Explaining offenders’ longitudinal product-specific target selection through changes in disposability, availability, and value: an open-source intelligence web-scraping approach

Table 3 Estimates of stealing counts-Google Trends search interest and stealing counts-mean price elasticities—2012–2019

	Log (Stealing counts) β	Standard Error	p value
Xbox One
Log (Google trends) lag1	0.846	0.384	< 0.05
Log (Mean price) lag1	0.067	0.671	n.s
Log (Other electronic stealing counts)	1.386	0.744	n.s
Lagged dependent variable	− 0.441	0.113	< 0.001
PlayStation 3	Log (Stealing counts) β	Standard Error	p value
Log (Google trends) lag1	0.389	0.181	< 0.05
Log (Mean price) lag1	− 0.226	0.191	n.s
Log (Other electronic stealing counts)	0.788	0.337	< 0.05
Lagged dependent variable	− 0.472	0.090	< 0.001
Xbox 360	Log (Stealing counts) β	Standard Error	p value
Log (Google trends)	0.528	0.193	< 0.01
Log (Mean price)	0.150	0.237	n.s
Log (Other electronic stealing counts)	0.282	0.410	n.s
Lagged dependent variable	− 0.447	0.089	< 0.001

The variance inflation factor (vif)—a measure of multicollinearity between model variables—was calculated for all model variables and values ranged between 1.00 and 1.12, where a vif above 10 indicates high correlation and is a cause of concern). Variables with the term ‘lag1’ indicate a time series variable where the values are lagged by one month. Β coefficients can be interpreted as a 1% change in the independent variable being associated with a Β% change in stealing counts. For example, a 1% change in Google trends values for the Xbox One console was associated with a 0.85% change in stealing counts of that console

ISSN: 2193-7680