Table G.1. The probability that classified information
about a classified project will reach an adversary through unauthorized
disclosures of that information, by persons working on that classified
project, for several scenarios S
P(S) = kS H NPS H NCOMS*
H COPS H CREVS H UARS H RTAS
|
|
|
Variables in the probability equation |
|
|||||
|
Scenario, S |
Constant in probability equation |
Number of persons knowing classified information,
NP |
Average number of communications by each person,
NCOM |
Communication's classifica-tion review factor, CREV |
Average number of copies made of each communication,
COP |
Probability that unauthorized recipient will get
communication, UAR |
Probability that recipient will transmit
information to adversary, RTA |
Overall probability of classified information
reaching adversary |
|
Deliberate
unauthorized disclosure |
k1,
disloyalty constant, about 10-5 |
NPS |
1 |
1 |
1 |
1 |
1 |
10-5 H NPS |
|
Inadvertent
direct unauthorized written disclosure |
k2(w),
carelessness constant, about 10-4 |
NPS |
NCOMS |
10-2
to 10-5 |
COPS |
1 |
10-2
(targeting) to 10-5 (random) |
10-8
to 10-14 H NPS H NCOMS
H COPS |
|
Inadvertent
direct unauthorized oral disclosure |
k2(o),
carelessness constant, about 10-3 |
NPS |
NCOMS |
10-1
to 10-2 |
COPS |
1 |
10-2
(targeting) to 10-5 (random) |
10-6
to 10-10 H NPS H NCOMS
H COPS |
|
Inadvertent
indirect unauthorized written disclosure |
k3(w),
carelessness constant, about 10-3 |
NPS |
NCOMS |
1 to 10-2 |
COPS |
10-2
to 10-3 |
10-1
(targeting) to 10-5 (random) |
10-6
to 10-13 H NPS H NCOMS
H COPS |
|
Inadvertent
indirect unauthorized oral disclosure |
k3(o),
talkativeness constant, about 10-2 |
kI H NPS(NPS
B 1)/2GS (footnote a) |
1 |
1 |
10-1
to 10-5 |
10-2
(targeting) to 10-5 (random) |
10-5
to 10-12 H kI H NPS(NPS
B 1)/2GS |
|
|
aFor the
inadvertent indirect unauthorized oral disclosure scenario, the total number
of communications, NP H NCOM, can be estimated by
considering the number of interactions I between the persons working
on the classified project, NP.
For this scenario, the total number of communications NCOM(total)
can be estimated by NCOM(total) = kI H I = kI
H NP(NP B 1)/2G, where k1
is a factor that takes into account that when NP is large then the
number of probable interactions is different from the theoretical number of
interactions and where G is a compartmentalization factor. |
||||||||