ABSTRACT
The data sources offered multiple measures of COP training and citizen interaction, permitting two first-order latent constructs. These fit the data well. Table 8.1 shows the training (α = .001) and citizen interaction (α = .05) constructs were both statistically significant based on their construct variances and z-statistics. With recruit COP training and group meetings serving as a scale to their respective constructs, 1 all the measures were statistically significant at the p<0.001 level and, as expected, positive. Overall, the relationships between the measures and their corresponding constructs were strong. The weakest was that of data accessibility for citizen interaction, but even here the standardized factor loading, which indicates the strength and direction of the construct-measure relationship on a scale of –1 to 1, was 0.4235, significant at p<0.001, and the construct explained about 18 percent of the variance, as indicated by the R-square value. Estimates of the Second-Order COP Measurement Model, 1997 https://www.niso.org/standards/z39-96/ns/oasis-exchange/table">
Component Fit
Standardized Factor Loadings
Z-ratio
R-square
Construct Variance
Z-ratio
First-Order
COP Training
0.3414 ***
5.9530
Recruit COP Training a
0.8309
0.6903
Sworn COP Training
0.9346 ***
15.4194
0.8735
Nonsworn COP Training
0.7870 ***
17.1993
0.6194
Citizen Interaction
0.0686 *
2.2354
Citizen Training
0.7068 ***
8.3839
0.4995
Group Meetings a
0.5342
0.2853
Website
0.4547 ***
6.3903
0.2068
Data Accessibility
0.4235 ***
7.2716
0.1794
Second-Order
COP Implementation
0.6428 ***
9.5641
COP Training
0.7110 ***
8.7791
0.5055
Written COP Plan
0.5583 ***
9.1577
0.3117
Fixed Assignment
0.4835 ***
9.2243
0.2338
Problem-Solving a
0.8017
0.6428
Citizen Interaction
0.8716 ***
7.9330
0.7597
Model Fit
χ 2/df (p-value)
59.3909/32 (0.0023)
RMSEA
0.0431
IFI
0.9784
NFI
0.9543
Standardized RMR
0.0614
CFI
0.9782
AGFI
0.9827
Served as the scale for the latent construct.
p<.05,
p<.001.