Beginning February 2018, Tenon began performing long-term research into patterns of accessibility on the web. Researching 0 unique domains (so far), we've gathered information on 0 technologies. We have categorized each technology into 0 distinct categories and 0 distinct vertical markets. The goal of this research is to determine what, if any, patterns exist in the data that can be used to predict the level of accessibility when those technologies are in use. We hope to answer questions like:
- What technologies are correlated with the most/ least accessibility issues?
- What type(s) of accessibility issues are correlated with specific technologies?
- How does each technology perform against others in the same category?
- Which vertical markets have the highest rates of error?
One of the more significant challenges in this type of research - beyond What can be tested and how - is that correlation does not imply causation. The existence of patterns that are correlated to specific error rates for a technology is not necessarily an indictment of that technology. Issues correlated to a specific technology may reside in its version, implementation, or configuration. We hope that ongoing data collection will allow us to gather enough data that the influences of such variations can be reduced.
This data gathering is a permanent ongoing process happening 24-7-365 using our own data sources as well as the logs from users who have consented to allowing us to gather anonymous stats.
This data is not yet ready to be used to draw any inferences. At the time of this writing there are approximately 1.3 billion websites on the Web. With that sample size, we would need to analyze 16,641 unique domains to have a 99% confidence level and confidence interval of 1. Given the extreme amount of variability in the data, we will not consider this data to be useful until we have analyzed at least 50,000 domains. Until then, regard this information as unreliable.
|Total||This Month||This Week|
|Total Logs Analyzed||29767||61904||8341|
|Total Distinct Domains||0||0||0|
|Total Distinct Technologies||0||0||0|
|Total Distinct Categories||0||0||0|
|Total Distinct Verticals||0||0||0|
|Average Alexa Rank||0||0||0|
|Average Quantcast Rank||0||0||0|
|Average errors per page||110||67||7|
|Average warnings per page||35||22||2|
|Average density per page||16||10||1|
|Average tests failing per page||23||14||2|
|Average level A errors per page||104||63||7|
|Average level AA errors per page||40||26||2|
|Average level AAA errors per page||50||29||3|
|Percent of pages error-free||0.02||0||0.02|
The overall goal of this ongoing research is to determine the role that certain technologies play in accessibility. Right now we are gathering data on 0 technologies. That number isn't expected to increase too significantly, however we also track their popularity. Each time we see the technology we are also increasing the sample size.
Although we have an index of all technologies, many of the technologies we track are unlikely to have a direct accessibility impact. We find the following technologies particularly interesting, from an accessibility perspective.