I started reading it but am out of time for now. I’m fairly sure that the 93% number is number of known shoplifting incidents and is unrelated to the value of the items. I’d need to read more to be sure.
I do wonder whether the dollar numbers are inflation adjusted or not. I’m sure that info is in the report.
An increase in the number of known shoplifting incidents would be conflated with increasing surveillance. It would be hard to distinguish observability vs actual increases.
Inventory would measure $ worth of goods missing, but wouldn’t ascertain the number of incidents that caused those losses. So the $/incident and incident count figures should be treated as if they have high uncertainty even if the $ figure is accurate.
Thanks, I continued looking further into it as well and found another article on it, which states:
Retailers reported a 93% increase in annual shoplifting incidents in 2023 compared to 2019.
So I believe you are right. Guess op proves that often you can have either interesting text with poorly supported numbers, or boring text with well supported facts. Too bad though, cause i did actually enjoy reading the article (so extra shout-out to op for sharing).
Here is the actual report.
I started reading it but am out of time for now. I’m fairly sure that the 93% number is number of known shoplifting incidents and is unrelated to the value of the items. I’d need to read more to be sure.
I do wonder whether the dollar numbers are inflation adjusted or not. I’m sure that info is in the report.
An increase in the number of known shoplifting incidents would be conflated with increasing surveillance. It would be hard to distinguish observability vs actual increases.
I feel like they would still know about it when doing inventory, no?
Inventory would measure $ worth of goods missing, but wouldn’t ascertain the number of incidents that caused those losses. So the $/incident and incident count figures should be treated as if they have high uncertainty even if the $ figure is accurate.
Thanks, I continued looking further into it as well and found another article on it, which states:
So I believe you are right. Guess op proves that often you can have either interesting text with poorly supported numbers, or boring text with well supported facts. Too bad though, cause i did actually enjoy reading the article (so extra shout-out to op for sharing).
deleted by creator