Designing an Assessment Solution that Fits for All Students is Different than Fitting All Students into an Assessment Solution
In 2019-2020, the requirement to administer state summative assessments was waived for all states. In the absence of this data, many states and school districts leaned on commercial interim assessment data to inform programmatic and instructional decisions to a greater extent than they had in the past. Given the significant role these assessments have come to play and the large number of states exploring the use of through course interim assessment designs for purposes of school accountability, it is important to ensure they provide valid information for all students, including students with disabilities.
Recently, NCEO convened a panel of experts to discuss issues related to the quality, availability, and use of interim assessments for students with disabilities and generated recommendations to help state agencies promote appropriate interim assessment practices. To inform these recommendations, technical documentation for several interim assessment programs was reviewed in search of evidence that suggested:
- the inclusion of students with disabilities was a consideration throughout the test design and development process (e.g., adherence to universal design principles; availability of accommodations and special forms) and that
- test scores could be interpreted and used in the same way for all students, including students with disabilities.
Current Interim Assessments are Not Designed for the Inclusion of All Students
The interim assessment review suggested that many commercial vendors do not offer the full range of accessibility features and accommodation necessary to ensure all students will be able to fully demonstrate their understanding of the targeted construct. In addition, none of the assessments reviewed currently offer alternate assessments for students with the most severe cognitive disabilities.
The fact that students with disabilities are not able to participate in these assessments in the same way as the general population is problematic. As noted in a recent NCEO brief by Browder, Lazarus, and Thurlow, “leaving a group of students out of a key component of an assessment system risks leaving them out of educational decisions and resources as well.” In addition, the Individuals with Disabilities Education Act (IDEA) requires that whatever assessment a state or district employs – consistent with ESSA or other state/local mandates – an alternate assessment aligned to alternate academic achievement standards must be developed for students with significant cognitive disabilities. Administering these types of assessments without a corollary is not only unfair, but it is also likely against the law.
Moving Beyond a One-Size-Fits-All Approach to Educational Assessment
Undoubtedly, we need to do better at attending to the needs of students with disabilities in our assessment systems. We cannot assume, however, that one-size-fits-all, or that filling gaps in the existing assessment system with parallel assessment solutions, is the best way forward, particularly as we move assessment for accountability closer to the classroom.
In a recent publication, my colleagues and I argue that the procedures and indicators used to inform decisions about school quality for schools supporting traditional students may not be appropriate for alternative schools due to differences in how quality is defined and the conditions perceived as most important to students’ success.
In the same vein, an assessment system design that assumes that the type and frequency of data necessary to inform educational decision-making is the same for all students may not be in the best interest of students with disabilities. Such an assumption is especially concerning for students with the most severe cognitive disabilities for which the time and resources necessary to design and administer assessments aligned to alternate academic achievement standards are significant.
For example, a commercial interim assessment that uses an adaptive design to benchmark student performance against end-of-year expectations may be an efficient way to monitor progress and identify instructional needs for the majority of students; but a parallel design (if possible) would likely be expensive and unwieldy for educators of students with severe cognitive disabilities. I am not saying that progress monitoring does not represent a shared information need; only that the solution must consider the unique factors associated with this population when weighing the costs/benefits of different designs, like the interim assessment options presented by Browder and colleagues.
A Solution Begins With Stakeholder Involvement
In previous publications, my colleagues and I have discussed the value of logic models or theories of action as a tool to support assessment system design (See: Dadey & Gong; Marion, et. al.; D’Brot). One of the first steps in the design process typically involves convening stakeholders to articulate the problem to be solved and the role of the assessment system in supporting the solution.
Stakeholders representing students with disabilities must be included in these discussions to ensure the identified problem and proposed solution are shared or at least appropriately inclusive of this population. Stakeholders can also serve to:
- highlight practices that are not well supported with existing tools (e.g., using commercial interim assessments to define IEP goals);
- flag assessment designs or products that may be problematic for students with disabilities due to feasibility, inclusiveness or underlying assumptions about how students acquire knowledge; and
- ensure assessments provide useful instructional information that is worth the time necessary for administration.
Once stakeholders have reached agreement on the shared problems, it may be determined that the ideal solution differs for students with disabilities or students with severe cognitive disabilities. For example, all parties may agree that a shared problem to be solved is that students are not able to apply what they have learned in novel settings. They further agree that a coherent system of curriculum, instruction, and assessment that prioritizes authentic demonstrations of student performance will promote the thinking and learning necessary to support transfer of learning across context and setting.
Working through the theory of action, however, clarifies that the role of assessment within this solution (i.e., with respect to type, frequency, and use of results) can and likely should differ across student groups. The development of summative performance assessments coupled with changes to the way academic expectations are modeled and evaluated may be considered necessary for students measured against the state’s academic content standards. On the other hand, teachers of students with the most severe cognitive disabilities who routinely administer individualized classroom performance tasks may suggest that opportunities that require students to apply knowledge outside of the classroom should be prioritized.
In order to support equity through assessment system design, we must recognize when unique assessment solutions are necessary to collect valid, useful information for different student groups. In the same way that instruction must be adapted to meet students where they are, the assessments that inform instruction must be selected and used in consideration of all the students they are intended to serve, including students with disabilities.