ABCDEFGHIJKLMNOPQRSTUVWXYZAA
1
Effort (to create/to test)Impact on users (ref: https://docs.google.com/spreadsheets/d/1sKSXs62C8Q2t5Ek8SrDC6ajBGucBzAVH0ElZpN9KajY/edit#gid=0)Overall Severity of Existing SCBase ScorePrevelance and Quantity/QualityMethods Ranking
Calculation (?)
Final points
2
Low (Multiplyer X1) (e.g. Language of Page - which can always be machine tested alone)Success Criterion 3.1.1 Language of PageFor each user group that is specifically impacted by the requirement = 1 multiplyer. (For example, Headings benefit both non-sighted users and users with cognition issues, thus multiplyer there would be X2. For futher discussion. This also presumes 4 basic groups: visual, auditory, mobility and cognition. We may also want to consider other scenarios, such as input modes - voice input for example)Visual = 1
Auditory = 0
Mobility = 0
Cognition = 1

Total score= 2		This starts with the assumption that not all SC are equal in severity; for example "Language" of page is far less severe than providing captions on videos, or ensuring every target on the page is keyboard accessible. I'm thinking here that we assign specific point values to each existing SC, which are then calculated with the multiplyers. This "value assignment" will require some discussion and thinking, and is (I will suggest) a critical part of our moving existing SC into the new format.All things being equal, I'd give this a 2/10, so the maximum score point = 2 - but then it goes through the multiplyers…4 PointsSome 'failures' can be of one item or requirement failing mulitple times, or a percentage of the time. Examples include number of images on a page, or number of duplicate ID values, etc. This will actually be a subtractive multiplyer: for example if 95% of the images have a good alt text, multiply score by .95, or for each duplicate ID value = minus 10%. More discussion required.)True = 1
False = 0		Another aspect to consider is when there are multiple methods to achieve the goal - do we want to apply a ranking that rewards better methods over less better methods? For example, alt text can be supplied using aria-label, aria-labeledby, alt, caption - with using @alt being the "best" method, followed by either aria method, and then trailing last is caption, which meets the functional need [sorta], but is the less preferred method. Again, for discussion, but for illustration, best method = 10, secondary method = 7, and subsequent methods = 5)This one is very simple, as it is a boolean question. None-the-less, the same algoritymn applies. However, there are still multiple techniques to achieve this SC, and so in my example it is presumed <html lang="EN"> as being the 'preferred' technique.Real Example:

t= 4
u = 1
v1 = 100% (1)
w1 = 10


(4 X 1) X ((1 X 10)) =
4 X ( 10 ) = 40

Final calculation: 4 X 10 = 40		40 (Max Score = 40)
3
Success Criterion 1.3.4 OrientationVisual = 1
Auditory = 0
Mobility = 1
Cognition = 1

Total score= 3		All things being equal, I'd give this a 4/10, so the maximum score point = 4 - but then it goes through the multiplyers…12 PointsTrue = 1
False = 0		This one is very simple, as it is a boolean question. None-the-less, the same algoritymn applies.Real Example:

t= 12
u = 1
v1 = 100% (1)
w1 = 10


(12 X 1) X ((1 X 10)) =
12 X ( 10 ) = 120

Final calculation: 12 X 10 = 120		120 (Max Score = 120)
4
Medium (Multiplyer X2) (e.g. Quality of ALT text or captions/audio description, where to verify requires human intervention)Success Criterion 1.1.1 Non-text ContentVisual = 1
Auditory = 0
Mobility = 0
Cognition = 1

Total score= 2		All things being equal, I'd give this an 8/10, so the maximum score point = 8 - but then it goes through the multiplyers… Medium = 2X32 pointsFor this example, I'll use the 95% of all images have alt text, so the multiplyer here = .95this one can get tricky, but presume that 60% of the images use @ alt, and the other 40% use aria-label.Real Example:

t= 32
u = .95
v1 = 60% (.6)
v2 = 40% (.4)
w1 = 10
w2 = 7

(32 X .95) X ((.6 X 10) + (.4 X 7)) =

Final calculation: 30.4 X 8.8 = 267.52		267.52 (Max score = 320)
5
High (Multiplyer X3) (e.g. the types of cognitive walk-through proposals under discussion, which require a higher effort to evaluate.)(Made Up) Users can save a form that is partially completed, so that they may return to it later. The ability to save any form data is also directly linked to SC2.2.1 Timing Adjustable. (I am attempting to construct a possible requirement that will necessitate an actual attempt at partially filling a form and then trying to save it - not sure if I've captured the idea of cognitive walk throughs accurately, but...Visual = 1
Auditory = 0
Mobility = 1
Cognition = 2*

Total score= 4

(* We might also consider "micro-weighting" if the TF/WG feel that is of benefit...) 		All things being equal, I'd give this an 7/10, so the maximum score point = 7 - but then it goes through the multiplyers… High = 3X84 PointsTrue = 1
False = 0		This one is very simple, as it is a boolean question. None-the-less, the same algoritymn applies. We might want to add an additional variable to this - "date since last tested". We could consider subtracting a percentage of the score based off fixed milestones, for example 3 months old = 10% 'discount', 6 months old = 20% discount, etc. For further discussion.Real Example:

t = 28
u = 1
v1 = 100% (1)
w1 = 10
a = 80% (.8) (Illustration of a 6 month old 'report')

(84 X 1) X ((1 X 10)) =
84 X 10 = 840

Additional calculation:
840 X .8 = 672

Final calculation: (84 X 10)/.8 = 672		672 (Max Score = 840)
6
Based solely on illustrated samples, current maximum score = 1320 Points
Based on current simulated evaluation demo, existing score = 1099.52 Points
1099.52/1320 = .8629 (or 86.29%)
IF Diamond = 100%
AND Gold = 90% - 99%
AND Silver = 80% - 89%
AND Bronze = 70% - 79%

Than current score for this example = Silver Level
7
How I did the calculation:
Base Score = Severity X Impact on Users

IF Base Score = t
AND Prevelamce = u
AND Methods scoring = v (possible that this will generate more than one value, which are expressed as a percentage; those values will need to be added together to represent a total percentage rank not to exceed 100%)
AND Methods Ranking percentage = w (possible that this too will generate more than one value, which are expressed as a percentage; those values will need to be added together to represent a total percentage rank not to exceed 100%)
AND Effort Multiplyer = e
THEN z = Total

[(t X u) X e] X [(v1 X w1) + (v2 X w2) + (? X ?)] = z
8
Best Possible Score (Example based on SC 1.3.4):

t = 12
u = 1
v1 = 100%
w1 = 10

(12 X 1.0) X [(1.0 X 10)] = 120
( e X u ) X [(v1 X w1)] = z		Best Possible Score (Example based on SC 1.1.1):

t = 32
u = 1
v1 = 100%
w1 = 10

(32 X 1.0) X [(1.0 X 10)] = 320
( e X u ) X [(v1 X w1)] = z		Best Possible Score (Example based on Made Up example):

t = 84
u = 1
v1 = 100%
w1 = 10

(84 X 1.0) X [(1.0 X 10)] = 840
( e X u ) X [(v1 X w1)] = z
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100