Abstract Purpose – This paper claims to identify the behavioral and cultural features that push to use, or not, cryptocurrencies for electronic commerce. Indeed, despite the use of cryptocurrencies for electronic commerce spreading worldwide at a fast and growing pace, there are supporters and detractors among their users. The analysis of what distinguish these two groups of users is fundamental for understanding their different intention to use cryptocurrencies for electronic commerce. Design/methodology/approach – A survey has been administered to 2,532 cryptocurrencies’ users across the USA and China, collecting data on their behavioral predispositions and cultural features. Results were then analyzed through structured equation modeling. Findings – Results showed that while attitude, subjective norms, perceived behavioral control and herding behavior have a positive impact on the intention to use cryptocurrencies for electronic commerce, financial literacy has no influence. Cultural dimensions amplified or reduced the discovered relationships and caused different effects: positive for the USA and negative for China when considering illegal attitude and perceived risk. Originality/value – Theory of planned behavior, financial behavior and cultural factors can, all together, represent a useful framework for envisioning the behavior of users in adopting cryptocurrencies for electronic commerce purposes through a test of all its elements. To the best of the authors’ knowledge, this is the first study considering behavior and cultural variables on the intention
5. Results 5.1 Measure validation they were initially About investigated through an exploratory principal component factor analysis with a Varimax rotation (Table 2). Component factor analysis is a statistical approach usually implemented for data reduction by creating one or more index variables (often referred to as factors, components and dimensions) from a larger set of measured variables (Field, 2013). the scales, All the items of the questionnaire signicantly loaded on the constructs; using the 0.40 rule-of-thumb, all cross-loadings are low, while the resulting solution explained 78% of the total variability. Then, through the implementation of the PLS, a conrmatory factor the discriminant and convergent validity of Electronic commerce 353 MRR 46,3 354 Table 2. Description of sample data Characteristics Total Average in total the USA (%) China
id: 23a409a2840caa403dca8ae64e8ff039 - page: 14
Gender Men Women 1,483 1,049 59 41 737 580 56 44 746 469 Age 1828 2838 3848 Above 48 810 1,170 389 163 32 46 15 6 397 593 246 81 30 45 19 6 413 577 143 82 Education High school College Bachelors degree Masters degree PhD 173 416 1,666 277 0 7 16 66 11 0 63 195 926 133 0 5 15 70 10 0 110 221 740 144 0 Income Less $10,000 From $10,000 to $30,000 From $30,000 to $50,000 From $50,000 to $70,000 Above $70,000 279 1,277 557 228 191 11 51 22 9 8 120 632 336 119 110 9 48 26 9 8 159 645 221 109 81 Frequency of cryptocurrencies use in e-commerce (times in a month) 0 0 15 349 610 1305 112,015 358 1620 267 >20 253 0 14 51 14 11 10 0 185 696 175 152 109 0 14 53 13 12 8 0 164 609 183 115 144
id: 1ffdc2d4d53223f70a819675e3ac4564 - page: 15
As usually done, discriminant validity is shown when the square root of each constructs average variance extracted (AVE) is larger than its correlations with other constructs. Tables 3 and 4 report that the square root of the AVE is much larger than all other crosscorrelations for both the USA and China samples. Tables 3 and 4 report the: (cid:1) (cid:1) reliability coefcients and AVE values; correlation matrix and descriptive statistics of the studys principal factors for the USA and China; and (cid:1) Cronbachs alpha coefcient for each construct as to measure reliability (for both countries they are above 0.70).
id: 386d3cc1417dd0b370d092b30bc39e71 - page: 15
5.2 Hypotheses testing The hypotheses developed in the prior sections have been tested by recurring to a PLS. PLS is able to specify the relationships among the principal construct, as well as with their (%) 61 39 34 47 12 7 9 18 61 12 0 13 53 18 9 7 0 13 50 15 9 12 a x i r t a m t n e n o p m o c d e t a t o R t n e n o p m o C 8 7 6 5 4 3 2 1 s m e t I s e l b a i r a v g n i t a i t n a t s b u s s n o i t s e u Q 7 7 6 0 . 1 T A y m e s a e r c n i l l i w s e i c n e r r u c o t p y r c g n i s U e d u t i t t A e m r o f s l a o g t n a t r o p m i e v e i h c a o t s e i t i n u t r o p p o 3 2 6 0 . 3 T A y m e v e i h c a e m p l e h l l i w s e i c n e r r u c o t p y r c g n i s U y k c i u q e r o m s l a o g l 7 0 6 0 . 5 T A f o d r a d n a t s y m e s a e r c n i l l i w s e i c n e r r u c o t p y r c g n i s U g n v i i l 9 0 7 0 . 2 N S I t a h t k n h t i l l i w e m o t t n a t r o p m
id: a1aa661173ea88a088e9de3a85d61f97 - page: 15