Can AI help K-12 educators overcome the burdens of assessments?
by
Many K-12 educators have resisted using artificial intelligence (AI). But a new national survey from Pearson and K-12 Dive’s studioID shows a new eagerness to use AI-powered tools in one area: classroom assessments.
Seven in 10 (72%) respondents say they’re already using AI with assessments or plan to do so soon.
Why? The Pearson survey suggests they see AI as a way to ease some of their assessment burden.
Assessment pain points
The assessment burdens on teachers are real. The biggest concern, according to survey respondents, is that assessment takes too much time at every step of the process — from creating the assessments, to giving the tests, to scoring the assessments, to using the results to guide future instruction. On top of this, educators seem particularly dissatisfied with the data they get from their assessment tools. Only 42% of respondents strongly agree that assessment data is easy to understand.
Little wonder, then, that the four most valuable AI features identified in the survey save teachers time at each stage of assessment. Educators see AI as a tool well suited to:
- Generating assessment questions
- Scoring open-ended responses
- Reducing the time they spend studying assessment data
- Making instructional suggestions based on the results
The survey findings show that educators want AI to make assessment more accessible and actionable, says Trent Workman, Senior Vice President of Pearson’s School Assessment division. “By providing an easy way to analyze various data points, AI can help educators tailor their instruction to better meet student needs and support school goals.”
The actionable piece is essential, and it speaks to another takeaway from the survey: Educators are concerned about assessment effectiveness, not just efficiency.
Where’s the benefit?
Assessments should (a) show what students have learned and (b) guide teachers on where they need to focus instruction. In other words, assessment should connect with teaching.
Unfortunately, most educators say it doesn’t.
The vast majority of survey respondents struggle to use assessments to pinpoint student needs, measure students’ progress toward specific learning targets and personalize instruction based on results. They also question how well assessments measure critical thinking — the skill they say is most important to assess. However, many educators don’t see how assessments benefit them or their students.
“If we don’t see the benefit of doing something, we want to stop doing it,” says Amy Reilly, Pearson’s Vice President of Assessment Product. “That’s what I see with a lot of teachers. They tend to feel like a lot of these assessments are done to them, not for them and their students.”
Time shouldn’t be the only concern
AI can certainly save teachers time. In that sense, technology can ease some of the assessment burden. But AI alone won’t solve all the problems educators identify in the survey.
Even the most advanced AI-powered tools can’t overcome flawed assessment techniques. Equally important, using AI to generate assessment questions runs the risk of creating more problems.
Assessment data must be valid and reliable at a detailed level to guide instruction. You need to have confidence in your assessments in many areas. Is the content appropriate to the grade level? Does it align with your state’s standards? Does it reflect appropriate rigor — and can you gauge how easy or hard the questions are?
If you can’t answer such questions to your satisfaction, then it’s a good time to explore best practices for improving your assessment data.
