- the 'general aviation census'. A campaign of postcards sent to random airmen which are promptly discarded by 90%. The 10% that are returned contain unverifiable information.
Good summary. I like the intent behind the GA survey. The trouble is, the conclusions are gossamer-thin, and when others stretch them to provide some additional data, they tend to break.
One of the survey results is an estimate of what percentage of aircraft on the FAA registry are active. These values vary only slightly, year-to-year. They do this for various groups, such as recip aircraft, recips with less than four seats, four or more seats, turbine aircraft, homebuilts, etc.
Others take that percentage, multiply it by the survey's estimate of total annual flight hours, and then multiply it by the number of aircraft on the FAA registry to estimate the total annual flight hours. They then compare the results to previous years, and across aircraft groups.
But starting in 2010, the FAA started a process to help eliminate inactive/non-existent aircraft from the registry. Over the next three years, they sent, basically, bills to all registered aircraft owners. If the owner didn't exist any more, or if they ignored the letter, their aircraft was removed from the registry.
Notice the similarity to the Survey process? The same kind of people who had ignored/didn't receive the Survey questionnaire were the same folks who ignored/didn't receive the FAA reregistration mailing.
So over the next several years, the Survey's prediction of the number of active homebuilt aircraft didn't change significantly. It rose from 55% to almost 60%, while the homebuilt registrations actually dropped by almost 25% (about 7800 aircraft).
The overwhelming proportion of those 7,800 were already inactive. Less than 3% of owners had canceled registrations restored. But that wasn't reflected in the FAA Survey results.
So if someone doesn't UNDERSTAND the limitations of the Survey data, and how deregistration isn't reflected, they can come to some erroneous conclusions.
For instance, that the homebuilt accident rate took a huge jump. It's about the same number of accidents, but (using the Survey's unchanged percentage of active aircraft in a deregistration era) the number of aircraft dropped drastically. And thus, the computed accident rate skyrocketed. It's a false indication...there are actually the same number of *active* homebuilts.
And you try to talk to those generating the bad results, and all they say is, "But my data is based on the FAA registry and the official FAA Survey....."
Gahhh.
Ron Wanttaja