2

I asked "What kind of margin of error can I expect in figuring distance?" My issue and concern is that I've seen a lot of discussions about which method is best for measuring distance, but I found it quite difficult to find out how accurate each method is.

That's the focus of the question: How accurate are different methods. I even ask if it's possible to know what causes issues of accuracy in the different methods.

I even state, in my question, that I've seen many discussions on which is better, but they do NOT say what kind of accuracy the different methods have. For instance, is the Law of Cosines accurate to within .34%? Or less? Or more? What about Haversine? What kind of accuracy does that have? Or Vincenty?

I'm not asking which is best. I'm asking for rates of accuracy on the different methods and if it's possible to know what factors impact that accuracy. That's not at all what the other question asks.

While the question it's marked a duplicate of ("Why is law of cosines more preferable than haversine when calculating distance between two latitude-longitude points?") does have an answer that discusses the accuracy of the Law of Cosines, it never gives a clear number to say just what the margin of error is. It does, at least, say, that small distances might be an issue, but this answers only part of my question.

I had read that answer on that question while doing research, but it did not give me a full answer and the focus is on comparing two methods.

I'm not a math major. In fact, honestly, I haven't used trig in many years. All I'm trying to find out is what margin of error the different formulas have and what impacts that margin of error so, if possible, I can allow for it. Maybe the problem is, since I haven't used trig since high school, I just don't have the trig and advanced background to understand everything that one answer says about accuracy.

Which is one reason I tried to ask, "What is the accuracy of these methods?"

I feel like I'm in school and I have asked the teacher, "How big is an elephant?" and rather than getting an answer, I'm referred to someone using logs and advanced biology to compare the size of a baby elephant to the size of an adult elephant without ever saying, "Oh, and a big elephant is about 11 feet high."

I'm sorry I'm not as advanced as everyone on this Q&A site, but I'm just trying to get an idea of what kind of accuracy I can expect I know if I'm going to be off by inches, feet, yards, or miles.

The other question (and the answers) either do not provide that information or provide it in an advanced format that not all of us can follow. I'm a beginner at this and I'd like to learn more, but if I have to start by knowing it all on this Q&A site so I can get answers and start learning, then that's quite a frustrating situation.

0

2 Answers 2

7

The best way to avoid this sort of closure is to include a paragraph in your question saying something like - I read the following answers and they help but are missing the exact measurement I need.

While this is the most important question on the site for you, others (and especially the moderators) are reading 10's of questions a day and may spend only a minute or two considering if yours is a duplicate of another one. So all the help you can give is useful to them.

1
  • @PolyGeo made that suggestion yesterday, so I've added that at the start of the question, so I hope the marking can be undone. I also felt I had specified, within the original wording, exactly what I was looking for and why comparisons were not appropriate. Maybe, in a quick read or skim, that was easy to miss since it was a few paragraphs down into the question.
    – Tango
    Sep 22, 2018 at 18:28
2

I have not voted on the duplicate status for your question on Main because I do not have the skillset to determine whether or not it is a duplicate of Why is law of cosines more preferable than haversine when calculating distance between two latitude-longitude points?

However, even if it is not, then I think that it should be closed as too broad because it seems to be talking around an issue to which you seek a simple solution, when that issue appears to be a series of complex problems that necessitates complex answers.

The question you ask there is:

Can someone either point to a good resource or explain what kind of accuracy I can expect from the different methods I've mentioned and if there's a way to know what the conditions are that will give me more or less accurate answers?

This is asking for a list of resources and/or a list of personal explanations for the accuracies to be expected from each of a number of methods for calculating distances between two latitude/longitude coordinate pairs, and then for a simple summary of all of that information to be provided.

You mention and have tagged GeoPy in your question so one way to focus your question would be to heavily revise it so that it asks something more like:

Using GeoPy and working with distances of 100 miles or less [in a particular part of the world] and the [your preferred candidate] method to compare coordinates of two locations will the accuracy be within a few feet?

Looking behind the scenes at your question I can see that you need to edit it some more because the current decision (after Review) to leave it closed as a duplicate has been made by three users and not just one moderator:

enter image description here

3
  • 3
    Thank you for your help, but at this point, I'm fighting for an answer. I'm not new to SE, but I am new to this SE site. While you, personally, have been a big help, this makes me feel that I just shouldn't be asking questions in this board because I simply don't know the subject well enough to target my questions in a way that they'll please the board. The Tour page for GIS says, "Your most important question is important to us," but I've never had to fight, so hard, on any SE site to try to get what would seem to be a simple answer. (Cont'd)
    – Tango
    Sep 24, 2018 at 16:28
  • I had spent about 3 hours researching this topic, including trying to find anything clear about accuracy of the different methods before asking this question. Now I've spent well over an hour or two on this - trying to word the question on a topic that's new to me, that discusses formulas I don't clearly understand, and trying to figure out wording to edit the question. (Cont'd)
    – Tango
    Sep 24, 2018 at 16:30
  • 1
    At this point, I feel my time spent on this question would have been better spent on Quora or just spending another few hours on searching to see if I found anything. I've had to ask questions on other SE sites when I was new to the topic and felt like I wasn't fighting the site itself to just get a simple answer. Yes, they needed editing at times or there were issues, but you are the only moderator who has shown an indication of wanting to help me at all. I notice all I got from the moderator who marked it was, basically, "We had a lot to do, so we don't have time to do much."
    – Tango
    Sep 24, 2018 at 16:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .