8
Assessment as a “Caring Practice”—Systems of Assessment
The assessment journey for the NGSS, said Aneesha Badrinarayan (Learning Policy Institute), started at the very beginning; one of the immediate consequences of three-dimensional (3-D) standards is that they “pose a real disruption to the way that we have traditionally thought about assessment.” However, assessment can serve as a guide for moving forward and thinking about meaningful teaching and learning, what information is useful to surface, and how to support teachers. “Assessment ultimately is about supporting student learning toward the goals that we have, and students have for themselves,” she said.
Badrinarayan asked workshop participants to think about what they value most about the experiences that students have, the performance that they can produce as a result of their educational experiences, and what role assessment plays in meeting this vision. When people are asked these questions, she said, they tend to focus on students seeing themselves and their identities as essential components of the scientific endeavor, on students building confidence and growing, and on students becoming critical thinkers who can navigate complex novel situations. What people generally don’t focus on, said Badrinarayan, is content memorization or knowing rote steps for a procedure without understanding the context or how to apply their knowledge. A major part of the assessment journey for the NGSS has been reconsidering how we match up what we want to cultivate and support in students with what our assessments are monitoring.
Since 2014, when the Board on Testing and Assessment (BOTA) and Board on Science Education (BOSE) released the report on Developing Assessments for the Next Generation Science Standards, there has been “a
real journey” of figuring out what assessment should look like and identifying the kinds of tools and resources that teachers and students need in the classroom. As time has gone on, criteria have been released to help us think about what high-quality assessments look like, she said, and there has been an evolution toward looking at assessment through an equity lens. Most recently, there has been a call to action to elevate science assessments in local and federal accountability processes, as well as to start to think about the use of assessment for students and teachers in the classroom. This represents a shift away from thinking about assessment as large-scale tests and toward assessment as a “wraparound support for all the teaching and learning goals that we have for students.” Resources have been developed, such as learning communities, implementation and design tools, and models to learn from. This work has led to some common “non-negotiable” features of 3-D science assessment, said Badrinarayan, to include the following: focused on figuring out phenomena and problems, requiring sensemaking with ideas and practices, and equitable and fair to students.
Through all this work, “we know a lot more about what we want assessments to look like and do,” she said. This panel will continue this work by considering the successes we have had in implementing the vision, what tools have been useful, and what opportunities there are to keep progressing. Badrinarayan cautioned that the discussion would not focus on which assessment is the “best one” but would instead look broadly at where we are and how to move forward. Like in the other sessions, Badrinarayan asked the following five panelists a series of questions.
- Daniel Alcazar-Roman, The Lawrence Hall of Science at the University of California, Berkeley
- Alec Barron, San Diego Science Project
- Tamara “TJ” Heck, Michigan Department of Education
- Angela Landrum, Colorado Department of Education
- Jim Pellegrino, University of Illinois at Chicago
VISION FOR ASSESSMENT
One of the challenges of developing assessments that are aligned to the NGSS, said Pellegrino, is that a lot of priority is given to classroom assessment. That is, teachers need tasks to give students to monitor their progress toward meeting performance expectations. Pellegrino shared a
video of three seventh-grade girls engaged in solving an assessment task that was developed by the Next Generation Science Assessment group, a collaborative that has been working to develop tasks for the classroom. This task was designed to get the students to construct a scientific explanation using evidence that describes the patterns of organisms interacting in different ways, he explained. The students have previously had instruction related to this performance expectation in class and received a prompt to help them solve the task collaboratively. These types of tasks are freely available to educators in an online portal;1 he noted that there have been more than 1 million page views, nearly 10,000 teachers have registered on the site, and the site has been accessed by users from 153 different countries.
“I am one of those users,” said Barron, saying that the San Diego Science Project has been supporting collaborative inquiry and sensemaking through tasks such as the one that Pellegrino shared. The key, he said, is making these performances visible not just to the teachers but also to the students so that they can monitor their progress with learning. Although there has already been quite a bit of work on long cycle assessments and end-of-learning tests, the next frontier of work is to focus on short cycle assessment processes that help support students with moving toward goals. In Barron’s former role as a content specialist, he used three core questions to guide student assessment: Where am I with my learning, where am I headed, and how am I going to get there?
Badrinarayan asked the panelists to elaborate on how teachers would use tasks such as the one Pellegrino shared. Pellegrino said that these tasks were deliberately designed to be used by teachers embedded in the context of their ongoing instructional plan. They are not stand-alone projects but instead are used purposefully to get evidence about how the students are doing. A teacher can also build off these tasks to stimulate conversation about the thinking process and the context of how the task fits into the larger curriculum. Pellegrino said that it is really important to understand that assessment is intended to be linked to ongoing curriculum and instruction, “not apart from it.”
CLASSROOM-LEVEL ASSESSMENT
___________________
In Barron’s work in Escondido, California, the teachers asked a novel question: “If the NGSS is framing for us a new vision for student learning, then shouldn’t that mean we have a new vision for assessment to support these learning experiences?” One of the great areas of progress that the field has made in this area, he said, is in creating tools like the Achieve Science Task Screener and protocols from the Center for Collaborative Education. These developments have supported the work of re-professionalizing educators to engage in classroom assessment of students. For example, said Barron, one powerful tool that came out of the work of professional learning communities was tuning protocols with rubrics to support students with more specific and actionable feedback. In terms of challenges, Barron said there is sometimes too much focus on longer cycle assessments rather than shorter cycle assessments. There is a need to develop tools and resources to support short-cycle formative processes in actual learning contexts, he said. Barron shared that he and his partners have been working to develop an NGSS assessment toolkit to bring greater coherence among curriculum, instruction, and assessment at the classroom level.
In Colorado, said Landrum, graduating students are required to pass one or more assessments from a menu of options; the menu includes traditional assessments such as the ACT and SAT but also innovative student-centered options such as capstones and performance assessments. The vision for the performance assessment component, she said, is to create a pathway for all students to connect what they learned in the classroom to their experiences as individuals in a way that leads to a more authentic and sophisticated application of knowledge and skill. Assessment is an essential component of effective implementation of the NGSS, and Landrum and her colleagues have been building resources and guidance for high-quality performance assessment. One of the first resources that they developed was a set of design elements to serve as criteria for what is considered high-quality performance assessment. The design elements include features common to NGSS implementation, such as alignment of content, skills, cognitive complexity, accessibility for all learners, authenticity, relevancy, and transferability. One unique feature, said Landrum, is that performance assessment be iterative, to connect it closely to the classroom. The collaborative protocols for this work are critical and serve to elevate the quality of the assessment by establishing “what it looks like when students are successful.”
Landrum listed several challenges that have been identified through this work. High school is “notorious” for being siloed, she said, so establishing coherence both vertically and horizontally can be difficult. District and school leaders sometimes express support for this work, but they don’t help remove the barriers for teachers or are unable to articulate how the work aligns and supports school and district goals. This lack of support from
leaders can leave teachers feeling isolated and undervalued; consequently, the professional learning efforts that teachers are involved in may not have the desired impact on educator practice and student outcomes. Landrum said that recent research has surfaced the need for systematic approaches to professional learning for teachers. In the absence of high-quality examples, “many educators really believe that the work that they’re doing is meeting expectations when in fact it is not.” These challenges are being addressed through work with the State Performance Assessment Learning Community, which provides educators with curriculum-embedded or curriculum-anchored assessment supports, and helps teachers design and use assessment with support, coaching, and connection with educators across the country. This is particularly helpful, she said, for educators in rural areas; feeling a sense of coherence and camaraderie with teachers in other areas is “really valuable.”
Alcazar-Roman gave his perspective on where progress is being made in this area. Educators are now collaborating to design assessments rather than working in isolation, and more principled approaches, such as evidence-centered design, are being used to guide assessment and development at the classroom and local level. For example, he said, in Washington, DC, modules from the Access Project were used to help teachers understand the NGSS and assessment; at the same time, teachers were engaged in the process of developing learning performances for every performance expectation that is included in state-level science assessments. Teachers spent many months deeply engaging in understanding the standards, unpacking the dimensions, and building maps to guide the development of assessments. As a result, groups of teachers in DC are getting together to use the same evidence-centered design to build their own assessments. In this process, said Alcazar-Roman, it was critical to go beyond teachers and involve the local education community. For example, the Smithsonian Science Education Center helped with engaging teachers in looking at claims for assessment, reporting, and figuring out local phenomena to include. There is great value in teachers, practitioners, and researchers collaborating to develop tools to work toward better assessments, said Alcazar-Roman.
ROLE OF THE STATE
One of the biggest projects undertaken in Michigan, said Heck, is an effort to develop a new state science assessment with a variety of partners.
Michigan has always involved local educators in the writing and reviewing of assessment tasks, but with the NGSS, there was a need to bring on and partner with people who had different kinds of knowledge, including classroom teachers and science education researchers. This collaboration brought together multiple perspectives of practice and research in the design space for large-scale assessment, she said. In addition, the collaboration offered opportunities for networking, professional learning, and building relationships between researchers and practitioners. Another state-led effort, said Heck, has been facilitating a shift away from rote knowledge practices toward phenomenon-focused assessment structures; this allows students to use their sensemaking skills and bring together their knowledge of various dimensions of knowledge on the assessment. Heck joked, “If you can Google it, it doesn’t belong on my test.” The vision is to move the large-scale assessment toward what we want to see happening in the classroom, she said. Although people “cringe” at the idea of teaching to the test, Heck said that her goal has been to create a test that is worth teaching to.
Colorado only recently adopted NGSS-aligned standards, and implementation began in the 2021 school year, said Landrum. Most of the educators in the state are still figuring out what it looks like to teach and assess in ways that are aligned with and appropriate for the NGSS, and educators are working to build consensus on what will be expected of students. Although the implementation of the NGSS is still very much in progress, Colorado is working to enable meaningful systems of assessment conversation. Landrum said that the state is providing guidance, examples, and guardrails for how classrooms and districts can generate evidence to move student learning forward. The state assessment is considered only one small piece of the comprehensive assessment system, she said, and it is not expected to capture all learning in science. Landrum and her colleagues at the state level are working tirelessly to bring teachers and leaders into the conversation about appropriate and intentional use of assessment across the system.
PROGRESS ON ASSESSMENT SYSTEMS
We always knew it was going to be a challenge to develop assessments for the NGSS, said Pellegrino, due to the nature of the standards, as well as the state of assessment design. Other speakers, he said, have noted the need for a system of assessments that are coherent with each other, aligned to the
same vision and goals, and designed for particular purposes. The BOTA/BOSE report on NGSS assessments discusses the need for both formative and summative classroom assessments, and the fact that state monitoring assessments cannot be “drop-in from the sky” large-scale assessments. Landrum and Heck both illustrated the need for assessments that are tuned to the particular context and purpose, he said. Even at the state level, it is important to not rely entirely on year-end large-scale assessments but to instead use smaller-scale classroom assessments that involve performance tasks. One of the biggest challenges, said Pellegrino, is supporting teachers in understanding exactly what these kinds of assessments look like in practice and how to support students in reaching the goals that are set. The large-scale state test is oftentimes the “boogieman” in the process, he said, but it can lead to positive change by signifying the kinds of things that we want students to be able to do.
EQUITY AND ASSESSMENTS
“When I first came into this position,” said Heck, “I started digging around” and asking colleagues what was happening regarding equity and large-scale assessment. The response, she said, was “crickets in the room.” We have come a long way in the past several years in considering equity and assessment by prioritizing assessments where students need to use sensemaking rather than rote memorization or high-level language skills. These are the kinds of deterrents, she said, that can prevent students from having the opportunity to show us what they know and can do in science. Another important step being taken is ensuring that phenomena that are presented on large-scale assessments are relevant and palatable to students.
At the state level in Michigan, a bias and sensitivity review committee examines each assessment item, and stakeholders with specific lenses (e.g., special education, English language learner) look for potential issues. Although this process is important, said Heck, it is not enough. It is critical to get people on these committees who are representative of our students. For example, in Michigan, meetings tend to be held on the lower peninsula, which results in a lack of representation from the rural upper peninsula. In addition, said Heck, there is a need to find a way to involve students in the assessment development and design process to get the perspective of the people for whom the assessment is designed. Finally, Heck said that as we bring more diverse groups of people into the space of assessment develop-
ment, we need to offer professional learning opportunities around issues like microaggressions, anti-racist training, and cultural diversity education.
Alcazar-Roman agreed that there is a need for more action on equity at the state level; in particular, he noted that bias and sensitivity reviews are the last step in the process rather than embedding equity along the entire path of assessment development. Pellegrino agreed with this point and said that we need to “think about equity from the very outset of design.” He added that equity be considered when identifying performance expectations, relevant phenomena, and potential linguistic challenges; waiting until the assessment has been developed is too “late in the game.” Although improvements are being made in this area, large-scale assessments are by their nature going to have a degree of decontextualization because of the need to homogenize assessments for a large group of students, Pellegrino said. Despite this challenge, he said “we still have to strive to make it as fair and equitable as possible.”
On the bright side, Alcazar-Roman said, there is a lot of investment in the world of large-scale assessment, which means the opportunity to leverage funds to meet the vision of equity. In DC, they have undertaken several efforts in this area, including identifying more local phenomena, reconsidering how to design problems, and building models of universal design. There is further work to do, said Alcazar-Roman, such as looking at social and environmental justice issues in the assessments. State- and federal-level assessments tend to have a number of restrictions, but at the classroom and district levels, there is more flexibility for adapting assessments to the context of the learner. We need tools to help educator networks contextualize phenomenon and problems and develop tasks that address local issues so that assessments can be more culturally relevant and equitable, Alcazar-Roman said. Doing so will enable students not only to demonstrate their science knowledge but also to see how they can make a difference and solve problems in the community. Furthermore, said Alcazar-Roman, there is a need to diversify the people involved in education and assessment—we need the people making decisions and development assessments for millions of students to match the diversity of those students.
DISCUSSION
When using assessment in a formative way, said Pellegrino, scores and grading should not be part of the process. Instead, he added, the feedback
should be aligned with the kinds of thinking that we want students to be able to do. However, part of the challenge is that grading is embedded in teachers’ assessment, literacy, and ongoing practices, he said. Heck added that we need to consider “who is grading serving” and how we can create a system that will actually be of value to the parents and students in providing feedback. One approach, she said, is to create a student learning story—how are the students learning and growing, and what can they do now that they couldn’t before?
This page intentionally left blank.
