BRAND LOYALTY AT SMARTPHONES MARKET: LINKING BETWEEN BRAND PASSION, HEDONIC AND UTILITARIAN VALUES

This paper summarizes the arguments and counterarguments within the scientific discussion on the factors that influence consumers' brand loyalty to smartphone brands. The main aim of the study is to analyze the relationship between hedonic value, utilitarian value, brand passion, and brand loyalty based on consumers' loyalty at the smartphones market. Accordingly, the study focuses on brand loyalty as a consequence of brand passion and reveals it in a holistic framework as it emphasizes the direct relationship between the two variables. In this context, this study is different from others in literature in a way that it focuses on smartphones, which are at the upper ranks of the technological product category. The fact that this study only deals with mobile phones makes it different and essential as studies on electronic and technological products often focus on the general situation. Researchers used positivist research as a quantitative research design in this study, which deals with factors that influence brand loyalty to smartphone brands. The study universe involves 18-year-old and older consumers with a purchase capacity. In this context, the study sample comprises smartphone users at or over the age of 18. The conceptual model and associated hypotheses are tested with a sample of 330 consumers. Researchers collected the study data with a convenience sampling method with the help of an online survey. In the study, data were analyzed through structural equation modelling. The results demonstrate strong relationships between the two antecedents (hedonic value and utilitarian value) and brand passion and between brand passion and its consequence (brand loyalty). Study results indicate that hedonic value (β=0,506; p<0,001) and utilitarian value (β=0,202; p<0,001) have a positive influence on brand passion. Study results also show that brand passion (β=0,683; p<0,001) has a positive influence on brand loyalty. On the other hand, the study also reveals that brand passion mediates the relationship between hedonic and utilitarian value and brand loyalty. Study results point out that hedonic value, utilitarian value, and brand passion have a positive influence on the development of consumers' loyalty to smartphone brands.


276
Marketing and Management of Innovations, 2020, Issue 1 http://mmi.fem.sumdu.edu.ua/en Huber et al. (2017), suggested that perceived utilitarian value and perceived hedonic value influence the passionate love of the brand. In their study, Huber et al. (2017) found that hedonic value influenced passionate brand love positively while there was no relationship between utilitarian value and passionate brand love. Bauer et al. (2007) found out in their study that hedonic value had a positive influence on brand passion. In this context, we might suggest that brand passion is positively related to both utilitarian value and hedonic value. Consistently, the following hypotheses result from these considerations: H1: Hedonic value has a positive influence on brand passion. H2: Utilitarian value has a positive influence on brand passion. According to Kotler and Keller (2016: 78), those who succeed in marketing are the ones who attach importance to developing customer satisfaction and loyalty. Therefore, brand loyalty ensures demand security for the companies, helping them to predict this demand. At the same time, it also sets barriers to restrict the entry of other companies into the market (Kotler and Keller, 2012: 242).
According to Moorman, et al. (1992: 316), «commitment is enduring, and it reflects a positive valuation of a relationship». Carroll and Ahuvia (2006: 82) have defined brand loyalty as «the degree to which the consumer is committed to repurchasing the brand».
Hemsley-Brown and Alnawas (2016) suggested in their study that brand passion, brand affection, and self-brand connection were positively related to brand loyalty. Study results revealed that brand passion has the most potent influence on brand loyalty compared to the other two components. Pourazad et al. (2019) revealed in their study that brand passion had a positive influence on attitudinal brand loyalty to sportswear brands. Study results showed that the positive influence of brand passion on attitudinal brand loyalty was significant. Albert et al. (2013) suggested that brand passion was positively related to brand commitment, and the results of their study revealed that brand loyalty influenced the attitudinal component of brand loyalty. Therefore, the study posits the following hypothesis.
H3: Brand passion has a positive influence on brand loyalty. Jones et al. (2006) supported the view that hedonic value and utilitarian value had a positive influence on brand loyalty with the study findings they obtained. Taking the conceptual framework and research model into consideration, they also suggested that brand passion would have a meditating influence. Therefore, researchers propose the following hypotheses.
H4: Brand passion mediates the link between hedonic value and brand loyalty. H5: Brand passion mediates the link between utilitarian value and brand loyalty. Methodology and Research Methods. The purpose of this study is to analyze the relationship between hedonic value, utilitarian value, brand passion, and brand loyalty based on consumers' loyalty to smartphone brands. In this context, researchers expected to reveal the mediating influence of brand passion in the relationship between hedonic and utilitarian values and brand loyalty.
Researchers measured all the variables in the question form (survey) with a 5-point-Likert scale (1: I strongly disagree, 2: I disagree, 3: I neither agree nor disagree, 4: I agree, 5: I strongly agree). They adapted the items in the scale from previous studies that were subjected to reliability and validity analysis. In this context, they adapted eleven items for hedonic value and four items for utilitarian value from Babin (1994). They adapted seven items from Albert (2009) for measuring brand passion and seven items from Carroll and Ahuvia (2006) for measuring brand loyalty. In this study, researchers tested a structural model by using AMOS 20 package.
According to the suggestions of Davis et al. (2009) andEdward et al. (2012), Researchers interviewed two academicians and two experts in order to evaluate external validity before passing onto the final study. Then, they carried out a pilot study with 25 mobile phone users to test the understandability of questions. They made several minor changes to make the items more understandable and clarify them as a result of the feedback obtained from the pretest. The following section presents the statements used for the http://mmi.fem.sumdu.edu.ua/en measurement of hedonic value, utilitarian value, brand fashion, and brand loyalty variables analyzed within the context of the study. Figure 1 displays the research model. The study sample comprises consumers at or over the age of 18 who are mobile phone users. Researchers collected study data from 330 consumers with a convenience sampling method, which is one of the non-probability sampling methods. They used an online survey in the study as the data collection tool and adopted a quantitative research pattern.
The researchers published the online survey of the study on docs.google.com. They delivered the survey to 330 persons at or over the age of 18, who voluntarily consented to respond in consideration of the study goal. Researchers evaluated the responses of participants who met the criteria such as «answering control questions correctly», «not giving the same response to every question on the scale», and «answering every question on the survey» (Arslan, 2017). In this context, they eliminated 20 surveys that failed to meet these criteria. Then, they evaluated the surveys of 310 participants who followed the relevant rules and finally carried out the analysis.
The demographic features of consumers at or over the age of 18 who were mobile phone users are as follows. 57.1% are women, and 42.9% are men. 67.7% are single, and 32.3% are married. Findings on education level show that approximately 40% of participants have a bachelor's degree, and the number of participants who have an associate degree and master's degree is equal (12.9%). 30.3% are high school graduates, while 5.2% are primary school graduates. Consumers between the ages of 18 and 31 constitute approximately 75% of participants. The rate of participants at 32-39 age range is 14.8%, and about 10% of participants are over the age of 40. The monthly income of 40% of participants is 2000 TL or less, and the monthly income of 20% of participants is between 2001 and 3000 TL. The rate of participants whose monthly income is between 3001 and 4000 TL (13.5%) is almost equal to the rate of participants whose monthly income is between 4001 and 5000 TL (11.9%) while the rate of participants who earn between 5001 and 6000 TL is 6.5%. The rate of participants who earn more than 6001 TL a month is 8.1%. Findings on participants' professions show that 13.2% are workers, 41.1% of students, and 10.3% of teachers. The rates of healthcare professionals and academicians (2.6%) are equal, and freelancers constitute 7.7% of participants. 2.9% of participants are shop owners, 5.5% housewives, and 6.1% engineers. The study sample also involves software developers, soldiers, public accountants, bank employees, and architects.
Results. In this section, researchers adopted the two-stage model that Anderson and Gerbing (1998) suggested. They analyzed thirty-nine items and 4-factor covariance structure measurement model of the study in terms of structural validity and reliability, and then tested research hypotheses through the structural model.
Researchers analyzed the convergent validity and discriminant validity in order to test the validity and reliability of the model.

Utilitarian Value
Hedonic Value

Brand Passion
Brand Loyalty Before passing onto the structural model, they evaluated the measurement model in terms of fit indices, regression weight, and modification indices (MI). Thus, fit indices obtained as a result of CFA (Confirmatory Factor Analysis) of the measurement model are χ²/df=4,420; GFI=0,671; AGFI=0,614; TLI=0,836; CFI=0,850; RMSEA=0,105. These results show that the model needs improvement. For this purpose, researchers evaluated modification Indices (MI) of the model, and Table 1 presents the variables that they removed from the analysis.  Table 2 are acceptable, which indicates the fit between model and data (Doll et al., 1994;Mishra and Datta, 2011). Researchers tested the structural validity of scales with convergent validity and discriminant validity of structures. They evaluated the factor loads of each structure in order to examine convergent validity. They found that factor loads of all structures were 0.50, which indicated the evidence of convergent validity was above suggested value (Hair et al., 2010:710). It is also possible that there is convergent validity as an average variance value (AVE) found for all structures are above 0.50, which is the critical value (Fornell and Larcker, 1981). The fact that AVE values are above the suggested value 0.50 also shows that there is internal consistency (Berthon et al., 2005;Fornell and Larcker, 1981).
On the other hand, researchers also tested convergent validity by utilizing CR and AVE to examine the relationship between each statement and the factor or structure that it belongs to. Hair et al. (2010) stressed out that AVE value must be higher than 0.5 (AVE>0.5), and CR value must be higher than AVE (CR>AVE) for convergent validity. Analysis of these structures shows that CR values are higher than AVE values for each structure, and AVE values are higher than 0.5. Therefore, it is possible to state that structures have convergent validity. Table 3 presents the correlations of structures. According to this table, the highest structure correlation of 0.46 is 0.85 below the suggested value. Accordingly, it is possible to state that the measurement model has discriminant validity (Kline, 1998:60). Moreover, the fact that the average variance (AVE) obtained for each structure is higher than the square of the correlation between each structure and other structures (Fornell and Larcker, 1981). Considering the values in Table 1 and Table 2, we might say that we provided discriminant validity.
Researchers tested the reliability of scales using Cronbach alpha and composite reliability (CR) statistics. Table 1 presents the reliability values. α and CR values for each structure are above the critical value 0.70 (Hair et al., 2010:710), which shows that all scales are reliable.

Structures
(1)  Table 6 shows that fit indices statistics of the model are acceptable. Table 2 presents the standardized estimations of the model (regression weights). Accordingly, the H1 hypothesis suggests that hedonic value influences brand passion. The structural model shows that hedonic value has a statistically significant influence on brand passion (β=0,506; p<0,001). H2 hypothesis suggests that utilitarian value influences brand passion. The structural model shows that utilitarian value has a statistically significant influence on brand passion (β=0,202; p<0,001). H3 hypothesis suggests that brand passion has a statistically significant influence on brand loyalty. Results show that brand passion has a statistically significant influence on brand loyalty (β=0,683; p<0,001). Accordingly, researchers accepted H1, H2, and H3 hypotheses according to the results of the hypotheses. Table 5 shows the results of the hypotheses.
Researchers calculated direct, indirect, and total influences in order to reveal all the influence of variables in the research model on brand loyalty. The influences that Table 4 presents stem from standardized structural coefficients.  Table  4 and their significance levels. As a Bootstrap method, «is a straightforward method to understand and use, which is far from intense mathematical formulas and has limited assumptions (Simon and Bruce, 1991) it yields reliable results in cases where known statistical methods and assumptions are insufficient» (Takma and Atil, 2006). Used in the analysis of mediating models, the Bootstrap method is taken as a part of the modern approach (Gurbuz, 2019). Analysis of the mediating model based on the contemporary approach focuses on calculating indirect effects and making inferences from the calculated values. Thus, the indirect effect is «the multiplication of the effect of estimation variable on mediating variable» and «the effect of mediating variable on the outcome variable». According to this, researchers consider that the mediating model is verified, and there is no need for another test if the indirect effect of the estimation variable is significant as a result of the Bootstrap test (Gurbuz, 2019: 55).
According to the findings that Table 4 presents, the effect of hedonic value on brand loyalty mediated by brand passion was β=.346; p<0,001. This result reveals that brand passion plays a mediating role in the relationship between hedonic value and brand loyalty. On the other hand, the effect of utilitarian value on brand loyalty mediated by brand passion was β=.138; p<0,001. Thus, paths in Table 4 that describe the indirect effects are statistically significant, which shows that indirect effects are significant. The fact that the indirect effects of estimation variables as a result of the Bootstrap test are significant also shows that the mediating model is verified.
Moreover, there is no need for another test (Gurbuz, 2019: 55). In this context, this result shows that brand passion plays a mediating role in the relationship between utilitarian value and brand loyalty. Table 5 presents the results of H4 and H5 hypotheses. 281 http://mmi.fem.sumdu.edu.ua/en

Figure 2. Results on Structural Equation Model
Note: ** p <0,001 Source: developed by the authors Table 5 presents the results of the hypotheses.  (Doll et al. 1994;Mishra and Datta, 2011).  Table 6 shows that brand passion variance is at 0.30 and brand loyalty at 0.47 levels.
Conclusion. Three essential theoretical contributions emerged as a result of this study that researchers carried out in order to analyze the mediating role of brand passion in the effect of shopping value (hedonic and utilitarian value) perceived by smartphone consumers on brand loyalty. The first is that this study is different from other studies in literature in the way that it focuses on smartphones, which are at the upper ranks of the technological product category. The fact that this study only deals with mobile phones makes it different and essential as studies on electronic and technological products often focus Utilitarian Value Hedonic Value

Brand Passion
Brand Loyalty 0,506** 0,202** 0,683** on the general situation. Considering the current marketing and brand literature, the second contribution is the analysis of hedonic and utilitarian values as antecedents of brand passion and the direct relationship between them. In this context, this is one of the pioneering studies that deal with the brand passion, brand loyalty, and value attributed to shopping together in the smartphone selection of consumers. Moreover, the study reveals its focus on brand loyalty as a consequence of brand passion in a holistic framework as it emphasizes the direct relationship between the two variables. Finally, the third noteworthy contribution is the analysis of the mediating role that brand passion plays in the relationship between brand loyalty and both hedonic and utilitarian value in addition to the direct effects. Thus, the study reveals the probable mediating role that brand passion plays in establishing brand loyalty in terms of hedonic and utilitarian value attributed by consumers to smartphone preferences.
This study found that hedonic value attributed by smartphone consumers to shopping has a statistically significant influence on brand passion (β=0,506; p<0,001). Results obtained in this study overlap with the studies of Bauer (2007) and Huber (2017). On the other hand, the study also reveals that utilitarian value attributed by smartphone consumers to shopping has a statistically significant influence on brand passion (β=0,202; p<0,001). This result differs from the findings of Huber's (2007) study that reveals that there is no significant relationship between utilitarian value and brand passion.
This study has several limitations, in addition to its theoretical contributions. The study sample involves only Turkish smartphone users at or over the age of 18. It is not possible to generalize the study results as researchers selected the study sample with a convenience sampling method. Therefore, study results are valid only for the study sample. To conclude these results, it is essential to carry out a study on smartphone purchase preferences of consumers from different profiles. This study also evaluates the relationship between brand passion, hedonic value, utilitarian value, and brand loyalty based on smartphone brands. Future studies might select another technological product or product category for evaluation. Considering the variables used in this study, future studies must deal with different roles of different antecedents. On the other hand, it would be useful for both researchers and implementers to carry out studies based on socio-cultural variables that might influence brand loyalty and brand passion. As it has become an essential parameter for businesses to know the value components of consumers in establishing loyal customer group, it might be a remarkable aspect in structuring market components.
Author Contributions: T. Y., M. S. prepared the outline of the paper and both authors wrote abstract, T. Y.; developed theoretical background and literature review, M. S.; provided data and prepared methodology, M. S.; performed the analysis and visualization of results, T. Y. and M. S.; wrote discussion, T. Y., M. S. ; was responsible for references control.