Digital Accessibility in Computer Science

My problem statement is the following.

Context: Over 1.3 billion individuals have significant disabilities, and according to the WebAIM 2023 survey, a staggering 96.3% of the world’s top one million websites are not fully accessible (similar state with mobile applications). In an era dominated by digital interactions, ensuring accessibility for ALL users, including those with disabilities, is imperative. Software engineers often attribute the lack of education on accessibility as a contributing factor to the prevailing issue of inaccessibility in digital artifacts. The CS curricula hardly covers topics on accessibility.

Question: How can we effectively teach digital accessibility skills (including Web content accessibility guidelines - WCAG) to computing students and software professionals ?

Problems:
(1) Instilling the “WHY” Component - Educating on accessibility goes beyond technical guidelines; it encompasses understanding the significance of accessibility, awareness of diverse disabilities, familiarity with assistive technologies, and fostering sensitivity to the subject. As computer science instructors, these aspects may lie beyond our typical expertise, leading some educators to advocate for their inclusion in K12 education.
(2) Effectively Conveying the “HOW” Aspect - The Web Content Accessibility Guidelines (WCAG), produced by W3C, is a technically detailed document that may not be beginner-friendly. Instructional materials for teaching accessibility are not readily available, adding to the challenge.
(3) Pedagogical Considerations - Should accessibility be presented as a standalone course? If so, should it be an elective? Alternatively, should it be integrated into existing courses such as web development or mobile development? Moreover, do instructors possess the necessary skills to cover these topics effectively?
(4) Exploring Game-Based Learning - I have tried approaches mentioned in (3) in academia and have found some success but for software professionals, having classes do not work effectively. Currently, my focus involves developing “Serious games” aimed at educating participants on various aspects of digital accessibility. The effectiveness of this approach remains uncertain (which is a part of my PhD to evaluate the effectiveness of games to teach accessibility).

Constraints:
(1) The interest and expertise of computer science faculty in this domain are currently insufficient, indicating a need for training. The scaling of teacher training initiatives in this area requires increased attention from larger organizations such as IEEE, ACM, AICTE, UGC, and others (not just at an institute level).
(2) Software engineers perceive digital accessibility as an additional burden amidst their demanding schedules. A change in mindset is essential. This issue is not primarily a complex technical challenge but rather a significant societal problem. I am actively exploring effective approaches to raise awareness and sensitize individuals to this critical area.

2 Likes

This is my PhD area and I would love to chat more about it and get your feedback during our meet. Though this is my main focus, I am interested in the below problems as well:

  • Restructuring Computer Science Education for Cultural Inclusivity - Many universities in India currently adopt same textbooks, pedagogies, and even presentation slides from Western universities :sweat_smile:. I believe that this may not be the most effective approach. What proves successful at, for instance, at an University in the US may not necessarily suit the context of a Tier 3 college in India. India faces distinct challenges, including linguistic barriers, prerequisites for a computer science program, and motivations for pursuing such a program. I propose the exploration of new, locally grounded pedagogies. (For instance, consider utilizing indigenous narratives such as the Ramayana to elucidate various computing principles. While this is merely a conceptual idea, extensive research and experimentation are necessary in this domain.)
  • Effective assessments - Many institutions (including state and private universities as well as tier 2/3 colleges), employ ineffective assessment methods. I will share specific examples of question papers during our meeting. It is concerning that a student can successfully complete a programming course without actually writing a single line of code (Yes, I am not kidding!). This poses a significant issue as students might falsely believe they are making progress when, in reality, they are not receiving the necessary training. Drastic reforms are needed in this regard (but of course step by step with more research on the underlying causes - are teachers not equipped to set effective assessments ? or is it to decrease dropouts from CS programmes ?)
  • Gender in CS - The status of women in Computer Science (CS) or STEM fields in India is noteworthy. Based on my observations, the participation of women in CS/STEM courses within higher education institutions in India is relatively low, and in certain institutes and programs (such as those at BITS Goa), it drops to less than 10%. I am eager to delve deeper into this phenomenon and uncover its underlying causes. If the number of applications is substantial but enrollment remains low, there may be research questions surrounding the root causes of this discrepancy. Conversely, if the applications themselves are limited, it indicates a systemic bias that necessitates attention and rectification at the K-12 level.

I am interested in the above problems and would love to chat more about these in our workshop and collaborate with folks who are interested in these problems.

3 Likes

This is a great, extremely important topic. I have done some work in accessibility so I am especially interested in this, and would be delighted to learn more about it from you. Definitely an important problem to work on.

3 Likes

These are also important questions. However, I worry that you may be spreading yourself a bit thin here. It’s fine to try to make parallel progress on these, but at some point you may have to drop some to make progress on the others.

1 Like

Thank you! Looking forward to it.

2 Likes

Great point. I am interested in these problems but have not invested any time on working on these yet. I would love to hear from you and others on these in the workshop but will work on them later (after the accessibility problem - which is my current focus).

2 Likes

My experience has been different, so maybe we should figure out the data about it too. In Tier 2/3 colleges, I have seen a lot of girl students in class (much more than 10%), approaching 30-40%, and anecdotally I have been told by people that there are a lot of girl students in these CS colleges. However, when I look at the distribution of freshers in product companies (which may not be a good yardstick), it is very skewed. My feeling is that our interview process focuses (unfairly) on evaluating skills that girl students may not be as strong for various reasons, and hence the skew starts.

1 Like

Interesting! What about Tier 1? What is the situation there at IIIT-H ? Does having entrance examinations play a role here ? I was discussing with a retired IIT-D professor and he had this hypothesis that any entrance exam in India, mostly results are dominated by Male (which in turn is connected to the accessibility and commercials of Coaching classes. Also, the number of Male students who take a year drop just to prepare for entrances are way larger than Female students). While I’m not actively working on this at the moment, I’m sharing it here to spark a conversation around the topic (with potential plans to explore it later) :stuck_out_tongue: Industry does take measures but as you said its very skewed. There are some interesting cases where some companies wanted to have a better D&I and have hired (unfairly) and that causes lot more trouble later within the organization. As Prof Shriram pointed out, addressing this issue requires focused effort (and, to be fair, we should not treat it as a side project). Let’s complete our current tasks and collaborate with like-minded individuals to actively work on these :+1:

Tier 1 situation continues to be pathetic, though I have heard it has improved a bit in recent years because some emphasis has been given. IIITH UG has 15-20% girl students (purely my guess based on campus experience), it may be better than other IITs and NITs slightly because it has many other routes of admissions which are quite diverse (including one through Class XII marks), small numbers but counts.

I agree with the coaching models skewing the mix… not only gender diversity, but also economic and social diversity. But it is a complex area to talk about and I don’t know enough to talk about it, haven’t seen much data either.

I feel CS has better chances of producing a diverse workforce than other engineering disciplines in India (but then other disciplines are getting decimated by CS/IT!).

1 Like

The moment you have any kind of exam, you automatically select for people who are likely to do well on that kind of exam and against people who will do poorly on it. For instance, if your exam is in English, you advantage English-medium students and disadvantage Hindi- or Kannada-mediums. Etc., etc.

In addition, something that may not be obvious: the very act of having an exam at all means you select for people who do well on exams.

I was a very good exam-taker. I actually liked the pressure of exams. But I also realized that there is no real value to exams in my courses. I want students to be able to demonstrate their understanding in authentic situations, and there are no real-world situations that are like exams. So I got rid of exams completely in my courses exactly 20 years ago.

We can talk about these things during “off hours” at CRiCKET. (-:

3 Likes

@shriram : Logic Model for my work.