so after talking about analytic confidence assessments,
is worth expanding a little bit into the idea of words
of estimated probability or wept. This relates to future events. And as we can see, there is a nice scale here, showing
a, uh, a certainty scale
with some very obvious, uh, labels
all the way down to you from 0% of the impossible level
to certainly not, probably not probable warm will start and certain
thes comm provide valuable input. When you're trying to make sure that the conclusion has the right waiting associated with it, you wouldn't want to make it seem like it's a simple yes or no consideration. So Sherman Kent in 1964 developed
this model so we'll have a look at that Wikipedia page,
and we see some examples here of policy and intelligence failures. Related
two words of estimated probability,
and this is gamble. Here we're seeing a mention of bin Laden terrorist attack information or attacks upon and embassies.
So because of the nature of when the information comes out and who says it
who provides it, what confidence or assurances provided by
their their actions and their reputation
that could affect the ability of the consumer of the information to interpret it correctly and be able to use it as as it was intended.
There's also some relevant used for medicine and so on.
But, uh, any of these methods would be useful.
Really. What we see here with with this chart
what I would I would I would term a semi quantitative measuring scale
because you've got words
like probable? Probably not, and so on.
But then there are ranges that are given numeric ranges in this case, percentages
between impossible and almost certainly not. There's, ah 7%
23% between Probably not, and almost certainly not.
So if you're at the lower end of almost certainly not, you can say OK, we want to give or take about 5% at this level.
As you get to the larger differences between the middle levels than the plus or minus, swing goes up to 10%
up all the way up to 12% at the probable level.
This way, there's a little bit of extra leeway for saying that this is the proper label.
if you were at 50%. And you thought, you know, plus reminds 10%. Maybe you're thinking plus 10%. So you might be higher at 60.
put you in the range of probable
because probable could go, You know, from 75 down to 63. So soon as you get a little bit above 60. Now, you're in the probable range to see how this work. It's a little bit of wiggle room
to make sure that your estimate or your words of estimated probability are not misinterpreted.
And having a guide like this to show the percentage of certainty would certainly be helpful to the intelligence consumer.
All right, that gets us to the end of module three.
I hope you'll stay tuned for Model four. Thank you.