Field Studies of RTI Effectiveness Albany Response to Intervention (ARTI) Model

Study Citation

Vellutino, F. R., Scanlon, D. M., Zhang, H., & Schatschneider, C. (2008). Using response to kindergarten and first grade intervention to identify children at-risk for long-term reading difficulties. Reading and Writing, 21, 437–480.

Program Description

The Albany Response to Intervention (ARTI) model is a standard-protocol model. Its purpose is to improve literacy development for young students through strategic instruction and interventions. Vellutino et al. (2008) identified several phases of the ARTI process:

1. Screening at beginning of kindergarten to determine at-risk students (at or below 30^th percentile).
2. Project-based or remedial services for at-risk group (Tier 2).
3. Assessment at beginning of 1^st grade year to determine a "continued-risk" group and a "no-longer-at-risk" group.
4. More intensive project-based intervention for continued-risk group (Tier 3).
5. Identification of normal reader controls for comparison and evaluation purposes.

Certified classroom teachers are responsible for both the kindergarten and 1^st grade intervention components. These teachers are trained and supervised by research project staff. Teacher training consists of a 5-day workshop followed by bimonthly group meetings and one-to-one supervisory meetings that are held approximately every 6 weeks. Fidelity of treatment is ensured through monitoring of tape recordings of treatment sessions by project staff.

Purpose of Study

Vellutino et al. (2008) conducted the study to discover the impact of the ARTI model on literacy outcomes and to determine if RTI procedures would more accurately distinguish between at-risk groups than psychometric measures alone. Specifically, the purpose of the study was to answer the following question:

1. What is the effectiveness of an RTI model on literacy development relative to normal reader controls?
2. Do RTI measures more effectively distinguish between continued-risk and no-longer-at-risk children than kindergarten-screening measures, measures of intelligence, or measures of reading-related cognitive abilities?

Study Method

A total of 462 students were identified as at-risk in kindergarten. To evaluate the efficacy of the kindergarten and 1st grade intervention program, effect sizes for group mean differences on the end of kindergarten and 1^st grade were calculated. For Question 2, logistic regression was used to assess the accuracy with which measures of response to kindergarten intervention would predict membership in the continued-risk and no-longer-at-risk groups identified at the beginning of 1^st grade. Four different models were compared. The predictors for the first model were measures from the psychometric screening battery administered at the beginning of kindergarten. The predictors for the other three models were RTI measures of incremental growth in emergent literacy skills, along with baseline measures of these skills to control for correlation with the composite measure used to classify the two groups. These variables were initially identified with hierarchical linear modeling (HLM) procedures.

Study Results

Question 1: Of the total sample of at-risk children who received intervention through the project and were available at the end of 3^rd grade, 84% were meeting grade-level expectations in reading by the end of 1st grade, either through kindergarten (Tier 2) intervention alone or through both kindergarten and 1^st grade (Tier 3) intervention. In contrast, only 16% of the children identified as at risk at the beginning of kindergarten demonstrated substantial difficulty with reading at the end of 2^nd and 3^rd grade.
Question 2: Results from the modeling analyses would seem to justify the conclusion that an RTI approach to classification is a more effective means of identifying children who may be at risk for early and long-term reading difficulties, and who may later qualify for reading disability status, than are more traditional psychometric screening approaches alone.