Learning First Alliance

Strengthening public schools for every child

New Assessments, Better Instruction?

obriena's picture

Advocates hope that the Common Core State Standards (CCSS) Initiative will lead to deeper learning by students – that the standards will result in students learning not only academic content, but how to think critically, work collaboratively, communicate effectively and more. And they recognize that for this to happen, classroom instruction has to change.

Many assume that new assessments aligned to the Common Core will serve as a key lever in its implementation, driving changes in instructional practice. But is that a reasonable assumption? Do large-scale assessment systems influence instruction?

While common sense and popular opinion hold that yes, they do, the research base on the issue is surprisingly thin. But in summarizing the little there is, New Assessments, Better Instruction?, a RAND literature review commissioned by the Hewlett Foundation, confirms what we already knew – testing does indeed influence instructional practice, particularly when high-stakes are attached to it.

The review (which included available research on high-stakes testing and performance assessment in U.S. public education, assessment in international settings, formative assessment, military and occupational testing, and processional certification and licensure testing) found that testing is associated with a variety of changes in educators’ practices. These changes occurred in three broad categories:

  • Curriculum content and emphasis, including changing the order in which content is covered, narrowing instruction to focus on certain subjects or content, and focusing on certain types of skills.
  • Instructional activities, including engaging in test preparation activities (particularly around test-taking strategies rather than content), adopting new instructional strategies (typically more traditional teacher practices, such as lecturing and worksheets), and changing assessment practices, often to mirror the format of a high-stakes test. 
  • Teachers’ interactions with students, including individualizing instruction to meet the needs of all students or focusing resources on students most likely to impact overall results – the “bubble” students. As a side note, one consistent theme throughout the literature is a gap between identifying student needs and changing instructional practices to address them – in other words, assessment data helps teachers target instruction, but does not impact how that instruction is delivered.

Of course, there was wide variability in the research – as one of the authors, Laura Hamilton, pointed out in a recent briefing, almost every study reviewed found huge variability in how teachers responded to assessments, which much of the variation unexplainable.

Still, overall, the review shows that assessment impacts instruction. And it points out that these changes are not inherently good or bad. If a testing program covers a broad range of skills and knowledge or a performance assessment leads to an expanded repertoire student-centered teaching strategies, the effect can be positive. The context of the assessment – both in terms of its role in an accountability system and the local environment in which it is given – matters. To that end, the report offers recommendations to create conditions that increase the likelihood of a positive impact, including:

  • Test content and format should mirror high-quality instruction.
  • Tests should be used only for purposes for which they were designed and validated.
  • Score reporting should be optimized to foster instructional improvement (for example, delivered in clear, accessible language in a timely manner).
  • Teachers should receive ongoing training and support to interpret and use test scores effectively.
  • Test scores should “matter” to ensure stakeholders take the assessment seriously, but important consequences should not follow directly from test scores alone to avoid a “teaching to the test” mentality that focuses on superficial items at the expense of deeper learning.
  • If there are externally mandated, annual high-stakes summative tests, they should be part of an integrated assessment system that includes formative components that can be used for instructional guidance.
  • Accountability metrics should value growth in achievement, not just status.
  • Assessment should be one component of a broader systemic reform effort (including standards, curriculum, professional development, community engagement and more).

The overall implications for Common Core implementation that I took from this report: New assessments will impact classroom instruction, and the policy decisions at the federal, state and local level that create the conditions under which those assessments operate will determine whether the changes are positive or negative. Given the current political climate, it is anyone’s guess as to how it turns out.

Moving forward, Common Core advocates face a big challenge. While we can expect CCSS assessments to influence instruction, we need to continue to remind the public – and ourselves – that the assessments are not the Common Core. Particularly given the controversy surrounding them, with concerns on the role they will play in high-stakes decisions for students (such as advancement to the next grade) and teachers (such as evaluation) currently unresolved, as well as uncertainty around the technological readiness of schools and districts to execute them as intended, the assessments could experience significant challenges and pushback in implementation. If so, and if the assessments are equated with the Common Core, support for the standards could plummet. 

Note: While in this post “assessment” and “testing” are used almost interchangeably, I recognize that “assessment” is a much broader term that can include a number of other strategies for determining whether learning has occurred.  

Views expressed are those of the author and do not reflect the endorsement of the Learning First Alliance or any of its members.

Image by Alex Morfin (Own work) [CC-BY-SA-3.0], via Wikimedia Commons