Responding to victimisation in a digital world: a case study of fraud and computer misuse reported in Wales

Crime Science

Table 3 Regression parameters for multinomial linear model age category ~ crime category

Age category	Parameter (crime category)	Estimate (β)	Standard error	Wald statistic	Odds-ratio
37–52	(Intercept)	− 0.09	0.06	− 1.4	0.92
	Card and banking	− 0.24	0.14	− 1.8	0.79
	Consumer fraud	0.14	0.08	1.9	1.15
	Hacking	0.02	0.12	0.16	1.02
	Investment fraud	0.98	0.98	3.2	2.7
	Malware virus DDOS	− 0.09	− 0.09	− 0.59	0.91
	Other fraud	− 0.01	− 0.01	− 0.12	0.99
	Services fraud	− 0.04	− 0.04	− 0.26	0.96
53–66	(Intercept)	0.18	0.06	3.0	1.20
	Card and banking	− 0.73	0.14	− 5.2	0.48
	Consumer fraud	− 0.16	0.07	− 2.2	0.85
	Hacking	− 0.80	0.13	− 6.06	0.45
	Investment fraud	1.31	0.28	4.6	3.7
	Malware virus DDOS	− 0.32	0.15	− 2.08	0.73
	Other fraud	− 0.28	0.09	− 3.04	0.75
	Services fraud	− 0.76	0.18	− 4.23	0.47
> 67	(Intercept)	0.43	0.06	7.6	1.54
	Card and banking	− 1.03	0.14	− 7.3	0.36
	Consumer fraud	− 0.60	0.07	− 8.6	0.55
	Hacking	− 1.47	0.15	− 9.98	0.23
	Investment fraud	1.02	0.28	3.6	2.8
	Malware virus DDOS	− 1.36	0.19	− 7.21	0.26
	Other fraud	− 0.45	0.09	− 5.04	0.64
	Services fraud	− 1.69	0.22	− 7.60	0.19

Model: \(\begin{aligned} &\text{log}\left( {\frac{{\Pr Y = \varvec{ }j}}{{\Pr Y = \varvec{ }j^{\prime}}}} \right) \\ & = \beta_{0} + \beta_{1} \;Crimecategory\left( {Card \, and \, Banking} \right) + \beta_{2} \;Crimecategory\left( {Consumer \, fraud} \right) \\ & \ + \beta_{3} \;Crimecategory\left( {Hacking} \right) + \beta_{4} \;Crimecategory\left( {Investment \, fraud} \right) \\ & \ + \beta_{5} \;Crimecategory\left( {Malware, \, Virus \, and \, DDOS} \right) + \beta_{6} \;Crimecategory\left( {Other \, fraud} \right) \\ & \ + \beta_{7} \;Crimecategory\left( {Services \, fraud} \right) \end{aligned}\) where j′ = reference category (age category = 0–36 years old)
\(\beta_{0} = Advance fee fraud\)

ISSN: 2193-7680