Australian Dept of Defense Scientific and Intel Aspects of the UFO Problem

Text

This text came with the file. It may have mistakes. When the text and the image disagree, the image is right.

4.
4 its existe-nce.
Public disl.rihut.ion vaa !ll~de, ho,.-ev{lr, of a so­
ee.lled 11 summa.ry" 1-hich in fa.ct did not Sul"lmariso, nor scarcely
&lludo to the 19·17-52 dc.ta 1 but conccntra.tcd or. 1953-55 rcportin.;
which vas cle~rly designed to reduce the residual unknowns to nn
insi{:nificant number, no matter how senseless tho identification
became.
10.
Vi t.hin U1u body of the diffi cul t.-to-oh ta.in .repo'r l thei-e
il!I an intores~ing diagram.
The product of the estimated obr.erv"r
relia.bil i ty and t.be rel)ort reliability be,ame the sighting
relinllility.
Tht! ,percenlng!! of reporls that had to be register"'J
a.s t•unkno1.1n 11 (i.e. incapsble of being .,ye,. approximately identif i ,.-d
aa a. known object) increased as the sighting reliability impi.·o·:ed.
Conver5eJy, the percentage listed as "insufficient info~naLion"
d"creased with improving reliability.
~ighting
No. of
Uoknovn
Insufficie111,
Relia.bili ty
Reports
(%)
Informs.ti en
(,t;)
Poor
435
16.6
21.4
Doubtful
794
13.0
14.0
Good
757
24.8
J.6
Excellent
213
33.3
'1.2
11.
Throughout the years of the UFO phc,nomenon, there has been
e. persistent for:n of official pronouncements ',/hich state thut th-,
percentage of unknnwn5 would bA redu~ed if more data were c~ailnble.
The tshove table contradicts that statement.
Reports of excellent
reliability g1merally s tom from a.strono:nerE:, pilots, scientists,
surveyorn 1 mctcorolo;:ists, radar operators etc, complete with
instrumented valueR and Rccund,ely cllit.fdlt~d uc:counts.
The
introduction o~ good raliablo reporting pr•vnnts the roady prosaic
int.eq>ri~tation,
In all probability the overall average percentr.g!J
of llnkno..,ns ( 19. T~) would h:~.ve be ,m subs len I, ia l ly increased if thE.'
dita hnd been more reliablo.
12,
Project BLUE BOOK ccmsul tants sta.ti s t:i cal ly tos ted the
unknovn object popula~ion to determine the likelihood tbat it wn3
similar to the popult\tion of identified objects and found tha.t the
probability was less than one in 1028 (i,e. using the American
~ational Archives of Australia
NAA A 13693, 3092/2/000