Does gender affect who is frisked during a search?

Analyzing Police Activity with pandas

Kevin Markham

Founder, Data School

ri.search_conducted.value_counts()

False    83229
True      3307

ri.search_type.value_counts(dropna=False)

NaN                               83229
Incident to Arrest                 1290
Probable Cause                      924
Inventory                           219
Reasonable Suspicion                214
Protective Frisk                    164
Incident to Arrest,Inventory        123
...

.value_counts() excludes missing values by default
dropna=False displays missing values

Examining the search types

ri.search_type.value_counts()

Incident to Arrest                           1290
Probable Cause                                924
Inventory                                     219
Reasonable Suspicion                          214
Protective Frisk                              164
Incident to Arrest,Inventory                  123
Incident to Arrest,Probable Cause             100
...

Multiple values are separated by commas
219 searches in which "Inventory" was the only search type
Locate "Inventory" among multiple search types

Searching for a string (1)

ri['inventory'] = ri.search_type.str.contains('Inventory', na=False)

str.contains() returns True if string is found, False if not found
na=False returns False when it finds a missing value

Searching for a string (2)

ri.inventory.dtype

dtype('bool')

True means inventory was done, False means it was not

ri.inventory.sum()

Calculating the inventory rate

ri.inventory.mean()

0.0050961449570121106

0.5% of all traffic stops resulted in an inventory

searched = ri[ri.search_conducted == True]
searched.inventory.mean()

0.13335349259147264

13.3% of searches included an inventory

Let's practice!

Analyzing Police Activity with pandas