PPRuNe Forums - View Single Post - VH-YTM final report
View Single Post
Old 27th Aug 2019, 22:46
  #132 (permalink)  
andrewr
 
Join Date: Apr 2008
Location: Australia
Posts: 490
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Sunfish
You aren’t a pilot
Why do you say that? Because real pilots believe you can eliminate VFR into IMC accidents if only you do enough education?

Originally Posted by Sunfish
and you still don’t understand statistics.
Maybe, but I think in this case we are trying to answer different questions with the statistics.

You are asking whether Angel Flight itself increases the danger of the operation. Yes, that question requires you to find the most similar private flights (or at least control for the differences). That would probably be not just private flights with passengers, but private flights with passengers where there were specific plans to be somewhere. And if you are that specific there might be no difference. There are plenty of examples of regular private VFR flights into IMC under those circumstances e.g. VH PXD. But is it useful to compare Angel Flight with some of the more dangerous private flights?

Why compare with other private flights at all? What we want to do is to understand the risk associated with the flights. The comparison with private flights suggests that private flights overall are not an accurate reflection of the risk of Angel Flight - which I think is the same as what you are saying, but approached from the opposite direction.

Other questions you could try to answer with statistics:
  • Is Angel Flight more dangerous than private flights (the original statement)
  • Is Angel Flight more dangerous than private flights carrying passengers
  • Is Angel Flight more dangerous than charter
  • Is Angel Flight more dangerous than RPT
  • Is the Space Shuttle more dangerous than Angel Flight (just to show that yes, you probably can draw valid conclusions based on only 2 accidents)
For each you select the data that best reflects what you want to compare, not the subset where the 2 groups are most similar.

Have YOU read the actual report yet? What statistics in the report do you disagree with?
andrewr is offline