This post continues with the analysis of response times for CAD immediate incident calls on a London borough. This analysis is carried out with the purpose of discovering geographical variations in police performance, trying to discover the reasons for these variations and trying to link them to confidence in the police.
The second cluster I have selected are those grid squares where the average response time is greater than 10 minutes. My initial hypothesis for this cluster is that the grid squares contain mainly residential areas which are off the main roads, in places that police do not so regularly patrol and locations that police are less familiar with. For instance, an address that is flat 1032a, Sampson House, Peabody Estate will inevitably have a longer average response time than the Red Lion Public House, High Street.
There are two grid squares that meet the criterion of the first and second clusters. Using the data presented so far I would suspect that any residents in living in these grid squares will have a higher level of dissatisfaction with police performance than residents of other grid squares.