Optimized Hot Spot Analysis - Analysis Field Variation

[AlexanderSwain]

Hello everyone,

I was wondering if anyone might be able to provide insight on how much variation is required in the input field when running an Optimized Hot Spot Analysis? When reviewing ESRI's'How Optimized Hot Spot Analysis works' documentation, the following information is provided: 'The Optimized Hot Spot Analysis tool will check the Analysis Field to confirm that the values have at least some variation.'

For context, I have a dataset for which I would like to run a hot spot analysis to look for statistically significant clustering, and the analysis field is an integer ranging from -3 to +3. When running the Optimized Hot Spot Analysis tool, I don't see any warnings or errors indicating that this level of variation is insufficient; however, I would love to find some additional resources that might address statistical considerations pertaining to variation in the analysis field.

Thanks for your help!

[RonaldHaug]

Hi Alexander,

I think it's pretty neat you're using integer values between -3 and +3 for your values in the analysis field.

It's a divergent field with normal at zero, kind of like cool, cold, freezing on the left side and warm, hot, burning on the right side.

Depending on how you classify it,  your data can be:

Ordinal: the data can be categorized and ranked.

Interval: the data can be categorized and ranked, and evenly spaced.

Why don't you run it and see what happens, report back, and enlighten us.

You're still going to need a minimum of 30 samples. I'm hoping you have lots more.

What I like about "optimized" hot-spot is it adjusts the neighborhood distance so you get meaningful results. 

Please just don't break the internet, ok?

RSVP - Thanks!