Table 3.

Hierarchical modelling: Associations between environmental characteristics and customer intoxication ratings.

				Multivariate
		Bivariate		Block analysis		Model 1		Model 2
	Variable	Estimate	P	Estimate	P	Estimate	P	Estimate	P
Contextual variables #	>100 customers	0.945	***			0.037	ns	0.139	ns
	Later visit	1.223	***			0.483	*	0.740	***
Venue entrance	Door staff	1. 017	**	0.496	ns
	Queue	0.715	*	−0.229	ns
	Entrance fee	0.823	*	0.124	ns
	House rules (entry)	0.201	ns	0.142	ns
Physical environment	Seating	0.240	***	0.062	ns
	Noise level	0.282	***	0.060	ns
	Movement/Crowding	0.191	***	0.087	*	0.025	ns	0.056	ns
	Ventilation/Lighting	0.280	***	0.092	ns
	Temperature	0.380	***	0.058	ns
	Clearing/Cleanliness	0.139	***	0.017	ns
	Glass on floor	0.296	***	0.030	ns
	Toilets	0.316	***	0.128	*	0.097	*	0.103	*
Bar activities	Dancefloor	1.252	***	0.993	***	0.269	ns	0.557	*
	Pool tables	−0.046	ns	−0.181	ns
	TV screens	0.282	ns	0.569	*	0.107	ns	0.266	ns
	House rules (inside)	−0.132	ns	−0.093	ns
	Rock/heavy music	−0.312	ns	−0.026	ns
	Rap/hiphop music	0.080	ns	−0.217	ns
	Pop/dance music	0.115	ns	−0.286	ns
Alcohol and food service	Alcoholic drink promotions	0.297	ns	0.336	ns
	Low drink prices	−0.350	ns	−0.344	ns
	Soft drink promotions	0.888	**	0.833	**	0.631	*	0.690	**
	Plastic glassware	0.706	**	0.818	**	0.602	**	0.614	**
	Table service	−0.936	**	−0.882	**	0.031	ns	−0.090	ns
	Food service	−1.183	*	−0.394	ns
Venue staff	Fewer bar staff	0.345	ns	−0.027	ns
	Young staff	−0.084	ns	0.020	ns
	Male staff	0.406	ns	0.202	ns
	Glass collectors	0.539	*	0.235	ns
	Staff monitoring	0.209	***	0.163	**	0.071	ns	0.081	ns
	Staff coordination	0.024	ns	−0.113	ns
	Staff attitude	0.206	*	0.181	ns
	Staff boundaries	0.130	*	0.052	ns
	Permissiveness	0.526	***	0.425	***	0.160	*	0.298	***
Customer factors	Male clientele	−0.017	ns	−0.018	ns
	Young clientele	0.886	**	0.590	*	0.316	ns
	Single sex groups	0.089	ns	−0.081	ns
	High alcohol drinks	0.181	ns	0.047	ns
	Dancing	0.276	***	0.126	**	0.073	ns
	Sexual activity/competition	0.237	***	0.085	*	0.065	*
	Rowdiness	0.460	***	0.243	***	0.125	ns

Analysis uses hierarchical modelling. # These two variables were included in all block analyses. ns = not significant; * P < 0.05; ** P < 0.01; *** P < 0.001. For significant associations in multivariate analyses, slope direction indicates whether the variable was associated with an increase or decrease (-) in intoxication rating.