Question about variable with hours

[#1]

I have an end-of-work variable of the respondents. I want to set so that people who finish their workday at midnight or 3 AM will be considered as those who finish later, at the moment it presents me as someone who finishes their workday earlier.
How can I set this up?

The problematic observations are highlighted in the attached image:

[#2]

Hi,

You will need to include the date (at least the day) in your timestamp; otherwise, JMP will not be able to figure out that the 3:00 AM is on the following day.

Best,

TS

[#3]

The problem is that this is an hour without a date, the end hours of the work represent the hours that the respondents usually finish their workday, it is not an hour of any specific day. There is no way to define the beginning and end of the day?

(E.g. the day starts at 5:00 AM and ends at 4:59 AM)

[#4]

Maybe add inhours(24) to any value < 8AM. You could do that in a formula column. I think you'll be trading one problem for another; the new axis will start at 8 hours and run to 32 hours.

[#5]

 You can use the Date Difference function (a TimeDate function) even without a date.  I tried it with hourly data formatted the way you have it, and Date Difference does work correctly.  Just decide what hour you want to measure the differences from and enter that as one of the times to measure the differences relative to.

[#6]

I'm quite an amateur at JMP I would very much appreciate it if you could explain in more detail how to do it

[#7 · Accepted solution]

I have attached a data table with a column formula that measures 24 hourly periods starting at 5AM and ending at 4:59AM.