A Data Variance Approach to the UFO Phenomenon
Why UFOs are real because of variance.
Variance is the amount of spread in a dataset. A high variance indicates a high level of spread in the dataset meaning data points are spread out whereas a low variance implies they are close together. Some phenomenon tends to be low variance, for example, summer temperatures in Houston, Texas. This is because there is a high level of humidity and consistent air pressure that keeps temperatures around the same. Weather as a whole tends to have a low variance because it doesn’t change too much except because of a freak storm. The averages for football players for the Dallas Cowboys over a season tend to be similar as well and usually have a lower variance.
Some phenomena are highly variant over time. The temperature on the moon for example can vary wildly because there is no atmosphere to spread the heat. The orbits of some comets can be highly erratic and have a high variance because of its trajectory. The equation for variance is deduced from the sample mean.
This means that the variance is just the squared differences between the data point and the mean of all data points. Therefore, the variance is a measure of how much every data point varies from the mean. High variance phenomenon will tend to have other factors affecting the spread of the data compared to data points with low variance. For example, if suddenly in summer the temperature drops below 80 degrees, there might be a cold front coming in or somebody is spreading chemtrails with reflective metal that reflects sunlight. If the UFO phenomenon were merely misidentification and nothing more than coincidence we would expect low variance in the data because people in different states and different countries shouldn’t be any worse or better at misidentification.
There are probably not significantly more or less trained observers per capita in Texas compared to Maine or Montana. One issue we will get to is dark skies but we will see why that’s not as important an issue for this article. Using data from NUFORC and Census data we can find the number of sightings per 100k population for each state. We can do this using this equation by dividing the number of sightings by the population estimate for 2023 and then multiplying by our factor for the population per capita.
The variance for all the states is 420.17, this is quite high, however, a better measure for understanding the spread of the data in the context of the order of magnitude we’re looking at in the data is to use the standard deviation. This is the square root of the variation and will give a measure of how spread out the data is but at the same order of magnitude as the data itself. The standard deviation is around 20.50 which is still quite high considering the lowest state is around 21 sightings per 100k and the highest state is around 95 sightings per 100k. This means the average deviation from the mean for each state is around 21 sightings per 100k people.
Looking at a state map you can see there are a high number of sightings per capita in some states, just not the states you would normally expect.
Some states have a high number of sightings and others are quite low. This is why the variance and standard deviation are quite a bit higher than zero. A standard deviation of zero would indicate there was no variability in the sighting totals for each state whereas a much higher standard deviation would indicate far more variability. Even the variance within some states can be quite high. Take, for example, South Carolina and the coast of South Carolina near Myrtle Beach.
According to the county-level visualization from Axios, most counties in South Carolina have below 40 sightings per 100k residents but Horry and Georgetown have way over 100. This implies there is something off the Atlantic coast near Myrtle Beach that is causing a much higher increase of sightings, and probably is on the same latitude. How could there be more misidentification per person in Horry County vs. Beaufort which is also on the coast? The misidentification hypothesis for all UFOs doesn’t make sense because why would some states and counties have much higher misidentification rates than others?
Another argument made is that dark skies impact the number of sightings and this is probably correlated. However, many counties have dark skies and have much lower sighting rates per 100k people than neighboring dark-sky counties. For example, most of North Dakota is small towns with largely darker skies compared to say New Jersey. Yet, Barnes County has a disproportionate number of sightings per 100k residents than any other surrounding county or even any county in the state. Misidentification can’t explain why specific counties have much higher sighting counts per resident than any other county.
I dare any skeptic or debunker to explain away the variance in the sighting reports. Misidentifications or natural phenomena would not cause the data from NUFORC or any public UFO dataset. We can see this more clearly when we look at the ratio of each state to the lowest state in terms of sightings per 100k residents. Texas has the lowest sighting count per capita at around 21 sightings per 100k residents which is quite low and on par with countries like Canada and a bit higher than Australia. The ratio of sightings per 100k residents for each state compared to Texas can be quite high considering this phenomenon according to “skeptics” is just misidentification.
Vermont has nearly five times the number of sightings per 100k residents as Texas. Washington state has nearly the same number and New Hampshire has four times as many sightings per 100k residents as Texas. You’re nearly five times more likely to see a UFO in Vermont as you are in Texas and only two times as likely if you live in Arkansas. Why do Vermonters have nearly five times the rate of misidentification as Texans? That doesn’t make any sense if you only subscribe to the hypothesis that all UFOs are misidentifications. Even if you assume many are misidentifications of drone sightings. Why are drones flying much more frequently over Vermont than Texas which has multiple military bases and lots of aviation enthusiasts who fly drones? Vermont and upstate New England really have no reason to have more misidentifications or drones than Texas. Some states are more “UFO prone” and therefore are unique compared to other states. It can’t be their intelligence because Vermonters on average are around 4 points higher in IQ tests than Texans.
The argument here is simple, misidentifications can’t account for the NUFORC data. There’s no good reason why they should. The standard deviation is also quite high, around 21, meaning on average a state will vary from the mean by around 21. There’s no reason why 21 more people on average per 100k people in the state should see a UFO or not if you assume they must all be misidentifications. Unless people in those states are all just bad observers and terrible compared to say people in Vermont or Montana. Something else is causing the increase in sightings and it’s not bad observers. “Skeptics” are debunked because they’re not really skeptics, as a real skeptic would look at the data.
