DIAGRAM Survey on Teaching with Accessible Image Description

DIAGRAM Survey on Teaching with Accessible Image Description

Report of Findings

Date: March 15, 2013

Respectfully Submitted By: Yue-Ting Siu, TVI & Educational Consultant; UC Berkeley

Other contributing project members:

• Anh Bui, Benetech
• Lore Schindler, Los Angeles Unified School District
• Laurie Vasquez, Santa Barbara Community College
• Lucia Hasty, Rocky Mountain Braille Associates

Executive summary

The DIAGRAM center was established tomake it faster, easier and more cost effective to create and use accessible digital images for students with print disabilities. This work focuses on two broad areas: research and development; and implementation and production. In order to support this mission, we first needed to determine the current climate of use of, and need foraccessible descriptions in the educational environment. We focused first on image descriptions as part of secondary and post-secondary education.

This survey gathered information from personnel who are tasked with providing accessible materials to students with visual impairments. The purpose was to illustrate how and if image description is currently handled in schools, and to identify needs in the field. The results should be considered to inform development and refinement of tools for describers (such as Poet), and refine the scope of future training programs for improved implementation.

All of the respondents preferred embedded image descriptions in textbooks as long as they were of high quality, and appropriate for the subject matter. The use of the term “embedded” refers to image descriptions that are within the digital framework of a textbook. A recurring theme in responses reflected priorities for flexibility in meeting different students’ needs, and to account for the range of descriptions from those that did not need to be described, to those that required supplementary tactile representation. All of the respondents attributed the benefit of image description to enhancing students’ independent learning.

Methods

Participants

A total of 59people responded to an online survey using SurveyGizmo. The survey was disseminated through Benetech’snetworks, and included respondents from 23 of the United States, Canada, and Norway. Data were gathered throughout the fall and winter of 2012. Respondents included personnel working in k-12 (37 people), postsecondary programs (15 people), other educational programs (i.e. in a museum or an alt. media designer), and assistive technologists (5 people). Most of the respondents reported serving less than 10 students who required accessible educational materials in a typical semester. Four of the respondents served 16-20 students per semester, and five of the respondents served over 20 students who needed accessible educational materials.

The survey consisted of eight questions,and was purposefully brief with the goal of providing a snapshot of needs in the field to target relevant training programs in the use of Poet. The following questions were included in the survey:

1. In a typical semester (fall or spring), how many students do you serve (on average) who require accessible educational materials?
2. In what capacity do you teach or provide alternate materials?
3. What subject matters do you typically teach or provide alternate materials for?
4. Currently, how do you typically represent visual images in educational materials to students who are visually impaired?
5. Of all the materials you need to adapt for accessibility, how often do you feel the need to describe images?
6. Do you think students who are visually impaired could benefit from having images in course materials described so they may independently access the descriptions on their own?
7. Do you think descriptions of images in a digital textbook can enhance student learning?
8. To provide better accessible materials to students who are visually impaired, how could image description be most effectively used?

Answers were provided as a multiple-choice response, with an option to fill in a short answer for most questions. Please refer to Appendix 1 for a copy of the survey.

Results

Providing alternate materials for different subject matter (question #3)

Over half of the respondents reported teaching or providing alternate materials for the following subject matters: Physical/Earth/Biological/Life Sciences, Math, History/Social Sciences, and Language and Literature. Less than 30% of the respondents reporting teaching or providing alternate materials for engineering/computer science, and visual/performing arts.Other subjects that some respondents reported using alternate materials for included: web accessibility, health, vocational, business education, and tests. Since respondents may have had differing caseloads with varying student needs, these subjects happened to be those that students were currently enrolled in, and should not be interpreted to mean that certain subjects should be necessarily prioritized over others for providing alternate materials.

Current methods of representing visual images in educational materials (question #4)

The respondents were asked to rank the methods they used most often to represent visual images in educational materials. “1” represented the most used method, “2” for the next most used method, and so on with “5” representing the least used method. “n/a” denotes the number of respondents who did not rank the method at all. The following four bar charts summarize howeach method was ranked according to how often it was used. The data are aggregated in a single table.

Table of bar chart data, summarizing how respondents ranked how often they used each method:

The method “images described verbally and informally by a teacher/peer/auxiliary aide as the material is covered” was ranked number “1” by the largest proportion of respondents (22). The method least used was “image descriptions embedded into a digital textbook” (22). It should be noted that of all the methods suggested, the least used method is one that currently cannot be manipulated by the respondent. All the other methods rely on instructional materials that may be constructed by anyone.At the time of this survey, embedded image descriptions in textbooks were dependent on the availability of packaged descriptions that accompanied a digital text.

Time spent describing images (question #5)

When asked how often the respondents described images given all the materials they needed to adapt, the majority (66.1%) reported spending less than half of their time describing images. In contrast, only 13.6% of the respondents reported spending most of their time describing images.

Different images serve different functions, and these statistics may be affected by the subject matter, and whether or not the instructional materials necessitated description. Some images do not require description at all, and others are better suited with a tactile representation. The data here should therefore be interpreted with caution, as it does not necessarily correlate to whether or not image descriptions are valued in the field, and cannot support any definitive conclusions regarding respondents’ needs.

Perceived student benefits of image descriptions (question #6)

The majority of respondents (64%) reported that image descriptions would indeed benefit their students. Of these respondents, roughly one-third expressed wanting the option to edit/annotate given descriptions to better suit their students. 36% of the respondents were more cautious, and could perceive image descriptions benefitting their students only if they were of high quality, and appropriate. Given the option of having images in course materials already described, none of the respondents preferred having a teacher/peer/auxiliary aide describe the image for the student as it is covered in class.

These responses indicate an overall positive attitude toward use of image descriptions with students, and reflect the need for appropriate, high quality descriptions. Since none of the respondents preferred having another person describe an image for a student, a possible benefit of having ready image descriptions in course materials may be that the student who is visually impaired can achieve another level of independence in the learning process.

Enhancing student learning with image descriptions in digital textbooks (question #7)

A majority of the respondents (81.4%) believed image descriptions in digital textbooks could enhance student learning. No respondents thought otherwise. Of those who reported “sometimes”, they commented that image descriptions would only enhance student learning if the information was appropriate for the student’s learning level, and if the description itself was appropriate for the image. In other words, for some images where description would not sufficiently convey the information, it would be important to have the image represented otherwise such as in tactile format. Some of these respondents expressed concerns that more complex images would require tactile representation in addition to the description. Other respondents maintained that a description would only enhance student learning if the image was relevant/important to the course material. One respondent expressed a need to turn image descriptions on or off according to the students’ need and learning style.

Most effective use of image description (question #8)

The respondents were asked to rank a variety of applications of image description, with “1” being the most effective method, “2” for the next most effective method, and so on. “n/a” represents methods that were not ranked at all. Those who chose “other” all expressed concerns that description was not always the most appropriate given an image. These respondents stated that alternatives should be considered based on a student’s preference, and whether or not an image was better conveyed via other media, or in conjunction with other media such as tactile, if at all. The method that ranked highest in being the most effective was “Image descriptions paired with a tactile representation”, followed by “Image descriptions embedded into a digital textbook”. The method that ranked lowest in effectiveness was “Images represented tactually”, followed by “Images described verbally and informally by a teacher/peer/auxiliary aide as the material is covered”. The method “Images described verbally and informally by a teacher/peer/auxiliary aide as the material is covered” was also noted in question #6 as the least preferred method that would benefit students.The following five bar charts summarize how each method was ranked according to the perceived effectiveness of each method. The data are aggregated in a single table.

Table of bar chart data, summarizing how respondents ranked the effectiveness of each method:

Wrap-up

The survey respondents all served students with visual impairments in a variety of educational capacities who were studying a range of subjects. The use of alternate materials for various subjects likely differed due to the range of students served and their needs. Overall, respondents felt that image descriptions could benefit their students, but the potential benefit hinged upon the quality and appropriateness of description for a given image. Respondents ranked highest methods and approaches that allowed for flexibility in tailoring accessibility to images depending on student needs, and how/if information from the image needed to be conveyed. All of the respondents agreed that image descriptions in digital textbooks would enhance student learning, and everyone preferred this method to having the student depend on another person to describe images as they came up in the classroom. Although the highest proportion of respondents ranked the most frequently used method as “images described verbally by a teacher/peer/auxiliary aide as the material is covered” (question #4), none of the respondents thought this method was the best for having students independently access course materials (#6). All of the respondents preferred embedded image descriptions in textbooks as long as they were of high quality, and appropriate for the subject matter.

Limitations

Since the survey was disseminated through Benetech’s networks, it may have been biased toward people who are already connected to uses of alternate media and assistive technology. Therefore, this sample of respondents may be more open to the sort of technology that image descriptions in digital textbooks require. Although respondents represented 23 of the United States (55 respondents), Canada (3 respondents), and Norway (1 respondent), more specific information regarding the demographics of their teaching environment was not gathered. It is unknown within what kind of infrastructure they served students, and what kind of resources they had access to in providing alternate media.Other personnel not included in this survey may have varying access to different types of infrastructure and resources that affect use of alternate media. Information from this survey provides an anecdotal snapshot of the population sampled rather than support statistical conclusions that can be directly applied to the general population. Respondents may not have been truly representative of all personnel who work with students with visual impairments due to possible sampling bias.

Thoughts and considerations for future directions

1. Image descriptions are relevant for a wide variety of subjects, and providing descriptions across different subjects depend on students’ needs. Efforts to expand the use of image descriptions should encompass a breadth of subject matters, and not be limited to the ones reported here.
2. Respondents preferred students use image descriptions embedded in digital textbooks rather than rely on another person to describe an image. Having embedded image descriptions in course materials were perceived to increase overall independence in student learning. Efforts to provide image descriptions in textbooks should therefore be viewed as a way to improve independent learning in students who are visually impaired.
3. Image descriptions enhance student learning if they are appropriate to the content material, and to the type of representation needed
4. Some respondents noted that if descriptions were not necessary to the subject matter, they would rather have an option to “turn descriptions off”. Image descriptions not relevant to conveying content material should be avoided, or else risk distracting the student from learning goals, and discouraging educators from using them at all.
5. Some respondents reported a need for tactile representations in addition to image description depending on the complexity and/or type of image. Image description is not an all-encompassing panacea for accessible educational materials. Tools should allow for flexibility to account for instances where images are not optimally represented by description, and may need additional tactile representation.
6. Some respondents reported that different students might benefit from more or less information regarding an image. They noted that flexibility in adjusting descriptions to student needs would be very important.Relevant use of image description is therefore dependent on student learning needs, and tools should allow for editing descriptions as necessary to suit the student.
7. Although all respondents preferred some use of embedded image descriptions in textbooks to having images described by another person, not all of the respondents reported using embedded image descriptions as the most often usedmethod. This indicates that there is indeed a need in the field to expand the availability of embedded image descriptions in textbooks, and personnel working with students with visual impairments would prefer this method to having students lose independence in the learning process.
8. This survey did not ask who these respondents thought would be best qualified to provide descriptions, nor did it ask how the respondents would ideally like to be involved in producing image descriptions. Future research, development, implementation, and production should consider:
9. How to standardize descriptions so they adhere to guidelines and ensure the highest quality possible?
10. How to deliver tactile representations to supplement descriptions for those images that require both?
11. How to limit the number of images described so that only those that are relevant to conveying content are described?
12. How to deliver image descriptions within digital textbooks using pre-existing infrastructures/resources that are accessible to personnel and students?
13. How to build enough flexibility into image description tools to meet student needs?
14. How to design image description tools and train personnel in using them so that they may incorporate this method into their teaching toolkit?

As always, comments, feedback, and ongoing discussions are welcome and appreciated.