Over the Airwaves - The Journal for the Proficient Pilot
Shooting in the dark
Each year, the AOPA Air Safety Foundation publishes a statistical report of general aviation accidents. This is called the Nall Report. It's named after Joseph P. Nall, an NTSB Board member who died as a passenger in an airplane crash in Caracas, Venezuela, in 1989.
Regrettably, based upon the paucity of new information contained in each annual Nall Report, I'm likely one of only a dozen or so people who actually read the 2010 edition cover-to-cover. I did so this in the vain hope of discovering something other than the positive spin on our chronic fatal accident rate that its editors and statisticians so dutifully apply year after year. Instead, it was the same findings, rationales, and defense of an industry that should be ashamed of its abhorrent safety record. Thus, no surprises were found.
If it is the GA industry's intent to convince the non-flying community that general aviation is safer than it really is - and that our safety record is improving with each passing year, the Nall Report rates an 'A plus'.
On the other hand, the Nall Report's usefulness in helping to formulate improvements in the way we train pilots, in uncovering the mysteries behind our chronic fatal accident rate, and in providing guidance to us veteran pilots on how best to optimize our proficiency, it deserves little more than a 'D minus'.
The faulty denominator
Hold the e-mail. I'll admit that the problem is not entirely in the hands of the Nall Report editors and statisticians. Much of the problem lies in the inherent weaknesses of the data they use to compile the report. The most serious offender in this regard is the FAA's "estimate" of annual aircraft hours flown.
This "estimate" is derived from a simple FAA-administered survey of a sampling of GA pilots - about one percent of the total pilot population. Curiously, I suspect that only active pilots actually complete and return these survey forms. The pilots who have curtailed or stopped flying likely never return their survey forms.
Given the fact that this "estimate" forms the basis of all reported accident RATE information, we can reasonably conclude that the Nall Report is little more than an "estimate" of what is happening in the real world of general aviation. This makes for a nifty way to spin the data whichever way one likes.
Equally problematic is the fact that the "estimate" of annual hours flown is non-aircraft type specific. For example, we really have no clue whether single engine operations are actually more precarious than multi-engine operations. Nor do we have an objective basis to determine if glass cockpits afford us any greater safety advantage than steam-gauge aircraft.
Does anybody really know, based upon actual accident data - by aircraft type, if Cirrus SR22 aircraft are more prone to fatal accidents than Cessna 206s? Does anybody really care? Those who pay insurance premiums might like to know.
"One size fits all" flight training curriculum
Since we lack any objectively derived accident data based upon aircraft type, make, or model, those of us in the flight training business are constrained to offer a "one size fits all" training curricula. On the other hand, if low wing singles are more prone to fatal spin encounters than high wing aircraft, based upon actual data, then training curricula should be adjusted accordingly.
Are we seeing more landing accidents in glass composite aircraft like the Cessna Corvalis and Cirrus SR22 than we are in metal aircraft? If so, let's tailor the training curriculum accordingly.
Do flight schools using Diamond DA20 aircraft have a better safety record, again based upon actual data, than schools using Cessna 172s?
Curiously, we've been seeing a big spike in the number of fatal accidents involving homebuilt aircraft. From this, can we conclude that homebuilt aircraft are inherently more dangerous than production aircraft? This big spike could be attributed to a huge increase in the number of homebuilt aircraft hours flown. This would certainly spike the numbers.
Right now, of course, there is no way to correlate the number of homebuilt aircraft accidents with the actual number of homebuilt aircraft hours flown. Thus, any current criticism of the homebuilding industry is entirely without merit.
The Sport Pilot debate
If anybody suggests to you that the Sport Pilot program is either safer or more risky than conventional flying, don't bother engaging in the debate. Why?
The reason is, nobody - including the FAA and the NTSB - really knows. Sure, we know how many Light Sport fatalities there are in a given period, but we have no idea how many Light Sport hours are actually flown. Thus, it is impossible to develop any reliable data on the true risks of Sport Pilot flying.
Imagine that. We create an entirely new kind of flying, design and build a whole new classification of aircraft, authorize a new pilot certificate, allow pilots to fly without an FAA medical, and we have no objective way to measure relative safety of the Sport Pilot program itself! Only the federal government would allow such a debacle to occur.
The aging fleet dilemma
We're beginning to be increasingly concerned about the aging of the GA fleet. Do aircraft with 7,000 or more hours total time suffer a higher fatal accident rate than, say, aircraft with 3,000 or fewer hours total time? Right now, nobody knows for certain.
To be fair, we cannot lump all GA aircraft types, makes, and models in this age-related safety determination. If Beech builds a more durable aircraft than Piper or Cessna, the accident data should bear that out. Similarly, insurance premiums should reflect that finding.
Training reform efforts
The FAA recently formed the Aviation Rulemaking Committee (ARC) that will review everything from the way we train pilots to the way that we retain our pilot certificates. "Everything is on the table," said Babbitt. "This effort is part of a five-year plan for transforming GA safety."
Everything is on the table, alright. Trouble is, nobody knows where the table is. How can we reform an industry that has no true handle on the extent of its problem? How will the FAA know if its reform efforts are working when we have no valid way of measuring the extent of the problem in the first place? This is yet another example of a big waste of government time and money.
Banish the naysayers
"Whoa," say the naysayers. "We can't possibly gather that kind of data. The costs would be prohibitive. The burden on the industry would be staggering. Besides, it's nobody's business how much I fly my airplane."
Predictably, AOPA would be standing at the head of the line protesting any such privacy infringement on its member pilots. The halls of Congress would be blanketed with form letters of opposition artfully crafted by AOPA for signing and mailing by its 400,000 member pilots. In the end, another positive step forward in improving our chronic fatal accident rate would be successfully thwarted.
In truth, gathering objective annual hours flown by aircraft type, make, and model would be a snap. Since every GA aircraft must go through an annual inspection or, in the case of home built aircraft, annual condition inspections, it would be a quick and easy step to record and submit to the FAA the aircraft type, make, model, year of manufacture, and number of hours each inspected aircraft has flown in the previous 12 months (or whatever period covered since the last inspection). Bingo, that's it.
Curiously, the federal budget required to implement this procedure would likely be less than the money it spends sending out and analyzing pilot surveys.
As for an infringement of pilot privacy, we pilots waive any such right to privacy whenever we sign a medical history form each time we see the aviation medical examiner. Recall, we report the number of hours we have flown in the past 90 days. We also waive such right to privacy whenever we sign an application for a new pilot rating or certificate.
Could this be a conspiracy?
Perhaps the reason we're not collecting actual aircraft usage data is due to fear on the part of the FAA that the total number of GA hours flown is substantially BELOW current estimates? Maybe GA is already standing on "death's door" and nobody really wants to have that fact objectively revealed. Well, sad as this may be, we deserve to know the truth. And once the truth is known, we can take REAL steps to fix it.
It's time we get real
We live in a brave new world. It's a world where fatal airplane accidents should no longer be occurring. Sure, there will be the occasional screw up or dumb pilot trick that results in a fatality. Let's say that those type of accidents account for about 50 a year. But what about the remaining 200 or so 'fatals' each year? Why are they occurring?
There is currently no way of knowing for certain what impact ANY changes we make in the GA system have or will have on our fatal accident rate. None, nothing, nada. We're only guessing - and that ain't no way to run an airline.
So, naysayers, and that includes you, AOPA, let's get behind this effort to objectively quantify our annual hours flown by aircraft type, make and model. When you do so and this system is finally implemented, the Nall Report will finally offer genuine benefit to the industry. Until then, it's simply another way to spend member pilot dollars in a totally non-effective way.
Flying has risks. Choose wisely,
Bob Miller, CFII, ATP
http://overtheairwaves.com