The statistics for the years 1947 through 1952 are contained in Project Blue Book Special Report #14. This report made use of one of the first large computers to analyze UFO data. The Battelle and Air Force (ATIC) analysts started with about 4,000 reports from June 1947 through December 1952. They eliminated about 800 as being not sufficiently detailed for worthwhile analysis. The remaining 3,201 reports were analyzed according to sighting details and descriptions of the objects and according to the reliability and credibility of the observer(s).
There are 240 statistical tables and about 30 graphical depictions of the data in this 1955 publication. One on these shows how object sightings varied with time. In order to understand these graphs if is necessary to know the definitions of "object" sightings and "all" sightings. "All Sightings" refers to reports made by individual witnesses, whereas "Object Sightings" refers to the number of objects seen. Because there often is more than one witness, the number of "all sightings" exceeds the number of "object sightings." (There is another definition also, "Unit Sightings" which is sort of an intermediate category.) Thus, there were 3,201 sightings reports which make up the category, "All Sightings," and the analysts concluded that these 3,201 witnesses saw 2,199 objects.
In the above Figure 7 from the report we see the number of "Object Sightings" as a function of time, month by month, for the 5 1/2 years. Note that the vertical scale is sixty sightings per month, which is satisfactory for representing the sighting variations until 1952, when the number per month reached over 500 in July (hence the break in the vertical peak. Note carefully that there are two graphs here: all Object Sightings (a total of 2,199, corresponding to 3,201 reports) and Unknown Object Sightings (a total of 434, corresponding to 689 reports).
The Battelle investigators placed sighting reports into three basic categories: Known (K), Insufficient Information (I.I.) and Unknown (U). The K category was, itself, divided into various explanations categories such as astronomical, (man-made) aircraft, weather related and other natural phenomena. Insufficient Information was a category by itself for sighting reports that were not sufficiently detailed to allow a decision that it was either definitely or probably identifiable or definitely not identifiable. The presence of this category is important because, as the report emphasizes, that means that the U sightings were not simply sightings without enough information for identification. Rather, the U sightings contained some details that prevented identification as a known phenomenon.
The overall statistics for the five years of sighting data are presented in the pie graph below. The outer annulus is for "All Sightings," the inner circle is for "Object Sightings" and the "Unit sightings" lie in between. Not that, rather than grouping all the explanations together as Known, the data for the Knowns are presented as classes of explanation (astronomical, aircraft, balloons and other). Also segregated out are the Insufficient Information cases and the Unknown.
From the chart it is easy to see that 689 reports ("All Sightings"; 21.5% of the reports) corresponded, in the opinion of the investigators, to sightings of 434 objects ("Unknown"; 19.7% of the objects). This large percentage of unknown cases was not reported to the press when SR14 was published, with a restricted circulation, in 1955 (at the same time as Ruppelt's book; possibly an attempt to counteract the effects of his rather positive history of the Project Blue Book?). The so-called "Summary" of the report was used as the press release. It did not state the summarized statistics for 1947 - 1952, but rather referred to the statistics for the 3 years that followed 1952 and emphasized that after 1952 the percent unknown was less than 10%. In other words, the press release did not inform the American people about what actually appears in the report (in data table A-1): 1947 - 24% of 117 reports ("All Sightings") were U; 1948 - 13% of 205 were U; 1949 - 14% of 395 were U; 1950 - 23% of 305 were U; 1951 - 32% of 160 were U; and 1952 - 23% of 2018 reports were unknown. In other words, except for 1948 and 1949, over 1/5 of the sighting reports was unexplained. This is an astounding percentage incorrect reports, if there was actually nothing new and novel being seen! On the other hand, if there really was some new phenomenon flying around, possibly even "interplanetary craft," then this is understandable.
With regard to the quality of the sighting reports, the Battelle investigators divided the Object reports into four classes: Excellent (E), Good(G), Doubtful (D) and Poor (P). The Excellent reports (Objects) were made by the most qualified observers who had sufficient time to make good observations and presented a self-consistent report with many details. By contrast the Poor reports (Objects) were those for which the observer was inexperienced and the report itself was lacking detail or was not totally self-consistent. The Good and Doubtful cases fell in between. The chart below illustrate the very surprising (if there were no true UFOs flying around) result that the better quality the UFO report the more likely it was to be unexplainable.
The key result regarding sighting reliability and quality is most easily seen by comparing the top two pie charts in the above picture. Note that for the Poor Object sightings, 16.6 % were Unknown and 21.4% had Insufficient Information with the rest (63%) being explained. However, for the Excellent Object sightings, a surprisingly large 33.3% were Unknown and only 4.2% had Insufficient Information (with, again, about 63% explained). The low number of I.I. sightings in the Excellent group is logical, since the best observers tend to provide the most credible information. The result presented above is consistent with the hypothesis that credible observers have rather accurately reported seeing incredible objects. The alternative, hypothesis, that that credible observers have made the most mistakes in observation, leading to the largest fraction of Unknown sightings, contradicts the defined meaning of the "Excellent" classification. This fact of the data is further emphasized in the next chart which shows the breakout of military vs. civilian sightings. Many of the military sightings were by witnesses on duty at the time of the sighting. Note that this is based on "All Sightings" rather than "Object Sightings."
Although it is difficult to determine from the above chart, the data table (Table A58) in the Appendix of SR14 shows that fully 37.7% of the Excellent military sightings were listed as Unknown, whereas "only" 29.8% of the Excellent civilian sightings were Unknown. One can also see that most of the Excellent and Good reports were made by military observers, whereas most of the Doubtful and Poor sighting reports were made by civilian observers.
Special Report #14 also presented numerous statistical comparisons between the characteristics of the Knowns and the Unknowns, such as color, sighting duration, environment of witness (indoor, outdoor, in car, in plane, etc.), number of objects per sighting, shape, brightness and speed. Although one can argue over whether or not the types of comparisons they made were appropriate to the data, the fact is that they found statistical differences, not large, but clear differences, between the Knowns and the Unknowns, differences that were not reported in the Summary and hence were not told to the press.
The above statistics cannot be compared directly with the official "UFO FACT SHEET" published in 1970 after Blue Book Closed. This summary provides a list of the "TOTAL SIGHTINGS" and "UNIDENTIFIED" for each year from 1947 through 1969. The numbers of TOTAL SIGHTINGS and UNIDENTIFIED for the first five years are, respectively: 1947 - 122 sightings, 12 unidentified; 1948 - 156 sightings, 7 unidentified; 1949 - 186 sightings, 22 unidentified; 1950 - 210 sightings, 27 unidentified; 1951 - 169 sightings, 22 unidentified; and 1952 - 1,501 sightings, 303 unidentified. Comparison of these numbers with the SR14 tabulated data shows that these are what SR14 calls the "Object Sightings" for the first five years. Thus one might assume that the numbers listed for the remaining years (1953 - 1969) are also numbers of objects sighted and the number of objects unexplained. Next
ONE THE LEGACY OF 1952:
YEAR OF THE UFO
TWO 1952, THE YEAR OF THE
THREE THE INTERPLANETARY
FOUR THE WASHINGTON MERRY-GO-ROUND
FIVE A WAR OF WORDS or A WAR OF
PAGE SIX SHIPS
FROM ANOTHER PLANET
PAGE SEVEN SHIPS
FROM ANOTHER PLANET (con't)
PAGE EIGHT MORE
PAGE NINE THE CIA INVESTIGATES BLUE BOOK
PAGE TEN UFOS
AND THE SOVIET THREAT
PAGE ELEVEN LEGACY
OF THE ROBERTSON PANEL
PAGE TWELVE STATISTICAL POSTSCRIPT - Project
Blue Book Report
PAGE THIRTEEN APPENDIX
The Rogue River Sighting
PAGE FOURTEEN APPENDIX
The Rogue River Sighting (con't)