RTI運用在國小學生閱讀障礙鑑定之研究

Abstract

本研究目的為：(1)以閱讀障礙類型之診斷為模擬效標，鑑別「無教學反應」(non-responsiveness,NR)的RTI分組指標為何；(2)比較不同教學反應組別在智力、語文能力、需求特徵之差異。本研究對象為曾接受語文有效補救教學一年的27位新北市某校五年級學生。本研究RTI分組候選指標有三種─成長(growth)、後測水準(level)、及雙重差距(dual discrepancy)。本研究以閱讀障礙類型之診斷作為RTI候選指標之模擬效標，而模擬效標之適切性分別由閱讀障礙學生出現率、閱讀障礙各類型之比例、及診斷是閱讀障礙學生之外部效標(縣市學障鑑定)作為依據。 本研究結果為：(一)最佳RTI分組指標為後測水準常模參照識字(正確性或流暢性)-1SD切截，其優點為敏感性(81.3％)與精確度(77.8％)高、犯偽陽率與偽陰率低，尤其對識字有困難的讀寫型與語言型閱障生之偵測可達100％，且犯偽陰性之缺點可藉由後續鑑定程序之其他工具協助修正；(二)本研究閱讀障礙診斷標準為-1SD，診斷結果符合文獻調查之閱讀障礙出現率及各類型比例，且診斷是閱讀障礙學生之外部效標良好；(三)最佳RTI分組指標所區分的R與NR兩組在智力方面無差異；兩組在語文方面除了受分組指標操作影響之項目(識字與朗讀文章的正確性與流暢性)呈自然差異外，閱讀理解、聽覺理解、聽寫等並無顯著差異；兩組在需求特徵之各題項雖無顯著差異，但NR組在「閱讀不流暢、理解抽象符號或詞彙差」較R組弱。此外，R組學生中是否具「不會寫出完整通順句子」特徵，可提供避免遺漏偽陰性學生鑑定之參考。 基於運用RTI在閱讀障礙鑑定之發現，提供學校實務及未來研究之建議。

The purpose of this research is to:identify what the response to intervention (RTI) criteria of non-responsiveness is by utilizing the simulated validity diagnosed within classification of reading disabilities; and comparie the responsived(R) with the non-responsived(NR) groups in intelligence, basic reading skills, and special needs. There are 27 fifth-grade students participanted in this study, who attended a one-year evidenced-based literacy remedial program, in an elementary school in New Taipei City. The classifications of candidate indicators of RTI in this study are discrepant score of pretest and posttest, post-treatment level of performance, and dual discrepancy. The stimulated validity of candidate indicators of RTI in this study came from the classification of reading disabilities, and the relevance of simulated validity depending upon the prevalence of reading disability students, the ratio of types of reading disabilities, and the external validity of diagnoses of reading disability students identified by the city government.

There are three majory findings. First, the optimal RTI classification indicator is the accuracy or fluency of word recognition in norm referenced approach with -1SD cutoff. The advantages of this indicator are high sensitivity (81.3%) and accuracy (77.8%) with low false positive and false negative ratio; more important, this indicator is able to completly detect dyslexia and language learning disabilities, and the flaw of false negative can be modified via follow-up evaluation of identification. Second, the diagnostic criteria of reading disabilities in this research is -1SD cutoff, the final results are accordance with the prevalence and the ratio of types of reading disabilities within literature reviews. Furthermore, the external validity of diagnoses of reading disability students is good. Third, there are no significant differences between R and NR groups in intelligence test. Except the accuracy and fluency of word recognition and text, there are no significant differences in reading comprehension, listening comprehension and dictation in the two groups. Furthermore, NR group represent lower performances than R group in the characteristics of “reading influent” and “hard to comprehend abstract symbols or phrases”, whereas there is no significant difference within the two groups in these characteristics of special needs. Additionally, the R group with the characteristic of “difficult to write down a complete and clear sentence” provide a clue to avoid the neglect of the false negative students to be unidentified.

In light of the findings of the RTI operation in reading disabilities identifications, the researcher provid some suggestions for practice in schools and future research.