CENTER FOR PUBLIC SAFETY INITIATIVES, ROCHESTER INSTITUTE OF TECHNOLOGY
Demographics and Crime Clearance Rates in Rochester Variable #1: Unemployment Rate
Variable #2: % of Single Parent HH
Variable #4: Mean HH Income
Variable #3: % of Immigrant Population
SnglPrntHH Vs Burglary Clearance Rate
UnempRate Vs Burglary Clearance Rate
Variable #5: % of Non-White HH
MeanHHI Vs Burglary Clearance Rate
ImmigPct Vs Burglary Clearance Rate
50
45
NonWhtHHPc Vs Burglary Clearance Rate
35
120000
120
30
100000
100
45
40
40
35
35
30
25
30
25
R² = 0.0133
20
R² = 0.0061
25
UnempRate
Linear (SnglPrntHH)
15
10
10
5
5
0
5
10
15
20
25
30
5
10
Linear (ImmigPct)
60
MeanHHI
Linear (MeanHHI)
40000
NonWhtHHPc
Linear (NonWhtHHPc)
40
20000
15
20
25
30
20
0
0
35
0
5
10
Clearance Rate Percent
Clearance Rate Percent
UnempRate Vs Robbery Clearance Rate
15
20
25
30
0
0
35
5
10
ImmigPct Vs Robbery Clearance Rate
50
15
20
25
30
35
0
5
10
Clearance Rate Percent
Clearance Rate Percent
SnglPrntHH Vs Robbery Clearance Rate
45
R² = 0.0069
60000
ImmigPct
5
0
35
80
R² = 0.0179
R² = 0.0092
15
10
0
0
80000
20
SnglPrntHH
20
Linear (UnempRate)
15
20
25
30
35
Clearance Rate Percent
NonWhtHHPc Vs Robbery Clearance Rate
MeanHHI Vs Robbery Clearance Rate
35
15
120
120000
45
40
30
40
35
35
R² = 0.0172
30
UnempRate
20
Linear (UnempRate)
15
R² = 0.007
SnglPrntHH
20
Linear (SnglPrntHH)
ImmigPct
15
Linear (ImmigPct)
R² = 0.0227
10
60000
60
Linear (MeanHHI)
40000
5
0
0
0
10
20
30
40
50
60
70
0
Clearance Rate Percent
10
20
30
40
50
60
70
0
20
0
0
0
10
20
Clearance Rate Percent
30
40
50
60
70
Clearance Rate Percent
Linear (NonWhtHHPc)
40
20000
5
NonWhtHHPc
MeanHHI
10
5
R² = 0.0077
80
80000
20
25
15
10
100
100000
25
R² = 0.0073
30
25
0
10
20
30
40
50
60
0
70
10
20
30
40
50
60
70
Clearance Rate Percent
Clearance Rate Percent
Descriptive Stats:
Burglary Clearance Rate
Mean: 12.9% Min: 5.3%
Max: 30%
Std. Dev.: 4.4
National Average : 11.95%
Average for Cities similar in popu-­
lation size: 11.2%
Descriptive Stats:
Robbery Clearance Rate
Mean: 30.6%
Min: 0%
Max: 60%
Std. Dev.: 11.3
National Average : 25.95%
Average for Cities similar in popu-­
lation size: 25.95%
Variable #7: % of Owner Occupied HH
Variable #6: Population Density
PopDensity Vs Burglary Clearance Rate
OwnOccPct Vs Burglary Clearance Rate
16000
90
14000
80
12000
70
60
10000
R² = 0.0137
8000
PopDensity
6000
Linear (PopDensity)
4000
2000
0
0
5
10
15
20
25
30
35
16000
14000
12000
R² = 0.0006
8000
PopDensity
6000
R² = 0.0037
50
OwnOccPct
40
Linear (OwnOccPct)
30
20
10
0
0
Clearance Rate Percent
PopDensity Vs Robbery Clearance Rate
10000
The geographical patterns for the clearance rates for both crime categories were found to be normally distributed and no existence of spatial autocorrelation was substantiated through a Global Moran’s I test.
This bolstered the hypothesis that clearance rates vary ran-­
domly across space and thus the are less likely to be explained by variations in population demographics, which happen to have clustering through space as evident from the demograph-­
ic maps above and the correlation matrix for the explanatory variables below
5
10
15
20
25
30
35
Clearance Rate Percent
OwnOccPct Vs Robbery Clearance Rate
90
80
70
60
R² = 0.0014
50
OwnOccPct
40
Linear (PopDensity)
Linear (OwnOccPct)
30
4000
20
2000
10
0
0
10
20
30
40
50
60
0
70
0
Clearance Rate Percent
The purpose of this study was to explore the strength of the relationship between crime clearance rates and demographic variables and how well the clearances rates in a geography could be explained by the demographics of the population residing in the same geography. We chose the City of Rochester, NY for our study. The crime categories considered were Burglary and Robbery. Seven demographic variables were considered : mean household income, percent of owner occupied houses, percent of single parent households, percent of non-white households, unemployment rate, percent of immigrant popula-­
tion, and population density. The level of geographic resolution was at the census tract level. The results of the linear regression analysis showed that clearance rates for either robbery or burglary could not be explained by demographic variables alone at all. None of the explanatory demographic variables on their own had any significant explana-­
tory power in the explaining the variation in the de-­
pendent variables –the clearance rates of robbery and of burglary. Also, ordinary least square models as a whole, comprising all the demographic varia-­
bles could not explain for any significant variation in the clearance rates for either of the crime categories. The addition of a “workload” variable didn’t bring about a change in the explanatory power either. 10
20
30
40
50
60
Clearance Rate Percent
OLS Model #1:
Burglary Clearance Rate = f (all demographic variables)
OLS Model #2:
Robbery Clearance Rate = f (all demographic variables)
R2 : 0.035
Pr>F : 0.923
No explanatory variables were statistically significant
R2 : 0.058
Pr>F: 0.743
No explanatory variables were statistically significant
Arindam Ghosh, CPSI, RIT
John McCluskey, CPSI, RIT
OLS Model #3:
Burglary Clearance Rate = f (all demographic variables, Number of Burglaries)
OLS Model #4:
Robbery Clearance Rate = f (all demographic variables, Number of Robberies)
R2 : 0.037
Pr>F : 0.951
No explanatory variables were statistically significant
R2 : 0.059
Pr>F: 0.829
No explanatory variables were statistically significant
70