Abstract
Although federal regulations require testing students with severe cognitive disabilities, there is little guidance regarding how technical quality should be established. It is known that challenges exist with documentation of the reliability of scores for alternate assessments. Typical measures of reliability do little in modeling multiple sources of error, which are characteristic of alternate assessments. Instead, Generalizability theory (G-theory) allows researchers to identify sources of error and analyze the relative contribution of each source. This study demonstrates an application of G-theory to examine reliability for an alternate assessment. A G-study with the facets rater type, assessment attempts, and tasks was examined to determine the relative contribution of each to observed score variance. Results were used to determine the reliability of scores. The assessment design was modified to examine how changes might impact reliability. As a final step, designs that were deemed satisfactory were evaluated regarding the feasibility of adapting them into a statewide standardized assessment and accountability program.
Notes
1There are limitations in treating tasks in this manner. First, results will not be produced that allow for interpretations in variability due to differences in task difficulty because task differences cannot be disentangled from task by student interactions in this design. In addition, the task variability that is confounded with the interaction might also be influenced by the fact that task is not a truly nested facet. It might be slightly underestimated as compared to a design where students were administered a purely unique pair of tasks.
2An interaction between rater type and student indicates that students are rank-ordered differently by the two types of raters. Another way of thinking about it is that the way in which the two types of raters differ in their ratings depends on the student.
3These percentages are based on total variance of scores had the scores for students been averaged across a single rater type, single task, and single attempt. This is not to be confused with the observed total variance of scores where the scores for students were averaged across two rater types, two tasks and three attempts.