Sajay Arthanat and I continue organizing the available research evidence on text entry rates (typing speeds) for people with disabilities. I shared an overview of the findings in an earlier post. Here, I’ve added two new studies to the dataset and created an infographic describing the distribution of text entry rate across 177 individuals.
Want to share this infographic?
Feel free to download a one-page PDF of this infographic, suitable for printing or sharing.
To publish it on your own site, copy and paste this code into your blog post or web page:
Some key points
The lowest speed in the dataset was 0.5 words per minute. That’s 2 or 3 letters in a minute, illustrated in the infographic by the word “Hi”. Imagine the challenge in communicating, writing, doing schoolwork, or texting at that speed. And this was not an isolated case. 17 people in the dataset had a text entry rate below 1 wpm. That’s about 10% of the total dataset of 177 people.
Looking at the mode and the median statistics tells a similarly sobering story: current approaches to computer accessibility may not be optimally addressing users’ needs. The most common text entry rate was only 1.5 wpm. This may partially reflect the way researchers select their study participants; perhaps individuals with the most severe motor impairments are more likely to be identified and invited for studies of this kind. But even if that is the case, the conclusion is still valid for that population.
As a field, assistive technology has provided access but perhaps fallen short on performance, at least for many people. Having access at all does open doors, but going through those doors and beyond may require more productive levels of performance. While it’s unclear exactly how to enhance performance across all individuals and scenarios, a good first step is to measure and understand what the performance actually is.
Where the data came from
Our published papers on the effect of interface and the effect of user characteristics give a full description of how we conducted the systematic review of the literature. Briefly, we searched 10 databases and included studies in which: typing speed was reported in words per minute or equivalent; the access interface was available tor public use; and individuals with physical disabilities were in the study population. This yielded 39 studies for our initial systematic review, and we’ve since added 3 more studies published since 2015.
To get the 177 data points in this dataset, we pulled the individual subject data out of all studies where it was provided. For example, if a study involved 5 subjects, and reported the specific text entry rate for each subject individually, that yielded 5 data points for our dataset. Of the 42 studies that met criteria, 35 of them provided separate data for each subject. Data from those 35 studies were pooled to yield 177 data points, each representing a unique individual.
In a future post, we’ll look at the actual histogram of this distribution and delve into things a little bit deeper.
See more information at kpronline.com/ter-review.
Search the data yourself at kpronline.com/atnode.
What questions do you have about these data? How might this information affect what you do?