How many incidents have occured so far?

There have been 394 incidents reported (227 to the Everett Police (EP) and 167 to the State Police (SP)) since the opening of Encore on June 23. There was no noticeable increase in the number as you can see below.

everett_date <- everett_police_data %>%
  count(date = date(incident_dttm))

state_date <- state_police_data %>%
  select(date, time, name, result) %>%
  unique() %>%
  count(date)

date_number <- full_join(everett_date, state_date, by = "date") %>%
  mutate(total = n.x + n.y) 

date_number %>%
  ggplot(aes(x=date, y=total, label=total)) +
  geom_line() +
  geom_label()

Which day of week sees most incidents?

Unlike my expectations, there was no huge difference between the number of incidents happened on weekdays and that on weekends. In hindsight, persons who get arrested might not have regular nine-to-five lifestyles.

date_number %>%
  group_by(day = wday(date, label = TRUE)) %>%
  summarise(count = n()) %>%
  arrange(desc(count))
## # A tibble: 8 x 2
##   day   count
##   <ord> <int>
## 1 Sun      18
## 2 Tue      18
## 3 Sat      17
## 4 Mon      16
## 5 Thu      16
## 6 Wed      15
## 7 Fri      15
## 8 <NA>      1

Common types of charges

I created a table of the most common types for each data, but neither is complete as both data are dirty. To make matters worse, the two agencies have different data-keeping methods so the same misconduct could have been reported under different categories. So if you want to know the total number of a specific type of incidents, I recommend you open two excel files - Everett Police and State Police data - type a keyword in the universal search bar, add results from each file.

Please notice: In EP data, each row represents each incident. On the other hand, in SP data, one incident can take up multiple rows if the incident involves multiple violations - for example, if a person trespassed while possessing cocaine.

For your information, I hand-searched some noticeable types: + Disorderly Conduct: 66 cases (33 from EP w/ keyword “Disturbance”, 33 from SP w/ keyword “Disorderly”) + Assault & Battery: 53 cases (9 from EP, 44 from SP) + Larceny: 46 cases (11 from EP, 35 from SP) + Gaming-related: 27 cases (7 from EP w/ keyword “Gaming”, 9 from SP w/ keyword “Cheating”, 11 from SP w/ keyword “Minor Gambling”)

If you are interested in other cases, please skim the two tables below first, and then strategize your search.

everett_police_data %>%
  group_by(incident_type) %>%
  tally() %>%
  arrange(desc(n)) %>%
  datatable()
state_police_data %>%
  group_by(charge) %>%
  tally() %>%
  arrange(desc(n)) %>%
  datatable()

Results

SP data had the “result” column, unlike EP data. To see the percentage of each result, I created a table based on SP data as below.

state_police_data %>%
  select(date, time, name, result) %>%
  unique() %>%
  group_by(result) %>%
  summarise(total = n(),
            percent = n()/167*100) %>%
  arrange(desc(percent))
## # A tibble: 4 x 3
##   result  total percent
##   <chr>   <int>   <dbl>
## 1 arrest    101   60.5 
## 2 summons    55   32.9 
## 3 warrant     7    4.19
## 4 <NA>        4    2.40

I also searched “arrest”, “summons” and “warrant” from EP data. There were 24, 0, 6 cases, respectively. If you are to use result data, you may want to mention SP only as EP doesn’t have complete result data.

When is the peak time?

Most incidents happened after dark.