一般智能優異幼兒鑑定模式之研究－以臺北市一所公立幼稚園為例

Ａ Study on Using Different Models for Identifying Preschool Gifted Children--The Experience of A Public Kindergarten in Taipei, Taiwan

范成芳(Chen-Fang Fan)

指導教授 ： 郭靜姿

國立臺灣師範大學/教育學院/特殊教育學系/碩士(2000年)

摘要

本研究旨在編製一份幼兒資賦優異特質檢核表，以供學前資優鑑定初選之用，並比較在初選中運用不同鑑定模式對於鑑定一般智能優異幼兒結果之差異情形。本研究並蒐集幼稚園教師對於一般智能優異幼兒鑑定模式之看法，以提供未來教育決策單位規劃辦理一般智能優異幼兒鑑定之參考。
本研究以臺北市立東門國民小學附設幼稚園110名四至六歲的幼兒為研究對象，以幼兒資賦優異特質檢核表、托尼非語文智力測驗幼兒版（甲氏）、簡易個別智力量表及魏氏幼兒智力量表修訂版（中文版）為研究工具，對於幼兒進行各項施測工作，請其教師及家長填寫檢核表，之後，比較運用各種不同評量方式及工具對於鑑定一般智能優異幼兒結果的差異情形。
研究結果發現如下：
一、幼兒資賦優異檢核表的內部一致性信度介於.85~.96，效度方面，各 題通過專家審核比率介於56﹪~100﹪，信度、效度均相當合乎理想。此外，幼兒在檢核表之得分，性別和年齡變項的交互作用未達.05顯著水準，性別差異亦未達.05顯著水準，但是，幼兒在教師版檢核表得分之年齡差異達到顯著水準（p<.01），大班組幼兒的得分高於小班組幼兒。
二、幼兒在三種智力測驗得分的相關均達顯著水準（p<.01），且在作業 量表之間的相關較高。
三、教師和家長在幼兒資賦優異特質檢核表上的觀察評分與幼兒的簡易個別智力量表智商之相關達到顯著水準（p<.01），然而教師觀察和家長觀察與幼兒的語文智商之相關較高。教師觀察與家長觀察之間的相關也達顯著水準（p<.05），且教師與家長在幼兒實際能力表現部分觀察的相關較高。
四、教師觀察在資優幼兒與普通幼兒兩組之間的評分差異達顯著水準 （p<.01），對資優幼兒的評分高於普通幼兒（p<.01）；家長觀察在資優幼兒與普通幼兒兩組之間的評分差異則未達.05顯著水準。由此可知，教師觀察能夠鑑別出資優幼兒，家長觀察則不能。
五、不同鑑定模式之鑑定結果在入選幼兒人數、名單、及各項評量工具表現上均有差異。在初選階段，以結合團體智力測驗和教師觀察推薦的篩選模式，最能減少遺珠之憾；而在複選階段，採用魏氏量表為評量工具時，入選的幼兒人數較採用簡易個別智力量表時入選的人數為多。
六、參與座談的幼稚園教師建議一般智能優異幼兒鑑定應採多元的評量方式，在初步篩選階段採用團體智力測驗和教師觀察推廌兩種訊息，且標準放寬為平均數正一個標準差以上；在複選階段，採用分量表較多的個別智力測驗，以從更多元的向度發掘更多資優幼兒。
最後，研究者綜合本研究之結論與研究過程中的省思，提出在行政及研究上的建議，以作為未來一般智能優異幼兒鑑定及研究之參考。

關鍵字

一般智能優異幼兒 ； 鑑定模式

並列摘要

The purposes of this study were to develop ”The Gifted Traits Checklist for Preschool Children” to screen the intellectually gifted preschoolers, and to compare the results of using different models for identifying preschool gifted children. The researcher also collected teachers' opinions on the screening standard and assessment tools.
The subjects were 110 four to six years old kindergartners. Three tests were administered for identifying the gifted children. They were TONI，Easy-and-Quick Intelligence Scale for Children（EQISC）, and WPPSI-R. The Gifted Traits Checklist for Preschool Children（GTCPC）was filled by parents and teachers separately to screen the gifted children. Three different models of identification were compared.
The results were as follows：
I. The inter-item consistency （Cronbach α）of the GTCPC is .85 ~.96. No significant difference of the scores getting on the GTCPC between boys and girls was found. But there was significant difference of the scores getting on the GTCPC（Teacher Verse）between group age 4 and group age 5（p<.01）.
II. Significant correlations among the scores getting on TONI，EQISC, and WPPSI-R（p<.01）were found. The correlations among verbal scales were more significant than those among performance scales.
III. Significant correlations between teachers and parents’ rating on the GTCPC and IQ scores（p<.05）were found.
IV. Teachers’ rating for preschool gifted children on the GTCPC were higher than those of their non-gifted counterpart（p<.01）. But no significant difference between the two groups on parents’ rating was found. Thus, teachers’ nomination on the gifted was more accurate than parents’.
V. Different identification models for preschool gifted children produced different results. At the screening stage, combing group intelligence test result and teachers’ observation could increase the opportunities entering to the candidate pool. At the identifying stage, using WPPSI-R as the assessment instrument can find more gifted preschoolers than EQISC does.
VI. Teachers suggested imply multi-assessments to identify gifted preschoolers. To find more gifted children, combing group intelligence test result and teachers’ nomination and lowering the standards to +1SD at the screening stage, and using individualized intelligence test with more sub-scales at the identifying stage was believed to be the most reasonable approach among three different identifying models.

並列關鍵字

preschool gifted children ； identification model