| This dissertation describes the development of an instrument to assess students' attitudes toward science. The rationale for developing a new instrument is presented in the context of a review of existing instruments. As discussed in the literature review, many existing instruments are based on ill-defined theoretical constructs, and include statements that do not appear to be assessing the single construct of attitude toward science. In addition, existing instruments do not distinguish between biological and physical science. Thus, the purpose of this study was to carefully define the construct, attitude toward science, and develop an attitude instrument to reflect that construct and to distinguish between biological and physical science. For this study, biological science was defined as any of the branches of science dealing with living things, such as genetics, entomology, or anatomy. Physical science was defined as any of the branches of science dealing with non-living things, such as physics, geology, chemistry, or earth science.;Once the Likert-scale instrument was developed, it was piloted on a sample population of students in science classes at a suburban, southwestern high school. Their responses were used to do item analyses and to calculate validity and reliability. Principal-component analysis was also performed to try to identify sub-scales. The results of the data analysis were used to select items for a final scale. The final instrument consists of two parallel forms of 20 items each. For each form, 12 items refer to science in general, four to biological science, and four to physical science.;The test-retest reliability coefficients of the final instruments are 0.768 for Form A and 0.788 for Form B. The parallel-forms reliability coefficient of the instrument is 0.93. The construct validity coefficients of the final instruments, compared to Germann's Attitude Toward Science in School Assessment, is 0.82 for Form A and 0.85 for Form B. Inter-item consistency, measured by Cronbach's alpha coefficient, is 0.91 for Form A and 0.89 for Form B. |
