Table 5

NCA Results: necessary condition effect sizes and significance tests on the perceptions PLS-SEM latent variable scores

Necessary condition effect sizes and significance tests
TangiblesReliabilityResponsivenessAssuranceEmpathy
# observations234234234234234
Scope23.44431.30027.87623.88825.480
Xmin−4.222−6.097−5.628−4.634−4.738
Xmax1.6391.7281.3411.3381.632
Loyaltymin11111
Loyaltymax55555
 ce_fdhcr_fdhce_fdhcr_fdhce_fdhcr_fdhce_fdhcr_fdhce_fdhcr_fdh
Ceiling zone2.9741.4878.2844.1425.2702.8604.7443.1655.1302.565
Effect size (d)0.1270.0630.2650.1320.2050.1030.1990.1330.2010.101
# above0000000600
c-accuracy100%100%100%100%100%100%100%97.4%100%100%
p-value0.0060.0130.0240.0390.0180.0190.0030.0000.0180.018
p-accuracy0.0020.0020.0030.0040.0030.0030.0010.0000.0030.003

Note(s): CE-FDH = Ceiling Envelopment – Free Disposal Hull. CR-FDH = Ceiling Regression – Free Disposal Hull. We consider the general benchmark of the necessity effect size (d): 0 < d < 0.1 “small effect”; 0.1 ≥ d < 0.3 “medium effect”; 0.3 ≥ d < 0.5 “large effect”, and d ≥ 0.5 “very large effect”. The p-values reported were estimated with 10,000 permutations and are treated as significant if < 0.05 P-accuracy: Accuracy of p-value estimated by approximate permutation as a function of number of permutations and estimated p-value. 95% confidence that exact p value = estimated p value ± p accuracy

Source(s): Authors’ own creation

or Create an Account

Close Modal
Close Modal