I make my living by solving problems. I work in electronic communications (commmercial 2-way radio) and have done so for 37 years.
Very often, in troubleshooting a 'system' problem, you have to take a 'reverse' approach, by ruling out the "possibilities", in order to 'narrow down' the "probabilities". This has become more and more necessary as electronics and communications systems have become more and more complex. And sometimes, the 'solution' or 'answer' is not necessarily what one initially thought it could even be. I also do beta-testing of new communications products for a major communications equipment manufacturer, working directly with development engineers, where again, very often, the 'solution' or 'answer' to a problem is evasive and the only way to find it it is to "rule out possibilities" until the remaining "probabilities" are small enough that the solution or answer becomes more evident.
After 37 years of doing this, I would even say that considering "negative evidence" is very often the ONLY way to eventually solve an especially 'tough' problem. Otherwise, the 'haystack' is just too big to ever find the 'needle'.
Considering the difficulty of solving the Earhart mystery, as a professional 'problem solver', I would say that using "negative evidence" is absolutely necessary to "rule out possibilities", in order to "narrow down the probabilities" that then can be looked for and checked.
As to 'skip conditions', if accurate details about the solar cycle conditions in July of 1937 can be found, then based on my 40 years of experience using HF communications (including the use of some HF marine band communications), I think that a reasonably accurate 'prediction' of where Earhart might have been (if her "Q5" transmissions were heard via skip propagation) can be developed. This would be a 'range' around Howland, not a specific distance or location, but it would be information that could be used to further "narrow down the probabilities" as to where she was located at that time and where she ended up.
I just thought of a way to sum up what I'm talking about. You can look for "positive evidence" (look for the needle in the entire haystack), or you can use "negative evidence" to rule out areas of the haystack where there is no probability of finding the needle, and then search the remaining part of the haystack for "positive evidence" (the needle). Thus, using "negative evidence" can greatly reduce the time needed to solve a 'problem' (find the needle). I think using information regarding radio propagation can help to 'rule out' some areas of this 'haystack'.
I thought that someone had already done some studies on this subject, but I've never seen any such information released to the public, so I don't know how detailed anyone ever got with anything like this.
John Rayfield, Jr.
Now we are both in to negative evidence, perhaps better covered in theorizing about theories. I believe I must concede I can not say how skip behaved on July 2 1937. In the end I am looking for a positive result, something like Smithy's landing gear. If you are unfamiliar, when the Lady Sothern Cross went down the only thing they ever found was one landing gear which had washed up on the beach. It was apparently floated by the buoyancy of the Tyre. It had brake gear with a serial number on it.
Search under Lockheed Files Altair
Neff