國小身心障礙及社經文化地位不利資優學生鑑定評量項目調整之研究：WISC-V調整性指數之運用

Using WISC-V indices for identifying giftedness in elementary school: Twice-Exceptional students and those grappling with socioeconomic disadvantage or cultural differences

身心障礙及社經文化殊異資優生常因內在認知能力顯著差異，致使無法通過傳統以全量表智商為單一標準的資優鑑定。為發掘高層次推理潛能資優生，本研究參採美國資優教育學會（NAGC）倡導運用多元WISC-V 調整性指數於資優鑑定之建議，以WISC-V 中文版臺灣標準化樣本（N=1,034）建立「擴展版語文理解指數」、「擴展版流體推理指數」及「擴展版一般能力指數」等三項新調整性指數；比較一般資優、雙重特殊需求及社經文化殊異等三組國小高智商學生（N=162）之WISC-V 主架構認知表現；並分析在三類鑑定標準下，運用調整性指數前後之鑑出率差異。研究發現：一、新增三項調整性指數之臺灣常模分數分布符合常態分配，信度在.94-.97 間，具良好效度。二、三組高智商學生認知特性同中存異，相同處為：高層次推理均優於基礎認知效能表現、處理速度均屬相對弱勢、且內在差異均高於臺灣常模表現，顯示高智商學生認知能力內在差異值得重視。三、鑑出率部分，若採標準一「現行標準FSIQ≥128分」，三組鑑出率為39-54%：若採標準二「NAGC 建議六項調整性指數中，任一分數≥128 分」，三組鑑出率為61-77%，提升幅度均在22-23% 間；若採融合前二標準精神之標準三「兩步驟鑑出」，三組鑑出率為50-71%。鑑出率由高至低依序為標準二、標準三、標準一。 四、「一般能力指數」與「擴展版一般能力指數」在高智商學生群展現相對較佳區辨意義。研究發現若在資優鑑定運用多元WISC-V 調整性指數分數，將更能顧及資優生內在差異並發掘其獨特認知潛能。資優鑑定標準如何選用涉及行政與實務多元因素考量，本研究提供重要的常模表與資料供各界討論以期凝聚共識，並對後續研究與應用提出建議。

Rationale and Purpose: Gifted students with twice exceptional and socioeconomic disadvantage or cultural differences often show large cognitive discrepancies. The presence of learning or cognitive discrepancies in children can limit the effectiveness of the Full Scale IQ score in the identification of reasoning potential. Recognizing this, the National Association for Gifted Children (2018) has advocated for a broader approach to assessment, including the consideration of six WISC-V index scores. These index scores would encompass not only the traditional FSIQ but also the Nonverbal Index (NVI), General Ability Index (GAI), Verbal (Expanded Crystallized) Index (VECI), Expanded Fluid Index (EFI), and Expanded General Ability Index (EGAI). Although the norm tables for FSIQ, NVI, and GAI indices have been developed and reported in the published Taiwan WISC-V manual, the norm tables for VECI, EFI, and EGAI indices are absent, hindering the collection and comparison of real-world data. Thus, this study 1) addressed the absence of norm tables for three WISC-V indices (VECI, EFI, and EGAI), as recommended by the National Association for Gifted Children (NAGC), and provided a reliability and validity report, and (2), analyzed and compared the cognitive characteristics of three distinct groups of students with high IQ to identify strengths and weaknesses across the groups. Three identification criterion were employed to examine variations in the occurrence of giftedness. Methods: This study used WISC-V scores from two samples: (1) The Taiwan WISC-V standardization sample aged 6 to 16 (N=1,034), and (2) 162 elementary school students with high IQ aged 6 to 8. The latter group included three subgroups: students with FSIQ above the 85th percentile (N= 83), high IQ students with disabilities (N=51), and high IQ students with socioeconomic disadvantages or cultural differences (N= 28). In Study 1, data from the Taiwan WISC-V standardization sample were used to develop norm tables for VECI, EFI, and EGAI, and assess the internal consistency reliability and construct validity using Pearson correlation coefficients. In Study 2, a comprehensive set of statistical analyses, namely profile analysis, ANOVA, dependent t-tests, and χ2 tests, were employed to compare the cognitive characteristics of the three groups with high IQ on the basis of the WISC-V five-factor model main structure. This study evaluated the effectiveness of three giftedness identification criteria: 1) FSIQ ≥ 128, the traditionally used criterion; 2) a score of ≥128 on any one of the six WISC-V index, which is the standard adopted by the NAGC; and 3) a two-phase identification approach that incorporated elements of criteria 1) and 2), which is described as follows. In the first phase, the standard of FSIQ ≥128 was applied. In the second phase, students with FSIQ < 128 (i.e., those who did not pass the first phase) but with a significant cognitive ability discrepancy were considered to be gifted if they had a score of ≥128 on any one of the six WISC-V indices. For each of the three groups, the number of students who were defined as being gifted (termed “gifted incidences”) were compared across the three criteria, and the discriminant validity of each of the six indices were examined. Results: The key findings were as follows. 1) The newly established Taiwan norms for VECI, EFI, and EGAI were found to be appropriate; each index had a normal score distribution and robust reliability and validity. 2) The cognitive profiles of these three high IQ groups were different in some respects and similar in others. All groups demonstrated strengths in higher-level reasoning, contrasted by lower cognitive efficiency performance, particularly in processing speed. The extent of cognitive discrepancies among the WISC-V main structure was significantly larger in these groups than in the Taiwan standardization sample. 3) The first criterion of FSIQ ≥128 had pass rates of 39%-54% in the three groups. The second criterion, of a score of ≥128 on any one of the six WISC-V indexes, had pass rates of 61%-77%, which was a 22%-23% higher relative to those for the first criterion, and the increasing rate was about the same across three groups. Lastly, the third criterion, which was based on two-phase identification, had pass rates of 50%-71%. In summary, the second criterion, recommended by NAGC, had the highest pass rate, whereas the first criterion, the traditional criterion, had the lowest pass rate. Among the six indices analyzed, GAI and EGAI had the highest level of discriminant validity. Conclusions and Implications: The present study indicated that the NAGC’s approach expands the field of gifted assessment, enabling the identification of a greater number of gifted students with substantial cognitive discrepancies when applying the new standards (criterion two or three). The findings also highlighted that all three groups of students with high IQ exhibited cognitive discrepancies significantly larger than the those of the base rate reported by the Taiwan norm sample. We advocate for the recognition of unique cognitive strengths and weaknesses in each gifted student, not only 2e students, in gifted education. Our study, the first of its kind to use empirical Taiwanese data on this topic, is valuable but not without limitations. We drew our sample of 162 students with high IQ solely from Taipei city, and the size for each of the three groups with high IQ was relatively small. Notably, our sample contained a relatively higher male to female ratio for majority of the 2e students recruited were with autism, ADHD, or EBD. Future research should include larger, more representative samples from various areas in Taiwan. Through collection of such samples across different gifted groups, more sophisticated statistical analyses, including tests of measurement invariance or latent profile analysis, could be conducted. Consequently, finer-grained differences in intellectual structure and performance across different gifted student populations could be better identified. Although choosing which criterion to use in practice involves multiple considerations, our work provides essential norm tables and data for administrators, academic researchers, and practitioners to help students with special needs reach their full potential.