The study was conducted in kindergarten and first-grade classrooms in public schools in poor rural counties in Nebraska, New Mexico, North Carolina, and Texas.
Sixteen rural schools were assigned to matched pairs based on district, school size, school participation in Reading First, and the percentages of students who were minorities and receiving free or reduced-price lunch. One school in each pair was randomly selected for the intervention. One intervention school later left the study due to problems with Internet connectivity. All kindergarten and first-grade classrooms in the remaining 15 schools were included in the study. Teachers identified struggling readers using state assessment data, classroom observation information, and input from the TRI literacy coach. Within each classroom, five struggling readers were randomly selected for the study, for a total of 385 students (220 in the intervention group and 165 in the comparison group). The analytic sample included 247–250 students (158–160 in the intervention group and 87–90 in the comparison group, depending on study outcome) across 15 schools (seven intervention and eight comparison). Over half of the sample was male (63% in the intervention group and 54% in the comparison), and less than half were White (49% in the intervention group and 39% in the comparison group).
Teachers in TRI schools attended a 3-day summer workshop on TRI strategies. During the school year, TRI teachers received biweekly observation and feedback from TRI literacy coaches, and met biweekly with other TRI teachers and their TRI literacy coach to reinforce strategies and problem solve. TRI teachers also participated in workshops every few months to obtain support with understanding of the TRI process, models, and strategies. All school-year support was provided via webcam. The TRI program also provides a website with instructional resources and the ability to interact with coaches via email.
During the school year, teachers delivered ongoing one-on-one reading instruction for struggling readers in 15-minute sessions. Each session included the following three components: (1) re-reading selected texts for fluency; (2) “word work” using letter tiles to demonstrate the alphabetic principle, teach phoneme–grapheme relationships, support phonemic awareness development, and improve student recognition of sight words; and (3) guided oral reading (more detail on the three components is provided in the study). Teachers worked individually with a student for an average of 14 sessions over the course of the year. When students made progress, they were placed in a small group, and another struggling reader began receiving one-on-one instruction from the teacher.
Students in the comparison condition continued to receive normal reading instruction, without the use of the TRI, via their regular classroom teacher or other school staff.
Support for implementation
All of the TRI coaches had experience as teachers and/or reading coaches in early elementary school. Most were doctoral students in education. The coaches received feedback from the intervention director by providing videotapes of their own teaching of individual students. The coaches received additional feedback throughout the school year, with a particular focus on how to motivate teachers to implement the TRI well.