PARCC Assessments in the Making: A Principled Assessment Design Approach

Thursday, June 28, 2012: 1:30 PM-3:00 PM
Marquette 1 (Hilton Minneapolis)
Content Strands:
  1. Assessment Design and Purposes
  2. Innovation in Teaching, Learning, and Assessment
Next-generation assessments will require a principled assessment design in which assessment development is accomplished with the ultimate inferences one wants to make about students in mind. With this understanding, the Partnership for Assessment of Readiness for College and Careers is committed to an evidence-centered design approach. This session will address how claim and evidence statement, derived from the Common Core State Standards, and related task models are being used in developing tasks for PARCC English language arts/literacy and mathematics assessments.

The first presentation in this session will address how evidence-centered design (ECD) principles are applied to item, task and assessment development by the Partnership for Assessment of Readiness of College and Careers (PARCC). Key concepts such as claims, evidences, and task models will be explained. The role of claims, evidences, and task models in assessment development will be illustrated with examples. The presenter will make the case that ECD is not a prescribed process; it is a set of activities and artifacts that help assessment designers focus on the observable evidence required to support claims about student proficiency. In fact, this is one of the main challenges in adopting this approach to development of a large scale assessment. There is not an easy-to-follow, one size fits all recipe for implementing ECD. The approach does not, for example, assert that you should construct an evidence model of any particular kind; rather, it highlights that the ideal path to move from a task to an inference is to use an evidence model to reason from observation to conclusion (Behrens, Mislevy, DiCerbo and Levy, 2010). ECD model relies on judgment on behalf of assessment developers that is driven by a focus on the purpose and use of the assessment and the constraints. All of the many terms, concepts, representations and structures in ECD are aimed at constructing a coherent assessment argument and building machinery to implement it.

Some of the questions that will be addressed in this presentation include the following:

  • How are claims and evidence statements related to Common Core State Standards?
  • At what level of granularity are claims written in PARCC assessment development?
  • What is the relationship among claims, reporting categories and performance level descriptors?
  • How are claims and evidence statements related to Common Core State Standards?
  • How do evidence statements and task models interact?
  • What is the level of specificity of task models?

The presenter will describe the entire assessment development process, and the lifespan of items and tasks in relation to this process from the birth of a claim to the piloting of an item.

The second and the third presentations in this session will report on the use of ECD in 2011 and early 2012 in creating prototype tasks for PARCC ELA/literacy and mathematics assessments. In particular, the presentation will address the use of ECD: (1) to inform the development of performance-based tasks;(2) to ground measurement of student performance in observable products; (3) to develop tasks that draw on multiple standards where appropriate; (4) to inform decision making regarding how to distinguish between partial and full expressions of the knowledge and skill(s) embedded in a given standard; and (5) to develop a wide variety of useful tools for administrators, curriculum developers, classroom educators, and other interested stakeholders.These presentations will illustrate various uses of evidence statements with sample ELA/L and mathematics tasks developed for PARCC. In addition, the relationship between evidence statements and task models and the use of task models in developing tasks will be illustrated.

Lead Presenter: