 # International Journal of Security and Terrorism •

Yüklə 194,05 Kb.

 səhifə 5/7 tarix 14.09.2018 ölçüsü 194,05 Kb.
 30 International Journal of Security and Terrorism • Volume: 4 (1) (OLR) model. This model is “estimated by a method called Maximum Likelihood Estimation  (MLE)” (Aldrich & Nelson, 1984: 49) that deals with “picking parameter estimates that imply  the highest probability or likelihood of having obtained the observed sample Y” (p. 51; see  also Agresti, 2002).  An ordered response model can be developed as a linear probability model with the  use of a continuous latent 5  variable (Long & Freese, 2006). One assumption of the ordered  response model is that an unmeasured (latent) variable, y * , that ranges from  – ∞  to + ∞  exists, and is “mapped to an observed variable y” (Long, 1997: 116), where the mapping  from the latent variable is done with the response categories of not utilized, utilized some,  and highly utilized in the current study. This division of y* into three “values of the observed  y” (Long, 1997: 117) is done by thresholds, or cut points, that are denoted as  τ . Here a  linear equation model is created by using log odds, where standardized coefficients are used  the same way that the coefficients are used in a linear regression model (LRM).  Another way of interpreting the results of ordinal-level outcomes is to use a non- linear model—specifically, odds ratios—rather than standardized coefficients. In that regard,  one  assumption  of  the  ordinal  regression  model  is  the  “parallel  regression  assumption  [italics original] and, for the ordinal logit [logistic regression] model, the proportional odds  assumption” [italics original] (Long & Freese, 2006: 197) where the intercepts may change,  but the coefficients for the independent variables are unchanged for each equation (see also  Powers & Xie, 1999). When “the assumption of parallel regressions is rejected, alternative  models  should  be  considered  that  do  not  impose  the  constraint  of  parallel  regressions”  (Long, 1997: 145). All of the three models in this study (i.e., command, patrol, and detective  models) were tested with the likelihood-ratio test, particularly with a model 6 command in  Stata (Long & Freese, 2006; Wolfe & Gould, 1998). It is “an omnibus test that the coefficients  for all variables are simultaneously equal” (Long & Freese, 2006: 199), and it evaluates  “how the log likelihood of the ORM would change if the constraint…was removed” (Long,  1997:  143).  The  test  results  also  showed  that  all  three  models  in  the  current  study  are  appropriate for the data. 3.2. Findings 3.2.1.Descriptive Statistics The  descriptive  statistics  of  all  variables  (i.e.,  dependent,  explanatory  independent,  and  control) used in the models are presented in the Table 2. 5   In this section, the “latent” concept does not represent the same thing as it represented in the factor  analysis section.  6   This is not an official Stata command. 31 Intelligence - Led Policing: How the Use Of Crime Intelligence Analysis Translates in to the Decision-Making  Table 2. Descriptive Statistics for Dependent, Explanatory Independent, and Control Variables 3.2.2. Multivariate Analysis As indicated in the previous sections, the results from nonlinear models are not easy to  interpret.  The  ordered  logit  is  mostly  “interpreted  in  terms  of  odds  ratios  for  cumulative  probabilities” (Long, 1997: 138). In the current study, the results of ordered logit models are  presented in terms of the percent change in odds. In other words, the percent change of the  metric values in the odds of “higher versus lower outcomes” (Long & Freese, 2006: 218) in  the dependent variable will be provided in this section.  Variable N Measure  Min. Max. Mean SD Dependent  Command-level use CAa efforts 519 Ordinal 0 2 1.28 0.621 0 = not utilized Detectives use CAa efforts 521 Ordinal 0 2 1.25 0.627 0 = not utilized 1 = utilized some 2 = highly utilized Patrol officers use CAa efforts 518 Ordinal 0 2 1.05 0.607 0 = not utilized 1 = utilized some 2 = highly utilized Explanatory Factor 1:  Statistical analysis 423 Continuous -2.02 3.94 0 1 Factor 2:  Crime analysis 423 Continuous -2.29 3.3 0 1 Factor 3:  Intelligence analysis 423 Continuous -2.21 3.64 0 1 Factor 4:  Survey analysis 423 Continuous -1.94 4.21 0 1 Factor 5:  Patrol strategy analysis 423 Continuous -2.56 3.12 0 1 Factor 6:  Displacement analysis 423 Continuous -3.06 3.74 0 1 Control Crime analysis unit  517 Dummy 0 1 0.65 0.479 (yes = 1, no = 0) Unions in the agency 535 Dummy 0 1 0.61 0.489 (yes = 1, no = 0) Agency size  493 Continuous 0.09 6.32 1.637 0.983 (# of sworn x 1,000  ⁄ population) Total operating budget  535 Continuous 4,554 1,492,567 153,366 112,401 (dollars in 12-month period) Organizational hierarchy b  535 Continuous 0.19 4.65 1.33 0.627 [(min - max salary)  ⁄ min salary] Crime rates  500 Continuous 0 421.47 61.27 39.17 (# of crimes x 1,000  ⁄ population) a CA = crime analysis.   b Unit in dollars. Dostları ilə paylaş:

Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©genderi.org 2017
rəhbərliyinə müraciət Ana səhifə