchers also mentioned the possibility of effective implementation by higher functioning peers (Simmons, Fuchs, Fuchs, Mathes, & Hodge, 1995).

In a review of the reading research literature, the National Reading Panel (National Institute of Child Health and Human Development, 2000) linked instructional procedures emphasizing repeated oral reading practice, including RR, with consistent positive effects on word recognition accuracy, reading rate, and reading comprehension, as measured by a variety of test instruments and at a range of grade levels (i.e., Grades 2-9). This meta-analysis yielded an overall weighted effect size (ES) average of 0.41, suggesting a moderate effect. Gains were found to be highest for reading accuracy, smaller for reading rate, and lowest (but still significant) for comprehension. Results also indicated that RR had a clear effect on the reading ability of readers without disabilities through at least Grade 4 and on the performance of students with reading problems through high school.

In the most recent meta-analysis of RR research, Therrien (2004) sought to examine the effectiveness of RR in terms of fluency and comprehension (including students with cognitive disabilities) and to identify the critical components of RR-based procedures. Therrien concluded that RR improved reading fluency and comprehension on practiced passages for students with and without learning disabilities (as indicated by a moderate mean increase in fluency, ES = 0.83, and a somewhat smaller mean increase in comprehension, ES = 0.67). In addition, analyses suggested that RR improves students' ability to read and comprehend untrained passages, despite the time-limited nature of the procedures utilized in re viewed studies (with most involving 45 or fewer intervention sessions). The critical components of RR-based procedures identified by Therrien included (a) providing students with cues as to their purpose in rereading, (b) having students read aloud to adults who could provide error correction and feedback, and (c) requiring students to reread passages three or four times. Specifically, Therrien (2004) noted that RR procedures involving three or four readings of a single passage were associated with average fluency effect size increases (ES = 0.85 and 0.95, respectively) that were significantly larger than those associated with RR involving only two readings of a text (ES = 0.57).

Limitations to Extant Research on RR

Despite strong empirical support for RR within both behavioral and cognitive scientific literatures (e. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.