Calculate correlation between time of day and number of created cases (call center)

Question

I have a data set showing all incoming call during a month and the cases created by the call center agents. The cases are related to a specific call.

My hypothesis is that the agents create less billable cases in the system during the busy hours, meaning that they answer the call and help the customer, but without creating a case in the system.

I know basics statistics, but nothing advanced. My approach would be a correlation test to see if there is a correlation between the time of day and if a case is created (yes/no).

I would prefer to use Excel for this, but I am also familiar with RStudio. I know how to perform a correlation test, but I am not sure how to incorporate the time element. Do I need to separate the calls into time periods such as from 8-9, 9-10 etc. or is there a function to use the time of day directly from the data sheet (e.g. 10:07:47)? And is a correlation test even the right approach?

Thank you very much in advance.

Answer 1

There are some options for you:

• convert time to a numeric value, there are various tools in R to do that, you can also check library lubridate.
• if you care more about which period of day, like 9-10 or 10-11, has correlation with your response, then go ahead with your binning approach. But now bins become category, and to measure relationship between a categorical variable and a numeric variable, you should use association, check here.