Can I use no-normal distributed data with spatial statistics tools?

[ThanassisSPIGGOS]

Hello,

I have gathered from Tripadvisor the reviews number/hotel, for all the hotels of my region (population and not sample).
The distribution of that review's number is not normal
Can I directly apply spatial statistics through Arcgis like spatial auto-correlation (Moran's I), Hot-spot Analysis, etc or should I first transform the data so it is normal before proceeding?
Thank you very much!!

[DanPatterson]

What is your hypothesis?

It probably goes without saying that hotels aren't spread out uniformly throughout the landscape.

Perhaps you can clarify your initial thoughts.

[ThanassisSPIGGOS]

Dear Mr Pattersonthanks for your interestHotels (many small and several large units) are mainly located throughout the coastline of the island.Through the number of reviews I want to examine the existence of  clusters of tourists who visit these tourism areas and hotels ... or that the distribution of reviews (and tourists) in the area is random (Ho)Thanks againT.S.

Στις Παρασκευή, 12 Ιουνίου 2020, 08:44:54 μ.μ. EEST, ο χρήστης Dan Patterson <geonet@esri.com> έγραψε:

#yiv7949301613 * #yiv7949301613 a #yiv7949301613 body {font-family:Helvetica, Arial, sans-serif;}#yiv7949301613 #yiv7949301613 h1, #yiv7949301613 h2, #yiv7949301613 h3, #yiv7949301613 h4, #yiv7949301613 h5, #yiv7949301613 h6, #yiv7949301613 p, #yiv7949301613 hr {}#yiv7949301613 .yiv7949301613button td {}

|

GeoNet, The Esri Community

GIS and Geospatial Professional Community

|

Re: Can I use no-normal distributed data with spatial statistics tools?

reply from Dan Patterson in Spatial Statistics - View the full discussion

What is your hypothesis?

It probably goes without saying that hotels aren't spread out uniformly throughout the landscape.

Perhaps you can clarify your initial thoughts.

Reply to this message by replying to this email, or go to the message on GeoNet, The Esri Community	GIS and Geospatial Professional Community
Start a new discussion in Spatial Statistics by email or at GeoNet, The Esri Community	GIS and Geospatial Professional Community
Following Re: Can I use no-normal distributed data with spatial statistics tools? in these streams: Inbox

This email was sent by GeoNet, The Esri Community | GIS and Geospatial Professional Community because you are a registered user.

You may unsubscribe instantly from GeoNet, The Esri Community | GIS and Geospatial Professional Community, or adjust email frequency in your email preferences

[ThanassisSPIGGOS]

Dear Mr PattersonIs this the answer to my problem with the use of no-normal distr. data?"Especially if the values for the Input Field are skewed, you want features to have about eight neighbors each" (https://desktop.arcgis.com/en/arcmap/10.3/tools/spatial-statistics-toolbox/h-how-spatial-autocorrelation-moran-s-i-spatial-st.htm)Or something else?Olease help meT.S.