Saturday, April 2, 2011

Looking for patterns in the data

Dear readers

I was out walking through parts of the River Torrens linear park, in Adelaide, this morning. It was a glorious day, 17 degrees c, blue sky, with some of the trees in autumnal splendour.

My eyes were drawn to such things as two aircraft condensation trails in the sky; a spider's web in the branch of a tree; and a number of ducks on a pond. This got me thinking about patterns in the UFO data! Then later in the day, at my local library I came across the following article in issue 271 of the English Fortean Times (click here) magazine, page 27. Synchronicity.

Hidden patterns:

Jenny Randles (click here) is the author of the article titled "Hidden Patterns."

She examines two UFO sightings. The first occurred at a place in Staffordshire, England, called Tittensor, in December 1978. At 9.56pm a farmer saw two bright lights approaching. He stopped his motor vehicle and got out, and watched a soundless, low flying object which came within 37m of him. It had "...a long oval base, tapered at both ends and a pillbox-like dome on top. It was about 35 feet long and 120 feet wide (11 by 3m.) The dome was about 10ft by 10ft..." It left the area.

At 9.30pm on 24 November 1981 "...again on a dark country lane surrounded by open land, hedges, trees and power lines," at Darco, Texas, USA a man and a woman in a car saw two bright lights approaching them. They stopped the vehicle to watch. It was soundless and came very close to them. It banked in the same way as the UK object had done. "The two craft were also physically alike. The Darco object was another cigar shaped oval (about 25ft/7.6m wide) with tapered edges and a small cupola-like dome on top."

"Overall, the impression of studying these files is of the same object, repeating its encounter in similar locations, at the same time of day and year...However, there were some minor differences..."

Jenny suggests that "...we should be searching for further examples in the extensive UFO records-they may form a pattern revealing further pieces of a puzzle that belongs together."


  1. Hiya Pauline, last year I spent a couple of months focused on seeking patterns. There appears to be patterns there, but the data is skewed and screwed by the human elements reporting and recording it. If we look at the larger databases (NUFORC, U, MUFON, UFOCAT, Rutkowsi's and NICAP) sightings tend to occur more frequently in the 11th hour in either hemisphere. There are also apparently two peaks that occur around June/July and October/November. Sturrock, Swords, Teodarani, Poher and Vallee have analysed the data and the patterns they display. Chris Rutkowski is still doing it...

    This being the 'UFO enigma,' nothing is ever as simple! The databases are flawed and contaminated by multiple entries of the same sightings. Added to that is the assumption that all the reports they account for were both honest and accurate. Furthermore, are the people who seed ideas of MJ-12, Serpo and Dulce bases also instrumental in these sightings databases?How about those people convinced aircraft landing lights are shape-shifting alien craft and reporting them? Generating a 'clean' database would be needed to falsify the findings of Teodarani et al and give these scientists better information to work with.

    How the hell do we do that? One way that springs to mind would be a network of vetted, scrupulous individuals to personally investigate and record reports: time, date, weather, background, circumstances, eyesight. They would be objective and actually require witnesses to prove their identity....especially reports over the phone. This labour-intensive, resource-sucking process would minimise discrepancies in the database. It would also generate data at a much slower rate than MUFON or NUFORC and thereby take much longer to yield a dataset large enough to apply analyses to. For example, 30-40 accurate reports annually could show any pattern imaginable; we'd need hundreds per year and many years of collection. Rather than re-inventing the wheel, any sightings that can be authenticated (Blue Book unknowns for example) could be suitable for inclusion in the 'new' database.

    No doubt there are other ways forward too. What's appealing in this approach is how it removes much of the human tendency towards seeing patterns in everything. Statistical analyses don't even have a symbol for 'demons' and are also unlikely to conclude 'light beings.' If a combination of factors pointed to increased and predictable UFO sightings, it wouldn't necessarily explain it all, but we'd be way closer to more answers.

  2. Hi Kandinsky

    As you say, a 'clean' database is essential for seeing accurate trends.

    I would think that the biggest problem would be the number of IFOs which have not been weeded out by the data input process. My reading of the UFO literature puts the number of IFOs at perhaps 95% of all incoming raw reports, or higher.

    The idea of building a database of sightings which have been vetted and well investigated seems to be the way to go, but as you say it would take much longer.

  3. Hi again Kandinsky

    Had a thought that I recall a mention of patterns, in Keith Basterfield's book.

    In 1972 in South Australia, there was an investigation team which looked at all the incoming raw reports for South Australia for a whole year. They did exactly what you suggested and vetted all the reports, thus ensuring the end result of "unknowns" was as free of IFOs as possible. They ended up with about 10 good "unknowns." There was a pattern, even in this low number. The good cases came from rural locations during one specific quarter of the lunar cycle. At the time of the sightings, the Moon was up but very low to the horizon.

    Don't know what to make of this pattern, but to my knowledge no one in South Australia has ever repeated this work.