Trust Mediates Religious Commitment and Halal Cosmetics Consumption: A SEM Analysis

The main purpose of this study is to contribute to the literature by focusing on the relationship between religious commitment and halal cosmetics consumption and identifying the mediating effect of middle-class Muslim females’ trust on halal cosmetics consumption. We used structural modeling (Lisrel) in investigating these relationship and applied non-probability convenience sampling to administer questionnaires among 450 middle-class Muslim female respondents currently living in South Sulawesi, Indonesia. The results showed that the direct effect of religious commitment, ability, benevolence, and integrity on trust showed T-Values of 1.71, 2.61, -1.92. These indicate that only ability has a significant and positive effect on trust, but the direct effect of trust, religious commitment, ability, benevolence, and integrity on halal cosmetics consumption showed T-Values of 4.11, 1.71, 2.60, -2.12. These indicate that only trust and ability have significant and positive effects on halal cosmetics consumption. Meanwhile the mediating role of trust showed T-Values of 1.41, 2.79, -2.12, and 1.86. These indicate that trust was significantly and positively found in mediating the relationship between ability and halal cosmetics consumption, but in the relationship between benevolence and halal cosmetics consumption, it was also significantly found but negatively.


INTRODUCTION
Halal cosmetics are in great demand by people in Indonesia.This condition is evidenced by the large absorption of Indonesian halal cosmetics in various segments of the beauty industry as the main driver (AHK Indonesien, 2022).Muslim consumer spending on cosmetics in Indonesia was the second highest globally in 2020 (Zhulkarnain, 2022).This proves that the COVID-19 pandemic was not a barrier for consumers to buy cosmetics (Zhulkarnain, 2022).Indonesian halal cosmetic industry worth $4.19 billion is expected to grow at almost 8% per year until 2025 (Indonesia Halal Lifestyle Center, 2021).From January to August 2020, the exports of Indonesian cosmetic products, including halal cosmetics, reached US$ 135.67 million (AHK Indonesia, 2022).The large demand for halal cosmetics can also be seen from the income of the cosmetic industry in Indonesia reaching Rp. 199 trillion in 2022 and is predicted to reach Rp. 135 trillion in 2024 (The Conversation, 2022).This shows that halal cosmetics products attracts the attention of Muslim female consumers in Indonesia.
Our study incorporates Model of Trust (MoT) developed by Mayer and his colleagues (Mayer et al., 1995).MoT has been applied in various fields (Fricker Jr et al., 2013;Utami, 2015;Wong, 2017).According to MoT, trust as a mediator of an actual behavior, is formed by three components; they are ability, benevolence, and integrity (Mayer et al., 1995).This study retained the variables of ability, benevolence, integrity, but it seeks to extend MoT by adding religious commitment as a predictor in that model.Religious commitment is religiosity at the individual level, reflected in the commitment to a religion and its teachings, and the individual's behavior reflected in that commitment (Sobari et al., 2022;Yener, 2014).Religious commitment can effect the basic components of behavior, such as interpreting, identifying, and feeling phenomena that occur.It means that such commitment can leave a big mark on one's beliefs (Dinh et al., 2022;Sutrisno & Dularif, 2020).
According to earlier studies, professionals haven't given the use of MoT in the context of using halal cosmetics enough thought.In fact, MoT describes a certain action taken by an individual on the basis of the beliefs he has, including halal cosmetics consumtion.Previous studies pertaining to halal cosmetics tend to address three issues.First, the studies focussing on consumers' perceptions.For example, Bhutto et al. (2022) extended the Planned Behavior Theory (PBT) in the context of halal cosmetics adoption, while Handriana et al. (2020) analyzed the behavior of the millennial generation in buying halal cosmetic products, and Ali et al. (2019) examined the influencing factors on halal cosmetics adoption.Briliana & Mursito (2017) investigated investigated the history and outcomes of halal cosmetics among young Muslim female consumers in Indonesia.Second, the studies focussing on the certification of halal cosmetics.For example, Annabi & Ibidapo-Obe (2017) concentrated on the UK, Zulkifli & Rahman (2019) focused on Malaysia, and Widjaja & Sijabat (2021) researched in Indonesia.Third, the studies focussing on halal cosmetic industry.For example, Sugibayashi et al. (2019) did a formulation of halal cosmetic manufacturing technology guidelines, while Tukiran & Anuar (2022) investigated the concept of halal in beauty soap production, and Malini et al. (2022) researched halal labeling in the cosmetic industry.These show that only this study used MoT in investigating halal cosmetics consumption, and this is its originality.
In general, the purpose of this study is to complete earlier research that did not sufficiently examine the mediating impact of trust on on halal cosmetics consumption among middle-class Muslim females.Meanwhile, in particular its purpose is to examine the effect of religious commitment, ability, benevolence, integrity on halal cosmetics consumption through the mediation of trust of middleclass Muslim females.These purposes will be used as the basis to formulate a postulate for the behavior of middle-class Muslim females to purchase halal cosmetics.Therefore, this study is the first that aims to examine the mediating role of trust in shaping the behavior of middle-class Muslim women who purchase halal cosmetics.Overall, this study contributes to the literature by focusing on the relationship between religious commitment and the consumption of halal cosmetics and identifying the impact of middle-class Muslim females' trust on their consumption of halal cosmetics.

Model of Trust (MoT)
In 1995 Mayer and his colleagues developed a model called MoT (Fricker Jr et al., 2013).Trust in MoT, is a result of trustworthiness.Trustworthiness in this case is a function of the trustee's perception of his competence, good faith, and integrity.These trustworthiness components, along with an individual's propensity to trust, theoretically predict the level of trust that a colonist has in a fiduciary (Mayer et al, 1995).MoT is a feedback loop in which reliability is tested at risk, results are observed, and then the trustee's rating is updated via the trustee (Mayer et al, 1995).The model was originally developed to explain trust in business relationships.For example, if employees perceive their managers as sufficiently competent, benevolent, and honest, they are more likely to trust their managers, leaving them vulnerable to their instructions, guidance, etc. (Wong, 2017).This model was then used in various fields (Fricker Jr et al., 2013;Utami, 2015;Wong, 2017), including being used in halal cosmetics consumption among middle-class Muslim females.

Halal Cosmetics
Cosmetics are substances or preparations intended for application to the exterior of the human body, such as the epidermis (skin), hair, nails, lips, external genitalia, teeth, and the lining of the mouth.Even if cosmetics are not eaten and drunk, halal (permitted in Islam) is important because these products are used in worship, such as praying and reading the holy book.Alcohol content (ethanol) in cosmetics can be classified as halal only when it comes from natural or synthetic fermentation processes, not from the alcohol industry (The Conversation, 2022).In the Indonesian context, halal cosmetics are products that have been recognized as halal by the Ministry of Religion's Halal Assurance Administration based on a written halal recommendation issued by Indonesian Council of Ulama (Majelis Ulama Indonesia, 2013).There are 15 halal cosmetic brands produced and marketed in Indonesia and even exported abroad, namely: Wardah, Emina, Oriflame, Somethinc, MAKE OVER, Madame Gie, Avoskin, ERHA, BLP Beauty, Garnier, Maybelline, Sariayu Martha Tilaar, ESQA Cosmetics, Rollover Reaction, Tropical Studio (Aura, 2022).This is what halal cosmetics mean in this study.

Religious Commitment and Trust
Religious commitment is an individual-level religion reflected in a commitment to a religion and its teachings, and individual behavior is reflected in commitment (Jianfeng et al., 2009).This means that religious obligations reflect an individual's adherence to particular religious values, beliefs, and practices (Abu-alhaija, 2018).(Abu-alhaija, 2018).Handriana et al. (2020) found that religion plays an important role in shaping consumer behavior and attitudes toward the consumption of products, both in the form of goods and services.Religiousness is how devoted one is to the teachings of Islam.If religion is the values and doctrines that make up a belief system, then diversity is the empirical form of that belief (Mustaqim et al., 2023).Therefore, it is recommended to only consume goods that are permitted in Islam (Suleman et al., 2021).The knowledge presented above demonstrates the high correlation between religious commitment and trust in halal cosmetics consumption.Thus the hypothesis that can be proposed is: H1: Religious commitment effects trust and halal cosmetics consumption.

Ability, Benevolence, Intergrity and Trust
There are three components that make up trust in others.they are ability, benevolence, and integrity (Cheung & Lai, 2022;Khuwaja et al., 2020;Maham & Bhatti, 2019).Ability alludes to the capacity and characteristics of a vendor/ organization to impact and underwrite a specific range (Mayer et al, 1995).In this case, how the dealer can give, serve and secure the exchange from impedances by others (Wong, 2017).This implies that shoppers get affirmations of fulfillment and security from vendors when conducting exchanges.Ability incorporates competence, involvement, organization mindfulness, and logical capability (Mayer et al, 1995).Benevolence is the seller's eagerness to form commonly useful fulfillment between himself and the buyer.Maximizing benefits for dealers and increment client fulfillment.In expansion to seeking after the greatest benefit, sales representatives put awesome significance on accomplishing client fulfillment (Mayer et al, 1995).Benevolence incorporates consideration, compassion, belief, and acknowledgment.Integrity alludes to a seller's behavior and habits in doing trade.The data given to customers is really honest (Mayer et al, 1995).Integrity can be seen in terms of decency, fulfillment, devotion, trustworthiness, unwavering quality, and unwavering quality (Mayer et al, 1995).The above understanding shows that ability, benevolence, and integrity have a strong relationship with trust in halal cosmetics consumption.Thus the hypotheses that can be proposed are: H2: Ability effects trust and cosmetics consumption H3: Benevolence effects trust and halal cosmetics consumption H4: Integrity effects trust and halal cosmetics consumption

Trust and Halal Cosmetics Consumption
Trust is the foundation of business (Azizi et al., 2017).A business transaction between two or more parties will occur if each trusts each other (Nguyen & Ha, 2021).Trust cannot simply be recognized by other parties/business partners, but must be built from scratch and can be proven (Yuliati et al., 2020).Trust is is the establishment of commerce (Azizi et al., 2017).Commercial exchanges between two or more parties happen when each trusts the other (Nguyen & Ha, 2021).
Trust is known to positively influence online consumer purchase intentions and it is always rated as the most important factor (Rokhman & Andiani, 2020).
In the meantime, according to Gefen & Straub (2004), trust is an evaluation of a person's relationship with other individuals who will carry out certain exchanges agreeing to the desires of their trusted individuals in an environment full of vulnerability.Based on the over definition, it can be expressed that trust is the certain parties to others in conducting value-based connections based on a conviction that the individual they believe will fulfill all their commitments appropriately as anticipated, including halal cosmetics consumption.Thus the hypothesis that can be proposed is: H5: Trust affects halal cosmetics consumption.Fishbein, & Ajzen (1975) in the Reasoned Action Theory (RAT), explained that belief will form one's attitude, so that it will effect one's intention and behavior.Based on this theory, a person's belief in halal cosmetics will effect their intensity to use them.Mayer et al. (1995) clarified that trust is the eagerness of a party to be defenseless to the activities of another party based on the desire that the other will perform a specific activity imperative to the trustor, independent of the capacity to screen or control that other party.Trust is shaped by three components; ability, benevolence, and integrity (Cheung & Lai, 2022;Khuwaja et al., 2020;Maham & Bhatti, 2019).While in Islam, every Muslim is required to only consume halal goods (Suleman et al., 2021).Thus the hypothesis that can be proposed is: H6: Trust can mediate between religious commitment and halal cosmetics consumption, ability and halal cosmetics consumption, benevolence and halal cosmetics consumption, and integrity and halal cosmetics consumption.

Conceptual Model of Study
Figure 1.

Conceptual model of study
The proposed structure model, shown in Figure 1, consists of three different types of variables and is based on the MoT extension that includes "religious commitment".The endogenous variable is "halal cosmetics consumption", and the exogenous variables include "religious commitment" (RC), "ability" (Ab), "benevolence" (Be), and "integrity" (In), as well as one mediating variable ("trust"; Tr).In conclusion, this way of thinking adds to the body of research by emphasizing the link between "religious commitment" (RC) and "halal cosmetics consumption" (HCC) and highlighting the mediating role of middleclass Muslim's "trust" (Tr).

Research Design
This study used structural modeling (SEM) in investigating the relationship between religious commitment and halal cosmetics consumption and identifying the mediating effect of middle-class Muslim females' trust on halal cosmetics consumption.We applied SEM because it allows us to simultaneously see relationships between variables in this study and consider the direct and indirect influences between them.So that it possible to create path models that can better explain the relationships between variables we investigated.

Measurement
The questionnaire was constructed with two main sections in order to validate the conceptual model (Figure 1) suggested in this study.The main section briefly describes the rationale for the consideration, the instructions for completing the questionnaire, and its relationship to socio-demographic information.The age, marital status, gender, occupation, education, and income of the respondent are all asked.The current section includes a multiple-choice item scale that adopts "a five-point Likert Scale" in order to improve the model structure.There are 37 questions in this part, including 7 about "religious commitment" (RC), 6 about "ability" (Ab), 6 about "benevolence" (Be), 6 about "integrity" (In), 6 about "trust" (Tr), and 6 about "halal cosmetics consumption" (HCC).

Data Collection
In order to collect data, a non-probability convenience sampling technique was used to distribute questionnaires to middle-class Muslim female respondents who were currently residing in South Sulawesi, Indonesia, during the final three months of 2022 (October to December).Through WhatsApp, the questionnaires were delivered to them online in Google Forms.There are 450 records in all of the responses from respondents that were successfully recorded on Google Forms.This suggests that 450 middle-class Muslim women in Indonesia took the test, according to the data obtained.The minimum test requirement for using multivariate analysis, according Hair et al. (2010), is ten times the number of disobedient questions.This requirement has been met (Kyriazos, 2018;Schermelleh-Engel et al., 2003).

SEM Analysis
The data that has been gathered is divided into two categories: the identities of the respondents and their responses.The identities of the respondents from the distributed questionnaire are represented in organizational shape, whereas the respondents' responses are initially examined while recently being represented, through three processes, as follows: (1) Factor analysis.For this stage, the IBM SPSS 25 application is used to indicate the key relationship between indicators and factors.Bartlett's Test of Sphericity ( 5%) and the KMO esteem (KMO 50%) are used to determine whether the test was successful.Additionally, Williams et al. (2010) mentions the value of the figure stacking (70%) and MSE (5%) for each indicator, (2) Measurement model assessment: In this step, the Lisrer 8.70 application (Joreskog & Sorbom, 1989) is used to determine the suitability of the proposed model for use in terms of the "Chi Square" (value [df; Sig.5%]), "Probability" (5%), "CMIN/DF" (2), and "RMSEA" (5%).In addition, the values of the following files (90%) show the fitting test of the show: "NFI", "NNFI", "CFI", "IFI", "GFI", and "AGFI" (Widarjono, 2010;Hair, et al., 2010),.To further emphasize that the indicators used are significant and the justifications used are sound (Yang & Montgomery, 2011), and (3) Structural model assessment.In this step, the "Weighted Slightest Square" (WLS) strategy is used with the Lisrer 8.70 application (Joreskog & Sorbom, 1989) to calculate the "R-Square" value and to determine the significance of the intervening impact (≥ 1.96; Rigdon & Hoyle, 1997) of exogenous factors on the endogenous factor.Age, marital status, occupation, level of education, monthly spending and pocket money, as well as the use of halal cosmetics products, are just a few of the demographic details listed in Table 1.The overall dominance of the respondents are students (56.70%) and lecturers (12.40%).Those who are not married are 65.30%, while those who are married are 32.40%.Respondents with the highest education level are Senior High School (38.70%), and those with a bachelor's degree are 34.90%.Pertaining to age, most of our respondents are in the "16-26" category (64.90%) and "27-36" category (16.20%).In the aspect of spending/ pocket money, it is in the "< IDR three million" category (78.40%) and the "> IDR three million to five million" category (13.60%).While in the aspect of halal cosmetics products used, Wardah is 50.24%, while Emina is 13.81%.Table 2 shows the variables used in this study.RC-coded variables consist of 7 items,6 Ab,6 Be,6 In,6 Tr,and 6 HCC.The results of the factor analysis shown in Table 2 are: (1) the values of Kaiser-Meyer-Olkin (KMO, ≥ 50%) for all variables are > 75%, it means that the sample used in this study is sufficient, (2) the values of Bartlett's Test of Sphericity for all variables are < 5%, so the factor analysis applied in this study can be continued to the next analysis (Confirmatory factor analysis), and (3) the values of MSE (≥ 50%) and Extraction (≥ 50%) for most indicators are > 60%, so most of them meet requirements for further processing and can be used to explain the factors applied.But the values of RC.4,RC.7, HCC.4, and HCC.5 are < 50%, so they must be removed for the next analysis.Source : Primary data processed, 2023

Test for Fit Goodness
According to Table 3, the "probability" is 0.0% (> 5%), the value of "Chi-Square" (CMIN) is 2551.48,and the value of "RMSEA" is 9.1% ( 5%).It was determined that these data did not fit the suggested model (Williams et al., 2009).As a result, the measuring model needs to be modified.Whereas according to Table 4, the "probability" is 35.9% (> 5%), the "Chi-Square" value is 41.57, and the "RMSEA" value is 1.2% ( 5%).After adjustment, it was discovered that these were suitable for the suggested model (Hulland et al., 1996).The values of "CMIN/DF", "NFI", "NNFI", "CFI", "IFI", "GFI", and "AGFI" can also be considered as fitting tests of the model (Widarjono, 2010;Hair, et al., 2010), and they were found to be fit.Table 5 demonstrates that the loading values of the 12 investigated indicators are around 60% with T-Values around 12.66.All indicators were thus determined to be genuine and meaningful (after adjustment).While Table 6 displays "CR" and "AVE" values of 71% and 56%, respectively.Also determined to be dependable were all variables.According to Hair et al. (2010), a construct is dependable if its "Construct Reliability" (CR) value is greater than 70% and its "Variance Extracted" (AVE) value is less than 50%.

Structural Model Assesment
Findings of R-Square 78% and 97% are two of the obtained R-Square values.The first is that, according to the suggested model, "religious commitment", "ability", "benevolence", and "integrity" make a significant contribution to "halal cosmetics consumption".The second is that its proposed model also shows a robust contribution from "trust" (Tr), "religious commitment" (RC), "ability" (Ab), "benevolence" (Be), and "integrity" (In) to "halal cosmetics consumption" (HCC).

Hypotheses test
Table 7. Table 7 shows that all direct effects (H1.a-b,H2.a-b, H3.a-b, H4.a-b, and H5) were not statistically found significant and positive, but only H2.a, H2.b, and H5 were found significant and positive.H3.b was also found significant but negative.Whereas all indirect effects (H6.a,H6.b, H6.c, and H6.d) are also found the same, only H6.b was statistically found significant and positive, while H6c was also found significant but negative.They were significant as they followed the t-value of more than 1.96 (Joreskog & Sorbom, 1989).

RESULTS AND DISCUSSION
The direct effect of religious commitment, ability, benevolence, and integrity on trust shows T-Values of 1.71, 2.61, -1.92, and 1.78 with B-Values of 0.28, 0.95, -0.86, and 0.46.These values indicate that ability has a significant and positive effect on trust.These results supported H2.a only.Therefore, increasing the ability level of halal cosmetics sellers will increase the trust of middle class Muslim females to purchase halal cosmetics.This finding is in line with previous studies by Cheung & Lai (2022), Khuwaja et al. (2020) and Maham & Bhatti (2019).Furthermore, it was found that religious commitment had no significant effect on trust.These results did not support H1a.This means that religious commitment of middle-class Muslim females does not form their trust to purchase halal cosmetics.This finding is not in agreement with the previous studies by Lina et. al. (2022), andMohamed &Daud (2012).Likewise, benevolence and integrity have no significant effect on trust.These results did not support H3a and H4a.This means that benevolence and integrity of halal cosmetic sellers do not form the trust of middle-class Muslim females to purchase halal cosmetics.This finding is inconsistent with previous studies by Cheung & Lai (2022), Khuwaja et al. (2020) and Maham & Bhatti (2019).
Whereas, the direct effect of trust, religious commitment, ability, benevolence, and integrity on halal cosmetics consumption shows T-Values of 4.11, 1.71, 2.60, -2.12, and 1.88 with B-Values of 0.69, 0.31, 1.05, -1.06, and 0.55.These values indicate that trust and ability have a significant and positive effect on halal cosmetics consumption.These results supported H2.b, and H5.Therefore, increasing the ability level of halal cosmetics sellers and the trust level of middle class Muslim females to purchase halal cosmetics will increase their halal cosmetics consumption.This finding is in line with previous studies by Cheung & Lai (2022), Khuwaja et al. (2020), andMaham &Bhatti (2019).Furthermore, benevolence has a significant effect on halal cosmetics consumption, but negative.These results supported H3b.This means that the benevolence of halal cosmetics sellers is not taken into consideration by middle-class Muslim females to purchase halal cosmetics.This finding is not in line with previous studies by Cheung & Lai (2022), Khuwaja et al. (2020), and Maham & Bhatti (2019).Meanwhile, religious commitment and integrity have no significant effect on halal cosmetics consumption.These ruesults did not support H1.b and H4.b.This means that religious commitment and integrity do not form the consumption of middle-class Muslim women toward halal cosmetics.This finding is not in line with the previous study on religious commitment by Lina et al. (2022), Mohamed & Daud (2012), and with the previous study on integrity by Cheung & Lai (2022), Khuwaja et al. (2020), andMaham &Bhatti (2019).
However, when testing the mediating role of trust on halal cosmetics consumption, it showed T-Values of 1.41, 2.79, -2.12, and 1.86 with B-Values of 0.19, 0.66, -0.59, and 0.32 for religious commitment and halal cosmetics consumption, ability and halal cosmetics consumption, benevolence and halal cosmetics consumption, and integrity and halal cosmetics consumption.The significant and positive mediating effect of trust was only found in the relationship between ability and halal cosmetics consumption.These results supported H6-a and H6.c.Furthermore, in the relationship between benevolence and halal cosmetics consumption, the mediating effect of trust was also found significant but negative.These results supported H6.a and H6.c.However, trust did not mediate the relationship between religious commitment and halal cosmetics consumption, and integrity and halal cosmetics consumption.This finding did not support H6.b and H6.d, but consistent with the previous study on religious commitment by Lina et al. (2022), Mohamed & Daud (2012), and with the previous study on integrity by Cheung & Lai (2022), Khuwaja et al. (2020) and Maham & Bhatti (2019).
The significant (positive and negative) results of mediating role of trust toward halal cosmetics consumption in this study are used as the basis for formulating a postulate through synchronization with the results of the Goodness of Fit test.The synchronization showed that all variables that have full mediation are both positive; namely the mediating effect of trust in the relationship between ability and halal cosmetics consumption, and negative; namely the mediating effect of trust in the relationship between benevolence and halal cosmetics consumption, left only two indicators each, they are Ab: Ab.2, and Ab.3, Be: Be.3,and Be.6,Tr: Tr.2,and Tr.4,and HCC: HCC.3,and HCC.4 (see,Table 5).If these indicators are elaborated, it can be formulated a postulate for middleclass Muslim females in purchasing halal cosmetics that middle-class Muslim females purchase halal cosmetics, namely obediently consume them and keep buying even though their brand are not popular, formed by their belief in the products, namely feeling satisfied to use them, and also on the basis of obediently consuming halal products.Their trust is formed because the sellers care about the wishes of consumers, in addition, the products have often been used by the public.However, it is not a consideration for them to purchase halal cosmetics if only the sellers are kind to them, namely only the sellers who serve well and well known in the public without considering other important aspects, such as quality, price, income, and others (Handriana et al., 2020;Ali et al., 2019;Briliana & Mursito, 2017).

CONCLUSION
The model put up to study Muslim females of middle class who use halal cosmetics did not perform well.This indicates that the MoT being used cannot properly accept it.A different hypothesis is therefore required to back it up.In any event, the study's key finding is that trust was able to significantly and positively moderate the relationship between ability and the use of halal cosmetics.Additionally, it was able to considerably but unfavorably buffer the link between benevolence and the use of halal cosmetics.Thus, the consumption of halal cosmetics by Muslim females of middle class is influenced by their belief in halal cosmetics, which is influenced by the concern of halal cosmetics sellers toward their desires and the products sold are frequently used by the public.However, they do not take into account the shopkeepers' generosity in allowing them to buy halal cosmetics.
Although this study was successful in demonstrating a limited mediating influence of trust on middle-class Muslim females' consumption of halal cosmetics, it appears that there are still some limitations that can be taken into consideration for future research that will hopefully address these issues.One of these is the geographical strategy; we only examined a general group of Muslim women in Indonesia's middle class.As a result, it is important to emphasize that the outcomes of research in this field depend on the geographic characteristics of the study population and may have comparable or dissimilar outcomes for other study population segments.

Validity and Realibility Test Table 5. Findings of the Validity Test
Trust Mediates Religious Commitment and Halal Cosmetics Consumption: …..