The following is an experimental result of notifying 200 people that their "facebook time" is to be monitored. In the plot below:
- Value 1 on the boxplot is a before test. The monitoring started 3 months prior to a notification being issued. A general notice was in the HR files, but no followup ever occurred with this. These values represent the average time (in minutes over the allowed hour) that (on average) the 200 employees exceeded the amount of time they are allowed to access facebook at their place of employment.
- Value 2 on the boxplot is the month's results following an announcement of a logging system being implemented to record the user and the details. This is the month directly after the monitoring was announced.
- Value 3 is the point 6 months after the monitoring system was implemented and no action or follow-up occurred. This is recording only, but no reminders and no action against the violators.
What we can see from this is that monitoring and logging does have an immediate effect, but if no action is taken, the effect quickly erodes leaving a state of lowered security.
Worse than this, the mean violation times did not return to the status quo of what they had been
prior to the announcement. They returned to a state double that before the issue of the monitoring announcement.
In this, we see the employees moving from a social context (the idea that they are a part of a "family") in their employment to more of a market context. The results are they feel less quilt from exceeding the time that they have been allotted and are more prone to violate the agreement that has been made to allow them some leaway in viewing the facebook site.
This demonstrates graphically that if you are to tell people that you are doing something you need to followup on what you state you are proposing or the results could be far worse than doing nothing at all.