我國資優教育研究之回顧與後設分析研究

The content-analysis and meta-analysis of the researches in gifted education of Taiwan, R. O. C.

彭瓊慧(Chiung-Hui Peng)

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(2001年)

摘要

本研究旨在探討自民56至民90共計35年間，我國資優教育實徵性研究之研究變項累計情形，包括作者背景、研究相關特徵、樣本特性、評量工具及方法、資料處理方式等五大面向，以瞭解歷年來研究成果之全貌。另抽離出創造思考教學研究進行後設分析，以了解資優教育創造思考教學之成效及可能的影響變項，進而提出教學上之具體建議。
本研究採用內容分析法及後設分析法。在內容分析方面，研究者以自編之登錄表甲式登錄441篇實徵性研究之內容，再以次數分配及百分比進行資料處理；在後設分析方面，以自編之登錄表乙式登錄15篇創造思考教學實驗研究，運用Hedge與Olkin（1985）的技術計算加權平均效果量d+值，並採用「類別模式」尋找可能影響創造思考教學的調節變項。綜合歸納結果如下：
一、 內容分析方面
（一）作者背景
1. 資優教育領域有一群核心研究者經常性發表研究
2. 男女性研究者比例相當，近年來有女性多於男性之趨勢
3. 我國資優教育研究偏向單一著作，除博碩士論文外，單一與聯合著作比例約為2：1
4. 第一線教學人員從事研究的比率低落，僅佔一成左右
（二）研究相關特徵
1. 期刊為實徵性研究的主要來源，其次為博碩士論文，且數量均隨年代而增加
2. 特殊教育學術性期刊刊登資優教育研究的比率偏低
3. 研究主題隨年代趨於多元化，「情意特質與社會適應」、「認知思考特質」、「生涯及追蹤」及「資優教育政策、制度及現況」為最常見之主題
4. 不同資料來源所著重的主題不盡相同，
5. 調查法與相關研究為最常用之研究方法，發展研究最為缺乏
（三）研究樣本
1. 取樣區域集中在都會區，且由北向南遞減
2. 研究對象偏重一般智能優異資優生，且以國小高年級學生為主
3. 不同主題之研究對象分布不均，「身心特質」研究對象多為國小資優生；「生涯及追蹤」、「資優教育政策、制度及現況」則集中在高中階段。
4. 缺乏對學前資優、身心障礙資優、文化殊異資優及低成就資優生等特殊族群的研究
5. 立意取樣為主要的取樣方式
（四）研究方法及工具
1. 研究者傾向以紙筆評量方式蒐集資料，近年來蒐集方式逐漸多元化
2. 問卷回收率大多良好，回收率多達七成以上
3. 現成的評量工具不足，且集中在特定主題
（五）資料處理方式
1. 資料處理方式以量化分析居多，多偏向初等或中等的統計
2. 不同資料來源均常用「描述性統計」、「t/z考驗」、「變異數分析」統計方法，然運用比重不同
3. 研究主題有集中特定統計方法之情形
二、 後設分析方面
1. 創造思考教學對提升資優生創造力具中小程度的效果
2. 創造思考教學對提升不同面向之創造力有不同的影響，依成效大小為變通力、獨創力、流暢力及精進力。
3. 創造思考教學可提升資優生高層次認知思考及學業成就的表現
4. 在調節變項之尋找上，教材來源、測驗工具種類兩因素為影響教學實驗效果之調節變項。
本研究最後依據研究結果對教育當局即學校、學術研究單位、教學研究提出若干點具體建議。
關鍵字：資優教育研究、內容分析、後設分析

關鍵字

資優教育研究 ； 內容分析 ； 後設分析

並列摘要

The main purposes of this study are: (1) to investigate the outcomes of experimental variables of researches in gifted education from 1967 to 2001. The features of this analysis are included researchers’ background, related features, the traits of subjects, assessment materials, and the analysis methods of data; (2) to investigate the overall weighted mean- effect sizes of creativity, cognitive thinking abilities, the performance of achievement in creative thinking instruction with meta-analysis method, and to find out moderator variables that may influence the effects of creative thinking instruction.
After systematic searching for the literature from bibliography and electronic database, 441 researches were coded for content analysis, and the related data were calculated with frequency and percentage. Besides, 15 researches on creative thinking instruction were coded for meta-analysis, and the effect- sizes were calculated by Hedges and Olkin’s(1985) procedure. Furthermore, this study was tested all coded features by ‘Categorical Model’ to find out moderator variables. The followings are major results of this study:
1. In content analysis
A. Researchers’ background
a. One group of core researchers conducts studies frequently on gifted education.
b. The number of female researchers has been increasing during those years, and almost catches up with that of male researchers.
c. Most researches on the current gifted education are almost finished by only one author, except for the dissertations/ theses. The number of working-alone researcher nearly doubles than that of group-working ones.
d. Only few teachers engaged in research.
B. Related features
a. The quantity of journals that dominates the sources of empirical researches conducted by dissertations/theses has been increasing those years.
b. Few researches on gifted education are issued in the journals of special education.
c. Topics keep changing every year. Some popular topics are: “Emotion Characteristics and Social Adaptation”, “Cognitive Thinking Characteristics”, “Carrier and Tracking”, and “Policies and System of Gifted Education”.
d. Different sources focus on different topics.
e. Survey and correlational researches are the methods used by the researchers most frequently.
C. The traits of subjects
a. Sampling areas focus on cities, and decrease from north to south.
b. The majority of samples are general ability gifted, especially for high- grade students in the elementary schools.
c. The different topics are not in a normal distribution.
d. Lack of studies for pre-school gifted, handicapped gifted, culturally disadvantage gifted, and underachieving gifted students.
e. Purposeful- sampling is the major method of sampling.
D. Assessment materials
a. Many researchers are willing to collect data with paper-writing instead of related instructional methods. But, it’s good to see that the following researchers use one more methods afterwards.
b. The 70% return rate of questionnaire is pretty high satisfactory.
c. Standardized assessment instruments are not enough, and they only focus on some specific topics.
E. The analysis methods for data
a. Most data processing methods rely on quantitative analysis. And statistics methods are usually adopted by basic and intermediate skills.
b. Statistics methods adopted more frequently are: “descriptive statistics”, “t/z test”, “analysis of variance“.
c. Most research topics used “descriptive statistics” statistic methods frequently.
2. In meta-analysis
A. The overall weighted mean- effect size of creative thinking on instructional studies was between middle and small magnitude.
B. The Instruction for creative thinking promotes different creativities. The descending orders of the weighted mean- effect size of creativities were flexibility, originality, fluency, and elaboration.
C. Instruction for creative thinking promotes the performance of high-level cognitive thinking and achievements of gifted students.
D. The moderator variables of creative thinking studies were sources of teaching materials, and many kinds of assessment instrument.
Base on the results of study, the researchers also provided a discussion and made some recommendations to upgrade the reliability and efficiency of gifted education.
Keywords: empirical researches, content analysis, meta-analysis

並列關鍵字

empirical researches ； content analysis ； meta-analysis

參考文獻

林邦傑（民76）：整合分析的理論及其在國內的應用。教育與心理研究學報，10，1-38。

連啟舜（民91）：國內閱讀理解教學研究成效之統合分析。國立台灣師範大學心理教育與輔導研究所碩士論文，未出版。

陳永慶（民90）：國內心理治療與諮商輔導效果的整合分析研究。高雄醫學大學行為科學研究所碩士論文，未出版。

賀德潤（民86）：領導行為與工作滿足的整合分析:以國內博碩士論文為例。高雄醫學院行為科學研究所碩士論文，未出版。

黃寶園（民87）：柯氏性格量表效度概化之統合分析。國立台灣師範大學教育心理與輔導研究所碩士論文，未出版。