CSE 590w #1: Why is data on disability so hard to collect and understand?
Talk by Richard Ladner, Professor Emeritus in UW CSE
This article is the first of a series. Next: 590w #2
As I mentioned a few articles ago, all five of my classes this quarter are either directly design-related, or else cousins to design, so I’ve been slow to decide which classwork I wanted to upload to the blog. My initial intent was for my blog to track my supplementary learning, but as time has passed, I’ve began to believe in its value as a public document detailing my academic progress. This started last quarter as my HCDE 210 class had writing requirements that could be met by blogging, and has continued this quarter with INFO 200’s writing requirements. Right now, I’m not extending my scope to ALL of my academic work, but in particular, this class fits the bill.
CSE 590w is a 1-credit seminar on accessible technologies. Many weeks, class consists of a conference-style presentation of related research, a format which looks exactly like many of the guest lectures I attended and wrote about last quarter. For this reason, I will include my notes for presentations in this class on the blog moving forward, though not every week will see a new presentation. For example, next week promises “Small Group Discussions” with more details forthcoming. I believe the website should be publicly available here: https://courses.cs.washington.edu/courses/cse590w/20sp/.
Our first presentation was given by Richard Ladner, Professor Emeritus in CSE, in week 2. His topic was very practical and interesting to me. Notes are below:
I believe that the paper is protected to UW netIDs only, so I’ll link it here. If you’re reading this and don’t have a netID, it’s probably possible to find another version somewhere on the internet by searching the title. (If anyone believes it is inappropriate to list this here, let me know and I’ll remove it)
This paper caught my interest for two particular reasons. I intended to be a psych major for about a semester sometime earlier in college, and one of the big takeaways I had from my Research Methods class, and something I’ve generally learned over a few years of projects, is that how you word survey questions has a massive impact on survey outcomes. This turned out to be a big issue in this paper, and it was exacerbated by the sensitive nature of the topic. This was the second major point to me, and something I identified with. The paper mentions specific language that was used by survey writers, designed to be socially/politically correct, that ended up completely skewing the results of surveys. The paper showed a super interesting graph that I don’t want to take a screenshot of since the paper is somewhat privately accessible, but I can explain the gist of it. The graph shows the percentage of doctorate students with a disability YoY, and the percentage is largely stagnant for years at a time until taking two major jumps from 2009–2010 and 2011–2012. It’s unlikely that the proportion doubled overnight — the explanation here is that the way questions of disability were worded in collecting demographic data changed, and it completely changed how many people self-reported as having some form of disability. This complication is super interesting to me, because as an outsider to many communities, I am always conscious of choosing the right terminology and phrasing, since I don’t naturally have a good understanding of the right terms. I think this is true for all people and the groups to which they don’t belong, but it’s a very important point to hold in mind moving forward in this class and through research in general.
