【正文】 erely emphasizing the teaching of the history of science in science classrooms to sequencing in instruction in science lessons and promotion of better teacher preparation programs in the universities. Introduction Science curricula vary widely among countries, states, school districts, and individual schools. The most vivid differences are concerned with the particular science topics or concepts to be included. Such differences in course and curricular content are unavoidable, as each course must present 第 2 頁 only a small sample of the scientific generalizations and principles drawn from a consistently and rapidly expanding discipline (Lederman, 1992). There is no consensus among science educators concerning the specific content to be included in contemporary science courses or even the methods and strategies of instruction to be used. However, there appears to be strong agreement on at least one of the objectives of science instruction. The development of an “adequate understanding of the nature of science” or an understanding of “science as a way of knowing” continues to be convincingly advocated as a desired oute of science instruction (American Association for the Advancement of Science (AAAS), 1989, Lederman, 1992). Although the “nature of science” has been defined in numerous ways, it most monly refers to the values and assumptions inherent to the development of scientific knowledge (Lederman amp。 understandings of nature of science to the reports of the Central Association of Science and Mathematics Teachers (1907) in which a strong argument was presented for increased emphasis on the scientific method and the processes of science. Concerns for the development of adequate understandings on the nature of science “have 第 4 頁 worn many hats” through the years (Lederman, 1992). In the early 1900s the nature of science objective was expressed in terms of increased emphasis on the scientific method “so as to better train students39。 National Science Teachers Association, 1982). Clearly, science educators and scientists have been extremely persistent in their advocacy for improved student understanding of the nature of science. Indeed, Kimball (1968) has referred to this objective as one of the most monly stated objectives for science education and Saunders (1955) went so far as to describe it as the most important purpose of science teaching. Research related to the nature of science can be conveniently divided into four related, but distinct, lines of research: (a) assessment of student conceptions of the nature of science。 and (d) identification of the relationship among teachers39。 conceptions did not start until 1954 (Wilson, 1954). Initial assessments of students39。 understandings of NOS (Lederman, 1992). Unfortunately, the results of such attempts were ambiguous and the specific variables contributing to improved conceptions of NOS remained unknown. The above mentioned lines of research were informed by two basic implicit assumptions: a teacher39。 attentions on the testing of these assumptions and attempts to derive those classroom variables related to changes in students39。 Rogan, 1990). Additionally, science educators39。 conceptions. Although the results of recent qualitative investigations have not outwardly contradicted the results of prior quantitative approaches, which were using cognitive science learning theory perspective, they have provided more in depth and valid assessments of teachers39。 conceptions of the nature of science. Therefore, it is assumed that this learning theory might have greatly influenced the early research on the nature of science. To highlight studies on nature of science during the time when cognitive science learning theory was popular, this paper discusses several studies. In 1961 Klopfer and Cooley developed the Test on Understanding Science (TOUS) which was to bee the most widely used paperandpencil assessment test of students39。 (c) the roles of theories and their relation to research。 (g) what constitutes a scientific explanation。 conceptions of the nature of science. The larger sample size used in this study gave it much credibility and it was followed by widespread curriculum development regarding the nature of science in science textbooks. As illustrated by the above mentioned studies, researchers who conducted studies on the nature of science during the times when cognitive science learning theory was popular, focused on developing instruments which measure students39。Malley (1990) has called into question prior approaches to assessing students39。 preand posttest responses were categorized as 第 10 頁 exhibiting absolutist or tentative views, a stratified sample was selected for a followup videotaped interview. During the interview, students were asked to clarify their questionnaire responses and provide information concerning the sources of their beliefs and factors causing beliefs to change. Although responses on the questionnaires indicated that students possessed absolutist view of science, the interviews indicated that the students actually were quite clear in their beliefs that scientific knowledge is tentative. Additionally, students did not view laboratory activities or any other science activities as specifically related to their present views of science. The researchers concluded that the use of interviews to assess students39。 conceptions of the nature of science did not involve interviews, the implications about the accuracy of over three decades of data were called into question by the researchers in this study. As illustrated by these studies, researchers working from a constructivist learning theory perspective emphasized the importance of using qualitative techniques, such as classroom observations and interviews, for gathering data about the students3