Millennial Intention to Pay Zakat: The Effect of Religiosity and Attitudes

: This study investigates religiosity's influence on paying zakat intention for millennials with attitude as a mediation variable. A Structural Equation Model (SEM) approach tests the measurement and structural model. The sampling technique is judgmental sampling to get 100 respondents through questionnaires distributed according to the characteristics of predetermined respondents. This study explains that religiosity and attitude positively and significantly affect paying zakat intention for millennials. In addition, this study also shows that attitude mediates the relationship between religiosity and paying zakat intention for millennials. This study provides a managerial implication for zakat institutions to have a better understanding that religiosity and attitude are essential to encourage Muslim millennials to pay zakat and the significant role of millennials in alleviating poverty in Indonesia.


INTRODUCTION
Indonesia has the largest Muslim population globally. Based on Pew Research Center (2020) data, Indonesia has 229,620,000 Muslim or about 87% of the total population.  (2020) Being a country with the largest Muslim population and ranked 10th country with the highest economic levels globally, Indonesia has enormous growth potential in zakat (Badan Amil Zakat Nasional, 2020). However, the data states that the collection of zakat funds in Indonesia is still not optimal (Mukhibad et al., 2019). The percentage of national zakat in 2017 is only 2.9% of the potential zakat of 217 trillion rupiah (Badan Amil Zakat Nasional, 2018). Based on the results of the Potential Indicator of Zakat Mapping conducted by the BAZNAS (National Zakat Board) Center of Strategic Studies, the potential of zakat increased to 233.8 trillion rupiahs in 2019 (Badan Amil Zakat Nasional, 2019). With this vast potential, it can be one of the solutions to reduce poverty. Until now, Indonesia is facing rampant poverty and social inequality. The total poor population in Indonesia is about 53,973.71 thousand people, or about 19.97% of the total population (BPS, 2020a). One way that can be done to reduce the poverty level in Indonesia is through equal distribution of income between those who can afford it and those who cannot (Canggih et al., 2017). Prophet Muhammad SAW. has implemented the zakat collection mechanism and distribution of the zakat assets from those who can afford it to those who cannot (Ahmed & Md Salleh, 2016;Aziz & Mohamad, 2016).
Zakat is an obligation for Muslims (Wahab & Rahman, 2013). According to Martono et al. (2019), in one's wealth, there is zakat that must be paid. Zakat is the third pillar of Islam and has been arranged in the Qur'an, in Surah Al-Baqarah, verse 43: "And establish prayer, pay zakat, and bow ' be with those who bow ". According to Wulandari & Kassim (2016), zakat is also called Islamic tax obligatory for a Muslim who has reached (nishab and haul) to spend a certain amount of his wealth and give it to 8 asnaf (zakat recipient), which have been regulated in the Qur'an.
Although Muslims must pay their zakat, it turns out that the amount of annual zakat collection in Indonesia is relatively small compared to tax collection. However, tax heaven is given to those who pay zakat (Saad & Haniffa, 2014). This implies that Muslims have neglected the obligation to pay zakat, or they may think that zakat is less critical than taxes. Otherwise, paying zakat is not subject to sanctions by the Indonesian government (Abdullah & Sapiei, 2018). Even some studies show that many Muslims avoid the obligation to pay zakat (Abdullah & Sapiei, 2018;Andam & Osman, 2019). Even with a Muslim population of more than 229 million, Indonesia certainly has a relatively large potential for millennial muzakki (Al Athar & Al Arif, 2021).
The Central Statistics Agency (BPS) explained in the 2020 Population Census that the millennial generation was born from 1981-to 1996. In 2020, the pre-boomer generation was only 1.71% of the total population of Indonesia. The boomer generation is 12.03%, generation X is 20.93%, generation Z is 26.46%, post Z generation is 13.07%, and the millennial generation (generation Y) is 25.8% (BPS, 2020b).
Millennials mean those who will be 25 to 40 years old in 2021. The millennial generation in 2021, on average, work and already has their income. From this millennial generation, the potential for zakat can be realized by the potential for income zakat. Income zakat becomes mandatory if it meets the nishab and haul standards (Andam & Osman, 2019;Owoyemi, 2020). Since February 2021, the millennial generation's main job is earning Rp. 5 million rupiahs per month, according to the professional zakat nishab, leadership and management staff. The average income for this main job is Rp. 6,739,618 (BPS, 2021). With the millennial generation population who dominates more than other generations, there is a great potential for zakat from this millennial generation (Badan Amil Zakat Nasional, 2020). This potential will be achieved if the millennial generation understands the importance of donating money such as paying zakat or giving alms (Fahrullah et al., 2020).
The millennial generation also has excellent potential to give alms, infaq and waqf (Al Athar & Al Arif, 2021). Charities Aid Foundation (2021) states that in 2020, Indonesia will become the first country on the list of the most generous in the world.   (2021) With a large Muslim population in Indonesia which is ranked first as the most generous country globally, people's interest in donating money is very high, including paying zakat. For this reason, this study examines what factors must be considered to influence paying zakat intention so that zakat potential can be realized. Religion forms the motivation of individuals to carry out all the rules contained in it (Hassan & Pandey, 2020). Religiosity is important to determine the intention of zakat payment (Mukhibad et al., 2019). Muslims who have high religiosity will increasingly realize that zakat is an obligation (one of the pillars of Islam) and must be implemented.
In addition to religiosity, which is one of the variables in predicting paying zakat intention, attitude is also one of the determinants of paying zakat intention for millennials. Aji et al. (2021) prove that attitude has a role in determining online infaq intention. This shows that social pressures and relatives' influence can significantly encourage people's behaviour, especially in paying zakat. Those who can pay zakat are aware of the needs of people in need, and the impact of giving zakat will alleviate poverty (Andam & Osman, 2019). In this study attitude is a mediating variable. The research of Kaawaase & Nalukwago (2017) uses mediating variable which is attitude and uses a variable of religiosity and intention but the object is Islamic banking. Meanwhile, this study used zakat and millennial object. These two objects have not yet been fully explored in the literature.
Previous research analyzed several factors that could influence paying zakat intention. One of them is the religiosity of the Muslim community. Research from Aji et al. (2021) stated that the Theory of Reasoned Action (TRA) model found religiosity did not affect attitude towards online infaq intention. However, several other studies found that religiosity has a positive and significant effect on muzakki and the strength of their paying zakat intention. Muslims who have high religiosity will increasingly realize that zakat is an obligation for a Muslim and must be carried out as stated in the pillars of Islam (Mukhibad et al., 2019). Heikal et al. (2014), in their findings, concluded that attitude has the weakest relationship with paying zakat intention. Bananuka et al. (2020) prove that attitude influences determining paying zakat intention. This research was supported by Aji et al. (2021), who show that the effect of attitude on online infaq intention is proven to be significant. This proves that social pressure and relatives' influence can encourage people's intention, especially in paying zakat. Ahn et al. (2018) show that attitude significantly influences people's donating intention. Thus, the fact that attitude affects intention is widely accepted.
Another study conducted by Kaawaase & Nalukwago (2017) found that attitude is a mediator in the relationship between religiosity and using Islamic banking intention. In various studies conducted by Amin et al. (2014), Lujja et al. (2016) and Kaawaase & Nalukwago (2017), attitude is the most influential factor in explaining intention.
This study used several variables from the previous study, namely religiosity, attitude and paying zakat intention. The attitude became a mediator variable in this study. In addition, this study also uses research objects that are quite different from previous studies, namely millennials who are the generation earning at this time.

DEVELOPMENT OF HYPOTHESES
The above hypothesis is illustrated in the following research model that has been developed for research instructions. Figure 1. presents the relationship between the religiosity variable, paying zakat intention, and attitude. In addition, the mediating effect of attitude was also examined in this study.  Mukhibad et al. (2019) examine the factors that determine muzakki to pay zakat intention and provide an analysis that religiosity has a significant positive effect on muzakki to pay zakat intention. In line with Muslims who have high religiosity, they will increasingly realize that zakat is an obligation for Muslims and must be carried out as stated in the pillars of Islam (Mukhibad et al., 2019). Kaawaase & Nalukwago (2017) proves religiosity positively and significantly affects intention. Religiosity will have an impact on paying zakat intention. Because religiosity is a view related to the teachings of their respective religions and implementing these teachings will influence their paying zakat intention. Thus, this study proposes a hypothesis: H1. Religiosity has a positive effect on paying zakat intention for millennials. Felix & Braunsberger (2016) explains that religiosity positively affects attitude toward any object. The theory of consistency in attitude-behaviour research by Millar & Tesser (1989) also confirms that religiosity has a positive primary effect on attitude. Religiosity will be able to influence attitude. This is because a person's religiosity based on his understanding of each religion's teachings will cause a decision to take a positive or negative attitude. Thus, this study proposes a hypothesis: H2. Religiosity has a positive effect on attitude Aji et al. (2021) prove that the effect of attitude on online infaq intention is significant. This shows that social pressures and essential people in life can significantly encourage people's behaviour, especially in paying zakat. Bidin et al. (2009) also confirm that attitude strongly influences determining paying zakat intention. Attitude will be able to influence paying zakat intention. This is because people's perceptions of an attitude object are their perspectives on negative or positive attitudes. People who have a good attitude will have positive feelings about what they are doing. Therefore, this study proposes a hypothesis: H3. Attitude has a positive effect on paying zakat intention for millennial It is a strong possibility that believers if close to religious or religious influence their attitude to intention. Religion functions as a social identity that is guided by religious beliefs, and therefore greatly affects the psychology of people to cause a positive or negative attitude (Ysseldyk et al., 2010). Kaawaase & Nalukwago (2017) stated that the study of religiosity positively influenced attitudes towards using Islamic banking intention. The other research results by Bananuka et al.
(2020) stated that religiosity positively influences intention with attitude as a mediating variable. But Amin et al. (2011) show that religiosity has no significant influence on intention. From the discussion above, it can be believed that the attitude affects paying zakat intention positively but it is not clear whether religiosity can affect paying zakat intention, especially in Indonesia, which has great zakat potential but is still not optimal. It is also questionable whether the attitude mediates the relationship between religiosity and paying zakat intention for millennials. Therefore, this study proposes a hypothesis: H.4 Attitude mediate the relationship of religiosity on paying zakat intention for millennial

RESEARCH METHODS
This type of research is conclusive causal research with the main objective of testing specific hypotheses and obtaining evidence of causal relationships using a quantitative approach. The data source used is primary data from field studies obtained by distributing online questionnaires. The online questionnaire in this study was designed using a google form and distributed through social media platforms such as WhatsApp, Line, and Instagram. All questions were assessed using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).
Using a judgmental sampling technique, this study only selected Muslim respondents and the millennial generation, aged 25-40 years, to adjust to the characteristics set out in this study. The Structural Equation Model (SEM) approach was used to test the measurement and structural models. In the structural model, SEM was chosen because of its effectiveness in estimating direct and indirect effects in one model (Ghozali & Latan, 2015). The analytical method used is Partial Least Square (PLS). In the PLS analysis, there is a process of checking validity, instrument reliability, and multicollinearity, then testing the structural model, namely r-square, q-square, and hypothesis testing (resampling bootstrapping).
The population size of this study is not known with certainty. the baseline sample size is ten times the number of arrows hitting the latent variable in the PLS-SEM model (Hair et al., 2017). In this study, the most significant number of arrows hitting the variable was 3, so the minimum sample was 30. Referring to the opinion, so the sample used was 100 respondents.
All item measurements were taken from the previous literature. The variables in this study were divided into independent variables (X), dependent variables (Y), and mediating variables (Z). The independent variable is religiosity (X), the dependent variable is paying zakat intention (Y), and the mediating variable is attitude (Z). The following are statement items used in the study: b. I will pay zakat because it follows Islamic teachings c. If my income has reached the nishab (limit of property ownership of someone who is required to pay zakat), I will pay zakat. d. I will pay zakat because I want to get a reward and blessing. e. I am willing to pay zakat Religiosity a. I believe that Allah SWT. who created the universe and everything in it (Worthington et al., 2003) b

RESULTS AND DISCUSSION
The primary data of this study was obtained from distributing online questionnaires to Muslim millennials. The data collected were 147, but the data that could be used in this study were only 100. This is because the other 47 samples did not match the characteristics of the study that is Muslim respondents and the millennial generation aged 25-40 years. The following are the results of the characteristics of respondents based on gender, last education, and occupation: This study was dominated by female respondents with a percentage of 62%, then dominated by the last education of high school/equivalent, as many as 66 respondents with a percentage of 66%, and dominated by respondents with private jobs with as many as 29 respondents with a percentage of 29%. It can be concluded that the research respondents are following the current research because the majority have jobs and of course, have sufficient income to fulfil the requirements to pay zakat income.

Convergent Validity
The indicator is considered valid if it has an extreme loading value of 0.50 to 0.60 (Ghozali & Latan, 2015). From the analysis results, it is known that all items on the questionnaire are declared valid because the loading factor seen through the outer loading value of each statement item from the variables in this study is more significant than 0.50.

Discriminant Validity
A measurement model meets discriminant validity if the AVE value is more significant than 0.50 and the AVE root of a construct is greater than the correlation coefficient with other constructs (Chin, 1998;Höck & Ringle, 2006). The results show that the discriminant validity value of a variable is greater than the correlation value between variables, and the AVE value is greater than 0.50. From the analysis results, it is known that the outer model of this research has met discriminant validity.

Reliability Test
The composite reliability and Cronbach's alpha are good when the value is above 0.70 (Henseler et al., 2012;Garson, 2016). In this study, the results obtained that all variables have a value of more than 0.70. Therefore, all measurement models used in this study have high reliability.

Collinearity Statistics
The general rule is that problematic multicollinearity may exist when the factor coefficient of variance inflation (VIF) is higher than 4.0 (Garson, 2016). Table 9. presents the results of the VIF value of each exogenous variable to the endogenous variable not exceeding the limit of 4.0.  (2022) In addition, it can also be confirmed through a tolerance value equal to 1.0 minus R² and a tolerance < 0.20 indicating the possibility of multicollinearity. This is equivalent to saying that R² > 0.80 indicates a possible multicollinearity problem. CS = 1 -R² attitude CS = 1 -0.509 CS = 0.491 CS = 1 -R² paying zakat intention CS = 1 -0.611 CS = 0.389 Based on these results, it can be explained that each exogenous variable (religiosity) does not experience multicollinearity problems as a predictor variable for endogenous variables (attitude and paying zakat intention).  Table 10. presents the value of R² for the variable of paying zakat intention is 0.611. This value proves that the variation of paying zakat intention, which is explained by religiosity and attitude, is 61.1%, while other variables explain the rest. The value of R² for the attitude variable is 0.509. This value proves that the attitude variable explained by religiosity is 50.9%, while other variables explain the rest.

Q-Square Analysis
Calculation of the value of q-square predictive relevance in this research model: The calculation results show the Qp² inner model value of 0.318, which means this research model has a low model fit. The model's accuracy is 31.8%, explaining that the model's contribution to explaining the structural relationship of the three variables studied is 31.8%. The rest is explained by other variables not involved in the model.

Resampling Bootstrapping (Test Hypothesis)
The results of testing the relationship between exogenous and endogenous variable are presented in the following figure:   (2022) The p-value is said to be significant if it is less than or equal to 0.05 (Garson, 2016). Religiosity to paying zakat intention has a coefficient with a positive value. The calculation results show that the p-value value is 0.008. It can be explained that religiosity positively affects a paying zakat intention. Thus, H1 is accepted. This result can be interpreted as the higher the religiosity of millennials. It is proven that religiosity have an increasing impact on the millennial paying zakat intention.
Religiosity to attitude has a coefficient with a positive value. The calculation results show that the p-value value is 0.000. It can be explained that religiosity has a positive effect on attitude. Thus, H2 is accepted. This result can be interpreted as the higher the religious nature of millennials. It is proven to have an increasing impact on the attitude of paying zakat millennials.
Attitude towards paying zakat intention has a coefficient with a positive value. The calculation results state that the p-value value is 0.000. It can be explained that attitude positively affects a paying zakat intention. Thus, H3 is accepted. This result can be interpreted that the higher the attitude of millennials toward paying zakat, it is proven to have an increasing impact on their paying zakat intention for millennials.
Attitude mediates religiosity on the variable of paying zakat intention for millennials with a p-value value of 0.000. It can be explained that religiosity positively affects a paying zakat intention through attitude. Thus, H4 is accepted. This result can be interpreted that the higher the religious nature possessed by millennials will have a higher impact on increasing paying zakat intention if it is related to the context of the attitude toward paying zakat of millennials.

The influence of religiosity on paying zakat intention
Based on the study results, it is known that religiosity positively affects a paying zakat intention. This means that if religiosity increases or decreases, there is an influence on the variable of paying zakat intention. Religion is essential for life. Religious millennials think that paying zakat is a good obligation and that it helps mustahiq meet their needs. In the Qur'an it is explained reads:  Mukhibad et al. (2019) conveyed that Muslims who have high religiosity will increasingly realize that zakat is an obligation (one of the pillars of Islam) and must be implemented. Muzakki's religiosity, reputation, trustworthiness of muzakki, and transparency of LAZ influence paying zakat intention for muzakki. Religiosity is a person's view and understanding regarding Islamic teachings and implementation of these teachings in every activity of his life so that religiosity determines paying zakat intention for millennials. According to Aji et al. (2021), a religious Muslim will have a good attitude and a strong paying zakat intention. The more religious a person is, the more motivated he is to help those in need, especially by donating money such as paying zakat (Aji et al., 2021). By increasing muzakki religiosity, reputation and transparency, the potential for zakat can be maximized (Mukhibad et al., 2019). In research Ridlwan & Sukmana (2017) stated that religiosity factors affect people's motivation in paying zakat.
These findings are supported by previous findings that have been carried out by Kaawaase & Nalukwago (2017), Bananuka et al. (2020) and Mukhibad et al. (2019) where these findings result that religiosity has a positive influence on intention. However, this study does not support other studies that show that religiosity does not significantly affect intention (Aji et al., 2021;Amin et al., 2011).

The influence of religiosity on attitude
Based on the study results, it is known that religiosity has a positive effect on attitude. This means that if religiosity increases or decreases, then there is an influence on the attitude variable. Their motivation to pay zakat is their belief in religion so that they carry out Allah's commands. In this context, millennials who have high religiosity tend to be willing to pay zakat. This is because they have good awareness and confidence to pay zakat following the law. In the Qur'an it is explained reads: Religion plays a role in all aspects of life and is believed to influence the attitude of donating to others (Kasri & Ramli, 2019). This result confirms the findings of Kasri (2013), which suggests an individual commitment to religion is expected to influence a person to have an attitude following the teachings of religion.
Religiosity is the view and understanding of the community regarding the teachings of Islam. So that they can carry out these teachings in every activity of their lives. Consumers with a higher level of religiosity tend to buy environmentally friendly products and have a high tendency of attitude toward the environment (Felix & Braunsberger, 2016). The higher the religiosity of millennials, the higher the attitude toward paying zakat. This condition is because people with high religiosity tend to carry out Islamic teachings in every activity of their lives. It has been recognized in the literature by Barro & McCleary (2003) and Lehrer (2004) that individuals' attitude, behaviour, and character are shaped by their religious beliefs.
The findings of this study support Kasri & Ramli (2019) and Felix & Braunsberger (2016), which explain a positive influence or relationship between religiosity on attitude. However, this study does not support other studies which show that religiosity does not affect attitude (Aji et al., 2021).

The influence of attitude on paying zakat intention
Based on the study results, it is known that attitude has a positive influence on paying zakat intention. This means that if the attitude increases or decreases, there is an influence on the variable of paying zakat intention. Millennials who have a good attitude think that paying zakat is an obligation so the desire to pay zakat comes from themselves, so they are aware of the law regarding paying zakat. Based on research, Andam & Osman (2019) and Bidin et al. (2009) attitudes strongly influence intention. Wiethoff (2004) argues that close family ties affect attitude.
Research conducted by Salazar-Clemeña (2002) concluded that close family relationships lead to a favourable attitude toward paying zakat. Those who can pay zakat are very aware of the needs of people in need, and the impact of giving zakat will ease their difficulties (Andam & Osman, 2019). In the Qur'an it is explained reads: Social pressures and essential people in life can significantly encourage people's intention, especially in paying zakat (Aji et al., 2021). Kasri & Ramli (2019) describe attitude as a person's tendency to evaluate certain behaviours positively or negatively. The more often an individual produces a favourable evaluation of his behaviour, the more likely he is to be positive about that behaviour. Theoretically, these findings align with the Theory of Planned Behavior (TPB) in research by Ajzen (1991), which indicates that attitude is considered a direct determinant of intention. In other findings, attitude affects intention positively (Ng et al., 2011). Thus, the fact that attitude affects intention is widely accepted.
According to the Theory of Reasoned Action (TRA), subjective attitude and norms significantly predict intention. Thus, the results of this study follow the theory. Additional support for this finding was provided by Amin et al. The effect of religiosity on paying zakat intention through attitude Based on the study results, it is known that attitude mediates the positive influence of the relationship between religiosity and paying zakat intention. In Islam, only those who are pious or religious show a good attitude towards payment and an paying zakat intention. As mentioned in the Qur'an, Allah the Exalted says: It means: "The believers, both men and women, are guardians of one another. They encourage good and forbid evil, establish prayer and pay alms-tax, and obey Allah and His Messenger. It is they who will be shown Allah's mercy. Surely Allah is Almighty, All-Wise" (Qur'an,9:71) In the perspective of Islamic Sharia, prosperity is not only realized by solely optimizing the fulfilment of material needs but also the fulfilment of spiritual needs equally (Anwar et al., 2020). As a society that has high religiosity generally respects religion as the primary goal in life (Allport & Ross, 1967), In this study, more and more millennial Muslims who have high religiosity and have a more positive attitude toward paying zakat to lead to a greater increase of paying zakat intention. These findings support the previous findings that have been carried out by Muhamad et al. (2018), Kaawaase & Nalukwago (2017) and Bananuka et al. (2020), where the findings state that religiosity has a positive influence on intention through attitude. However, this study does not support other studies which show no relationship between religiosity-attitude-intention (Aji et al., 2021).
This study indicates that religiosity will strongly influence paying zakat intention when mediated by attitude. Based on this, many Millenials are willing to pay zakat because their high religious level understands the law of paying zakat and practices it in daily life, of course, having a good attitude will increase their paying zakat intention.

CONCLUSION
From the research and hypothesis testing results, it is known that religiosity and attitude have a positive and significant effect on millennial paying zakat intention. Religiosity can encourage millennials to pay zakat intention because many think religion is important in their lives. Religious Millennials think that paying zakat is a good obligation and that paying zakat helps mustahiq meet their needs. Therefore, paying zakat, which is included in the pillars of Islam, will attract particular intention from Muslim millennials. Then, millennials who have a good attitude think that paying zakat is an obligation, so the desire to pay zakat comes from themselves, so they are aware of the law of paying zakat. In addition, this study also proves that attitude mediates religiosity towards millennials paying zakat intention. Many are willing to pay zakat because their high religious level understands the law of paying zakat and practices it in daily life, of course, having a good attitude will increase their paying zakat intention.
This study limits its focus on millennials. This study provides a managerial implication for zakat institutions to have a better understanding that religiosity and attitude are essential to encourage Muslim millennials to pay zakat and the significant role of millennials in alleviating poverty in Indonesia. From the research results, this study suggests that future research might observe the Z generation as a research object and consider other variables that influence them to pay zakat such as knowledge, perception, comfort, motivation, and social presence in which moderation tests can also be used.