NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED422404
Record Type: Non-Journal
Publication Date: 1998-Apr
Pages: 30
Abstractor: N/A
Reference Count: N/A
ISBN: N/A
ISSN: N/A
Performance Assessment in the Service of Evaluating Science Education Reform.
Ruiz-Primo, Maria Araceli; Wiley, Edward; Rosenquist, Anders; Schultz, Susan; Shavelson, Richard J.; Hamilton, Laura; Klein, Steve
This study explored the sensitivity of a multilevel approach in detecting outcomes of a hands-on science program, exploring whether the instruction of a hands-on unit had any impact on students' performance and whether the estimated magnitude of the impact was different according to the proximity of the assessments to the unit taught. Also studied were whether the impact could be detected at a distal level, and whether the differences observed across types of assessments could be replicated across curricular units. This pilot study was conducted in a medium-sized urban school district in the San Francisco Bay area (California). Five schools, 7 teachers, and 163 fifth graders participated. To implement the multilevel achievement assessment, two performance assessments were selected for each unit, one close and one proximal. Distal assessments included performance assessments developed by the California Systemic Initiatives Assessment Collaborative. Preliminary results suggest that instruction had no impact on student performance. As predicted, close assessments were more sensitive to changes in student performance, while proximal assessments did not show as much impact of instruction. It was not possible to assess the sensitivity of distal assessment because no pretest-posttest data were available. High between-class variation in effect sizes suggests the need to examine the instruction students are receiving. Results were not replicated across the two instructional units. Characteristics of the close and proximal assessments seem to have a higher influence on detecting students' improvement than originally thought. Ongoing studies are discussed. An appendix contains a chart of raw score descriptive statistics. (Contains 19 references.) (SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: National Science Foundation, Arlington, VA.
Authoring Institution: N/A
Identifiers - Location: California