The AIM Score is a combination of your Automated Accessibility Score and Manual Testing Impact Score. Let’s get into exactly how these scores are generated with examples.
Learn more about how your organization can use the AIM score.
Automated Accessibility Score
We use the WAVE API, a trusted accessibility testing tool, to find WCAG 2.1 issues.
We use the following measurements and weights to calculate the Automated Accessibility Score:
- 60% average page error count points
- 30% average page error density points
- 10% average page alert count points
These measurements are compared to the most recent WebAIM Million results. Then, the score is normalized from 1-10.
How normalization works
The amount of errors, amount of potential errors, and density of errors are normalized by aligning the results with deciles from the WebAIM Million dataset. A site is assigned 1 point per decile alignment.
The decile ranges are:
- 1st decile (10% of the 1M pages align with these values)
- 0 – 2 errors
- 0 – .0049 error density
- 0 – 2 alerts
- 2nd decile (the next 10% of the 1M pages align with these values)
- 3 – 5 errors
- .005 – .0124 error density
- 3 – 5 alerts
- 3rd decile
- 6 – 10 errors
- .0125 – .0204 error density
- 6 – 10 alerts
- 4th decile
- 11 – 17 errors
- .0205 – .0284 error density
- 11 – 17 alerts
- 5th decile
- 18 – 25 errors
- .0285 – .0391 error density
- 18 – 25 alerts
- 6th decile
- 26 – 37 errors
- .0392 – .0527
- 26 – 35 alerts
- 7th decile
- 38 – 53 errors
- .0528 – .0701 error density
- 36 – 49 alerts
- 8th decile
- 54 – 80 errors
- .0702 – .095 error density
- 50 – 77 alerts
- 9th decile
- 81 – 142 errors
- .0951 – .1394 error density
- 78 – 139 alerts
- 10th decile
- 143+ errors
- .1395+ error density
- 140+ alerts
Example
If a website has an average of 22 errors (5th decile), an average error density of .0212 (4th decile), and an average of 14 alerts (4th decile), it would get the following points:
- 5 for errors
- 4 for error density
- 4 for alerts
Next, we apply the weightings to get the Automated Accessibility Score: 11 – ((5 X .6) + (4 X .3) + (4 X .1)) = 6.4.
The Automated Accessibility Score for this example would be 6.4.
Manual Testing Impact Score
The Manual Testing Impact Score is four pages scored on 10 accessibility strategies. Each accessibility strategy gets a score from 1-10. The 10 strategies are:
- Accuracy of the document’s defined language
- Image alternative text
- Empty links and buttons
- Labeled or unlabeled form inputs
- Low contrast content
- Page title
- Animation and movement
- Keyboard focus indicators
- Keyboard accessibility
- Page reflow/responsiveness
This is not a complete list of possible accessibility issues, but it is some of the most common and impactful. The Manual Accessibility Impact Score gives a meaningful measure of how these issues impact users with disabilities on the four selected pages.
Next, the page’s overall accessibility is scored from 1-10.
Pope Tech then averages each of the four pages’ scores for the 10 different accessibility strategies (the accessibility strategies average).
Next, Pope Tech takes each of the four page’s overall accessibility score and averages it with the accessibility strategies average for that page.
Now, we have four scores – each page has a score that includes the 10 accessibility strategies average and overall accessibility score.
Lastly, Pope Tech averages the scores for each of the four pages to get the final Manual Testing Impact Score.
Example
Here are 10 accessibility strategy scores for four pages with their average:
- (7+3+6+4+4+2+7+5+5+3) / 10 = 4.6
- (4+5+8+7+9+3+6+7+6+5) /10 = 6
- (8+4+3+7+9+6+5+7+8+4) / 10 = 6.1
- (8+7+6+4+7+8+4+6+7+4) / 10 = 6.1
Next, we’ll average the page’s overall accessibility score with the accessibility strategy averages.
- (4+4.6) / 2 = 4.3
- (6+6) / 2 = 6
- (6+6.1) / 2 = 6.05
- (5+6.1) / 2 = 5.55
To get the Manual Testing Impact Score, we’ll average these four scores: (4.3+6+6.05+5.55) / 4 = 5.475
AIM Score
We now have the Automated Accessibility and Manual Testing Impact scores. These are averaged with a 50% / 50% weighting to get the AIM Score.
Example
Since the Automated Accessibility and Manual Testing Impact scores are weighted equally, we’d get the AIM score by finding the average of the two scores: (6.4+5.475) / 2 = 5.94.
The AIM Score for this example would be 5.94.