![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
This study describes and uses a knowledge framework of science achievement for describing science test scores and relating them to students’ learning opportunity. It is a secondary analysis that uses various data collected through science test items and questionnaires in the TIMSS-R study, a large-scale research project on mathematics and science education. First, we defined our conceptual framework according to four types of knowledge that are characteristic of competency or achievement in science: declarative knowledge or "knowing that," procedural knowledge or "knowing how," schematic knowledge or "knowing why," and strategic knowledge or "knowing about knowing." Second, we applied this framework to logically analyze test items and statistically model item scores. We used a coding system to analyze and map test items into the type(s) of knowledge they intend to measure. We performed confirmatory factor analyses using item scores to ascertain whether the knowledge-factor model accounts for the underlying pattern of test scores compared to alternative models. Third, we focused on one aspect of learning opportunity – instructional practices reported by students and linked instructional characteristics to student performances on different types of science knowledge.
College of Education, University of Washington
Box 353600 Seattle, WA 98195-3600
coe@u.washington.edu