Concept Map Of Research, Measurement And Evaluation Of High School Chemistry Teaching

The author introduces process, designs, models, hypotheses, and estimated variance component of Generalizability Theory (GT) at length on the basis of three kinds of educational test theories. Along with the research of concept map and its assessment technique theories, concept map assessment is investigated and discussed on three aspects which are Generalizability Study and Decision Study on several random crossed model designs based on Generalizability Theory, validity analysis based on Classical Test Theory (CTT) and case study on other elements. Three findings are embodied as follows. First, items including title and show, raters and marking methods have an effect on reliability of concept map at different levels. The most effect is marking method, then, is item of title or combination of title and show, finally, is rater. Second, concept map test is guaranteed well on validity and shows differences and merits in contrast with other tests. Moderate construct validity suggests that three different measurements (concept map, essay and multiple-choice) reflect different aspects of understanding. Low concurrent validity occurs if we consider multiple-choice as our edumetric criterion of validity. When card carding is regarded as psychometric criterion, high concurrent validity occurs. Third, concept map assessment can provide valuable information of student's knowledge stucture really. Different students construct different concept maps, which explain the difference of cognitive structure and impact on higher-level knowledge constucting. We can judge whether he or she understand the relation from the proposition and mark it. Several marking methods show different abilities and levels of students in many elements. Consequently, this article proves out feasibility of concept map as an alternative assessment tool in chemistry at high scool from quantitative and qualitative measurement in length and breadth.