Halal cosmetics and behavior of Muslim women in Indonesia: the study of antecedents and consequences

Purpose: This study aims to analyze the antecedents and consequences of the attitudes of Muslim women towards purchase intention of halal cosmetics. Methodology: This sample was taken by using the purposive sampling technique. In total, 232 Muslim women respondents are aware of halal-labeled cosmetics. Data is processed using Partial Least Square-Structural Equation Modeling (PLS-SEM) method with SmartPLS. Findings: The results of this study indicate that religiosity has a positive and significant influence on attitudes towards halal cosmetics, knowledge has a positive and significant influence on attitudes towards halal cosmetics, subjective norms have a positive and significant influence on attitudes towards halal cosmetics, and attitudes towards halal cosmetics has a positive and significant influence on purchase intention of halal cosmetics. Originality: Due to the inconsistent findings from past studies, the results of this study are expected to contribute to the literature in evaluating the behavioral intentions of Muslim women in Indonesia towards halal cosmetics. Cosmetics brand manufacturers can also use the results of this study for the halal labeling process of their products.


Introduction
For a woman, cosmetics represent a symbol of beauty. Cosmetics seem to be an immediate need for certain females, along with the times. Women not only use makeup to enhance the beauty of their faces but also to increase self-confidence. The Ministry of Industry of the Republic of Indonesia in 2018 stated that Indonesia is fertile land for the cosmetics industry with a population of around 250 million, making Indonesia a promising market for cosmetics companies. The existence of various cosmetics products on the market influences purchasing decisions and makes consumers more selective in choosing cosmetic brands. Consumers make many considerations before deciding to buy a product, such as seeing the label on the product packaging.
According to Thomson Reuters' report (2014), the amount of consumption of the world's Muslim population reached 17.4% of total world consumption. Indonesia takes the first position in the world's Muslim consumption because it has the largest Muslim population. For Muslims, consuming halal products is an obligation. Muslim women must be extra careful in choosing cosmetics and ensure that their cosmetics comply with sharia law by being halal. Yuswohady (2015) states that producers are busy embedding halal labels along with increasing awareness of the benefits of halal from Muslim consumers because it will also affect consumer purchase intentions of halal products. In choosing and using halal cosmetic products, consumers must have the intention to buy these cosmetic products. According to Ajzen (1991), behavioral intention is a person's tendency to choose to do or not do a job. To predict attitudes accurately, marketers must know consumer relationship interests because, according to Peter & Olson (2000), purchase intention is often the best way to predict future purchase behavior. In the consumer behavior model, purchase intention is formed from beliefs about a product or brand and its attributes, followed by a shift in attitudes towards purchasing action and is influenced by internal and external factors. Thus, purchase intention is influenced by consumer attitudes towards halal cosmetic products.
Consumer attitudes are only based on belief and knowledge about certain goods or commodities. It results from gathering information that affects individual goals to behave in a certain attitude (Putri et al., 2019). For example, consumers' awareness of halal cosmetics has led to the trend of hijrah, halal market, and halal cosmetics among Indonesian people. Therefore, the manufacturer must be able to meet the demand from consumers to buy halal products. In this case, it can be said that attitudes towards halal cosmetic products are influenced by religiosity, halal knowledge, and subjective norms as a function of belief that determines the thinking of individual consumers.
Religiosity is a factor influencing consumer attitudes towards halal cosmetic products. Religiosity is the extent to which an individual commits to his religion and to whom that religion is reflected in the attitudes and behavior of the individual (Briliana & Mursito, 2017). Religion is one of the factors that influence the lifestyle of consumers, which in turn will also influence decision-making behavior. The influence of religion on various aspects of human life has been investigated by many researchers in the social sciences (Lada & Amin, 2009;Ireland & Rajabzadeh, 2011;Mukhtar & Butt, 2012;Teng & Wan, 2013;Hassan & Harun, 2016).
Apart from religiosity, attitudes towards halal cosmetic products are also influenced by halal knowledge. It refers to the expertise and skills acquired by a person or group of people through a theoretical or practical understanding of a subject (Rahman et al., 2015). Rao & Monroe (1988) stated that product knowledge can influence how consumers access products. Furthermore, consumer knowledge is a fundamental concept for understanding consumer decisions in shaping and changing attitudes and adopting new technology (Oh & Abraham, 2016). Thus, the knowledge possessed by consumers about things that are forbidden and permissible in Islamic law will affect attitudes towards halal cosmetic products.
Not only do religiosity and halal knowledge influence attitudes towards halal cosmetic products, but also subjective norms. Lada & Amin (2009) stated that subjective norm is a function of beliefs that determine individuals' acceptance of certain behaviors by a group. Consumers can make purchases because they are influenced by the environment, which is often called subjective norms. Subjective norms determine how a person acts and behaves. Mukhtar & Butt (2012) revealed that subjective norms and intrapersonal religiosity positively affect attitudes towards halal products.
Halal cosmetics are starting to develop rapidly in Indonesia, and a growing number of researchers have conducted studies related to the attitudes of Muslim women towards purchasing intention of halal cosmetic products. Many studies have produced positive relationships between these variables, but some studies showed negative results. For example, Aji (2017) found that Islamic religiosity does not affect attitudes towards halal labeled products and intention to buy products labeled halal. In addition, Rahman et al. (2015) also found that knowledge did not significantly influence attitudes towards halal cosmetics. Likewise, the results of Nora's (2016) study found that inaction is an intermediary for the influence of religiosity on purchase intention.
Thus, empirically, the findings of previous studies related to the relationship between attitudes towards halal cosmetics and purchase intention towards halal cosmetics are still inconsistent. This becomes a problem and a research gap to be closed, so it is urgent why this research needs to be done. In general, this research model replicated Briliana & Mursito's (2017) by changing the context to halal-labeled cosmetics. The formulation of the problem in this study is as follows: 1. Does religiosity have a positive impact on attitudes towards halal cosmetic products? 2. Does halal knowledge have a positive impact on attitudes towards halal cosmetic products? 3. Do subjective norms have a positive impact on attitudes towards halal cosmetic products? 4. Does the attitude towards halal cosmetic products have a positive impact on purchase intention?

Literature Review and Hypotheses Development
This study replicates the research model of Briliana & Mursito (2017). Their research examines several variables such as knowledge, five dimensions of religiosity, subjective norms, attitudes towards halal cosmetics, and purchase intention with a sample of Muslim women in Jakarta. This study modifies the five-dimensional variable of religiosity into intrinsic and extrinsic religiosity variables in the context of Muslim women in Indonesia. This research model was also developed based on empirical studies on previous studies that raised the relationship between the four variables: knowledge, religiosity, attitudes, and purchase intention (Rahman et al., 2015;Aji, 2017;Nora, 2016;and Briliana & Mursito, 2017). The results of their research in more detail can be seen in Table 1. • Knowledge is significantly related to attitude towards halal cosmetic products • Ideological found to be a predictor of attitudes towards halal cosmetic products • Ritualistic has a negative influence on attitude towards halal cosmetic products • Intellectual has a positive and significant influence on attitude toward halal cosmetic products • Consequential has a positive and significant influence on attitude toward halal cosmetic products • Experimental has a positive and significant influence on attitude toward halal cosmetic products • Subjective norm is significantly related to attitude towards halal cosmetic products • Attitude towards halal cosmetic products has a positive and significant influence on purchase intention.

Religiosity and Attitude toward Halal Cosmetics
According to Souiden & Rani (2015), the level of religiosity is a determinant of the attitudes and behavior of Muslim consumers. Religiosity significantly affects a person's intention in choosing halal products (Balques et al., 2017;Mukhtar & Butt, 2012;Tabassi, 2012). Consumers with a high level of religiosity are very concerned about the halal label used to identify halal cosmetic products. This statement is supported by research by Adriani et al. (2020), which states that consumers who have high religiosity will be more concerned and aware of the importance of halal cosmetics. However, Aji (2017) found that religiosity does not influence attitudes towards halal-labeled products and intention to buy halal-labeled products. In this study, the researcher argues that high religiosity can affect consumer attitudes towards halal cosmetics. The more religious consumers are, the more likely they are to choose halal cosmetics. Thus, based on this explanation, the formulated hypothesis is as follows: H1: Religiosity has a positive impact on attitudes towards halal cosmetic products

Halal Knowledge and Attitudes toward Halal Cosmetics
Knowledge influences intention through attitude; hence, increased knowledge is likely to influence intention (Bang et al., 2000;Aji et al., 2020). According to Adriani et al. (2020), knowledge of halal products positively and significantly affects consumer attitudes and purchase intentions. Larasati et al. (2018) supported this statement, who found that knowledge positively influences consumer attitudes towards halal cosmetic products. However, Rahman et al. (2015) revealed that knowledge did not significantly influence attitudes towards halal cosmetic products. In this study, researchers argue that when someone has good knowledge of halal cosmetic products, it will affect attitudes towards halal cosmetic products. When consumers understand the context of halal and haram in cosmetic products, Muslim consumers will make halal cosmetics their choice. Based on this explanation, the hypotheses that can be formulated are as follows: H2: Halal knowledge has a positive impact on attitudes towards halal cosmetic products

Subjective Norms and Attitudes toward Halal Cosmetics
Mukhtar & Butt (2012) revealed that subjective norms, attitudes towards halal products, and intrapersonal religiosity positively affect consumer attitudes towards halal products. Their findings align with Endah's (2014) study, which shows that subjective norms greatly influence consumer attitudes about halal cosmetics. Briliana & Mursito (2017) also supported this statement, which found that subjective norms are significantly related to attitudes towards halal cosmetic products. A person will consider several things when making decisions, including references from his friends or family. Therefore, researchers argue that in this study, the individual decision to purchase halal cosmetic products will be influenced by family, friends, or important people in their lives. Based on this explanation, the hypotheses that can be formulated are as follows: H3: Subjective norms have a positive impact on attitudes towards halal cosmetic products

Attitude toward Halal Cosmetics and Purchase Intention
Based on the Theory of Planned Behavior (TPB), an individual's behavioral intention is influenced by attitudes, subjective norms, and perceived behavior control. Previous research has found that the more positive customer attitudes, the more positive their purchase intention for halal cosmetics (Balque et al., 2017). Lada & Amin (2009) found a positive relationship between attitude and intention to choose halal products. Alam & Sayuti (2011) reinforced this statement, who found a significant and positive relationship between attitude and intention to buy halal products. Rahman et al. (2015) and Divianjella et al. (2021) also found that attitudes towards halal cosmetics affect purchase intentions. In the context of this research, consumers' attitudes towards a product will encourage them to purchase halal-labeled cosmetics. Therefore, the authors assumed that when someone feels that halal cosmetics are good, they will have an intention to buy them. Based on this explanation, the hypotheses that can be formulated are as follows: H4: Attitude towards halal cosmetic products has a positive impact on purchase intention

Methodology Data Collection and Sampling Techniques
In general, this study used a quantitative approach. The object of this research is halal cosmetics. It is defined as cosmetics that are made from ingredients that do not contain haram elements and have a halal label issued by the Indonesian Ulema Council (MUI), namely Wardah, Mazaya, Safi, Zoya, Purbasari, and Sariayu. The research location was not limited to Yogyakarta because the questionnaires were distributed online via Microsoft Form, then distributed to social media. The questionnaire was distributed to several social media such as Instagram, Twitter, and TikTok. The distribution of questionnaires on Instagram was carried out by embedding the questionnaire link in the bio and through direct messages. On Twitter, the questionnaire was distributed through community accounts around schools and lectures. Then on TikTok, the questionnaire was distributed with live features carried out by the author. The authors also distributed the questionnaire via LINE and WhatsApp. The sample was selected using a purposive sampling method, with the criteria of Muslim women who knew halal labeled cosmetics. The sample size in this study uses the Roscoe formula (1975), which can be calculated with minimum sample size: 5 x number of observations (items) and maximum sample size: 10 x number of observations (items). In this study, there were 25 observations (items), so that the minimum size was 5 x 25 = 125 and a maximum of 10 x 25 = 250.

Operational Definition and Variable Measurement
Johnson et al. (2001) define religiosity as an individual's commitment to religion and acknowledge its teachings, such as individual attitudes and behavior. Islamic religiosity is defined by El-Menouar (2014) as Muslim faith in the pillars of Islam and the pillars of faith, as well as behavior in practicing religious teachings. In this study, religiosity is operationally defined as how far a Muslim woman is committed to her religion in carrying out the pillars of Islam and her fear of sin. The higher the religiosity of Muslim women, the more careful they will determine the cosmetics to be used. This study uses measurement items taken from the research of Alam et al. (2012), including "I always pray five times a day", "I always fast fully in the month of Ramadan", "I always take the time to read the Koran", " If my assets are sufficient, I will pay zakat "," I try to comply with Islamic teachings in all aspects of life "," I always try to avoid small and big sins ".
Maichum, Parichatnon, & Peng (2017) said that knowledge is essential for consumers in making product purchase decisions, especially halal products. Operationally, halal knowledge in this study represents how well Muslim women understand haram and halal in halal cosmetics. This study uses measurement items taken from the research of Abd Rahman et al. (2015), namely "I understand halal and haram related to cosmetics", "I have sufficient knowledge of halal and haram cosmetics", "I have sufficient knowledge to distinguishes between halal and haram cosmetics "," I know the issues related to the prohibited ingredients in cosmetics "," I understand the rules of halal certification for cosmetic products ".
Theoretically, Lada & Amin (2009) explain that subjective norms are a function of beliefs that determine a person's thinking about accepting specific behavior by a related group, for example consuming halal products. As for operationally, this study defines subjective norms with the influence of friends and family in making decisions to choose halal cosmetic products. This study adapted a 5-item measurement from Alam et al. (2012) including, "People who are important to me support me to use halal cosmetics", "Most of my friends who have used halal cosmetics agree that halal cosmetics is better", "In choosing cosmetics, I follow my friends' choices", etcetera.
In this study, attitudes are operationally defined by the behavior of Muslim women towards halal cosmetics. A positive attitude of Muslim women towards halal cosmetics will be formed when they prefer cosmetics labeled halal, but if not, it will cause a negative attitude. This study uses measurement items taken from the research of Abd Rahman et al. (2015), namely "I like cosmetics labeled halal", "I have a positive impression of cosmetics with a halal label", "I feel happy when using cosmetics with a halal label", "Using halal cosmetics is my own choice".
Furthermore, purchase intention in this study is operationally defined by the desire of Muslim women to buy and use halal cosmetics in the future. This study uses measurement items taken from the research of Alam et al. (2012), namely "I intend to choose halal cosmetics", "I will most likely buy halal cosmetics", "I hope to buy halal cosmetics in the future", "I predict that I will buy halal cosmetics in the future", "I plan to buy halal cosmetics".

Respondent Demographic Characteristics
The questionnaire in this study was created using Microsoft Form. Overall, there were 318 respondents. Of the total 318 respondents obtained, 232 were usable following research objectives that focused on Muslim women. The first classification based on gender was 232 respondents (100%). In terms of marital status, five respondents were married (2.16%), while 227 respondents were unmarried (97.84%). Based on religion, all respondents are Muslim, as many as 232 respondents (100%). It was found that 92 respondents were aged <20 years (39.66%), 139 respondents were 20-30 years old (59.91%), the rest was only one respondent aged 31-40 years (0.43%). In terms of educational background, the respondents were dominated by a senior high school graduate 72.41%, 41 respondents were a bachelor's degree (17.67%), and 17 respondents were junior high school graduates (7.33%). Lastly, six respondents were associated with degrees (2.59 %).

Measurement Model Test: Validity and Reliability
Validity testing is divided into two, namely convergent validity and discriminant validity. The criteria for good convergent validity are AVE> 0.5 (Hair et al., 2017). As for the discriminant validity, the thing that must be considered is that each variable's square roots AVE value is greater than the correlation value of other variables. As shown in Table 4, the square roots AVE value in the diagonal column is greater than the score below. Therefore, it shows that all data are free from discriminant validity problems. A reliability test is done by considering the value of composite reliability (CR). The CR value is considered good if it is more than 0.70 (Hair et al., 2017). As presented in Table 3, the CR value of all variables is more than 0.70. Thus, it can be concluded that the measurement items in this study are reliable.

Inner Model Testing: Structural Model
R 2 is used to see how much the antecedent variable affects the dependent variable. For example, figure 1 explains that the R2 of the purchase intention variable for halal cosmetics is 64%, so it can be interpreted that purchase intention can be explained by 64% by the antecedent variable; other variables explain 36%. Then, the variable attitude towards halal cosmetics in this research model is explained by 51% by religiosity, halal knowledge, and subjective norms. Another 49% is influenced by other variables that are not described in this study.   The predictive relevance is used to analyze how relevant variables predict a model by looking at the value of Q2. Q2 testing was conducted to determine the goodness of fit in this study. The higher the Q2 value, the better the data. Both variables in this study can be suitable and have predictive relevance for value> 0. In Figure 1, we can see that intention to purchase halal cosmetics has a Q2 value of 0.50 (50%) while attitude towards halal cosmetics is 0.37 (37%). It shows that both variables have predictive relevance. Purchase intention was predicted by 50% by the attitude variable towards halal cosmetics. Religiosity, knowledge, and subjective norms predicted attitude towards halal cosmetics by 37%. Thus, it can be concluded that the model used in this study has good predictive relevance.   Figure 1 shows the path coefficient test results for each variable. Religiosity has a positive and significant effect on attitudes towards halal cosmetics (t-statistic = 4,468; p-value 0.000 <0.05). Halal knowledge has a positive and significant effect on attitudes towards halal cosmetics (t-statistic = 2.925; p-value 0.004 <0.05). Subjective norms has a positive and significant effect on attitudes towards halal cosmetics (t-statistic = 9.765; p-value 0.000 <0.05), and attitudes towards halal cosmetics have a positive and significant effect on purchase intention (t-statistics = 29.415; p-value = 0.000). These results indicate that all hypotheses are supported and have a significant relationship. More detailed information can also be seen in Table 5.

Discussion
In general, the results of statistical testing show that all the hypotheses in this study are supported. The results of this study are supported by several previous studies, such as research conducted by Larasati et al. (2018), which concluded that the level of religiosity that consumers have has a positive influence on their attitudes towards halal cosmetic products. In addition, similar research results were also found in Rahman et al. (2015), which showed the influence of religiosity on consumer attitudes towards halal cosmetics. The results obtained in this study indicate relevance to the market conditions for cosmetics labeled halal in Indonesia. Many consumers have a good attitude and choose halal cosmetics, so that cosmetics manufacturers start competing to embed halal labels on their products to meet the increasing demand of Muslim consumers. If it is related to this research, there is a possibility that a person's high level of religiosity affects their behavior towards halal cosmetics as their choice. The higher the consumer religiosity, the more influential it will be on consumer attitudes towards halal cosmetics.
In addition, Clarita & Setiowati (2020), in their research, found a positive and significant relationship between halal knowledge and attitudes towards halal cosmetics. Aji (2017), in his research, also found that attitudes towards halal-labeled products are influenced by knowledge of halal processes and procedures if consumers have low skepticism about the halal label. Tsakiridou (2012) found that a good level of education influences knowledge and primary considerations in buying products. Of course, respondents who are dominated by students get knowledge of religion at school and campus so that they have obtained a basic knowledge of halal and haram. Such religious knowledge is implemented in everyday life, including when buying a product that first sees its contents, whether it is safe and lawful or vice versa. This explanation indicates that Muslim women in Indonesia have sufficient knowledge about halal cosmetics and form a good attitude towards halal cosmetics.
This study supports the research conducted by Mukhtar & Butt (2012), which revealed that subjective norms positively influence attitudes towards halal products. This study also supports Endah (2014). The respondents of this study are dominated by those aged between 20-30 years old. During that age category, people are actively seeking information before making a purchase decision. They can get information search from friends, product reviews on social media, and their families. The amount of information and references obtained raises his attitude towards a product, in this case, related to halal cosmetics, so this research proves that subjective norms have a significant positive effect on consumer attitudes towards halal cosmetics. It means that before purchasing halal cosmetics, a person will depend on the individual or group he believes.
This finding also supports Rahman et al. (2015), who states that attitudes towards halal cosmetics have a significant positive effect on purchase intention. The better consumer attitudes towards halal cosmetics, the better the intention to purchase halal cosmetics. In connection with the respondents' jobs, most of them are students who are more educated and have broader insight. They can search for information related to products purchased from various sources, especially from social media. It can form a good attitude towards a product and then generate purchase intentions.

Conclusion
In conclusion, all hypotheses are supported. However, based on the results of R2, other antecedent variables can explain the dependent variable. For that, it is suggested for further research to consider other factors that are the influence of the dependent variable. The findings in this study also can be used by cosmetics brand manufacturers as one of the bases for making halal labeling on their products in the future.