Learning Record Shows Promise For Accountability Uses

Status: 
Archived
Subject: 
K-12 Testing

By Sally Thomas

 

“The moderation acts as a professional development tool to improve the quality of the [student] record.”

 

“Teachers use the language of the scales to help them look for evidence. Teachers also cross check their knowledge/interpretations before arriving at a score.”

 

These comments from observers at the 2001 Learning Record Inter-Site moderation illustrate some benefits of the Learning Record Assessment System (LR), now slated for use in 35 schools across the U.S. in 2001-2. The LR was adapted from the English Primary Language Record to a U.S. context (see Examiner Spring 1999; Summer, 1995; Summer 1992). It is a detailed, classroom-based means of documenting and assessing students’ literacy and mathematical achievement and understanding. Ongoing data collection and evaluation helps teachers develop plans to improve teaching and learning. The record includes detailed descriptions of student learning, placement of a student on a “scale” of learning growth, and student work samples.

 

Under the moderation system developed by the Center for Language in Learning (CLL), educators review and “re-score” a sample of a teacher’s Records. This enables classroom assessment to provide information for external accountability purposes. Sharing interpretations of criteria for judging student work quality across classrooms and schools also helps to ensure fairness as teachers learn to hold each other to similar standards.

 

Teachers first join with other educators at their school site to read the sample records; then at an inter-site moderation, representative teachers from all LR schools participate in reading the same records. At the 200l moderation in Carefree, Arizona, 740 randomly selected records from 24 schools in 12 states were reviewed by 90 teachers.

 

From this process, schools and individual teachers receive a wealth of valuable information for instruction and for future in-school professional development. Teachers receive comments from other teachers at both the site and the inter-site moderations on the quality of the evidence provided in each LR.

 

Individual teachers and their schools receive reports on the reliability of teacher judgment,
i. e., the match between the classroom teacher’s rating and the final moderated placement. A study of previous years’ moderations concluded that the LR is a reliable and valid measure of student learning.

 

The school also receives charts reporting on the number of students meeting benchmarks on the LR scales that measure state standards set for particular grade levels. The growth of individual students from year to year and percentages of the total school population meeting standards is reported. Disaggregation of data helps schools analyze the progress of different groups of students by gender, race/ethnicity, language, and socio-economic status. Correlations between LR ratings and state-mandated assessment system scores are provided.

 

A summary of a survey of teaching practices shows the extent to which teachers are providing students with learning experiences supported by child development and learning research. Finally the school receives feedback on the consequences of using the LR. Moderation participants provide information to help answer to questions such as: Do teachers find it useful in supporting teaching and learning? Is it useful in communicating with parents? Does it guide school reform?

 

Schools find the reports both inform the practice of teachers individually and help school staff analyze school-wide strengths and needs. One American Indian boarding school in its second year of LR use discovered that its students need more opportunities to learn to “evaluate information from multiple sources” as part of the higher reading/writing demands in the final years of high school. This sort of information is not available from standardized tests, which fail to assess such complex work.

 

Inter-rater reliability quotients (the degree of agreement among different readers) of .77 for reading and .75 for writing show that teachers at this school have similar standards for documenting, analyzing, and measuring student performance. At more experienced schools, inter-rater reliability reached the 95% range. Combined with the strong validity of the LR, the reliability data provides evidence that the LR can be used in an appropriate public reporting system.

 

• Sally Thomas is co-director of CLL.
• CLL, 10610 Quail Canyon Road, El Cajon, CA 92021; 619-443-6320; http://www.cwrl.utexas.edu/~Syverson/olr/