It's not the same.

(i) I collected the data myself (ii) there are many more data points (more than 350 in all) (iii) I used the proper formula to compute the RE predicted distance between the locations. The other experimenter used a piece of string and a globe, from memory.

If the data is based on the spherical coordinate system of Latitude and Longitude, which you admit is based on the idea that the earth is a sphere, then the results are invalid until you can demonstrate that the system and model is correct.

I do not agree with your logic.

This type of experiment does not require that the Earth must be a sphere. Instead, this type of experiment supports the hypothesis that it is. The logic goes like this:

a) If the system and model is correct, this graph will make a straight line.

b) If the graph is NOT a straight line, then the system and model are incorrect.

So we try it out. If the line comes out curvy, we know that the system and model are flawed in some way.

It comes out as a line, so that means we do not have any cause to suspect the model based on this test.

This becomes a little sticky now. This test does not

*prove *the model is correct. What it does is

*fail to disprove* it.

This helps bolster the evidence for the model. As we say, we do not "prove" things in science, but we can "disprove" things.