Panel Summaries

Panel On Diverse Student Needs, Research, Tools

Panelists led discussions on (1) relevant research in diverse fields (e.g., learning design for science education, the psychology of reading and writing, and natural language processing) and how it could inform the creation of the next generation of cyberlearning tools; and (2) tools and strategies not specifically designed for people with disabilities that hold promise for addressing diverse student needs.

Panel Members:

• Becky Passonneau, CAP: Advancing Technology and Practice for Learning Reading and Writing Skills in Secondary Science Education
• Prasun Dewan, EAGER: Automatic Classification of Programming Difficulties by Mining Programming Events
• Ina Wanca, Cyberlearning Workforce Readiness

For EAGER, how do you identify and receive notifications of when a user is encountering problems?

• We use cameras, physical sensors, and see when editing is needed in each person’s programming. However, this isn’t always accurate, and sometimes we get a false positive. Not all people show the same patterns, but some certainly do.
• There is a difference between design difficulties and conceptual difficulties. People often stop using a program because they have an issue with the design, while others stop because they’re simply taking a break or struggling with another issue.

We have wanted to incorporate more writing and reflecting into computing classes. However, we don’t always know the accessibility of writing in computing classes or the tools involved. What are some unique accessibility challenges to programs that incorporate writing with programming?

• If you don’t want to penalize students for the time it takes to write or grammar skills, then you have to change the approach. There is a difference between learning to write and writing to learn, and we often need to find a middle ground for these issues. If a program is providing feedback for how someone writes or solves an issue, then we need to make sure the program is designed to look at the solution differently.
• While writing is important, students need to be provided the opportunity to engage and learn in a myriad of methods. Interacting with one another is another way to learn those skills and discuss the information. We’ve often have students grade each other’s work, which can be helpful way to let students see how others are answering the same questions.

How does automatic grading of coding work?

• Grading code is easier than grading writing—writing can be vague, and a human can be needed to search for the answer. Code is much less vague.

A big part of learning is the emotional investment. When we focus on technology, we can forget about the human aspects of students—how can technology be used to see where students are invested, and how is student investment included in your learning goals for computer-graded work?

• When students enjoy learning, that is the optimal situation. Getting frustrated can often disengage students, which happens at higher rates for students with disabilities since the systems often aren’t created accessibly. We are looking into what can frustrate learners and how we can find solutions to unstick students who get stuck. We want all of our students to enjoy learning and feel satisfied with their progress.
• Feedback is often highly individualized; however, a lot of learning can be very general, where people make the same kinds of errors. Computers can detect those issues and provide the support for what is missing from the lesson. While we can’t expect machines to be inspirational for a student to learn, they can allow for a lot of education to occur over a larger scale, especially when done in tandem to human feedback.

What demographics are you looking at when developing your programs?

• I don’t really know the demographics—the field study was just whoever was in the class, and at least one was a female student. We did try to be diverse when we did specific testing.
• It’s hard to know who is all in field studies, especially since those are real classrooms, where we can’t handpick who attends. I am sure there are people and groups being overlooked in the data.

Panel On Assessment of Student Learning and Maximizing Student Engagement

Panelists led discussions on (1) how technology can best adjust instruction to learning challenges and ultimately assess the learning of all students, including those with disabilities, and (2) strategies that can maximize the engagement of students who might not typically engage in cyberlearning.

How does Autism Glass work with people who are very sensitive to touch?

• We’ve worked with a lot of groups to find out what works and feels best to the largest number of people, and where having something customizable works well. The reality is for this product, something will always be on your face.

We know captioning can be expensive—what are all the benefits?

• Captioning is not only an accommodation, but can be helpful for all students. We’re finding ways to make captioning cost less, and have more uses, like downloadable transcripts and searchable text. Furthermore, by teaching students about captioning, those students go out in the world and pass on that knowledge to others and into their own work.

How are you involving teachers in creating a community of shared experiences? How do you design for a diverse group?

• While Number Stories is still in the prototyping phase, we are having teachers write reviews. We are hoping to be able to connect students and teachers in the future as well.
• As we design individual problems for our database, we have to think about both individual students and the commonality of all students. We’re hoping for different versions of the same problem, which would allow students choice in understanding an issue.

Is Number Stories connecting with Common Core?

• We are trying to break free of the word style problems from most textbooks. We want to give problems context so students see how these situations are in the real world. We’ve started with the content we thought was missing from standard curriculum.

If Google Glass is basically discontinued, how do you think future products can stay above the curve for rapidly evolving technology?

• The turn over for technology is crazy. And while yes, the model for Google glass isn’t being created anymore, sometimes you recognize that this piece of technology you’re designing can help influence the next technology. Currently, mobile is the way everyone is interacting with technology. Since technology is so important for people with disabilities, we have to keep this in mind when designing. We can’t just assume people all use technology the same way.

You each engage with diverse learners in different ways; what specifically will your project do to link to diversity and diverse learners?

• We are reaching out to as many users as possible, and we have a diverse set of designers on our team. When designing, we try to ask as diverse a group as possible for input.

• Students with disabilities can often not experience accessibility throughout their education, leaving them disengaged from school. We want to turn this around and show students they can be in control of their own learning by giving them different options and ways of showing their knowledge.

Panel on Lessons Learned and Promising Practices

Panelists led discussions on (1) how specific challenges of students with disabilities can inform all cyberlearning projects, and (2) promising practices for addressing these challenges.

Panel Members:

• Emily Moore, EM: Inclusively Designed Simulations for Diverse Learners (DRK-12)
• Raymond Rose, TxDLA Accessibility Certification Program
• Mike Wojan, University of Michigan Digital Innovation Greenhouse

How do you collaborate with the Disability Resources for Students office to encourage faculty to create accessible classes?

• Providing incentives is great—while we don’t want to encourage the idea that faculty should be paid more for accessibility, we do want to promote the idea that it can require extra planning and work.

How do you make the simulation work and visualizations accessible?

• When we say visualize, there’s often a lot more involved in that. We want to show information succinctly and easily. Students should be able to recognize relationships between different items, and information should be represented in multiple ways so different students can understand it.
• Text based simulators (like the Oregon Trail video game) can also be a helpful tool. Students who are blind can be visual learners as well if text is given to them in a descriptive manner.

How do you know that these designs are working for learners and how do you track and measure results?

• We do a lot of usability testing. We try to get user opinions from a very diverse group of learners. Getting feedback along the way is a lot better than retrofitting at the end of a project.
• Making something accessible doesn’t just mean considering people with disabilities. A program has to be accessible and engaging to people of different races, ages, genders, ethnicities, etc. We always try to bring in a very diverse testing group.
• We prefer a one size fits one, which can be very open ended and customizable per person. With this tool, we will be doing classroom pre- and post-tests and user testing (including screen reader testing).

What is the TxDLA Accessibility Certification Program about?

• It is a twenty hour certificate program with a set of online modules that teach about captioning and a variety of other accessibility issues. There is also a conference hosted to promote accessibility.