The Need

A major retail organization in the U.S. has faced allegations that its security personnel target primarily people of color when identifying and monitoring shoplifting activity in the store.  Aside from pending litigation, the store has suffered from negative media exposure, and has a concern about the practices of its security department and personnel.  The department store needs to defend itself in court, but it also wants to eliminate bias if it exists in the store’s shoplifting enforcement activities, while simultaneously maintaining a high level of security presence to prevent shoplifting and subsequent financial losses from those activities.

The Solution

This department store, like most department stores in the U.S., carries several brands of clothes in several different departments.  Marketing strategies vary significantly by department and brand, and subsequently, the people who visit and shop for merchandise in these departments varies as well.

The fundamental premise behind our work was to determine the racial and ethnic percentage of individuals in the store, and to further understand whether this percentage differs from department to department.  We conducted observation surveys at carefully selected points within the store to determine the racial percentage of the store as a whole, and of the racial percentage of the individuals shopping in the different departments. 

We also reviewed factors that could influence the stores detentions of shoplifters such as shortage areas and subsequent focus on those areas, detectives schedules, race percentages by time of day and day of week.

Outcomes

By factoring these data together, we were able to answer such questions as:

• How much do shopper demographics vary by department?
• Are detectives stopping more shoplifters in high shortage areas?
• Do shopper demographics vary according to time of day and day of week?
• Are more shoplifters stopped when more detectives are on schedule?

Answers to these questions helped paint a much clearer picture for the organization about enforcement activity and race.  With these data, we were able to demonstrate that the stores responses, security scheduling and detentions were much more highly correlated with shortage than with race.  This analysis has begun to demonstrate why making a simple observation of the percentage of detentions by race is a simplistic and faulty method to review any mature organizations security practices and procedures.