Increasing Assessment Accuracy and Accessibility through UDL
We have outlined numerous shortcomings inherent in traditional assessments. How can new media and the UDL framework support teachers' ability to improve the evaluation process? Our knowledge of the three brain networks helps us understand the kinds of flexibility needed for the most accurate and most informative ongoing assessments.
Flexibility in Presentation
Technology enables teachers to provide multiple representations of content in the context of ongoing assessment. Varying the media within the representation is one of the most useful options. For example, Mr. Costa can provide text-to-speech support for Sophia during a history or science test so that her visual recognition deficits will not confound the results. But the concept of multiple representations goes well beyond the idea of offering varied media. Different representations can incorporate different supports, such as links to important background information, context-sensitive vocabulary supports, or glossary items. They can also be displayed in different formats—such as a visual concept map, a set of key points, a timeline, or a diagram—to make the content more accessible to certain students.
How can these varied representations and supports help to create accurate, ongoing assessment? First, they ensure that every student's assessment is accessible to him or her. Second, they permit a teacher to select an assessment that provides the supports a student normally uses. For example, Mr. Costa could retain the "read-aloud" feature in the science material when assessing Sophia's science concept knowledge, and Ms. Chen could allow Charlie to use the supports that help him focus on a task. The result will be the truest picture of students' progress, based on what is being evaluated, the supports these students normally use, and the format that is most accessible to each.
Multiple representations also serve teachers by providing information about the learning process. For example, Mr. Costa could provide the test questions in two formats—one in a large font and the other as text-to-speech—and compare Sophia's performance in the two cases to determine which support is most beneficial. If Ms. Sablan wants to determine whether diagrams, images, word definitions, or short summaries help Paula to connect concepts and understand what she is reading, she can present Paula with multiple versions of the content, each with selected supports turned on or off, to find out where Paula is most successful. This is an example of dynamic assessment (see Feuerstein, Rand, & Hoffman, 1979; Lidz, 1987).
As these examples show, offering multiple representations enables teachers to disaggregate specific problems linked to students' recognition networks from the learning and achievement factors under evaluation. It gives teachers a much fairer, more accurate, and deeper understanding of student learning.
Flexibility in Expression and Strategic Supports
We have seen how individual differences in strategic networks interact with traditional paper-and-pencil assessments to skew the accuracy of results. Within the context of ongoing assessment, teachers can accommodate differences in strategic networks by providing students with multiple means for expressing what they know, such as the option to respond by writing, speaking, drawing, creating an animation or video, or developing a multimedia presentation. When students are using tools that are familiar and appropriate for their own styles, needs, and preferences, they are not hindered by the medium of expression and are more likely to be able to demonstrate what they know and know how to do.
d
|
|
- Figure 7.1 -
Options for Expression in Tenth Planet's Geometry Program
|
| From Tenth Planet's Introduction to Patterns. Sunburst Technology. |
|
Few would argue against letting a student with severe motor difficulties take a test in electronic form that allows for alternative keyboard or voice recognition software. Somewhat less obvious is the degree to which students with subtler difficulties would benefit from the multiple expressive options electronic media can provide. Multiple media options are tremendously useful to students who have trouble with writing mechanics and they are also helpful to the many students more accustomed expressing themselves on a word processor screen than on paper (Russell & Haney, 2000).
Technological tools make it easier for a teacher to provide every student with multiple options for expression. Speech recognition systems can record spoken responses and translate them to text. Software such as Hollywood can be used to create animated presentations. HyperStudio, PowerPoint, and other similar tools make it easy to develop multimedia presentations.
Technology also offers the opportunity to assess skill learning in a deeper and more meaningful way. For example, science students might conduct virtual lab experiments, in which their actual manipulations of data, technologies, and substances would demonstrate their understanding of processes, methods, and outcomes more clearly than any written or verbal response could. Tenth Planet's Geometry program, shown in Figure 7.1, illustrates some of the multiple options for expression digital tools can provide. The program's assessment piece offers a multimedia journal in which students can record and play back audio notes, type in comments, or draw images related to what they have learned. Throughout the assessment, students receive scaffolding in the form of review information and prompts to support them as they enter ideas into their journals. With this one program, students can develop a whole portfolio of work that demonstrates their learning in multiple ways and with a variety of strategic supports.
Options for monitoring varied strategic supports are one of the most complex and fascinating areas of UDL assessment. By examining where supports succeed and fail, a teacher can identify how students successfully learn how to learn. For example, throughout the school year, Ms. Chen could experiment with highlighting, underlining, and italicizing tools to identify which best promotes Charlie's ability to stay on track and digest a text.
With Anne Marie Palincsar and Jeanne Paratore, CAST is investigating how supports for strategic reading embedded in narrative text can help students learn and can help teachers and students monitor ongoing progress (Palincsar & Brown, 1984). For example, CAST's research prototype, Thinking Reader, uses embedded assessment to track and save all student responses to strategy prompts. When reviewing the student record, students and teachers can see the book and date, the prompts or questions, and the students' responses. Figure 7.2 shows how a Thinking Reader work log might look for Kamla.
d
|
|
- Figure 7.2 -
Thinking Reader Work Log
|
| Prompts embedded in a digital novel help guide Kamla's strategic reading. The software collects Kamla's responses in the electronic work log, and later, she and Ms. Abrams can use this record to evaluate her ongoing progress as a reader. |
|
In addition to collecting student input, the Thinking Reader prototype supports student self-reflection and self-assessment, offering rubrics and tools for determining whether they are satisfied with the predictions, summaries, and questions they have generated. The screen in Figure 7.3 shows a rubric for checking summary construction. Students can review their summaries to see whether they have addressed the key features and then either revise their work or send it on to their student work logs.
d
|
|
- Figure 7.3 -
Thinking Reader Self Check
|
|
These examples are just the tip of the iceberg—the products of only the early stages of development and testing in classrooms. Nevertheless, teachers can combine a variety of tools currently available to set up strategic supports that track learner progress. For example, Mr. Mitchell might use Inspiration to help Charlie set up a concept map of the steps to follow when completing an assignment. Each bubble on the map could include prompts and reminders to support Charlie's progress, and he could fill in the bubbles with color as he completes each step. Links to useful Web sites could also be embedded in the map.
Assessments featuring multiple, varied strategic pathways result in far more accurate and insightful evaluation. They ensure that no student's skills and knowledge are obscured by expressive difficulties and that each student can benefit from appropriate supports. Further, they foster a much deeper understanding of the teaching methods and materials that are helping students learn.
Flexibility in Engagement
The emotional valence of an academic task is critical in determining how well a student will succeed at it—or even how much effort he or she will invest. Students' unique affective networks manifest in their different responses to testing. For some, the stakes are set too high by testing in general and grading in particular, leading to a reduction in performance sometimes called test anxiety. For others, anxiety is a function of the particular task's difficulty.
The emotional valence of an academic task is critical in determining how well a student will succeed at it—or even how much effort he or she will invest. Students’ unique affective networks manifest in their different responses to testing. For some, the stakes are set too high by testing in general and grading in particular, leading to a reduction in performance sometimes called test anxiety. For others, anxiety is a function of the particular task’s difficulty.
Embedding assessment into ongoing work removes some of the emotional impact of testing and highlights its more positive aspects. For students like Paula and Kamla, who fear academic assessment, freestanding tests loom as an obstacle, a hurdle, a "failure detector." But when assessment is removed from its isolated stature and made a normal, constant part of learning, the feedback for both student and teacher is informative and helpful rather than intimidating.
In a digital environment, embedded assessment can offer additional flexibility to further accommodate students' affect. First, most students find the options available within a multimedia environment—images, sounds, animations, and simulation—fun and appealing. Second, teachers' ability to level and scaffold embedded assessments can ensure that every student is working at a comfortable and appropriate stage of difficulty. For example, the Thinking Reader research prototype offers five levels of support in the form of prompts. The prompts change with each level in response to information recorded as students work. This kind of flexibility means that as students become more skillful with reading strategies, they can be challenged to apply those strategies independently.
Figure 7.4 is a composite screen-shot of the five levels of "summarizing" prompts Thinking Reader might provide for the same section of digital text. Let's quickly examine a few:
- The Level 1 prompt provides a high degree of scaffolding; it is appropriate for the student who is just learning the skill and needs to be reminded of which strategy to use and what the key components of that strategy are. At this level, the digital document is an extension of the teacher who is teaching this skill.
- The Level 3 prompt provides a medium degree of scaffolding. It is for the student who has basic facility with the strategy but needs reminders and help self-evaluating.
- The Level 5 prompt provides minimal scaffolding. It is appropriate for learners who know how to use a variety of strategies and how to select among them according to the context, but who may still need reminders to apply strategies consistently.
d
|
|
- Figure 7.4 -
Thinking Reader Prompt Levels
|
|
A third way to increase student engagement with assessment is to vary the content within a particular assessment tool. Standardized tests rarely do this. A test of reading comprehension, for example, is likely
to present the same set of text passages for everyone, not taking into account whether each student will find the passages interesting or worth reading. Sophia, the music lover, Kamla, the basketball enthusiast, and Jamal, the expert on tanks and submarines, might all be assessed on the same passage about Mozart. Sophia would most likely be more attentive to the task than the other students, which would give her the best opportunity to show her actual reading skills. Providing multiple content options in a traditional print environment is costly and impractical. But in a digital environment, there is no reason why Kamla couldn't select a passage about sports for her reading comprehension assessment and Jamal, a passage about submarines, as long as both passages are of comparable difficulty.
Possibilities such as these are just beginning to be explored. The research community has paid less attention to issues surrounding affect than to issues relating to recognition and strategy, but this is slowly changing. Teachers and researchers now realize, for example, that student skills may look very different when working with different content. A student may summarize a passage about the Pilgrims poorly but provide an excellent summary for a passage about race cars. This is important information for guiding teaching. Flexible and ongoing assessment can inform teachers about what most interests their students and help them to enlist students' motivation—the essential engine of learning.
|
