A content analysis for quantitative master’s theses of science education from normal colleges and universities

• Main Authors: Li, Yung-jin, 李永進
• Other Authors: Hsu, L. R.
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/92799941792002067994

碩士 === 國立臺中教育大學 === 自然科學教育學系碩士班 === 94 === The major aim of this study is to analyze master theses of quantitative research in science education in recent years in Taiwan. It makes use of content analysis to arrive at an understanding of research tool development, methods of reliability and validity testing, and methods of statistical testing. Besides attempting to gain such understanding, another aim of this paper is to explore potential problems in quantitative science education research papers so as to provide future research with useful information. The content analytic method used in this study is based on four research tools independently developed by the author himself. They are the “Conditions for Quantitative Research Tool Development Scale”, the “Reliability and Validity Testing Methods Scale”, the “Statistical Testing Methods Scale”, and the “Potential Problems Detection Scale for Quantitative Research.” The corpus for analysis includes 134 master’s theses of quantitative research in science education written by graduate students from normal universities and colleges in Taiwan from 2001 to 2003. The major findings are as follows: 1. The average number of research tools used in these theses was 2, and these tools were usually independently designed by the researchers themselves. In terms of reliability and validity, 14.2% of the authors of the theses did not explain clearly the methods by which they test the reliability and validity of their data. 2. The hierarchy of frequency of use for reliability testing methods was: the Cronbach α ＞ test-retest reliability ＞ interscorer reliability. On the other hand, the hierarch of frequency of use for validity testing methods was: content-related evidence of validity ＞ construct-related evidence of validity ＞ criterion-related evidence of validity. With regard to content-related evidence of validity, the most frequently used method was expert validity. 3. With respect to statistical testing methods, the hierarchy of frequency of use was: t-test ＞ one-way ANOVA ＞ Pearson correlation. It was also found, from our analysis of the theses, that the authors of the theses tended to overuse statistical testing in their quantitative research. 4. Problems related to “consistence”, “accuracy”, “objectivity’, “completeness”, and “openness” were also observed in the theses with varying degree of frequency. On the basis of the above findings, recommendations and suggestions are made concerning the development of research tools, methods of reliability and validity testing, methods of statistical testing, and methods of potential problems detection for quantitative study in science education.