Originally posted by atlasastroYes, but misleading. atlasastro up to his same old tricks....
And on the statistical methedology [of SR14] ... "... on page 76 of Report #14 ... it is stated that the Chi Square Test neither confirms nor denies that the Unknowns are primarily unidentified Knowns".
He forgets to mention the following primary conclusion regarding the core data, reported in the very first sentence after the main chi square results, before inferences and speculation are piled on:
"In five of the six cases, the probability is less than 1 percent that the distributions [Known vs. Unknown] are the same." (p.68, emphasis mine)
That is BIG.
If the distributions of Knowns and Unknowns are not the same ... they're different. Statistically different, reliable to at least the 99% significance level. Basic statistical theory says this means Unknowns therefore do not differ from Knowns due to chance; they differ for reasons other than chance. And all that's left after chance is excluded is that Unknowns differ from Knowns because they are DIFFERENT.
We cannot assign a reason for that difference. It does not mean Unknowns = Flying Saucers. However, as soon as the sentence "the probability is less than 1 percent that the distributions the same" is written, the skeptics' argument that (Unknown) = (Known minus sufficient data) has the rug pulled out from under it.
That argument would be proper only if SR14's basic data found the two distributions to be the same. At least closer. But the core data only shows that Knowns and Unknowns are statistically different; NOT that Unknowns are likely to be, or could even reasonably be considered just hidden Knowns. Not even close. Even ambiguous results (and calling <1% "ambiguous" is a stretch) preclude that very common skeptical argument.
It's no secret, and it's been written about elsewhere, that the Air Force was VERY surprised by these results. They wanted Unknowns to be statistically the same as Knowns. In fact, in the 1955 public / press release, they announced that the Unknowns comprised only 3% of all sightings. The actual results of SR14 show that the real figure was 22% Unknown. 22% ... vs. 3%. Draw your own conclusions.
The skeptics are right about this: it's important to remember the quality and nature of the "data in". It had substantial subjective elements. But it's also important to keep in mind that these very things were accounted for by the scientists in the process of classifying cases as Excellent, Good, Doubtful or Poor. See pp. 1-10 of the Report.
Also interesting: On pages 69 and 76, yet another of the skeptics' favorite arguments falls:
"This would seem to lead to the conclusion that astronomical phenomena are easy to identify and there are very few left in the UNKNOWNS."
And later,"The results ... do indicate that relatively few of the UNKNOWNS are actually astronomical phenomena."
The Unknowns are especially not 'hidden' astronomical Knowns, just short on data. And that's a big problem for the Klass- and Menzel-like minds of the world. That's their "go to" mundane....
Originally posted by atlasastroOf course it can't determine the true nature of the Unknowns! If it could, we'd not be here. UFOs would be identified as [whatever].
So the test cannot determine the true nature of unknowns relative to all the knowns inputted into the analysis.
[Then it's more of the same 'no proof for flying saucers' talk]
Objections like the above, or like this:
"these tests do not necessarily mean that the UNKNOWNS are primarily 'flying saucers' "
do not extinguish the core statistical result. The distributions are significantly different.
The Air Force's interpretation of those results ... well, I doubt anyone would be surprised. Repeating them is like repeating that the Condon Report says there's nothing to UFOs.
One can argue, though, that SR14's "data in" was bad and therefore "data out" must be bad. (Would a debunker argue the same had the basic statistical results shown Unknowns = misidentified Knowns?) But can we truly discard all 'data in' as "bad", since it was "subjective", even when multiple scientists got together and addressed these very issues? Are 'Excellent' or 'Good' quality sightings (where 1 in 3 were ultimately 'Unknown', remember) likely to be insignificant?
These SR14 battles have been fought before, and pseudo-skeptics dis not fare well. (Which is why you'll very rarely see them introduce this topic.)
For more, see Maccabbe's report on SR14. His presentation on it is here. (Warning: PowerPoint!)