Academic journal article School Psychology Review

Generating Reading Interventions through Experimental Analysis of Academic Skills: Demonstration and Empirical Evaluation

Academic journal article School Psychology Review

Generating Reading Interventions through Experimental Analysis of Academic Skills: Demonstration and Empirical Evaluation

Article excerpt

Abstract. This article describes the application of experimental analysis methods for identifying reading fluency interventions for two elementary school students (fourth and fifth grade) referred for reading problems. For each student the experimental analyses examined use of rewards, instruction, and a treatment package containing both reward and instructional components across difficulty levels to determine the condition that led to the most efficient increases in reading fluency for each student. Based on the experimental analyses, individualized reading packages were developed that required minimal supervision. These interventions were carried out over time during normal instructional routines, and progress monitoring data were gathered to determine the effects of the interventions. Results are discussed in terms of how practitioners can use experimental analyses of academic skills within a broader response-to-intervention framework. Also, recommendations for future research in this area are made.

**********

At least 20% of the population in the United States has reading difficulties (Lyon & Moats, 1997). The reading proficiency of students with poor reading skills has worsened rather than improved since 1992, according to data gathered by the National Center for Education Statistics (2004). This crisis did not occur because educational researchers have ignored the problem. Over 100,000 research studies on reading have been conducted since 1966 (National Reading Panel, 2000). However, there are significant challenges to producing research that is readily translated into practice. To name just a few, complex methodological hurdles limit how well intervention studies can represent classroom conditions; heterogeneous samples that increase unaccounted-for variance defy precise definition of student characteristics in these studies; poorly defined interventions make it difficult to know which intervention components are responsible for change; and characteristically weak generalization of treatment effects fail to inspire enthusiasm for the interventions (Lyon & Moats, 1997).

For the practitioner, every application from a research study to a student's instruction involves generalization of some type. Regardless of the type of research methodology used in the study, the critical question is how confident the practitioner is that the results are applicable to his or her client(s), with successful student performance being the ultimate criterion of whether the generalization was truly appropriate or not. There is never a guarantee that the intervention will work and ongoing formative evaluation (Fuchs & Fuchs, 1986) is the only sure method of determining efficacy. The challenges with translating research to practice noted earlier add further ambiguity and uncertainty to the practitioner's task. Unfortunately, these challenges for reading research outline the very questions that are of the utmost importance to practitioners. Will the intervention be appropriate in the student's classroom? Is the intervention appropriate for this particular student (given his or her current skill level and learning history)? What procedures should be administered as a part of an intervention? Will the effects be positive and lasting?

There is an emerging area of study in which researchers have sought to develop methods for directly testing interventions with brief instructional trials. The experimental analysis of academic performance has an explicit intervention focus, borrowing single-case experimental design elements to determine the most effective intervention for a particular student. The logic of this approach is simple. Before recommending an academic intervention, the evaluator tries it out first (Daly, Witt, Martens, & Dool, 1997; Witt, Daly, & Noell, 2000). When adequate controls exist, the evaluator can determine the size of the effect as a part of the experimental analysis. …

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.