Violence Reduction in Joliet, Illinois: An Evaluation of the Strategic Tactical Deployment Program

44 

 

and .118, respectively).

 

On the other hand, the data indicate increases in the number of robberies for both the STD 

and STD+ components. The increase in robberies during the STD only period was significantly 

different from the pre-STD time period (57% increase, p = .018). However, the increase in 

robberies during the SDT+ intervention period did not differ significantly from the pre-program 

period.  

 

To further assess the potential impact of the STD strategy, we conducted a time series 

analysis for shots fired in the STD target areas only. Specifically, monthly shots fired and 

robberies were assessed for sectors 11, 16 and 22. Figure 6 indicates that the average number of 

monthly shots fired increased by 18% when comparing the pre-STD to the STD time period and 

20% when comparing the pre to the STD+ period.  

Figure 6     

Confirmed Shots Fired for STD Areas (Sectors 11, 16, 22),  

January 2005 – September 2012 

 

---

45 

 

The data in Table 6 indicate that both the STD only and STD+ program components were 

associated with decreases in the monthly number of shots fired.  According to the incidence rate 

ratios, the STD only component had a 22% percent decrease in the monthly number of shots 

fired events and the STD+ component had a 27% reduction.  However, these reductions were not 

significantly different from the pre-intervention period.  The data also indicate that the monthly 

counts of robberies increased during the intervention period.  Although not statistically 

significant, robberies increase 59% and 97% during the STD only and STD+ program periods, 

respectively.  It is important to note that the overall model for monthly robberies was not 

significant, thus indicating a poor model fit.  Most likely, the poor model fit was due to low 

monthly frequencies and too many zero counts.  

Table 6 

Poisson Regression Results for Shots Fired and Robbery 

 STD Areas (Sectors 11, 16, 22) 

   

 

 

 

   Shots Fired  

 

 

                            Robbery 

variable 

B (SE) 

IRR 

p-value 

B (SE) 

IRR 

p-value 

trend 

.003 

(.007) 

1.00 .659 -.008 

(.013) 

.992 .529 

unemployment 

.027 (.023) 

1.03 

.244 

.013 (.045) 

1.01 

.766 

% minority pop. 

1.39 (1.88) 

4.02 

.459 

.933 (3.04) 

2.54 

.759 

# police officers 

.000 (.003) 

1.00 

.993 

.002 (.005) 

1.00 

.703 

Monthly # drug 

arrests 

 

.006 

(.007) 

1.01 .358 -.006 

(.011) 

.994 .602 

STD only 

-.248 (.211) 

.780 

.238 

.464 (.366) 

1.59 

.205 

STD + 

probation/parole  

 

-.321 

(.360) 

.725 .372 .678 

(.629) 

1.97 .281 

intercept 

1.70 (.334) 

--- 

.000 

.539 (.489) 

--- 

.270 

---

46 

 

 

Shots fired:   

Deviance = 203.27, Pearson Chi-square = 191.24 Value/df = 2.39/2.25 

     

 

 

Likelihood ratio Chi-Square = 14.09, df = 7, p = .050 

Robbery:   

Deviance = 169.56, Pearson Chi-square = 159.75 Value/df = 2.00/1.88 

   

     

Likelihood ratio Chi-Square = 7.35, df = 7, p = .394 

The same analysis was conducted for shots fired in the non-STD target areas only. Figure 7 

indicates that the average number of monthly shots fired increased by 15% when comparing the 

pre-STD to the STD time period and 4% when comparing the pre to the STD+ period. 

Figure 7 

Confirmed Shots Fired for Non-STD Areas, January 2005 – September 2012 

 

The data in Table 7 show that neither the STD only nor the STD+ program components 

were associated significantly with reductions in the monthly number of shots fired. The data do 

indicate reductions in monthly shots fired, but these decreases were not significantly related to 

the intervention. This is not surprising since these sectors were not the primary focus of the STD 

intervention. In addition, there was a significant inverse relationship between the number of 

Joliet police officers and shots fired in the non-STD areas.  

 

---

47 

 

Table 7 

Poisson Regression Results for Shots Fired and Robbery, Non-STD Areas 

     Shots 

Fired 

   Robbery 

variable 

B (SE) 

IRR 

p-value 

B (SE) 

IRR 

p-value 

trend 

-.005 (.007) 

.995 

.529 

-.003 (.007) 

.997 

.633 

unemployment 

.002 

(.025) 

1.00 .939 -.037 

(.023) 

.963 .103 

% minority pop. 

2.89 (1.87) 

17.99 

.122 

1.29 (1.63) 

3.64 

.429 

# police officers 

-.005 (.002) 

.995 

.053 

.001 (.002) 

1.00 

.765 

Monthly # drug 

arrests 

 

-.004 

(.004) 

.996 .299 .006 

(.004) 

1.01 .107 

STD only 

-.084 (.217) 

.919 

.699 

.557 (.216) 

1.75 

.010 

STD + 

probation/parole 

 

-.047 

(.379) 

.954 .901 .518 

(.387) 

1.68 .181 

intercept 

3.13 (.231) 

--- 

.000 

1.33 (.304) 

--- 

.000 

  Shots fired:   Deviance = 177.40, Pearson Chi-square = 171.35 Value/df = 2.334/2.255 

     

 

 

Likelihood ratio Chi-Square = 26.946, df = 7, p = .011 

  Robbery: 

Deviance = 137.11, Pearson Chi-square = 129.20 Value/df = 1.61/1.52 

     

 

 

Likelihood ratio Chi-Square = 38.12, df = 7, p = .000 

The data in Table 4 also indicate increases in robberies for both the STD and STD+ 

components. There was a statistically significant increase (75%) in robberies during the STD-

only component as compared to the pre-STD time period.  Similarly, though not statistically 

significant, there was a 68% increase in monthly robbery counts during the STD+ period. 

Difference-in-Difference Poisson Regression Models 

The impact of program outcomes can be estimated by computing a double difference, one 

over time (before-after) and one across subjects (between beneficiaries and non beneficiaries). 

The basic logic behind this 

difference-in-differences (

DiD) estimation is one where outcomes are 

---