Hi all,

I need some help, please.
I'm conducting a research analysis on how different high school calendar selections can impact crime rate.

I've gathered my data and perfomed a high-level analysis and discovered that at least one portion of my sample data displays major differences when on one calendar as opposed to the other. It's very clear in the graph that I generated, but I'm not sure how to express this statistically.

Using a time-panel series model and plotting the sum of each crime type on a graph I can model how the number of crimes committed increases or decreases year to year. When the test school is on one particular calendar type one of these variables displays very little variance from year to year, but once they go to the other calendar type, the year-to-year variance increase drastically.

I can phsyically see the difference visually on the graph, but I'm not sure what the best statistical method would be to explain this data. Any advice? Thanks!