Table 7

Multilevel model specifications and equations

DescriptionSpecification
Equation 1 Intercept-only modelBATTij = γ00 + u0j + rij
Equation 2 Baseline modelBATTij = γ00 + γ10*PBGij + γ20*PBLij + γ30*BFAMij + γ40*CONij + u0j + rij
Equation 3 Cross-level modelBATTij = γ00 + γ01*BS(W/C)j + γ02*BOS(W/C) j + γ03*BBS(W/C) j + γ10*PBGij + γ20*PBLij + γ30*BFAMij+ γ40*CONij + u0j + rij
Note(s):

BATTij is brand attitude (dependent variable) for observation i in group j, γ00 is the fixed regression coefficient for the intercept of the regression equation, u0j is the random regression coefficient for the intercept of the regression equation for group j, rij is the observation- and group-specific residual, PBGij is perceived brand globalness (Level 1 predictor) for observation i in group j, PBLij is perceived brand localness (Level 1 predictor) for observation i in group j, BFAMij is brand familiarity (Level 1 predictor) for observation i in group j, CONij is representing a vector of controls – age and gender (Level 1 controls), γ10 is the fixed regression coefficient for the main effect of PBGij, γ20 is the fixed regression coefficient for the main effect of PBLij, γ30 is the fixed regression coefficient for the main effect of BFAMij, γ40 is the fixed regression coefficient for the effect of vector of controls CONij, BS(W/C)j represents brand warmth or competence for group j, γ01 is the fixed regression coefficient for the main effect of BS(W/C)j, BOS(W/C)j represents brand origin warmth or competence for group j, γ02 is the fixed regression coefficient for the main effect of BOS(W/C)j, BBS(W/C)j represents brand buyer/user warmth or competence for group j and γ03 is the fixed regression coefficient for the main effect of BBS(W/C)j

Source(s): Authors’ own work

or Create an Account

Close Modal
Close Modal