A period that featured standard deviations which were just small fractions of their averages.
For instance:
Dismissals averaged 13%; with just a 6% σ (ie, µ/σ ≈ ½)
Losses averaged 33%; with just a 6% σ (ie, µ/σ ≈ 1/5)
Settlement rates averaged 29%; with just a 4% σ (ie, µ/σ ≈ 1/7)
During that period, the rate of settlement has been 29%.
During that same timespan, the rate of losses has been 33%.
TBD contends that this is a function of a lack of data (ie, not a function of changed realities).
Between 1979 and 1999, there were only an average of 29 case dispositions per year - due to scarce amounts of digitized data (see disclaimer).
Since 2000, though, there have been an average of 168 case dispositions per year (see “Totals” graph).
Note #1: In 2000, the graph shows a large spike in discrimination charges. TBD contends that this is not a function of increased discrimination (at that timepoint). Rather, it is a reflection of increased data (borne out of technological advances in public recordkeeping).
Additional: These numbers can be further analyzed by:
(a) case type;
(b) charge (eg, age, col, etc.);
(c) determination impact; (ie, 'cause' vs 'no cause')
(d) legal representation; (ie, 'with attorney' vs 'without attorney')
(e) retaliation impact (ie, 'with' vs 'without' [retaliation]); and
(f) the plaintiff's demographics (eg, age, dis, etc.).
