To symbolize crimes by neighborhood median income using a feature class of crimes and a polygon feature class of neighborhoods, which association should be used?

To symbolize crimes by neighborhood median income using a feature class of crimes and a polygon feature class of neighborhoods, using a join is the most appropriate method. A join allows you to combine attributes from one table (in this case, the neighborhood median income data) based on a common field, which effectively associates the crime data with the corresponding neighborhood attributes.

When a join is performed, each crime feature can be enriched with the median income information from the neighborhood layer. This allows you to visualize the data effectively, as each crime record can easily pull in the relevant income data, facilitating an analysis of how crime rates relate to neighborhood economic conditions.

In contrast, relating data is suitable for situations where you need to maintain a one-to-many relationship without merging the datasets. A spatial join combines attributes of two layers based on their spatial relationship, which may lead to additional complexity that isn’t necessary for this task. Finally, a relationship class is utilized within geodatabase architecture to manage relationships between different datasets, but this is also more complex than needed for the straightforward goal of symbolizing crimes by neighborhood income.