February 2015: To Wait in Tier I or Intervene Immediately?


Al Otaiba, S., Connor, C. M., Folsom, J. S., Wanzek, J., Greulich, L., Schatschneider, C., & Wagner, R. K. (2014). To wait in Tier I or intervene immediately: A randomized experiment examining first-grade response to intervention in reading. Exceptional Children, 81, 11–27.

Summary by Dr. Elizabeth Swanson

Overview

The 2004 reauthorization of the Individuals With Disabilities Education Act allows states to use response to intervention (RTI) to provide early intervention to students who struggle in school and/or have a learning disability. Many RTI models comprise three tiers. In Tier 1, students receive high-quality general education instruction. In Tier 2, students receive more targeted instruction in small groups. Tier 3 instruction is most intensive and in some school districts is considered special education. Students move through these tiers based on their performance in comparison to classroom peers. For example, if students experience difficulty in Tier 1 instruction, they may attend additional Tier 2 instruction for a set amount of time. If they have “caught up” with their Tier 1 peers, they return to the Tier 1 setting alone. If not, they may receive additional instruction within Tier 2 or the more intensive Tier 3.

There is great variability across the nation in how RTI is operationalized. For instance, one situation that leaders in the field have warned against is keeping students in Tier 2 for too long when their needs are intensive enough to warrant Tier 3 instruction. These educators suggest that students with significant struggles should be placed directly into Tier 3 to best meet their needs in the most time-efficient manner.

Study Design

Studies that compare RTI models that allow for “fast tracking” versus those that follow a lock-step Tier 1, 2, 3 progression can be used to decide whether fast tracking students who struggle the most is beneficial. Al Otaiba et al. (2014) conducted a study that compared “dynamic RTI,” where first-graders with the weakest skills immediately received Tier 2 or Tier 3 instruction based on their reading profile, to “typical RTI,” where all students began in Tier 1 and progressed through subsequent tiers. The goals of the study were to learn (a) the effects of dynamic RTI and typical RTI on student reading outcomes at the end of first grade and (b) whether assignment to specific tiers predicts gains on standardized assessments and whether this prediction differs when comparing dynamic RTI and typical RTI.

The researchers used a randomized study design. Through randomization, student characteristics such as ethnicity, ability, and achievement are equally distributed across all study groups. In other words, neither group has an advantage at the outset of the study, making randomized control trials the most rigorous and sound test of scientific hypotheses. A total of 34 first-grade teachers and their 562 students participated in the study. These 34 classes were randomly assigned to either dynamic RTI or typical RTI conditions.

TierInstructorTimeGroup SizeCurriculum
Tier 1Classroom teacher90 minutes, dailyWhole classOpen Court
Tier 2Trained researcher30 minutes, twice per weekFour to seven studentsImagine It! and FCRR lessons
Tier 3Trained researcher45 minutes, four times per weekOne to three studentsEarly Interventions in Reading and FCRR lessons

Note. FCRR = Florida Center for Reading Research.

The dynamic and typical RTI conditions were identical in that teachers used the same core reading program in Tier 1, which was implemented effectively. Interventionists provided the same Tier 2 and Tier 3 interventions, which were implemented with fidelity. The only difference in conditions was when the Tier 2 and Tier 3 services began. In the typical RTI condition, all students started in Tier 1. At the end of 8 weeks, students were assessed and, if needed, were moved into Tier 2. Later, if needed, these students were moved into Tier 3. In the dynamic RTI condition, students were automatically assigned to Tier 2 or Tier 3 at the beginning of the school year based on need.

Key Findings

What are the effects of dynamic RTI and typical RTI on student reading outcomes?

  • Students in the dynamic RTI group had statistically significantly higher spring reading scores than students in typical RTI group.
  • The effect size was 0.31, which is considered a moderate effect size. (For more on effect sizes, see our November 2012 column.)

Does assignment to specific tiers predict gains on standardized assessments, and does this prediction differ when comparing dynamic RTI and typical RTI groups?

  • Tier 2 students in the dynamic RTI condition had significantly higher spring reading scores when compared to students initially eligible for Tier 2 in the typical RTI condition who received Tier 2 instruction after time spent in Tier 1 instruction alone.
  • Tier 3 students in the dynamic RTI condition (who received it immediately) achieved higher spring reading scores when compared to Tier 3 students in typical RTI who had to wait until they had moved through both Tier 1 and Tier 2 instruction.

Implications

Several practical conclusions from this study may be used to guide RTI design in schools. First, multitier models have potential to improve reading achievement. Second, it was possible to identify students who needed the most intensive intervention by using brief 3- to 5-minute screeners (i.e., AIMSWeb Letter Sound Fluency, Word Identification Fluency, and Test of Word Reading Efficiency) coupled with a teacher rating of reading severity. Third, because student groups were equal at pretest due to random assignment and received the same Tier 1 instruction and the same interventions at the Tier 2 and Tier 3 levels, confidence builds for the conclusion that fast tracking students through the dynamic RTI condition led to significantly higher reading outcomes. Not all students may need to go through Tier 1 or Tier 2. Instead, if need dictates, they could be assigned to the appropriately intensive tier in the beginning of the year. Finally, consider that this more dynamic approach was used with beginning readers. Findings may not generalize to older grades. In fact, interventions conducted in the earlier grades (i.e., K–2) typically are associated with higher effects than reading interventions in the older grades.