Influence of premium vs masked cigarette brand names on the experienced taste of a cigarette after tobacco plain packaging in Australia: an experimental study

Skaczkowski, Gemma; Durkin, Sarah; Kashima, Yoshihisa; Wakefield, Melanie

doi:10.1186/s12889-018-5200-8

Table 4 Regressions with objective (ratings from the masked cigarette) and expected ratings predicting ratings of experienced enjoyment, quality and harshness when the brand name was known

From: Influence of premium vs masked cigarette brand names on the experienced taste of a cigarette after tobacco plain packaging in Australia: an experimental study

	Perceived Enjoyment of the branded cigarette (n = 70)^a
	Unadjusted Model			Adjusted Model^{A, B}
Predictor variables	b	[95% CI]	β	b	[95% CI]	β
Objective Enjoyment (ratings of the masked cigarette)	0.20	[0.01, 0.39]	.24*	0.23	[0.05, 0.41]	.28*
Expected Enjoyment	0.32	[0.11, 0.52]	.34**	0.31	[0.11, 0.51]	.34**
Cigarette order				−9.29	[−17.59, − 1.00]	−.24*
Difference in number of puffs between conditions				7.58	[−.19, 15.36]	.20
	Perceived Quality of the branded cigarette (n = 71)^b
	Unadjusted Model			Adjusted Model
Predictor variables	b	[95% CI]	β	b	[95% CI]	β
Objective Quality (ratings of the masked cigarette)	0.06	[−0.17, 0.28]	.06	0.08	[−0.13, 0.30]	.09
Expected Quality	0.40	[0.13, 0.66]	.35**	0.46	[0.21, 0.72]	.42**
Cigarette order				−7.64	[−16.40, 1.12]	−.19
Difference in number of puffs between conditions				10.82	[2.43, 19.20]	.28*
	Perceived Harshness of the branded cigarette (n = 71)^c
	Unadjusted Model			Adjusted Model^{C, D, E, F}
Predictor variables	b	[95% CI]	β	b	[95% CI]	β
Objective Harshness (ratings of the masked cigarette)	0.08	[−0.18, 0.33]	.07	0.11	[−0.14, 0.36]	.10
Expected Harshness	0.18	[−0.05, 0.40]	.19	0.19	[−0.02, 0.41]	.21^†
Cigarette order				12.68	[.80, 24.55]	.25*
Difference in number of puffs between conditions				−1.62	[− 12.60, 9.37]	−.03

Note: There was no indication of multicollinearity in any model, with correlations between objective measures being low: r = .24 (p = .043) for enjoyment, r = .35 (p = .003) for quality, and r = .09 (p = .454) for harshness. VIF values from the adjusted regression models ranged from 1.02 to 1.18, further suggesting that multicollinearity was not a concern
^aUnadjusted model: R² = .22, F(2,67) = 9.19, p < .001); Adj. model: R² = .31, F(4,65) = 7.17, p < .001)
^bUnadjusted model: R² = .14, F(2,68) = 5.70, p = .005); Adj. model: R² = .24, F(4,66) = 5.32, p = .001)
^cUnadjusted model: R² = .04, F(2,68) = 1.54, p = .223); Adj. model: R² = .11, F(4,66) = 1.94, p = .115)
^ASensitivity 1: Objective enjoyment no longer predicted perceived enjoyment when the brand variant name was known (Adj. model: β = .18, t(52) = 1.38, p = .174)
^BSensitivity 4: Objective enjoyment no longer predicted perceived enjoyment when the brand variant name was known (Adj. model: β = .17, t(59) = 1.43, p = .157)
^CSensitivity 1: Expected harshness no longer tended to predict perceived harshness when the brand variant name was known (Adj. model: β = .19, t(53) = 1.41, p = .164)
^DSensitivity 2: Objective harshness tended to predict perceived harshness when the brand variant name was known (Adj. model: β = .22, t(56) = 1.73, p = .089). Expected harshness no longer tended to predict perceived harshness when the brand variant name was known (Adj. model: β = .14, t(56) = 1.11, p = .272)
^ESensitivity 3: Expected harshness no longer tended to predict perceived harshness when the brand variant name was known (Adj. model: β = .18, t(61) = 1.53, p = .132)
^FSensitivity 4: Expected harshness no longer tended to predict perceived harshness when the brand variant name was known (Adj. model: β = .19, t(60) = 1.56, p = .123)
** p < .01. * p < .05, † p < .10

Back to article page

ISSN: 1471-2458

Contact us

Submission enquiries: bmcpublichealth@biomedcentral.com
General enquiries: ORSupport@springernature.com

BMC Public Health

Contact us