Assessments have the power to shape educational outcomes, but are we truly measuring what matters? Ensuring that assessments are fair, inclusive and meaningful for all students is a growing priority for educators. Bias, whether systemic or unintentional, can affect accuracy, disadvantaging students from diverse backgrounds. This requires a critical look at both what and how we assess, ensuring the most important skills and knowledge are prioritized.

Educational leaders are addressing these concerns by creating assessments that are not only standardized but also equitable and relevant. Bringing together diverse stakeholders, including assessment creators, teachers and students, can help design tools that provide a more complete picture of learning.

Recently, EdSurge webinar host Matthew Joseph discussed with education experts the need for assessments to measure what truly matters and power human progress. Webinar panelists included Patrick Kyllonen from ETS, Candace Thille from Stanford Graduate School of Education, Eugene So from JFFLabs and Temple Lovelace from Assessment for Good.

EdSurge: How can schools and educational institutions ensure that assessments are equitable and inclusive for all students?

So: Participation is key. At JFF, we focus on coalition development. When discussing consensus and assessment goals, it’s important to consider who is at the table validating skills. A more diverse cadre of stakeholders collaborating around the table improves goal-setting processes and outcomes.

Lovelace: One group I’d like to add to this discussion about equity and inclusivity in assessments is the assessment creators themselves. We need to consider these issues from the very inception of the assessment tool.

At Assessment for Good, we review our tools multiple times, asking if the wording captures diverse experiences. We use collaborative design to ensure equity and inclusivity by understanding students’ current experiences and co-creating tools with educators and students that match those experiences.

Kyllonen: Equity has a lot to do with opportunities, and assessments can uniquely provide opportunities to learn. Assessment feedback is key to showing performance and areas for improvement.

Students must know what is being assessed. There should be no confusion! Otherwise, we are not assessing properly. Students can’t demonstrate their skills if they don’t know what’s being assessed. These issues are addressed in length in Charting the Future of Assessments.

What role does assessment play in personalized learning, and how can it be used to tailor educational experiences to individual student needs?

Thille: Personalized learning involves individualizing experiences to support learners’ goals. We must consider not just the learner but all human actors in the system and the decisions they need to make to support that learner’s journey. These actors include mentors and assessment creators. They need to be aligned on the goal and have insight into the learner’s current state relative to that goal. This is where assessment is key, providing real-time insights into the learner’s changing state throughout the learning process.

As learners engage, these activities provide evidence for assessment. The resulting insights can then be shared with all actors — instructors, mentors and learners — enabling them to make informed decisions about the learner’s next steps toward their goal.

So: As we compound learning, we’re moving away from a two-dimensional view based on transcripts or degrees. Instead, we capture unique experiences that provide a more holistic view of what we’re assessing and toward what goal.

Often, students see assessments as punitive — failing a test can be damaging — rather than performative. In industries like fitness, assessments gauge progress toward goals. How can we use this performative-based assessment approach in education? Other sectors’ assessment practices can inform new approaches in education.

---

Watch the full “Unleashing the Untapped Potential of Assessment to Power Human Progress” webinar on-demand now.

---

How can educators implement innovative assessment practices to enhance student learning?

Lovelace: We often ask learners to pause learning to be assessed. Ideally, we should think about how to assess them while they continue learning, whether individually, in groups or in their community.

In our work, we’re also looking at “power skills” — skills that power the learning process. Knowing fractions is important, but believing you’re a math learner is equally as powerful. We need to consider what we assess along with how we assess to provide more complete data to educators.

The speed of assessment is also important. As an educator, getting scores back after summer break wasn’t helpful. We should innovate to leverage emerging technology, getting data back at nearly the speed of teaching and learning. This allows everyone, including the learner, to make the best data-based decisions possible.

How can assessment data effectively inform instructional decisions and support professional development for educators?

Kyllonen: We can now go beyond traditional methods with rich process data, including student conversation data. Communication and relationship building have always been in the background, but technology allows us to bring them to the foreground. We can analyze conversations and actions in interactive simulations to understand students’ thinking.

As technology improves, we’ll be flooded with classroom information. We need to develop process analysis models to understand these conversations and interactions. Facial expressions, for example, can indicate whether a student is understanding, frustrated or happy.

This rich data will enhance our understanding of classroom dynamics. It’s up to us to capitalize on this and develop systems that can inform teacher professional development and improve student instruction.

Thille: This means disambiguating the signal-to-noise ratio. We faced challenges extracting meaning from early clickstream data due to low signal-to-noise ratios.

An advantage of new technologies is the ability to collect more data. However, this creates bigger challenges in identifying patterns within the data that truly represent the signal.

It is not just data that educators want — it’s insights. And we need to deliver the insight in a way that is actionable.

Lovelace: While we can gather richer data from educational experiences, we need to do more to make it truly meaningful for educators, families and learners. We must communicate this data in an understandable way.

Educators don’t want more disparate data; they want to understand its immediate importance, how it relates to what they’ve just taught and possibly receive recommendations for next steps based on their chosen curriculum or current unit. It’s great to have technologies providing more data, but if we can’t understand it at the point of teaching and learning, we have more work to do to embed it into daily educational practice.

What are some of the most common misconceptions about assessment?

So: One misconception is that assessment is punitive. We have an opportunity as practitioners and innovators to view assessment tools as nonpunitive. Instead of seeing them as penalties, we can use them to uncover human potential and identify pathways to opportunities. This shift allows us to build on individuals’ strengths and support their growth.

Kyllonen: Another misconception is that assessments take time away from learning. Tests can be part of the learning experience, just as games and recitals are. In cognitive psychology, we know this as the testing effect. Taking a test can be more powerful for learning than recitation or memorization. This puts assessment in a different light. We must use assessment opportunities to take advantage of developing methods, procedures and technologies.

What future trends do you see emerging in the assessment field, and how should educators prepare for them?

Thille: This isn’t just about AI in assessment, which we’ve used for decades. It’s about new forms of AI, particularly generative AI. We’re seeing that generative AI can score well on traditional assessments, and now that learners have direct access to these tools, we need to rethink our approach to assessment.

We can’t simply tell learners not to use these performance support tools; it’s like saying, “You can’t use a calculator.” Instead, we should focus on helping people build skills with these available tools. This shifts what we’re trying to assess.

The big challenge now is figuring out how to use these new capabilities to create interesting assessments and assess things that matter. The goal is to use these tools to increase the diversity of voices, not standardize, and provide evidence about what works for whom under what conditions to support human learning.