LOBO Studies Efficacy of the LOBO Training Syllabus

Forums: 

Tom Sullivan, LOBO Board member, undertook extensive research to determine whether pilots who have taken LOBO approved transition and/or recurrent training have less accidents than those who take other Lancair type training or no type training at all. The results show that LOBO training has had a significant effect; please read Tom's brief Here. LOBO, with the assistance of former LOBO President Jeff Edwards, has been taking this data to the insurance industry on behalf of the membership.

The data is quite surprising.

The data is quite surprising.  I would also be interested to find out the flying background of all the mishap pilots.  In my experience over the last 12 years in the Lancair community I am always surprised at how many potential Lancair owners state that they don't need training or that their 'vast' experience in other airplanes will just require a 'quick' checkout.  These people have been military pilots as well as airline pilots.  I will tell you that these are fine flying machines but they are unlike any other plane, one off builds, and non-standardized avionics.  Another point I like to reinforce is to the avionics and understanding what you have.  Between Cheltons, Grand Rapids, Avidyne, Advanced Flight Systems, Dynon, Garmin, and the mix of any of these to a partial 'steam gauge' panel can be a pilot's undoing until they learn their system.  I would submit there's a lot of 'heads down' time spent trying to figure out what the 'magic' is doing when it isn't doing what you thought it should be doing.  Case in point: Although the NTSB data is not official, it's my opinion of a causal factor in the Evolution that went down in CO last year that they likely activated the GPS approach within the FAF, were low on GS, and the plane did what Garmin does and went back to the IAF to start the approach.  Being low, maintaining their altitude the plane flew back to the IAF into rising terrain and hit the high power tension line.  Again, understanding what your system is doing is key.  

Thanks for posting and thanks

Thanks for posting and thanks to Tom for doing the research. I read his post and would really like to see the statistical analysis posted. This dataset needs to be cross referenced with pilot time, currency, time in time/type as well as training. If the other entries are statistically insignificant, then I think it will help validate this claim. If you need help crunching numbers, please let me know. 

I would also like to see the

I would also like to see the data.  When stating that LOBO training is associated with fewer accidents (which I wholeheartedly believe is true), it is important to say that this is true after controlling for other factors related to accidents, Pretty simple regression techniques can be used to control for other factors when ascertaining specific effects (here the effect of training).  Even apart from training the training effect, it is of interest to know what factors are important and the strength of the statistical statement that can be made, which depends on the sample size and other factors.  Also happy to help with this.  Sounds like Ryan and I both have statistics backgrounds.

A two tailed Z tail test (0

A two tailed Z tail test (0.05 significance) was performed on all pilots (127) trained by LOBO compared to all others (870+). There have been 101 NTSB defined accidents since the inception of LOBO training in 2009. 43 of those accident were fatal. Only six accidents have involved LOBO trained pilots-- most were engine failures. All LOBO pilots walked away. Zero fatal accidents involving LOBO trained pilots. The qualitative side of the research indicates why/ what factors are important. Bottom line is create a good training program with high standards, select and train great instructors and maintain those standards.  The paper awaits peer review and publication in an academic journal.

Best regards,

Jeff Edwards

Jeff, thank you for this

Jeff, thank you for this information.  It sounds very promising.

One issue raised in studies like this is potential selection bias that arises because training is a choice rather than an exogenously imposed treatment.  The basic idea is that if those that select training tend to be safer, then the results **could** be explained by whatever factors make them safer rather than the training itself.  This is the classic issue with uncontrolled experiments.

But there is a response to that criticism if peer reviewers or insurers raise the point. Even if it the findings are partly explained by selection bias (could be partly the case, almost surely not entirely), that shouldn’t matter for the insurance carriers.  What matters is for them to be able to identify risks regardless of the explanation.  I would assume that actuaries understand this.  The basic point is that the findings show that those who chose to be “treated” are safer.  That should be enough for the  “treatment” group to receive lower insurance rates than they would receive if the carriers did not have this information even if the treatment group was endogenously selected. And that in turn should increase the demand from Lancair pilots for LOBO training.

It would be nice to have a comparison of “treated” Lancair flyers with typical Cessna, Vans, CIrrus, Beech flyers, etc. I imagine that’s information carriers keep close to the vest, but maybe there are studies out there that would make for useful comparisons.

I imagine you have considered that if these points and the likely insurance response hold up, it **could** price some folks out of the insurance market — the sum of the training and insurance costs **could** be higher than the insurance costs without training for some people.  Or folks might choose liability only if they deem training plus full insurance too steep relative to no training and liability only.  I’m sure the hope is that the full price of insurance plus training will be attractive enough to bring more people in.  I certainly hope that’s the case.

it would be nice to see the paper, but I understand if you want review from the insurance folks or others first.

Happy and safe flying,

Dan

 

Dan,

Dan,

As a researcher in the area of GA safety I am aware of the biases and variables you mention, including self selection that you mentioned here. What is interesting in this mixed methods research is that the several LOBO trained individuals involved in NTSB defined accidents and incidents involving engine failures on takeoff (Tom Sullivan, Dave Taylor); engine failures at altitude (Don Gunn, John Romer); windshield failure at FL250 (Josh Herndon), all attributed their survival and success in dealing with the emergencies to their LOBO training. These were their written words describing their experiences. Additionally, in reviewing other type club organizations, the positive influence of type club (and other type specific training) is undeniable and statistically significant. For example, Tom Turner at American Bonanza Society reported similar results ten years ago with the ABS training program. The MU2 community recorded an amazing  decline (almost overnight) in MU2 accidents when the SFAR was introduced, same for Robinson Helicopter.  I discussed this phenomena with Mr. Turner at ABS and Pat Cannon (MU2 guru). fmi on the MU2 see https://www.ainonline.com/aviation-news/aviation-international-news/2010... In the MU2 and Robinson situation there is no self selection -- its is regulatory. Similar results were found when the FAA imposed the mandatory flight review requirement decades ago .. an immediate 15% decline in fatal accidents. These results negate the "self selection" bias.

I did similar research on the efficacy of type clubs comparing ours to COPA (Cirrus) and ABS (Beech Bonanzas) and discovered that pilots who join type clubs are significantly less likely to be involved in an accident. The qualitative portion revealed that members believe they are safer pilots because of their involvement with like minded pilots. Here is a link to that research. https://docs.lib.purdue.edu/jate/vol5/iss1/2/

If training had no effect on safety or accident rates then the FAA would not require any training to obtain a pilot certificate, or the states to require training to get a driver's license. The question becomes-- how much and what kind of training is necessary? IMHO the quality of the training program, the standards applied, and the selection and training of instructors is very important.

I understand your comments about insurance but I remind people that insurance is for when bad things happen. Training is to prevent those situations from arising in the first place.

As soon as the journal publishes the research I will post it here.

 

Best regards,

Jeff

Thank You for correcting me

Thank You for correcting me Jeff. I await the official report. That’s too bad because that looks like something easily overlooked on a preflight. 

georgerosel1's picture

Attention…

Attention…

Regarding the service bulletin mentioned here.  You need to place close attention to which bolt needs the safety wired!!! (Only one!!) Easy mistake!… and if wrong, it could cause very serious results when that bolt comes unscrewed. 
 

… George Rosel

Anyone firs knowledge if

Anyone firs knowledge if going through this will help with getting insured, seems most will not insure new lancair owners.

the site needs an edit

the site needs an edit feature, my last message should have read, "first hand"  sorry

webmaster's picture

Good news: there is an edit

Good news: there is an edit feature! An "Edit" button should be visible right below the body of your post. (If not, let me know!)