Table 3 Longitudinal multilevel model for total number of cigarette advertisements* per store predicted by time and neighbourhood characteristics, controlling for store type

	Coefficient	Robust standard error	p Value
Fixed effects
For intercept:
Intercept	21.77	1.54	0.001
For Slope (growth rate):
Intercept	−0.68	0.54	0.205
Convenience store with petrol	0.45	0.44	0.302
Convenience store without petrol	2.77	0.55	0.001
Large pharmacy	0.19	0.57	0.735
Petrol only	−1.63	0.55	0.003
Liquor store	2.26	0.51	0.001
Supermarket	5.26	1.30	0.001
Population density quartile 4	−1.37	0.41	0.001
SES†	−0.33	0.25	0.189
High proportion of African–Americans	2.45	0.48	0.001
High proportion of Asian/Pacific Islanders	−0.75	0.46	0.108
High proportion of Hispanics	−0.72	0.42	0.088
Random effects
Intercept variance (R₀)	540.43		0.001
Slope (growth rate) variance (R₀₁)	40.04		0.001

*Advertisments include interior and exterior signs, shelving units, displays, and interior and exterior functional items.
†SES (socioeconomic standing) is a composite measure; higher scores equal higher socioeconomic standing.