I am trying to analyze values in a spatial grid using the Getis Gi* hot spot tool. I started with point data for observed fishing tows, and each point has an amount caught. I aggregated the point data into the cells of the grid, summing the amounts caught in each cell. (For simplicity I have left out a few steps here). Because there aren't the same number of observations in each cell, I want to run the hot spot analysis several times, including only cells with a minimum number of two types observations in each run. I can easily query for the number of observations because I retained the counts for the two different types of observations.
The tool's help file says the tool will only run on selected features, so I tried selecting the cells I want using the select by attribues tool and using the definition query tool. My selection is simple: it is of the form "counttype1>0 and counttype2>0". Neither approach works - my output includes cells that do not meet my minimum number of events criteria. I don't really want to create a new layer or feature for every number of observatiosn I want to try.
Any ideas what the problem might be?
Thank you for your help.
Note: I case it matters: the file I am analyzing is in a personal geodatabase; I am putting the output in a shape file. There are no null records.
You are absolutely right, the Hot Spot Analysis tool should be working on your selection set. In order to figure out exactly what the problem is, I'm hoping you can try a few things and answer a few questions to narrow it down.
1. When you use "Select by attributes" on the dataset, do the correct features appear selected in the map? If you open the attribute table with the selection do you see those features are selected? --If the answers to these questions aren't "yes" (ie the selection isn't working) it sounds like it is a problem with the selection, potentially with your query. Try the same thing with your definition query, too.
2. With the data selected, can you please try running another tool? Something simple, like maybe copy features or buffer. We want to make sure that your selection is being honored by other tools that are supposed to honor selections. If it isn't, then we may have an issue with selections and/or your dataset. If the selection is being honored, then would it be possible for you to share your data with us so that we can try to reproduce the issue and figure out what's going on?
So sorry you're having trouble, and hopefully we can get it figured out quickly!