Identified or conflicted: a latent class and regression tree analysis explaining how identity constructs cluster within smokers

Table 3 Scores on identity variables in the two classes: Descriptive statistics and t-tests (N = 231)

Variable	M (SD)			Mean z-score		t-test
Variable	Overall	Class 1	Class 2	Class 1	Class 2	t-test
Smoker self-identity	2.67 (0.77)	3.10 (0.62)	2.15 (0.60)	0.56	− 0.67	t(223.54) = 11.82, p < 0.001, d = 1.56
Expected identity loss	2.27 (0.97)	2.83 (0.80)	1.60 (0.70)	0.57	− 0.69	t(228.41) = 12.39, p < 0.001, d = 1.64
Non-smoker self-identity	3.47 (0.70)	3.18 (0.51)	3.82 (0.74)	− 0.42	0.49	t(179.15) = − 7.42, p < 0.001, d = 1.01
Quitter self-identity	3.37 (0.70)	3.17 (0.50)	3.62 (0.81)	− 0.29	0.35	t(166.57) = − 4.97, p < 0.001, d = 0.67
Smoker group-identity	3.40 (0.77)	3.56 (0.66)	3.21 (0.85)	0.20	− 0.24	t(194.56) = 3.39, p = 0.001, d = 0.46
Non-smoker group-identity	3.45 (0.66)	3.26 (0.60)	3.68 (0.66)	− 0.29	0.34	t(212.26) = − 4.97, p < 0.001, d = 0.67

Class 1 refers to ‘Identitified smokers’ (n = 126) and Class 2 refers to ‘Conflicted smokers’ (n = 105)

ISSN: 2050-7283