Fashion Consumption of Naturally Dyed Products: A Cross-Cultural Study of the Consumption of Blue-Dyed Apparel Between China and Japan

Consumers’ preference for blue-dyed products embodies natural feelings and aesthetics; it is inseparable from the development of a country’s traditional handicrafts and a deep reflection on the pollution of printing and dyeing water [17,18]. In China, naturally blue-dyed apparel can be divided into tradition and ethnic styles. This clothing is represented by brands such as NAMSAN, Indigohood and A Life On The Left, which emphasise the authenticity of the handwork, based on ancient dyeing techniques, and are committed to promoting traditional culture. However, Chinese consumers’ acceptance of natural blue dyeing remains limited, and is manifested in a lack of a mature brand design and operational system. Current consumers are also limited to high-income earners [19,20]. Compared to blue-dyed garments, the blue dye industry in ethnic minority areas is more systematic, but mainly focuses on producing ethnic apparel [21,22].

Japanese natural blue-dyed brands have formed a symbolic feature that uses natural blue dyeing craftsmanship as a selling point. It also combines blue dyeing with contemporary and trendy lifestyles [20].

Japan’s industrialisation of natural blue products predates that of China and South Korea. Most Japanese natural blue-dyed product stores are located in Tokushima, where indigo plants are commonly grown and the production, research and sales are integrated, and its neighbouring regions Kyoto, Osaka and Okayama. Unlike Chinese brands, Japanese blue-dyed product stores do not focus on distributing clothing products in urban high-end shopping malls and on e-commerce platforms. Moreover, in Japanese shopping malls, natural dyed products account for 89.1% of available clothing, which is higher than that of China and South Korea. The price range is diversified and the product categories are rich, from initial retro tannins to t-shirts, sweaters and dresses [18,20]. Nevertheless, the popular patterns of blue-dyed products in China and Japan are similar [23].

Generally, China’s natural blue dye industry adheres to historical traditions and national culture, whereas Japan to the current lifestyle. Differences remain in the development of the trend of natural blue dye between the two countries.

TPB examines consumer intentions, attitude, subjective norms and PBC to make inferences pertaining to actual conduct [10]. It is an extension of the theory of rational behaviour (TRA) [24]. TPB incorporates measures to control beliefs and PBC. In TRA, behavioural intention is predicted through attitude and social norms [25,26]. Attitude refers to the evaluation of a specific behaviour; the more positive the consumers’ attitude, subjective norms and PBC, the more willing they will be to implement the behaviour [27]. In terms of sustainable fashion consumption, the impact of positive attitude on intentions has been fully demonstrated as an important predictor of purchase behaviour [28]. Attitude formation includes recognition of sustainable fashion ethics, brand culture recognition and the evaluation of brand ecological value. Among them, inner emotional value impacts attitude the most [29,30]. Natural blue-dyed fashion has ecological value, reflecting cultural concepts of harmonious coexistence and eliciting a strong value of craftsmanship emotion [22]. Therefore, in accordance with the logic of TPB, consumers’ moral and emotional recognition of ecological protection contribute to a positive attitude and lead to a higher intention to pay product premiums [31].

H1: The more positive the attitude toward natural blue dyeing, the stronger the intention to consume such fashion products.

Subjective norms, the second construct of TPB, is defined as the personal mandate to ‘perform or not to perform a certain behaviour because of social pressure perceived’ [32,25]. Subjective norms indicate consumer intentions derived from self-references to certain groups or individuals [33]. They also represent a dominant factor in the formation of consumers’ ethical purchasing intentions vis-à-vis fashion products [30]. Subjective norms emerge from two aspects of social pressure, namely, the construction of identity to meet the requirements of social life and an intrinsically motivated rationale to follow trends. The positive impact of subjective norms on sustainable consumption intentions has also been confirmed by many studies [34,35]. Given the external characteristics of fashion, consumption decisions are not only based on life needs but even more so on self-expression. Consumers insinuate tastes, values and social recognition through the cultural symbol of personal outerwear [36,37]; hence, compared to attitude, the perception of social pressure may have a stronger influence on consumption intentions [38]. Therefore, the generation of consumers’ intention to consume natural blue-dyed apparel is inseparable from the influence of social factors.

H2: The stronger the awareness of subjective norms, the stronger the intention to consume natural blue-dyed fashion products.

The third factor of TPB, that is PBC, has an impact on both consumption intention and actual behaviour. PBC can predict conduct that is not completely subject to volition and explain how individuals perceive their ability to engage in certain behaviours based on the discernment of constraints or favourable conditions that inform consumption behaviour [10]. PBC reflects consumers’ actual purchasing power, consumption resources and opportunities, as well as other actual control conditions. Therefore, when there are more convenient access opportunities and purchase conditions to buy natural blue-dyed fashion products, both consumption intention and actual purchase behaviour will be stronger. If the attitude and subjective norms belong to a consumer’s personal ideology, PBC is a factor of influence on consumer self-confidence through both internal experience and external conditions. PBC promotes the transformation of attitude into consumer intention and directly into enacted consumption behaviour [39]. Thus, we developed two hypotheses for PBC as follows.

H3: The stronger the PBC regarding natural blue dyeing, the stronger the intention to consume natural blue-dyed fashion products.

H4: PBC positively affects consumption behaviour directly regarding natural blue-dyed fashion consumption.

Since actual behaviour occurs under various circumstances, it would be difficult to assert that a single subjective identity-related factor, such as inner identity or self-image enhancement, solely promotes actual consumption outside of considerations of economic capacity, consumption inertia, resources and environment along with other factors [40,38]. Intention includes internal subjective and external objective factors, and evidently it is a critical controlling factor affecting the formation of consumer behaviours [41]. Although the performance of intent comprises the generation of behaviour, this transformation is affected by multiple factors, including economic standards, knowledge and lifestyle [42]. Subjective norms, behaviour attitudes and PBC also explain intention. Therefore, value recognition, belief norms and strongly identified self-concept regarding natural blue-dyed fashion products can only contribute to the formation of intention, but not to actual behaviour directly. Thus, intention could contribute as the complete mediator of such subjective factors to ensure purchase behaviour. However, PBC differs from attitudes and subjective norms that only represent consumers’ internal consciousness. It is affected by a wider range of factors and could become the only variable in TPB with a direct influence on behaviour, possibly regulating the shift from intention to behaviour [43]. Therefore, intention could contribute to behaviours by partially mediating PBC [26], from which the following hypotheses can be made:

H5: The higher the intention towards natural blue-dyed clothing consumption, the stronger the actual consumption behaviour.

H6a: Intention has a completely mediating role in the relationship between attitude and natural blue-dyed fashion consumption.

H6b: Intention has a completely mediating role in the relationship between subjective norms and natural blue-dyed fashion consumption.

H6c: Intention has a partially mediating role in the relationship between PBC and natural blue-dyed fashion consumption.

Cultural values are considered to be some of the most powerful factors influencing consumption motivation behaviour [44]. In fashion consumption, cultural values play a role that must be considered. As a social phenomenon, fashion can be an expression of traditional culture and novel concepts, which is entirely separate from the basic practical function of clothing itself [45]. Cultural values are affected by factors such as country size, population density, average education level, and per capita income, with various influential elements. Craig and Douglas identified three factors that influence cross-cultural consumption, namely, values and belief systems, material culture and artefacts, as well as language and communication systems [46]. Ceglia presented the CCSC framework combined with TPB [9]. All of these factors influence TPB constructs as well as influencing one another. Among them, values and belief systems are the basic elements which reflect national ideology and policies, as well as material consumption sources for sustainable fashion [10,13,30]. There is a clear relationship between beliefs, material culture and TPB. Attitudes are based on past personal experiences involving emotional issues and guided by the value of universalism [47]. Subjective norms are affected by the influence of social groups on individuals and the belief in a collectivist spirit. PBC reflects the convenience of resources for sustainable consumption and is affected by the availability of material cultural resources. Various cultural bases render different TPB constructs [9,48].

H7: Differences exist regarding all factors of the TPB between two samples with differing cultural backgrounds.

This research applied quantitative methods to analyse the relationships between attitude, subjective norms, PBC, natural blue-dyed apparel consumption intention and actual behaviour. It also analyses the differences of natural blue-dyed clothing consumption in relation to culture differences. A preliminary questionnaire was developed comprising 24-Likert scale questions with 4 items for each TPB construct.

The questionnaire was divided into three parts. The first includes demographic questions of age, gender, income and education level. The second examines cultural value orientation based on Ceglia’s CCSC [9], including belief orientation, blue-dye material culture and communication orientation. The third covers the TPB measures. All TPB measures were based on Ajzen’s research in 2002 [34] and adopted from relevant previous research [30,54]. The nouns were changed to natural blue-dyed apparel and items were answered on a 7-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’. Questionnaire items for TPB are presented in Table 1. The questionnaire was conducted with 50 students in both countries as a pre-test to verify whether the items were well-designed and understandable.

In the next step, English questions were translated into both Chinese and Japanese, and then scholars majoring in English in both countries were invited to back-translate the questions into English to reconcile any post-translation deviations in the articulation of the questions.

Moreover, to strengthen the respondents’ understanding of natural blue-dyed fashion products, pictures from relevant national brands in both countries, such as Visvim and A Life On The Left, were included.

The questionnaire was placed on the wenjuanxing (http://www.wjx.com) website, one of the leading online survey agencies in mainland China. The main survey was conducted online in mainland China in early February in 2021, followed by a survey in Japan in March 2021. Each of the participants received 5 RMB for their effort. To reduce the influence of common method bias, the order of questions in the online survey was randomly reversed, and there was no hint of construct connotation.

Overall, 301 effective Chinese questionnaire responses and 302 effective Japanese responses were collected. Among the samples, the major age in the Chinese sample was 25–35 and in the Japanese the same, 25–40; 56% men in China and 54% men in Japan; 55% of the respondents were married in the Chinese sample and 50% in the Japanese. In addition, the majority of the respondents in both countries were at the bachelor’s degree level. In summary, the samples share common characteristics that allow for the study of consumption differences from differing cultural backgrounds.

This study used SEM to demonstrate the latent variables. Before presenting the results of the TPB model, confirmatory factor analysis (CFA) was performed with AMOS21.0 to test whether each factor in the five constructs satisfied the correspondence between items. The model fit indices (CHI/DF = 1.970, CFI = 0.983, GFI = 0.932, AGFI = 0.919, RMSEA = 0.028) suggest an acceptable fit. Composite reliability (CR) estimates were between 0.802–0.863, exceeding the recommended threshold of 0.70 [55]. The average variance extracted (AVE) ranged from 0.757 to 0.875, above the recommended value of 0.5 [56]. These values indicate that the measures within the construct are consistent. The results of the CFA for all scale items are presented in Table 2.

Discriminant validity testing was also performed. Discriminant validity indicates that each latent variable represents a different construct [57]. This distinction was determined by comparing the item loadings associated with one factor and its cross-loading on other factors. Each load on the related factor must be greater than the cross-loading on other factors [58]. Table 3 presents the discriminant validity testing results, showing that all respective square roots of the AVEs were greater than the correlations between factors [59]. Therefore, the five variables in this research demonstrate acceptable discriminant validity.

Common method bias (CMB) primarily arises in survey research. It is caused by response tendencies, such as social desirability, dispositional mood states, similarities or the time and location of data collection [60]. CMB can also be affected by the structure of a questionnaire [61]. Although we reordered all the questions, we also applied CMB testing to ensure minimal impact.

The unmeasured latent method construct (ULMC) was used to test whether CMB existed in this sample data. ULMC was used to detect and eliminate variances in substantive indicators that could not be attributed to their substantive constructs or random errors [62]. The testing result is presented in Table 4. The variance of the indicators was computed at 0.781, while the average method-based variance was 0.038. The ratio of substantive to method variances was approximately 21:1. Additionally, all the method factor loadings ranged from 0.08 to 0.829 and were all above the 0.05 level, being not significant. All the variance inflation factors (VIFs) were in the range of 3.69–6.05, within the acceptable range of 1–10. Therefore, the model can be considered free of CMB.

Fig. 1.

After accomplishing CFA, the latent model for pooled-data (N = 603) was specified and indicated good fit: CHI/DF = 1.58, Comparative Fit Index (CFI) = 0.991, Goodness-of-fit Index (GFI) = 0.951, Adjusted Goodness-of-fit Index (AGFI) = 0.933, Root Mean Square Error of Approximation (RMSEA) = 0.022. These results were congruent with values (i.e. GFI > 0.90, AGFI > 0.90, CFI > 0.90, RMSEA < 0.06) recommended by Fornell and Larcker [60]. Figure 2 presents the results.

Fig. 2.

The results of pooled-data hypotheses testing are presented in Tables 5 and 6. Attitude, subjective norms and PBC were shown to significantly influence natural blue-dyed fashion consumption, thus H1–H3 are verified (H1: β = 0.585, t = 10.970, p < 0.001; H2: β = 0.114, t = 2.384, p = 0.017 < 0.05; H3: β = 0.267 t = 5.696, p < 0.001). PBC was also found to have a positive influence on blue-dyed apparel consumption behaviour, confirming H4 (β = 0.486, t = 11.049, p < 0.001). Although the conversion of intention into actual behaviour is affected by many complex factors, H5 is also confirmed (β = 0.465, t = 10.005, p < 0.001), reflecting previous sustainable fashion consumption results based on TPB, e.g. [13,63,30].

The results of pooled-data mediation testing are shown in Table 7. Since bootstrapping has proved more powerful than the Sobel test in testing the effects of intervening variables [64], this study used the AMOS bootstrap operation to analyse the total, indirect and direct effects. Because the variable subjective norms cannot lead to consumption intentions, it also cannot contribute to purchase behaviour through intention, and an insignificant result (Z = 0.413 < 2, P = 0.232) was rendered for the total effect; H6b was rejected. The two remaining mediation hypotheses were confirmed. Attitude had significant data on indirect and total effects; yet, it did not have direct effects on actual behaviour (Z = 0.193 < 2, P = 0.804); therefore, intention mediates the relationship between attitude and behaviour, and H6a was accepted. PBC had significant results in all the three paths (Z = 5.408 > 2, P = 3.433 > 2, P = 0.001; Z =4.464 > 2, P = 0.002). Therefore, intention partially mediates the relationship between PBC and behaviour; thus, H6c was accepted.

To assess whether there are significant differences between Chinese and Japanese consumers (regarding their attitude, subjective norms, PBC and intention towards engagement in consumption of natural blue-dyed fashion), we performed a multivariate analysis of variance. The results revealed that blue-dyed apparel material culture and national belief orientation exerted significant effects on all TPB constructs (material culture: all the F-values were significant at p < 0.001; values and belief orientation: all the F-values were significant at p < 0.05). However, communication systems presented an insignificant effect on all the TPB constructs (F-values significant at P ≥ 0.159). Therefore, two intercultural characteristics have significant effects on differences in blue-dyed fashion consumption; H7 was confirmed.

Figure 2 presents the sample differences. Generally Chinese consumers rate all TPB factors more highly, particularly in ATT and INT, indicating that Chinese consumers have more emotion identity and higher intention to consume natural blue-dyed apparel than Japanese consumers. However, the bias between the actual consumption and intention of Japanese consumers was smaller to that of Chinese consumers, indicating that the intention conversion rate of Japanese consumers is higher. This might be a result of the higher economic standard of Japanese consumers, as economic factors could be a main obstacle to actual sustainable behaviour.

To test whether the TPB model is stable in multi-group samples, a multi-group analysis was conducted. Table VII presents the model fit applying four different constrained conditions. All the bias between?? the model are below the level 0.05, thus TPB model l is not affected by various sample factors in the study of blue dye consumption behaviour.

TPB models for two group samples are presented in Figure 3. In China, attitude is the most influential factor for intention towards the consumption of natural blue-dyed fashion, followed by PBC, which was found to exert effects on both intention and actual behaviour. However, subjective norms have no influence on intention towards actual consumption, thus H8b was not supported.

Fig. 3.

In Japan, PBC was the most influential factor towards actual behaviour, and the impact was stronger than for Chinese consumers; thus, H8c was confirmed. However, attitude has less effect on intention towards behaviour than for Chinese consumers. Subjective norms have a higher impact than on the Chinese, but it also the weakest impact on intention; therefore, H8a was rejected.

The differences between the two sample groups are presented in Table 8. Z-scores were used to test the significance. The results indicated that all five paths differed in each culture. Among them, the effect of attitude and PBC in leading to intention elicits the most significant differences, whereas the contribution of subjective norms to intention showed a relatively small difference.

Our research results could bring inspiration to the development direction of naturally dyed apparel brands. Judging from the research data from China and Japan, the influence of subjective norms is relatively weak. Therefore, the promotion of naturally dyed fashion consumption should neither focus on the consumption environment nor social policy, but should pay more attention to education and training consumer attitudesas well as the strategic reform of product categories.

For the promotion of naturally dyed fashion consumption in China, consumers’ love of traditional dyeing should continue to be strengthened. In addition, adjusting the strategy to promote the creation of more naturally dyed brands, including high-, mid- and low-end brands, will construct a differentiated and positioned??? naturally dyed fashion consumption material resources to reach a wider range of consumer groups. Furthermore, these brands need to create products that are more practical and more in touch with fashion trends.

As for the Japanese market, many natural blue-dyed brands have become quite well-known and cover a wide range of naturally dyed products. However, consumers are less emotional about natural dyeing than in China. Therefore, the Japanese brand can delve deeper into Japanese traditional cultural heritage, strengthen the education of traditional material culture, and enhance consumers’ enthusiasm for traditional handicrafts.