An analysis of four ways of assessing student beliefs about sts topics

The study investigated the degree of ambiguity harbored by four different response modes used to monitor student beliefs about science-technology-society topics: Likert-type, written paragraph, semistrue tured interview, and empirically developed multiple choice. The study also explored the sources of those beliefs. Grade-12 students in a Canadian urban setting responded, in each of the four modes, to statements from Views on Science-Technology-Society. It was discovered that TV had far more influence on what students believed about science and its social, technological context than did numerous science courses. The challenge to science educators is to use the media effectively in combating naive views about science. Regarding ambiguity in student assessment, the Likert-type responses were the most inaccurate, offering only a guess at student beliefs. Such guesswork calls into question the use of Likert-type standardized tests that claim to assess student views about science. Student paragraph responses contained significant ambiguities in about 50% of the cases. The empirically developed multiple choices, however, reduced the ambiguity to the 20% level. Predictably, the semistructured interview was the least ambiguous of all four response modes, but it required the most time to administer. These findings encourage researchers to develop instruments grounded in the empirical data of student viewpoints, rather than relying solely on instruments structured by the philosophical stances of science educators.