Rethink Attributes Of Items In A Large-Scale Science Assessment

Items are important tools of testing evaluation. Items'attributes and functions can not only influence the result of a test and its analysis, affect the diagnosis of students' learning ability, but also impact on teaching effects and teaching strategies. We need to rethink the rationality of the recognition of items'attributes. This article couples cognitive psychology with the practice of a large-scale assessment. Design interview based on the method-"thinking aloud"--to collect data by selecting some typical physics items. Rethink the rationality of the recognition of scientific items'attributes by comparing the results from the date derived from a large-scale assessment and provide research evidence for an optimized process of attribute recognition. This full text includes five parts:The first part is talking about introduction and theoretical framework. At the beginning, I raise research questions and introduce this research's background and values, including research purpose, main contents, and definitions of those related concepts; and then I introduce theoretical framework that the research adopted. For conducting this research, that is, collecting data and accumulating the knowledge of measurement, ten physics items selected from a large-scale science assessment are taken by students.The second part is literature review and analysis. I introduce the theories regarding the dichotomy of educational objectives based on students'cognitive abilities that are stated by Bloom, Klopfer, and Anderson et al. Particularly, I compared those different cognitive demands between "understanding" and "application". Moreover, I analyzed students' cognitive demands in some national and international large-scale assessment projects, in order to provide references and insights for the practice of Chinese large-scale assessment. Finally I introduce the methods and process of rethinking items'attributes in this research.The third part is regarding the main design in this research design. From the psychology of information operation perspective, I hypothesize students'cognitive processing of operating on these three cognitive dimensions of "knowing", "understanding" and "application".I also infer to cognitive levels based on data analysis of difficulty from a large-scale assessment. Furthermore, I introduce the main research methods and how to select the subject in this research, and state the implementation of the study. Ten representative physics items were selected from the large-scale science assessment to interview students by using the cognitive method-- "thinking aloud". After collecting data, I organized those recording materials and do data analysis.The fourth part is the analysis results. I examined the attributes of items from the perspective of students'cognition and the results of the large-scale assessment. Discuss the rationality of recognition of items' attributes. I analyzed the representative items and rethink the original calibration of items'attributes. The research found that those items involved more solid preconception can be used to test students'cognitive level of "understanding".The fifth part discusses some thinking about recognition process of scientific items' attributes based on the conclusions in this research. The patterns of students' information operation are various and complicated. Implications are provided for item developer, teachers, and educators based on those research findings. Finally for the reflection of study, research insufficient was presented in this article, and I offered the outlook for some further research.