The Clash of Generations: Driving Forces Behind Charitable Giving Among Older and Younger Muslims

: In this exploratory, survey-based study ( n = 606), we examined whether Muslim perceptions about giving have changed due to the increased government scrutiny of charitable donations. We also examined age-based di ﬀ erences in preferences for cash versus non-cash donations, sending cash abroad versus giving within the U.S., and the likelihood of Muslims donating to causes that bene ﬁ t non-Muslims. Our empirical models controlled for demographic di ﬀ erences such as gender, marital status, income, education, race etc. Our statistical results suggest that in the U.S. (1) some Muslims feel that charitable giving to Muslim causes has decreased as a result of monitoring of such giving by the government; (2) older Muslims tend to prefer documented means of giving as opposed to cash donations; (3) Muslims who prefer to donate in cash within the U.S. tend to be similar in age to those who prefer to send money abroad; and (4) age has no bearing on U.S. Muslims ’ likelihood of donating Zakat to causes that bene ﬁ t non-Muslims. Implications are discussed.


Introduction
The attacks on September 11, 2001, had a significant impact on Muslim American philanthropy, with three prominent Muslim charities forced to relinquish their operations in September 2001 as a result of President Bush's vow to attack the financial foundations of an alleged worldwide terror network (Turner 2009).This focus on charitable donations and funding "inaugurat [ed] a new era of antiterrorism practices" which came about through the use of executive orders via the International Emergency Economic Powers Act (James 2019).Soon after, 30 more Muslim charities were placed on high-alert, and more were censured in the name of "terrorism funding" and "national security" despite the lack of public evidence (Turner 2009).
Since then, Muslim Americans have lived with the fear of government surveillance due to charitable donations made in fulfillment of their religious responsibility to provide zakat, or alms, and sadaqah, or "extemporaneous giving" (Abraham 2017).In fact, many American Muslims have stated that the climate due to the War on Terror has made it impossible for them to fulfill their religious obligations of zakat and sadaqah (Charity 2009).This hesitancy continues 20 years later because Muslim individuals (Fam, Hajela, and Henao 2021) and Muslim nonprofits still encounter discrimination and suspicion (Noor et al. 2021).
Muslims in the United States might be hesitant to provide documented cash or in-kind donations to Muslim charities and causes due to both a real and perceived threat of state surveillance and a perception that their financial transactions are being monitored.This hesitancy assumes that if the group or cause to which they donate is investigated, they will be implicated as well.Current research on the real effects of state surveillance and public scrutiny of Muslims and Muslim organizations (Elfenbein 2021;Kumar 2021;Nasir 2019;Rana 2011) rarely addresses how Muslims themselves perceive their own charitable acts of giving, especially how age impacts giving in an overly scrutinized demographic group.
In this exploratory survey-based study, we explore the impact of some important factors affecting Muslim American giving.We intend to see how race, age, gender, and legal status affect philanthropic behavior.We particularly look into the variations in giving by age, if any, and whether younger Muslims who have come of age during the Global War on Terror are more or less likely to give than their older counterparts.In addition, we also analyze the overall effect of age on the perception of discrimination and surveillance among U.S.-based Muslims.While a decrease in giving has been assumed to be an obvious outcome of state surveillance of Muslim charities, Siddiqui (2013) has claimed that the opposite is, in fact, true: Muslim American giving increased rather than decreased in the aftermath of the September 11 attacks.He suggests that the scrutiny on Muslims in America and their institutions and increased levels of islamophobia has prompted a greater need for Muslim Americans to give more to combat these community challenges.Muslim American giving may also be ways to show their patriotism or the value they add to society.
In considering age as an influence on giving, this study analyzes the question of Muslim American giving after 9/11 through the lens of generations.Demographic research has shown that younger 'Muslim Millennials' in the U.S. tend to hold different attitudes toward political and social issues than their counterparts with other religious affiliations (e.g.Christian Millennials) and older generations of Muslims (Diamant and Gecewicz 2017;Dorsey 2023).For instance, while all Muslim Americans equally value the role of religion in their lives, a significant number of older Muslims still hold conservative political and social views, as reflected in their voting patterns.On the other hand, younger Muslims tend to be more liberal in their political and social leanings; for example, younger Muslims are more accepting of LGBTQ+ practices than older generations (Diamant and Gecewicz 2017).In this article, we investigate what these generational differences mean and what effect they might have on donor behavior among Muslims of various age ranges.Young adults (18-29 years old) differ in their experiences, opinions, and religious practices compared to older members of the U.S. Muslim community (Pew Research Center 2007).Nevertheless, we suggest based on our findings, that there is little-to-no evidence that younger Muslims in the United States give less or prefer to give through documented means (donations with a paper trail) as compared to older Muslims.

Muslims and Muslim Philanthropy in the U.S.
Islam is the fastest-growing religion in the world as well as in the United States (Fardila, Imamah, and Dewi 2020;Lipka 2017).An estimated 3.45 million Muslims live in the United States (Mohamed 2018), and Muslims constitute the youngest faith community in the U.S. with around a quarter of Muslim Americans between 18 and 24 years of age (Mogahed and Ikramullah 2020).U.S.-based Muslims are incredibly diverse both in terms of race and ethnicity, with no single ethnic group constituting a majority.
About 58 % of US Muslims are foreign-born Americans, 18 % are first-generation Americans, 28 % are second-generation, and 24 % are third-generation or beyond (Cooperman 2017;Lipka 2017).In terms of denomination, 29 % of Muslims in the US prefer to be recognized as "simply Muslims," while 16 % identify as Shiʿi Muslims and 55 % as Sunni Muslims (Moore 2015).Even though American Muslims are educated on par with other Americans, they remain disproportionately poor (Mogahed and Chouhoud 2017).Moreover, studies find that Muslim Americans donate significantly more than the general population (Siddiqui and Wasif 2021;Siddiqui et al. 2022).Muslim philanthropy for both faith-based causes and non-faith-based causes is higher than non-Muslims.Muslims on average gave USD 1810 to faith-base causes compared to USD 1138 in the general population.Similarly, an average Muslim gave USD 1400 to non-faith-based causes compared to USD 767 in the general population.This highlights the significant role of U.S. Muslims and their philanthropic contributions.
Charitable giving is an important component of the Muslim faith, and Muslims give generously in many ways.In addition to sadaqah, Shiʿi Muslims are expected to donate khums or 20 % of their annual profits or extra income, whichever is greater (Awang 2017;Bakar 2008).Zakat is traditionally donated to Muslims and Muslimrelated causes especially for those in need, according to eight categories outlined in the Qur'an and in Muslim traditions (hadith).However, an increasing trend in recent years among Americans has been to provide financial assistance or volunteer services to non-Muslims (Khan and Siddiqui 2017;Mahajneh, Greenspan, and Haj-Yahia 2021).While most Islamic scholars agree that non-obligatory Islamic charitable donations like sadaqah can be given to non-Muslims, they have conflicting views on whether zakat can be used to aid non-Muslims in need (Ummulkhayr, Owoyemi, and Cusairi 2017).

Perceived Risk of Monetary Donations to Charities and Individual Causes
Even though Muslim Americans have a rich history of philanthropic efforts in the country, the 9/11 attacks had a significant impact on their community's ability to contribute and changed the course of their philanthropy for years to come (Wasif 2023).On September 24, 2001, President George W. Bush issued an executive order to strengthen the Treasury Department's unilateral ability to freeze the assets of groups it deemed terrorist organizations (Bush 2001).In the following weeks, federal officials announced investigations into over 30 Muslim charitable organizations.In December 2001, the federal government shut down three Muslim charities in the United States -Benevolence International Foundation, Global Relief Foundation, and the Holy Land Foundation for Relief and Developmentby freezing their assets during the holy month of Ramadan, which is the time when Muslims give away most of their charity (Turner 2009).As a result, Muslim Americans complained that the government's actions had stifled their ability to practice their faith freely and completely by giving to charitable causes (Charity 2009;Wasif 2023).At times, individuals wish to keep their donations anonymous for various reasons.Nevertheless, when donations are made through traceable methods like online transactions or checks, maintaining anonymity becomes impractical.Interest in anonymous giving might be especially strong for those considering donations to certain, more controversial causes or groups, and connected to the perceived risks of being affiliated with a charity.For example, Erica Caple James (2011) has traced an emerging post-9/11 paradigm of charitable giving and organizational presentation that she calls "risk-based philanthropy."Risk-based philanthropy is based on the government's suspicion that charitable funds are being diverted to so-called "terrorist organizations" or causes, and, as a result, produces both surveillance and anticipation of surveillance at the individual level as well as the community level.GhaneaBassiri (2017, p. 11) claims that focusing on material support for terrorism pathologizes Muslim giving: "Their pathologizing approach to Islam results in their identification of a doctrinal Muslim practice, zakat, as a possibly nefarious source of funding for terrorism, irrespective of the way individual Muslims decide to practice zakat."James (2011, p. 66;2019) further claims that, as a result of this pathologizing discourse, Muslim charities and non-profits adopt a risk-based philanthropic approach which (over-) emphasizes professionalized "best practices" in a manner that can have chilling effects on Muslim organizational spaces.Since much of US Muslim philanthropy occurs through informal means, this "risk-based" approach could thus unduly impact Muslim giving.Furthermore, Goede (2012) argues that post-9/11 antiterrorism finance laws like the International Emergency Economic Powers Act produced a "preemptive" and "speculative" security apparatus consolidating trends towards securitization emerging in the 1990s, the effects of which will outlast the "war on terror" and continue to shape charitable practices alongside development and humanitarian aid into the foreseeable future (Howell and Lind 2009;Shannon 2009).Moreover, increased securitization and surveillance of monetary donations, especially to Muslim charities, extends beyond the U.S., affecting Muslim populations in the U.K. and the European Union member states (Clunan 2007;Donohue 2008;Goede 2012;May 2019May , 2020May , 2021;;Napoleoni 2007).

Islamophobia and the Changing Course of Muslim
Charitable Giving Since 9/11 Islamophobia is often pointed to as one of the most significant forces shaping not only how Muslim non-profits are perceived, but also how they perceive themselves and act as a result.Scholarship on the origins, development, and spread of anti-Muslim bias has traced the long arch and transformation of anti-Muslim sentiments from medieval Christian discourses of 'holy war' against 'the Saracens' to the Spanish Inquisition and finally to immigration policies that negatively target Muslims and non-Europeans (Anidjar 2014;Bayoumi 2006;Mastnak 2002;Wasif 2023).As Hammer (2013) has argued, those contemporary forces of discrimination are frequently bundled together under the concept of Islamophobiathat is, a "fear of Islam"-are more often expressed through layered discourses of anti-Muslim racism, or active forms of bigotry and discrimination of Muslims as racialized religious subjects rather than as reactive fears (Hammer 2013).Hammer's position is echoed by Ernst (2013), who states that Muslim communities in the US are regarded as racialized The Clash of Generations minorities instead of a religious minority in the U.S. social context.Hammer describes the tension between racial and religious minoritization in the following way: "Islamophobia has been described as a particular form of racism, as creating unfounded fear of Islam, as an extension of Orientalism, as a response to multiculturalism, and as an ideology scapegoating Muslims in order to distract from or re-center other issues in contemporary societies.Each of the definitions advanced seems to hold a kernel of truth or to reflect a specific angle on the ways in which contemporary European and North American societies seem to have focused much of their attention on Islam and Muslim."(Hammer 2013: 29) Grewal ( 2013) further argues that Islam is not simply considered a religious identity or tradition in America, rather it is treated as a 'foreign country' itself (Grewal 2013).After 9/11, there were significant changes in regard to the perception of Islam.Since the events of 9/11, Muslims all over America have faced and continue to face negative stereotyping and stigmatization.Ahmed and Matthes (2017) highlight the role of media in portraying Muslims negatively.Misconceptions regarding Muslims were on the rise, but the election campaigns of the 2016 especially exploited the negative narrative against Muslims (Saleem et al. 2019).Furthermore, the increased funding of the Countering Violent Extremism Task Force (CVE) together with Muslims targeted as its main focus also triggered feelings of insecurity among Muslim youth (Shams 2018).Siddiqui (2013) points out that, in 2004, U.S. Muslim donors increasingly did not want receipts for their donations, or they chose to pay in cash.Some prominent Muslim American charities reported that more people were giving money without wanting receipts or by using cash instead of checks.These organizations claimed that donors did this to avoid being identified as contributors because they were afraid of being targeted.These donors worried about being on a list of donors to a charity that may be investigated.Although this fear generally affected a small number of Muslim American donors, it is important to note that some are willing to give up the benefits of tax deductions, credit card rewards, and recognition in exchange for anonymity (Siddiqui 2013).
By 2011 (Aziz 2011), the level of donations to Muslim charities had not returned to its pre-9/11 levels.Muslims continued to give to charities, but less to religious causes and more to secular causes, outside of their communities and to more mainstream philanthropic organizations (Khader and Siddiqui 2018).Jamal (2011) suggests that Muslims began emphasizing anonymous giving, such as cash donations, to ensure that there would be no legal consequences or prosecution if the organization was accused of providing financial support for terrorism.In this case, the perceived reluctance is tied more to the closing down of some charities or risks of not only being shamed but also being found legally culpable for material support of terrorism simply by donating to Islamic organizations.This shift in giving happened after 9/11, and the subsequent crackdown on Muslim charities had long-term effects.
Reputation has been found to be one of the driving motivations for charitable giving (Bekkers and Wiepking 2011b).The perception that one's reputation might be harmed by a donation and not enhanced could provide a logical deterrent from making donations openly for a cause that might be perceived badly or prove to have negative ramifications.A person who supports a controversial cause could choose instead to donate anonymously or give in cash but conceal their identifying information.Research from Muslim donors in countries like Indonesia suggests that donors favor religious causes as well as poverty relief, but also that many choose to give through informal channels, working to support their community through their community, and not necessarily through a third-party (Kasri 2013).
Considering the evolving history of Muslim American giving and changing preferences over the past two decades, our first exploratory research question is as follows: RQ1: Has the charitable giving practice of Muslim Americans changed due to the perceived surveillance of law enforcement and security agencies?

Age and Giving
As scholars have turned their focus to the changing course of Muslim philanthropy in the West, particularly in the U.S., limited attention has been given to an important demographic factor: age.Considering the generational differences among Muslim populations is crucial, given that more than half of Muslims in the U.S. are foreignborn (Lipka 2017) and may maintain strong ties to their country of origin.There is some evidence which suggests that average age of Muslims immigrating from other countries is higher than that of U.S. born Muslims.For example, using a 2019 survey Harjanto and Batalova (2022) reported that for immigrants from Muslims countries in the Middle Eastern and North African region that includes Iraq, Syria, Jordan, Saudi Arabia, Egypt and Sudan, among others, the median age was 42 compared to a median age of 37 for the U.S. born sub-group.Additionally, studies have shown that age, in general, impacts religious giving (Argue, Johnson, and White 1999;Lancaster et al. 2014;Opoku 2013).Within the Western context, studies indicate that older Muslims are more generous in terms of giving (Khan, Jain, and Cameron 2022;Noor, Schwoerer, and Siddiqui 2023;Siddiqui et al. 2022).However, younger Muslims are more open to giving to causes that may not directly benefit other Muslims (Morton, Ou, and Qin 2020;Noor, Schwoerer, and Siddiqui 2023).
In their extensive survey of the literature on predictors of charitable giving, Bekkers and Wiepking (2011a) found that age often has a positive relationship to philanthropy, particularly with religious giving.Furthermore, there are generational (cohort) differences in giving tied to decreased religious involvement.In their study of generational differences in Catholic and Protestant religious giving and involvement, Wilhelm Rooney, and Tempel (2007) demonstrated that Catholic and Protestant giving to religious causes decreased together with religious involvement from the prewar to the baby boomer cohort.
A study of generational trends in giving show that, as of 2003, gen-Xers were "both less likely to give and less generous when they do give" in comparison to prewar and baby boomer generations (Steinberg and Wilhelm 2003, p. 76).Bekkers and Wiepking caution that the relationship between age and giving decreases with controls for other life-cycle variables such as income, religious involvement, or the financial stability of one's children (Bekkers and Wiepking 2011a, p. 351).But what of the relationship between age, religion, and methods of giving?How do generations differ in terms of how they give their donations or what external factors, rather than internal motivations, influence their giving?In what follows, we analyze the impact of age and external factors such as the perception of security and surveillance on Muslim American giving.In terms of giving methods, several studies indicate that the majority of young donors prefer donating through online means (Harden, Jukes, and Joyce 2015;Martin 2022).
In this study, we aim to examine the influence of age on charitable giving preferences among Muslim Americans.Specifically, we seek to analyze whether there are differences between older and younger Muslims in terms of their preferences for donating via cash or documented means (cheque, online methods) and their inclination to send charity money abroad.Based on these objectives, we propose the following two exploratory research questions: RQ2: Are older Muslim Americans more likely to donate to Islamic causes through documented means than younger Muslims?RQ3: Do older Muslim Americans prefer sending their charity money abroad more than younger Muslims?
Based on the literature suggesting that younger generations of Muslims have shown greater openness to donating to causes irrespective of whether they benefit Muslims or non-Muslims, our aim is to empirically examine if this inclusivity in donor behavior is influenced by age differences.Specifically, we intend to investigate if younger Muslims exhibit a higher level of acceptance for benefiting non-Muslims compared to older Muslims.Consequently, we propose the following exploratory research question: RQ4: How do the charitable giving preferences differ among younger and older Muslim Americans regarding causes benefiting Muslims and non-Muslims?

Data Source and Participants
Data used in this study came from an anonymous and voluntary online survey of Muslim adults residing in the United States.We used snowball sampling for this study.The survey was administered between April 1 and May 10 of 2020, and coincided with the Muslim holy month of Ramadan in which charitable donations from Muslims generally increase.The survey was commissioned by the Muslim Philanthropy Initiative at the Indiana University's Lilly Family School of Philanthropy, and was supported by several nonprofit organizations and student associations.These include Islamic Relief USA, which shared the survey link with all U.S.-based Muslim donors in their database and staff members; the Center on Muslim Philanthropy (CMP), which shared a link with 400+ Muslim students, donors, and staff members; and several Muslim Student Associations from universities across the United States that helped circulate the survey.
We used Qualtrics, a popular cloud-based tool, to field the survey which enabled several quality checks.The participants initially responded to the main survey and upon completion were directed to an additional section requesting contact information for disbursement of a $5 Amazon card for the first 300 participants.As there was an incentive involved, the "Prevent multiple submissions" option was enabled to prevent respondents from completing the survey multiple times.In addition, the survey was not shared on open social media platforms.This option was enabled for both the actual survey and the follow-up section.Another Qualtrics feature "Bot detection" was enabled to ensure that bots did not complete the survey.
Survey participants could complete the questionnaire anytime and anywhere on their cell phones, tablets, laptops, or desktop computers.The participants were given the opportunity to complete the survey within a week of starting it.The original sample size was 743 and represented participants from all over the US.However, there were some incomplete responses, with missing values for demographic and other variables.In order to utilize only those cases that provided full information we used list wise deletion to drop cases with missing values on one or more variables used in this study.As a result we ended up with a reduced sample of 606 complete cases.This sample was used for all statistical results reported in this study.Since the original data file did not include sampling weights, readers The Clash of Generations should exercise caution in trying to generalize results to populations that are not represented by our sample.

Measures
We used five survey questions to evaluate the four research questions presented earlier in this paper.These survey questions are summarized in Table 1.Three of these questions (Q27-Q29) collected information on perceptions of Muslim Americans about change in giving due to surveillance of charitable contributions by law enforcement and security agencies.We asked Q29 with the understanding that donating cash through money transfer apps (such as Venmo, PayPal, and CashApp) leaves a digital trail as compared to handing over cash in person.In addition to these three surveillance-related questions, one item (Q30) asked respondents about their giving preference (U.S. vs. abroad), and another (Q15) measured the likelihood of a respondent giving Zakat to causes that do not primarily benefit Muslims.Descriptive statistics for all five questions are summarized in Table 2 while correlations for numeric questions are presented in Table 3.We examined these five primary survey questions (see Table 1) in conjunction with demographic factors listed in Table 2.

Analytical Procedures
We treated responses on Q27-Q29 and Q15, which were all measured on a sevenpoint Likert scale, as continuous.For these continuous outcomes our main analytical technique was a multiple linear regression model that predicted each outcome from a set of demographic characteristics.These demographic characteristics included religious sect, religious attendance, gender, education, race, immigrant status, marital status, income, whether they converted to Islam, annual monetary donation, and age.For cash donation preference (Q30) we employed a logistic regression model.However, we took care to ensure that the set of predictors in this model matched those in the multiple linear regression model.Finally, we used an ANOVA model to examine age differences across cash donation preference categories.SPSS was used for all computations.The Clash of Generations 3 Results

Descriptive Statistics
For the dependent variables, descriptive statistics by demographic factors and the correlation matrix are presented in Tables 2 and 3.The group means reported in  3).In order to see if there was any association between preference for cash donation (Q30) and age we performed a chi square test of association.Results from this test supported a significant association between categories of the two variables, χ 2 (9) = 82.06,p < 0.001 suggesting that cash donation preference is not independent of age.

Multiple Regression Results
The descriptive results presented in Tables 2 and 3 are preliminary and unadjusted for demographic characteristics.In order to adjust the average estimates for these characteristics we estimated a multiple regression model predicting each of the continuous outcomes (Q27-Q29, and Q15) from demographic factors.Regression slope coefficients are summarized in Table 4.There were several significant predictors (including age) for each of the four outcomes.In order to see if the average response on the dependent variables still exceeded 4 (i.e.neutral category) we have presented the adjusted overall means in Table 5.These results suggest that after controlling for demographic differences the average response was neutral (i.e.not significantly different from 4 on the 1-7 Likert scale) for all four dependent variables.However, we did observe some age-based significant effects (see Table 6).For example, there was a full 1-point difference in perception of change in giving to Muslim causes due to government scrutiny (Q27) between the 18-29 age group (Adj.M = 3.97) and the 50+ age group (Adj.M = 2.97) with the average response being higher for the younger group, p = 0.011.Thus, on average, younger respondents felt that there was no change in giving due to government scrutiny while older respondents felt that there was a slight decrease.We saw a similar result for Q28 with the mean for the 18-29 age group (Adj.M = 4.36) exceeding that for the 50+ age group (Adj.M = 3.56), p = 0.024.Overall, our regression results suggest that older (50+) Muslims feel that charity to Muslim causes as well as their own giving has decreased because of government scrutiny.For Q29 we observed a small but statistically significant mean difference between the 30-39 age group (Adj.M = 4.02) and the 18-29 age group (Adj.M = 4.38) in our multiple regression results, suggesting a relatively higher level of agreement in the former group that it is better to donate in cash due to government scrutiny.However, this difference disappeared with Bonferroni adjustment for multiple comparisons, p = 0.289. 1 A post hoc (Bonferroni-adjusted) comparison between 30 1 Such contradiction in results is due to two reasons: (1) multiple regression results allow group comparisons with only the reference category which can hide significant differences between groups The Clash of Generations   6 for pairwise group comparisons).Older and younger respondents had a similar level of agreement about their likelihood of donating Zakat to causes that do not primarily benefit Muslims.The overall adjusted average response to this question was 4.13 (see Table 5) which is not significantly different from 4 (neutral category on the scale), 95 % CI: (3.70, 4.56).

Logistic Regression Results
We estimated a logistic regression model with Q30 as the outcome and demographic factors as independent variables.In this analysis we compared respondents who preferred to donate cash to someone in the U.S. versus those who preferred to send charity money abroad.The odds ratios from this model are presented in the last represented by dummy variables, and, perhaps more importantly, (2) group comparisons in regression models are not based on a method of adjustment for multiple comparisons (such as Tukey or Bonferroni).Both of these factors can contribute to contradictory findings in multiple regression comparisons with reference group versus post hoc pairwise comparisons among all groups.Since post hoc pairwise comparisons are evaluated at a more conservative alpha level they provide a higher degree of confidence compared to dummy variable-based regression results.
The Clash of Generations column of Table 4.For the modeled age groups, none of the odds ratios were significantly differently from 1 suggesting that the odds of sending money abroad (as compared to giving to someone in the U.S.) for older Muslims were not different from similar odds for younger Muslims.

ANOVA Results
For our final analysis we evaluated the average difference in age across all four cash donation preference groups after controlling for demographic differences.Results are reported in Table 7, and suggest that there is no significant difference in average age of Muslims who prefer to give cash to someone in the U.S. and those who prefer to send money abroad, which is consistent with our logistic regression results.However, we found evidence of a significant mean age gap between Muslims who preferred neither method (i.e.do not prefer to donate in cash, M = 37.15 years) and those who prefer to use both methods (i.e.give cash in the U.S. as well as abroad, M = 33.41years), p = 0.001.This finding suggests that Muslims who are relatively more inclined to donate via documented means tend to be older than those who prefer to donate in cash.

Discussion and Conclusion
In this survey-based study we examined Muslim perceptions about how giving has changed due to increased government scrutiny of charitable donations to Muslim causes.We also examined age-based differences in the preference for cash versus non-cash donations, preference for sending cash abroad versus giving cash within the U.S., and the likelihood of Muslims donating to causes that benefit non-Muslims.We controlled for demographic differences such as gender, marital status, income, education, race etc. in all of our analyses.Our main findings are that in the U.S.: (1) some Muslims feel that charitable giving to Muslim causes has decreased as a result of monitoring of such giving by the government, (2) older Muslims tend to prefer documented means of giving as opposed to cash donations, (3) Muslims who prefer to donate in cash within the U.S. tend to be similar in age to those who prefer to send money abroad, and (4) age has no bearing on U.S. Muslims' likelihood of donating Zakat to causes that benefit non-Muslims.
The main finding in our study is the relationship of age with the perception of change in giving due to government monitoring.Our results suggest that after controlling for demographic characteristics, there are significant age-based differences in perception of change in giving due to such monitoring with older Muslims feeling that there has been both an overall decrease in charitable giving to Muslim causes as well as a decrease in their own giving due to government scrutiny.In addition, we found evidence that older Muslim Americans (who tend to feel that government scrutiny has affected charitable giving to Muslim causes) prefer not to donate in cash (instead preferring alternative methods such as check or online donations), whereas younger Muslim Americans have a relatively higher preference for cash donations.Taken together, these perceptions may be reflective of an overall desire among older Muslim Americans both to avoid government scrutiny of their donations (by decreasing giving) as well as to retain documentary proof of giving (by shifting away from cash donations).
In the United States, Muslim American nonprofit organizations use a variety of ways to collect funds.They receive donations in the form of cash, credit card, checks, appreciated stock, assets, and in-kind donations (Siddiqui 2013).Our expectation was that among Muslim Americans who prefer to donate in cash, older individuals would be more likely to send their charitable donations abroad compared to their younger counterparts who have a higher preference for giving cash in the U.S. One argument supporting this expectation is that since the majority of older Muslims in the U.S are firstor second-generation Americans, they may have stronger ties to their countries of origin through active social and familial networks.On the other hand, younger Muslims, who constitute nearly a quarter of the whole Muslim American population, are primarily U.S.-born Muslims (Mogahed and Ikramullah 2020).However, we did not find any significant difference in average age between the two groups.Younger Muslim Americans are as likely to send money abroad as their older counterparts.This result challenges the notion that older Muslims prefer to send cash donations abroad instead of giving in The Clash of Generations the U.S. due to their comparatively stronger links to the broader Muslim community outside the U.S. Our analysis provided additional evidence connecting age and preference for cash donations.Specifically, we found that the average age of Muslims who prefer to give in cash (both in the U.S. and abroad) is significantly lower than those who prefer non-cash means of giving.
Finally, we did not find any evidence that charitable giving preferences differ among younger and older Muslim Americans regarding causes benefiting non-Muslims.Our findings indicate that older and younger generations have a similar likelihood of donating to such causes.

Limitations and Suggestions for Future Research
Our study has several limitations.First, this study is based on a convenience sample rather than a nationally representative probability-based sample of Muslims.Therefore care should be exercised when drawing conclusions about broad groups of Muslim Americans.Second, given the voluntary nature of respondent participation, there is always a possibility that biases such as self-selection bias and confirmation bias exist in survey responses.Third, our findings are based on information collected from Muslim respondents only in the U.S. and thus should not be generalized to Muslims in other countries.Future research can focus on replicating our methods and research questions in large-scale nationally representative samples in the U.S. as well as other countries.Another line of inquiry is to extend this study by incorporating additional variables (such as recipient type: individual vs. organization), treating volunteered time as a type of donation, and by examining differences across various religions and traditions.

Table  :
Summary of outcome variables used in the study.

Table  :
Descriptive statistics for outcome variables by demographic characteristics.

Table  :
Correlations among outcome variables and age.

Table  :
Parameter estimates from regression models.

Table  :
(continued)Ref.= reference group for dummy coding.R  for the three regression models with Q, Q and Q as outcomes are .%, . %, . %, and . % respectively.a All estimates presented in this model are partial slope coefficients except those for Q (n = ) where exponentiated estimates (i.e.odds ratios) are given (reference category = cash donation in the U.S.).b The standard error on this dummy variable was very large suggesting collinearity with other predictors.However, elimination of this variable from the analysis did not change the logistic regression results in any meaningful way, therefore we decided to leave it the model.∼ = approximately zero.*p < ., **p < ., ***p < ..

Table  :
Unadjusted and adjusted overall mean estimates for the dependent variables.Note.n=.The response categories ranged from strong decrease in giving () to strong increase in giving () for Q and Q; from strongly disagree () to strongly agree () for Q; and from () extremely unlikely to () extremely likely for Q.Adjusted estimates control for all demographic factors listed in Table  (except cash donation preference[Q]which is an outcome in this study).

Table  :
Pairwise mean comparisons across age groups.groupand50+age group was significant, p = 0.012 with the mean for the 50+ age group (Adj.M = 3.44) being lower than that for the 30-39 group (Adj.M = 4.38).Thus, overall we have some evidence of age-based mean group differences on Q29 with younger Muslims (30-39) on average agreeing that it is better to donate in cash due to government scrutiny of Muslim charitable causes.For Q15 we did not observe any statistically significant age-based group differences (see Table4for multiple regression estimates; see Table

Table  :
(continued)Note.n = .Adjusted estimates control for all demographic factors listed in Table  (except cash donation preference which is an outcome in this study).Multiple comparisons adjusted with Bonferroni method.*p < ., **p < ., ***p < ..