과학 관련 태도의 타당한 측정을 위한 연구 II - "과학에 대한 태도"의 감정적 요소 측정을 위한 척도 개발-

A Study of Valid Measurement in Science Related Attitudes (II) - To Develop an Affective Component of Attitudes toward Science Scale-

The purpose of this study was twofold: (1) to design a system for constructing Likert attitude scales as supported by the sociopsychological and measurement literature, and (2) (using the design) to develop an affective component of "Attitudes toward Science" scale for high school students. The rationale for developing a new attitude scale is presented in the context of a review of existing attitude scales. As discussed in the literature review, many existing attitude scales are based on ill-defined theoretical constructs, and includes statements that do not appear to be assessing a single construct of attitude toward science. In addition, existing attitude scales do not distinguish between affective and behavioral and cognitive components of attitude toward science. Thus, this study was to carefully define the construct, subcomponents of attitude toward science, and develop an affective component of "Attitude toward Science" scale to reflect the construct and to distinguish between affective and behavioral and cognitive components of attitude toward science. The results of this study: (1) three-mains step for designing reliable and valid attitude scale were developed, and (2) 35 items(16 positive and 19 negative) for an affective component of "Attitudes toward Science" scale with the following characteristics were developed: (a) The internal consistency was estimated using Cronbach's coefficient ${\alpha}$, 0.9727. (b) The range of adjusted item-total correletion(${\gamma}$-value) were 0.58${\sim}$0.83. (c) The correlation coefficient ranged between 0.61 and 0.74 revealing a moderate relatedness between subcomponents. (d) The correlation coefficient for concurrent validity were 0.55 with TOSRA and 0.51 with SAl.