INVESTIGATING SOCIAL MEDIA INFLUENCERS’ CREDIBILITY AND RELIGIOSITY, AND THE MODERATING ROLE OF HEDONIC VALUE ON PURCHASE INTENTION OF MODEST FASHION BRANDS

: In recent years, modest fashion has grown significantly in Indonesia. Since the pandemic happened in 2020, some well-known modest fashion brands in Indonesia have started to change their focus from offline to online sales, using Instagram as one of the strategies. As a result, the role of social media influencers (SMI) in marketing strategies is expanding. This study's objective is to investigate experimentally how SMI can help brands in creating brand attitude and thus improve purchase intention. Building upon naïve theory, consumer socialization theory, and market signalling theory, this study wants to analyze the mediating role of perceived credibility and religiosity of SMI on modest fashion brands in Indonesia, as well as the moderating effects of hedonic value on the product. This research utilized 246 respondents who are active female Instagram users and used a single cross-sectional quantitative methodology. The data was processed using AMOS 24 with the SEM method. This research offers empirical evidence on the mediating effect of expertise, goodwill, trustworthiness, and religiosity between SMI power and brand attitude. Furthermore, the result of this study also shows that the influence of brand attitude on purchase intention is moderated by the hedonic value of the products.


INTRODUCTION
Modest fashion is described by many as an emerging trend, a cultural phenomenon, or a specific category in the world of fashion (Slater and Demangeot, 2021;Zainudin et al. 2020).It is, however, more than any of that.Modest fashion, rooted in a principle of modesty, embodies identity, culture, and mostly, religiosity (Lewis, 2019).Though it is strongly connected to the Muslim world (Hassan and Ara, 2022;Pemberton and Takhar, 2021), modest fashion is relevant to many other religions.Nonetheless, according to the State of Global Islamic Economy Report 2022, modest fashion's innovation and its subsequent growth are more prominent in Muslim-majority countries such as Turkey, Indonesia, Malaysia, and the Gulf Cooperation Council (Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates).
Indonesia, with the largest Muslim population in the world, is the world's third-largest buyer of Muslim clothes; the total spending was 20 billion USD or around IDR 300 trillion in 2017.Then, in 2019, modest fashion products recorded the highest growth in Indonesia (IPrice, 2021).The growth enjoyed by the modest fashion industry is largely contributed by the extensive growth in domestic mobile adoption and the e-commerce sector (Sumarliah et al. 2022;Yuliastuti, 2022).
Accelerated by the Covid-19 situation in 2020, several well-known modest fashion brands in Indonesia have shifted their strategies to online channels.Based on research by APJII (2020), up until 2020, there were 197 million Indonesians connected to the internet.Therefore, capitalizing on the online market is pertinent to growth; and social media has arisen as the chosen channel to tap on.Through social media, the interaction between the customer and the business performer will be closer, and customers can make private communications to order products that suit their needs (Moy et al. 2019).
The rapid growth of social media in Indonesia has moved both organic and paid conversation into the channel.Instagram is one of the social media platforms that is currently expanding significantly.In Indonesia, the number of Instagram users as of July 2021 has reached 93 million (Annur, 2023).With the growing usage of social media channels, some contributors attract a large audience, cultivate a fan base, and serve as a resource for their followers, thus becoming social media influencers (SMI) (Vrontis et al. 2021).SMI are third-party endorsers that attempt to shape consumer attitudes toward a brand.
According to Hjelmeseth (2018), an SMI is a person with more than 10,000 followers who brings more professionalism to the table by taking their social media presence much more seriously and actively promoting brands and offering information about their products.In addition to their product endorsements, the content they share engenders the loyalty of their followers (Nafees et al. 2020).
The use of SMI as a marketing tool increased more than threefold between 2017 and 2020, from three billion to 9.7 billion USD (Statista, 2023).Moreover, Instagram has more than 500,000 active influencer accounts with more than 15,000 followers which is 39% of all accounts on Instagram.Among these active influencer accounts, 81% have followers between 15,000 and 100,000 users (Droesch, 2019).According to Statista (2023), there were 3.8 million sponsored posts on Instagram.
Despite the rapid growth in the number of social media users, especially SMI, from the researchers' perspective, the impact of SMI power on brand attitude and purchase intention has not been extensively studied.Previous research focused on investigating dimensions of SMI credibility (Ha and Lam, 2016;Saima and Khan, 2020) and SMI marketing (Lou and Yuan, 2019).The past study combined many theoretical perspectives and provided additional insight into the significance of SMI power and perceived credibility, and their impact on consumers' brand attitudes (Nafees et al. 2021).However, to apply the previous framework to a modest fashion context, an adaptation is needed.Perceived religiosity is one of the elements in the identity of modest fashion users that emerge and can be examined.Silalahi et al. (2021) also found that SMI's perceived religiosity can also motivate consumers in making their consumption choices.Nafees et al. (2021) did not further examine the influence of brand attitude on purchase intention.Although past research has shown that such an effect would be positive (Ha and Lam, 2016;Lin et al. 2021;Wang et al. 2019).Nonetheless, a certain customer value is able to strengthen the effect, in which a higher level of consumers' hedonic value perceptions can lead to higher consumers' purchase intention as well (Amin and Tarun 2021; Liu et al. 2021;Su et al. 2019) With In total, 479 responses were gathered but only 246 responses were analyzed due to the screening process which included an Instagram account verification.The questionnaire was developed as the main research instrument in which most measurement items were adopted from previous studies.The indicators for SMI power followed the research of Nafees et al. (2021) and Kim and Kim (2021).For perceived expertise, perceived goodwill, and perceived trustworthiness, items were derived from the research of McCroskey and Teven (1999).Then, items on the perceived religiosity of SMI were adapted from Silalahi et al. (2021).The measure for brand attitude refers to Nafees et al. (2021).Meanwhile, the hedonic value was adapted from Dubois and Laurent (1994).Lastly, the purchase intention was from the research of Ha and Lam (2016).
Items in SMI power, purchase religiosity, hedonic motivation, brand attitude, and purchase intention were measured using a 7-point Likert scale ranging from "1-strongly disagree" to "7-strongly agree".Meanwhile, for items perceived expertise, perceived goodwill, and perceived trustworthiness using semantic differential scale.The data analysis was performed by using SPSS 26 and AMOS 24.
this justification, it is clear that SMI research is still expanding.
Therefore, in this study, the researchers aim to examine the effects of SMI power on brand attitude via SMI credibility which consists not only of perceived expertise, perceived goodwill, and perceived trustworthiness, but also perceived religiosity.
Additionally, this research considers the influence that the hedonic value of the promoted modest fashion brand products has on brand attitude and purchase intention.To conclude, theoretically and empirically, this study wants to examine why and how SMI power can benefit modest fashion brands' purchase intentions.

METHODS
This study employed a quantitative approach based on a cross-sectional study.Therefore, a set of hypothesis tests is carried out.The research model is shown in Figure 1.For the purpose of the study, data were collected from Indonesian female users of Instagram who follow a modest fashion influencer and were actively using the platform in the last 6 months before the data collection period (March -September 2022).H3: Perceived goodwill mediates the relationship between SMI power and brand attitude.
Wiedmann and von Mettenheim (2020) addressed trustworthiness as the "question of whether an individual is believable, does the source express his or her honest opinion, or whether is he or she influenced by third parties".Moreover, Instagram users that perceive an influencer as sound, reliable, dependable, honest, or trustworthy consider that influencer's trustworthiness (Ceyhan et al. 2018).Based on previous literature, this study suggests that the relationship between SMI power and consumer perceptions of a brand can be mediated by perceived trustworthiness.Therefore, the hypothesis formed is as follows: H4: Perceived trustworthiness mediates the relationship between SMI power and brand attitude.
Besides its credibility, the perceived religiosity of the SMI might also influence consumers' purchasing decisions (Silalahi et al. 2021).Moon et al. (2019) also confirmed that people trust religious people at various levels of religiosity.Then, there is a study by Mansour and Diab (2016) which found that the perceived religiosity of SMI influences brand attitude in the context of celebrity endorsements in television advertising.In this study, the perceived religiosity of SMI is correlated with the degree to which Instagram followers are able to perceive the religious aspect of the SMI they follow.Therefore, the hypothesis formed is as follows: H5: Perceived religiosity of SMI mediates the relationship between SMI power and brand attitude.
According to Trivedi and Sama (2020), purchase intention is defined as the consumer's propensity to buy the brand in the future.They also found that a positive consumer attitude toward endorsements made by SMI will increase purchase intention.Moreover, many studies also showed that consumer attitudes toward brands have a positive and significant influence on purchase intention (Ha and Lam, 2016;Lin et al. 2021;Wang et al. 2019) This is also supported by the Elaboration Likelihood Model which said that purchase intention is a function of brand attitude.In addition, the Technology Acceptance Model and the Hierarchy of Effects Model also mentioned that brand attitude leads to purchase intention.Therefore, the researcher came up with the following hypothesis:

Hypothesis Development
For this research, the impact of SMI on consumer brand attitudes is defined as the influencer's power to increase the likelihood that a consumer will purchase a brand endorsed by the SMI (Nafees et al. 2020).Trivedi and Sama (2020) also defined brand attitude as an established mediator between advertising stimuli and multiple marketing variables like purchase intentions.Moreover, increasing SMI power can support the formation of positive brand attitudes of consumers toward brands promoted by SMI on social media (Schouten et al. 2020).In addition, previous research conducted by Nafees et al. (2021) also showed that SMI power has a positive and direct impact on brand attitude.Therefore, the hypothesis formed in this study is: H1: SMI power has a positive and direct influence on brand attitude.
According to existing research, brands sponsored by SMI with high credibility will have a positive effect on brand attitude towards advertisements and brands compared to brands endorsed by SMI with low credibility (Schouten et al. 2020).In general, there are three sources or dimensions of credibility: expertise, goodwill, and trustworthiness (Jiménez-Castillo and Sánchez-Fernández, 2019; McCroskey and Teven 1999).Wiedmann and von Mettenheim, (2020) described expertise as "the source's level of knowledge.
It is defined in terms of peak or at least high levels of knowledge, experience, and problem-solving skills within a given domain".In the prior study by Nafees et al. (2021), perceived expertise was able to increase SMI ratings.Thus, the formed hypothesis is: H2: Perceived expertise mediates the relationship between SMI power and brand attitude.
McCroskey and Teven (1999) introduced a concept they called perceived caring or goodwill.The concept is composed of three components: understanding, empathy, and responsiveness.According to Nafees et al. (2021), goodwill is an evaluation of the extent to which consumers perceive that SMI prioritizes consumer interests.Similar to an earlier study, this study suggests that the relationship between SMI power and brand attitude can be mediated by perceived goodwill.

Descriptive Data
The demographic analysis shows that most of the respondents are women with the age range of 26 to 41 years (64.5%).Next, the education level of most respondents is a diploma with 54.5%.Most of them have a monthly income of less than Rp4,000,000 (47.1%).Lastly, most of the respondents are private employees (44%).

Measurement Model Test
Based on convergent validity and the loading factor method in

Structural Model Test
The Goodness of Fit findings from the results of the AMOS 24 program analysis are then presented in e3 is an error from SMI2 indicator, and e4 is an error from SMI1 indicator e1 ↔ e2 e1 is an error from SMI4 indicator, and e2 is an error from SMI3 indicator e26 ↔ e27 e26 is an error from HE1 indicator, and e27 is an error from HE2 indicator e22 ↔ e24 e22 is an error from BA4 indicator, and e24 is an error from BA2 indicator e10 ↔ e11 e10 is an error from PT4 indicator, and e11 is an error from PT3 indicator e1 ↔ e3 e1 is an error from SMI4 indicator, and e3 is an error from SMI2 indicator e2 ↔ e3 e2 is an error from SMI3 indicator, and e3 is an error from SMI2 indicator e2 ↔ e4 e2 is an error from SMI3 indicator, and e4 is an error from SMI1 indicator Ha and Lam (2016) also found that the perceived expertise of celebrity endorsements has a significant and positive influence on consumers' attitudes toward the brand.Additionally, Nafees et al. ( 2021) also found that perceived expertise can partially mediate the relationship between SMI power and brand attitude.

Hypothesis Test
After testing the influence of variables, a variety of results have arisen for each hypothesis studied.Table 5 presents the results of the test of the influence between variables.H1, H2, H3, H4, H5, H6, and H7 imply that each has a significant effect.Moreover, the researchers present further explanations for each hypothesis outcome and back it with prior study results.In detail, Figure 2 shows the SEM model.
In H1, it tests the effect of SMI power on brand attitude.
Table 5 shows a structural model path that illustrates the relationship between SMI power and brand attitude.It has a significant effect between SMI power and brand attitude (t = 4.226 > 1.96) and a significant score (0.000 < 0.05).Therefore, H1 is accepted.The results imply that in Indonesia, while promoting the products of modest fashion brands on Instagram, SMI is able to directly affect consumers' perceptions of the brand.
The results of H1 regarding SMI power on brand attitude are in line with the findings by (Schouten et al. 2020).They said that increasing the power of SMI can build a positive brand attitude on social media.
The study by Hussain et al. (2020) and Srivastava (2021) also proved that SMI supports brand attitude.
In addition, Ha and Lam (2016), whose study is about the credibility of celebrity endorsement in Vietnam, Figure 2. SEM Model of this study also found a positive impact of celebrity trustworthiness on consumers' attitudes toward the brand.Thomas & Johnson (2019) also indicated that celebrity trustworthiness has a positive impact on consumers' attitudes toward the advertisement.Furthermore, Nafees et al. (2021) proved that perceived trustworthiness is able to partially mediate the relationship between SMI power and brand attitude among Instagram users in the United States.
For H5, regarding the effect of SMI's perceived religiosity on brand attitude, the findings of the structural path show that there is a significant effect (t = 2.968 > 1.96) with a significant value (0.003 < 0.05).Hence, H5 is accepted.The research data show that SMI's perceived religiosity is able to mediate the relationship between SMI power and brand attitude when they promote modest fashion brand products in Indonesia.
The results of H5 regarding SMI's perceived religiosity on brand attitude are in line with the previous study by (Silalahi et al. 2021).It found that perceived religiosity is able to moderate the relationship between perceived credibility (trustworthiness and attractiveness) of celebrity endorsements and purchase intention.In addition, there is also research by Mansour and Diab (2016) which states that perceived religiosity is able to mediate celebrities' credibility with brand attitude and purchase behavior in the context of television advertising.
In H6, it is about the effect of brand attitude on purchase intention.There is a significant effect between brand attitude and purchase intention (t = 2.195 > 1.96) with a significant score (0.028 < 0.05).Therefore, H6 is accepted.The results imply that, in Indonesia, when In H3, the analysis focused on the effect of the mediating role of perceived goodwill between SMI power and brand attitude.A significant effect is found between perceived goodwill on SMI power and brand attitude (t = 2.432 > 1.96) with a significant value (0.015 < 0.05).Followed by the result of the analysis, H3 is accepted.
The result indicates that perceived goodwill has the ability to mediate the relationship between SMI power and brand attitude.
The results of H3 align with the research conducted by (McCroskey & Teven, 1999).They said that two factors can increase the perceived goodwill of SMI, namely as follows; (1) SMI's concern for their followers, and (2) SMI emphasizes followers' interests.
It means the increase in SMI's power which is caused by the accumulation of perceived goodwill makes the marketing value of SMI increase.However, these research results are not in line with (Nafees et al. 2021).They found that perceived goodwill was unable to fully or partially mediate the relationship between SMI power and brand attitude.
H4 states the effect of the mediation role of perceived trustworthiness on brand attitude and SMI power.
The findings of the structural path indicate that there is a significant effect between SMI power, perceived trustworthiness, and brand attitude (t = 2.697 < 1.96) with a significant value (0.007 > 0.05).H4 is accepted.The research results indicate that perceived trustworthiness is able to mediate the relationship between SMI power and brand attitude.
The hypothesis test results regarding the mediating role of SMI's perceived trustworthiness are similar to the research conducted by Ha Lam (2016) whose context is about celebrity endorsement in Vietnam.The results The findings of this research are also beneficial for owners and brand managers for any modest fashion products in terms of providing guidance in conducting social media marketing strategy.First, the selection of SMI should be based on perceived credibility that consists of expertise, goodwill, trustworthiness, and religiosity.Second, managers should focus on using SMI to promote products with a higher hedonic appeal.This is because the hedonic value is proven to be able to increase purchase intention for consumers who have a positive attitude toward the brand.Lastly, managers can consider using hedonic shopping motivation as part of profiling potential consumers.

Conclusions
The research successfully identifies the influence of SMI power, perceived expertise, perceived goodwill, perceived trustworthiness, perceived religiosity, brand attitude, purchase intention, and hedonic value using SEM.The data obtained from the research show that SMI power has a positive and direct effect on brand attitude.It means that when promoting the product of modest fashion brands, SMI is able to directly influence brand attitude.The underlying mechanism lies in several dimensions of SMI credibility, namely perceived expertise, perceived goodwill, perceived trustworthiness, and perceived religiosity.This can be interpreted that the perceived expertise, goodwill, trustworthiness, and religiosity of SMI are the essential elements of credibility that must be possessed by SMI in order to strengthen the brand attitude of the modest fashion brand products that are being promoted by them.The dynamic between SMI power and attitude is crucial to bridge consumer intention to buy modest fashion brands.Furthermore, this research delves deeper into the conditional analysis to determine that the relationship between brand attitude and purchase intention is moderated by the hedonic value of modest fashion brand products.A higher perceived hedonic value from modest fashion brands will further strengthen the positive impact that consumer attitude has on purchase intention.
modest fashion brands are promoted on Instagram, the brand attitude has a positive and direct relationship with purchase intention.
The hypothesis test results regarding the effect of brand attitude on purchase intention are similar to the results obtained by Ha and Lam (2016).It shows that purchase intention is significantly and positively influenced by consumers' attitudes toward the brand.Moreover, many previous studies also found a positive relationship between brand attitude and purchase intention (Ha and Lam, 2016;Lin et al. 2021;Wang et al. 2019).
For H7, regarding the moderating role of hedonic value on brand attitude and purchase intention, the findings of the structural path show that there is a significant effect (t = 3.492 > 1.96) with a significant value (0.000 < 0.05).H7 is accepted.The research data shows that the hedonic value of modest fashion brand products in Indonesia is able to moderate the relationship between brand attitude and purchase intention.
The results of the hypothesis test of the moderating role of hedonic value align with the research conducted by Martín-Consuegra et al. (2019), which found that hedonic values are capable of moderating the relationship between brand attitude (brand involvement) and purchase intention.Then, based on the study of Ling et al. (2009) hedonic value is a value that is able to influence the proportion of income spent on fashion with a higher percentage by consumers.

Managerial Implications
The research also can provide several implications that can be applied to business.As for SMIs, their credibility is not only formed by how knowledgeable they are about the products being promoted, how caring they look towards their followers, and how honest or trustworthy they are.More than that, when promoting a modest fashion brand, religiosity is an important factor in forming consumer judgment about a brand.More specifically, it is the external religiosity that matters the most; it represents how religious they look from the outside.Therefore, it is determined by how frequently they attend worship places, how frequently they donate to religious events, how involved they are in religious organizations as volunteers, and how active they are as members of a religious community.

Figure
Figure 1.Research Model

Table 1 .
Table 1, it is determined that all indicators have loading factors more than 0.70.The indicators can be categorized as meeting the convergence validity.Test of validity and reliability 70.According to the results, the indications are reliable.Any indices with AVE more than 0.50 yield the Average Variances Extracted (AVE) results.It meets the criteria for describing how to use the indicators.

Table 2
. Before conducting a hypothesis test between two variables, a few requirements must be met.According to Table2, the SEM model cannot be deemed acceptable

Table 1 .
(Hair et al. 2016and reliability (continue) Square, RMSEA, CFI, TLI, IFI, PGFI, and PGNI meet the existing cut-off value criteria.To evaluate the viability of a model, we can only employ 4-5 Goodness of Fit criteria.According toHair et al (2016), "the use of 4-5 Goodness of Fit criteria is considered sufficient to assess the feasibility of a model, provided that each criterion of Goodness of Fit namely absolute fit indices, incremental fit indices, and parsimony fit indices are represented".The next step is hypothesis testing.because it does not meet the Goodness of Fit criteria based on(Hair et al. 2016).Thus, it is important to modify the SEM model with steps to correlate between errors based on modification indices and the covariances table of the AMOS program's output findings.

Table 4 .
It shows the fitted SEM model.The Chi-

Table 4 .
Goodness of fit in the modified model

Indonesian Journal of Business and Entrepreneurship, stated
Nafees et al (2021) attitude toward the brand is influenced by perceived congruence with the brand, perceived trustworthiness, and perceived expertise possessed by the celebrity endorsement.Nafees et al (2021)also said that SMI power has a positive relationship with consumers' attitudes toward the brand among Instagram users in the United States.

Table 5 .
Hypothesis test result