Are Women More Generous Than Men? Evidence from the U.S. Consumer Expenditure Survey
February 8, 2010
CHU-PING LO
Department of Agricultural Economics National Taiwan University, Taipei 106, Taiwan
SANAE TASHIRO Department of Economics
Rowan University, Glassboro, NJ 08028, USA
Any errors are the sole responsibility of the authors. We thank anonymous referees and all participants at conferences for their comments on earlier versions. We also thank Mark Kolakowski for his editorial assistance. Address correspondence to Chu-Ping Lo, Department of Agricultural Economics, National Taiwan University, Taipei 106, Taiwan; Email: cplo@ntu.edu.tw, Fax: 886-2-2362-8496 or Sanae Tashiro, Department of Economics, Rowan University, Glassboro, NJ 08028; Email: tashiro@rowan.edu, Fax: 856-256-4921.
Are Women More Generous Than Men? Evidence from the U.S. Consumer Expenditure Survey
Abstract
The paper examines how gender, marital status, income, race and ethnicity affect giving.
Testable hypotheses are based on human and social capital theory. We employ Tobit estimates and compute the marginal effects of estimated explanatory variables on donations using the 2006 U.S. Consumer Expenditure Survey. Among the unmarried, women are more generous than men, while among the married, men are more generous than women. White and Asian women are less likely to donate, yet Hispanic men donate more than non-Hispanic men. Donations in dollars increase with income, but not necessarily with the share of income, presenting an inverse Kuznets curve.
Key Words: Donation, Gender, Race, and Ethnicity
I. INTRODUCTION
Approximately half of Americans claim to volunteer time and money to a cause (Eckstein, 2001). U.S. charitable giving was $295.02 billion in 2006, $314.07 billion in 2007, and $307.65 billion in 2008 (Giving USA Foundation, 2007, 2008, 2009). Charitable bequests are expected to grow, yet future trends are unclear given changes in the economy, society, and donor behavior.
Existing studies find that women are more likely than men to engage in philanthropic behavior, such as charitable donation (Piper and Schnepf, 2008; List, 2008, 2004; Eckel and Grossman, 1998; Anderson, 1993) and volunteering (Lammers, 1991). Other studies find that there are no gender differences in giving (Bolton and Katok, 1995; Okunade et al., 1994).
Researchers have extended analyses on gender differences to both the amounts given and the rates of giving to charities. Some studies find that women are more likely to donate, but give less (Belfield and Beney, 2000), others find that men give larger amounts than women (NSGVP, 2000; Kaplan and Hayes, 1993), and yet others conclude that women are more generous when the price of giving is high while men are more generous when the price is low (Andreoni and Vesterlund, 2001). Conversely, several studies show that women are more generous in both the frequency of giving and in the amount given to charities (Piper and Schnepf, 2008; Mesch et al. 2006).
Existing studies have also paid close attention to the effect of marital status on altruistic behavior by gender. Married men and women are more likely to donate and donate more money than single men (Mesch et al., 2006). Other studies find that married men and women are more likely to donate than single men, but do not donate more money (Bryant et al., 2003). Several studies find that single women give more than single men after controlling for other demographic
variables (Mesch et al., 2006; Andreoni et al., 2003). These studies are inconclusive and thus further research is warranted.
Gender differences in preferred charitable causes are also well examined. Hall (2004) asserts that “men tend to give to enhance their own standing or maintain the status quo, while women give to promote social change or help others less fortunate.” Existing studies confirm that men are more likely to denote to voluntary organizations (Mesch et al., 2006), religious organizations (Piper and Schnepf, 2008), and political organizations (Hall, 2004), while women are more likely to donate to health organizations (Mesch et al., 2006), animal welfare, education, and the elderly (Piper and Schnepf, 2008).
Are women really more generous - in rates of giving and in amounts given - than men? How does marital status, race and ethnicity affect gender differences in giving? This paper replicates the existing literature on gender differences in giving, yet it extends the study by taking the following new avenues. First, we uses two measures of donation: total donations in dollars, which is widely used in the literature, and total donations as a share of income, which takes into account the potential income or wealth gap that may exist by gender and by race or ethnicity. Second, we employ a nationally representative dataset, the U.S. Consumer Expenditure Survey, to examine the proposed questions empirically using Tobit estimates and marginal effects of estimated explanatory variables.
This study finds that female gender, marriage, income, age, education, and wealth increase giving. Women are more generous than men when they are unmarried, and men are more generous than women when they are married. We also find that income increases donations in dollars but not always as a share of income: middle-income respondents donate a lower share of income than low-income respondents, while high-income respondents donate the largest share of income, presenting an inverse Kuznets curve. Finally, we find racial and ethnic differences in
charitable behavior by gender: Asian men, White women, and Asian women are less likely to donate than their counterparts, yet Hispanic men donate more than non-Hispanic men.
This study confirms that age, race, ethnicity, education, income, and wealth independently affect philanthropic behavior and also interact with gender and/or marital status. The effects of gender and marital status on donations vary and are significant in determining donations in dollars and as a share of income. Our empirical results, derived from a nationally representative dataset, confirm the robustness of the existing findings and support the current literature on gender differences in philanthropic behavior. Our findings also can help donors, researchers, fundraisers, managers in non-profit organizations, and policymakers understand behavioral differences in giving and formulate appropriate strategies to maximize charitable bequests.
II. TESTABLE HYPOTHESES
We draw upon both the economic and sociological literature in developing the framework for philanthropic behavior. In particular, we consider both human capital and social capital as influences on philanthropic behavior (Bryant et al., 2003), and propose the following testable hypotheses.
Human capital theory predicts that age, education, skills, and experience of a worker increase productivity (Becker, 1964). The stock of human capital facilitates philanthropic behavior (Musick et al., 2000). Volunteering could be a form of investment in human capital. Hence, the more human capital one has, the more likely one is to engage in philanthropic behavior.
Hypothesis 1: Older people, highly educated people, and high-income people engage more in giving relative to young, less-educated, and low-income people, respectively.
Hypothesis 2: Minorities (non-Whites), who are likely to have lower human capital, engage less in giving than majorities (Whites) who have higher human capital.
Social capital refers to an endowment of social structures that include relations among actors (Coleman, 1988), and networks of civic engagement that foster certain norms and trust (Farr, 2004; Putman, 1995). The stock of social capital stabilizes mutual expectations through trust, enables collaborative actions (Spence et al., 2003) and lowers the transaction costs of volunteering or donating (Bryant et al., 2003). Hence, the more social capital one has, the more likely one is to engage in philanthropic behavior (Mesch et al., 2006; Musick et al., 2000). Social capital also includes prior social participation and one’s marital status (Janoski et al. 1998). Married people are thus more connected with social networks than single and divorced people (Bryant et al., 2003).
Hypothesis 3: Married people, who have high social capital, are more likely to engage in giving than single people, who have low social capital.
Gender differences in philanthropic behavior are present. Eckel and Grossman (1998) argue that women are more socially-oriented (selfless) and men are more individually-oriented (selfish). The predominant studies argue that women are more likely to engage in philanthropic behavior than men.
Hypothesis 4: Women are more likely than men to engage in giving in terms of rates as well as amounts.
III. DATA
This paper uses the Consumer Expenditure Survey (CES) data for 2006. The CES presents comprehensive data on the purchasing habits of American consumers and provides detailed
expenditure and income data.
The micro data used in this paper are from the interview component of the 2006 CES and have two sources: (1) the family (FMLY) file; and (2) the expenditure (EXPN) file. The FMLY file contains one record per household with information on its demographic status (such as age, sex, race, ethnicity, educational attainment, marital status, metropolitan living status, region, annual income, wage, weeks worked, occupation, and housing tenure, vehicle ownership). The EXPN file contains expenditure data and identifies distinct spending categories. We have a particular interest in expenditure categories related to contributions.
The 2006 CES data, gathered from the FMLY file and the EXPN file, initially contains 32,011 respondents and includes household members aged between 16 and 86. Annual income before taxes in the 2006 CES is top coded at $681,197.60. The data sample is restricted to adults under the retirement age (individuals aged 18-65 at the survey date) with annual income before taxes greater than $1 per year. After restrictions, the sample size is 23,454 for the year 2006.1
We consider two outcome measures: total donations in dollars per person and total donations as a share of income per person. Total donations refers to the monthly contributions reported by a respondent at the survey date and include donations to: (1) political organizations; (2) religious organizations; (3) charities and all other organizations; and (4) any and all other persons not in the consumer unit, which are all external, non-educational, non-obligated contributions.2 Total donations as a share of income are calculated by total donations divided by annual income before taxes.
We use several socio-economic variables. Age dummy variables are constructed for three age levels: age 18-25, age 26-45, and age 46-65. The “female” dummy variable equals one if the
1 One of 23,454 observations reported total monthly donation in share of annual income of 80 percent. We consider this observation as an outlier and exclude it from the analyses that use the total donation as a share of income as the dependent variable.
2 Total donation excludes contributions to: (1) college student living away from home; and (2) educational institutions, child support, and alimony.
respondent is female, and zero otherwise. The “education” dummy variables are constructed for five education levels: less than a high school diploma, at least a high school diploma (includes a high school diploma or GED with or without some college but no degree), Associate degree, Bachelor’s degree, and Advanced degree. Other variables of interest include race, ethnicity, marital status, annual income, housing tenure, vehicle ownership, metropolitan living status, region, and occupation. A detailed description of variables is in Appendix 1.
IV. DESCRIPTIVE ANALYSIS
This section summarizes individual characteristics and total donations of the entire sample, of men and women, and of married and unmarried respondents.
Table 1 shows selected characteristics of individual respondents by all respondents and by gender. Within the all-respondent sample, about 46 percent are in ages 26-45, about 44 percent are 46-65, and 10 percent are 18-25. The distribution is similar for both men and women. As for race and ethnic composition, approximately 79-84 percent are Whites and 16-21 percent are non-Whites (of which 12 percent in the full sample, 9 percent of men, and 16 percent of women are Blacks), and 13-14 percent are Hispanics. Across all samples, approximately 46-48 percent of the population has at least high school diploma, and 18-21 percent has a bachelor’s degree. About 9-12 percent holds an associate degree, 10-12 percent holds an advanced (master’s, professional, or doctoral) degree, and 12-13 percent holds less than a high school diploma. Approximately 22 percent of the population in the full sample reported annual incomes of less than $25,000, while 26 percent, 19 percent, and 13 percent of the population earned $25,000-49,999, $50,000-74,999, and $75,000-99,999, respectively, and the remaining 20 percent earned more than $100,000. About 44 percent of men and 53 percent of women reported annual
incomes of less than $50,000, while 56 percent of men and 47 percent of women earned annual incomes of more than $50,000. Men earned higher annual incomes than women. As a proxy for wealth, about 64-66 percent of the population owned a house and 86-89 percent owned at least one vehicle. About 45 percent of the population in the full sample, 40 percent of men, and 49 percent of women are unmarried. About 94-95 percent of the population lives in metropolitan areas.
Table 2 provides total donations for the all-respondent sample, men and women, married and unmarried people, separately. About 46 percent of the full population (men and women) engages in giving. By marital status, 55 percent of married respondents donate, while only 36 percent of the unmarried give. In terms of mean total donations in dollars, married people donated the largest amount, $308.13, followed by women at $298.65. In terms of mean total donation as a share of income, women donated the largest share, 3 percent, followed by 2.2 percent for unmarried people.
V. EMPIRICAL STRATEGY
The objective of this paper is to examine empirically how gender, race, ethnicity and marital status affect giving. Using a nationally representative dataset, the Consumer Expenditure Survey data, allows us to test the robustness of the existing findings in the literature.
A substantial portion of respondents does not give and thus reports a zero value in the dataset. In our dataset, approximately 54 percent (or 12,595 out of 23,454) of observations in the full sample equal zero. Since total donations is the dependent variable, these zero values lead to censored response bias. In this case, conventional regression methods fail to account for the qualitative differences between zero observations and continuous observations; hence, we
employ the Tobit model (Tobin, 1958), which accounts for observations with a zero value:
Yi = Xiβ + εi while εi = N ~ N(0, σ2ε) (1)
where Yi is a dependent variable implying the total donations. We consider two measures of
donation: total donations in dollars, which is commonly used in the literature, and total donations as a share of income to take into account the potential income or wealth gap that may exist by gender, by race, and by ethnicity (Conley, 2000). Since Yi is not always observable:
Yi = Xiβ + εi if Yi > 0
0, otherwise. (2)
Explanatory variables (Xi) include: (1) age; (2) gender; (3) race; (4) ethnicity; (5)
education; (6) marital status; (7) annual income; (8) housing tenure; (9) vehicle ownership; (10) metropolitan living status; (11) region; (12) occupation; and (13) a mean zero individual error term (εi). The subscript i refers to each individual.
The post-estimation analysis is recommended in the Tobit model. We also present marginal effects of all explanatory variables in the estimated specifications using the decomposition procedure developed by McDonald and Moffitt (1980).
VI. EMPIRICAL RESULTS
This section examines the results of the Tobit model and the marginal effects of estimated explanatory variables on total donations, which show the changes in total donations resulting from a one unit change in each of the explanatory variables. The analyses of two dependent variables, total donations in dollars (shown in Table 3) and total donations as a share of income (shown in Table 4) are conducted by examining marginal effects of selected variables. We focus
on the effect of gender, marital status, annual income, and race and ethnicity on total donations to examine gender differences in giving.
1. The Effect of Gender and of Marital Status on Total Donations
In the full sample, total donations in dollars by women are larger (by $11) than for men (Table 3), while total donations as a share of income do not differ by gender (Table 4). Gender differences persist in the amount given, but not in the share of income.
The effect of marital status on donations is positive, large and differs by gender. Table 3 and Table 4, respectively, show that married men donate more (by $37 and 0.018) than unmarried men, and married women donate more (by $17 and 0.001) than unmarried women. These results confirm that marriage increases giving, which is consistent with hypothesis 3.
Turning to gender differences by marital status, Table 3 and Table 4 show that donations in dollars and as a share of income do not differ between married women and men. On the other hand, unmarried women donate more (by $18 and 0.003) than unmarried men. These results confirm that gender differences in giving vary by marital status: women are more generous than men among the unmarried, men are more generous than women among the married, and men tend to be the decision makers in the family unit (Andreoni et al., 2003), leading married men to be the largest donors.
2. The Effect of Annual Income on Total Donations by Gender and by Marital Status
A striking finding emerges from the effect of annual income on donations. Table 3 shows that income increases total donations in dollars but not always as a share of income. Starting with analyses of the effect of income in giving by gender, total donations in dollars of both men and
women are larger (by $16-61 and $14-58, respectively) than those of their counterparts earning less than $25,000 (Table 3). However, the trends on total donations as a share of income show somewhat different pictures. In Table 4, men earning $25,000-$49,999 and women earning $25,000-74,999 donate less of their income (by 0.007 and 0.002, respectively) than their counterparts earning less than $25,000. These results suggest that middle-income respondents donate a lower share of income than low-income respondents, while high-income respondents donate the largest share of income, presenting an inverse Kuznets curve.
Table 3 also shows that the effect of income on donations differs by marital status. In Table 3, total donations in dollars among married people increase with income and the effects are more prominent for those with higher incomes. These trends persist for unmarried people except for those earning $25,000-$49,999. Table 4 shows that total donations as a share of income for married people earning $25,000-74,999 are less than for married people earning less than $25,000, and those of unmarried people earning $25,000-50,000 are also less than unmarried people earning less than $25,000. Married people earning more than $100,000 increase their donations as a share of income (by 0.008) relative to their counterparts. These results show that the allocation of income to giving varies by marital status and income level.
3. The Effect of Race and Ethnicity on Total Donations by Gender and by Marital Status
Interesting findings emerge from analyses of the effect of racial and ethnic differences in giving by gender. In Table 3, total donations in dollars by White and Black men are not different from those of “other” races, while those of Asian men are less (by $40) than their counterparts. However, there are no differences in total donations as a share of income among men across races (Table 4). As for women, Whites, Blacks, and Asians donate less than those of “other”
Further, White women contribute smaller shares of income (by 0.004) relative to their counterparts. These results show that Asian men, White women and Asian women are less likely to donate. Hispanic men donate more (by $10 and 0.007) than non-Hispanic men, while there are no ethnic differences in giving among women. Our existing findings do not fully support hypotheses 2 and 4.
Table 3 confirms that race and ethnic variations exist by marital status. Married Asians contribute less ($33 and 0.013) than married persons of “other” races, while total donations of married Whites and Blacks are not different from their counterparts. Conversely, total donations in dollars of unmarried Whites, Blacks, and Asians are lower (by $54, $31, and $37, respectively) than among unmarried persons of “other” races (Table 3), and total donations as a share of income for unmarried Whites is also lower (by 0.011) relative to those of “other” races (Table 4). These results indicate that married Asians and unmarried people across all races are less likely to donate than their counterparts. Married Hispanics contribute less (by $9) than married non-Hispanics, but their donations as a share of income are not different from their counterparts. Unmarried Hispanics, on the other hand, donate more (by $17 and 0.005) relative to unmarried non-Hispanics. These results indicate that unmarried Whites contribute less than unmarried non-Whites and that unmarried Hispanics are more generous than married Hispanics, findings that are inconsistent with hypotheses 2 and 3. Our findings suggest that race and ethnicity interact with gender or marital status to affect giving behavior.
VII. CONCLUSION
This paper examines how gender, marital status, income, race and ethnicity affect donations. Testable hypotheses are developed based on human and social capital theory. We
employ Tobit estimates and compute the marginal effects of estimated explanatory variables on total donations using the 2006 Consumer Expenditure Survey.
Estimates yield several important findings. First, female gender, marriage, and income increase giving, as in the charitable giving literature. Second, gender differences in giving vary by marital status; women are more generous than men when they are unmarried, and men are more generous than women when they are married. Third, income increases total donations in dollars but not necessary as a share of income; middle-income respondents donate a lower share of income than low-income respondents, while high-income respondents donate the largest share of income, presenting an inverse Kuznets curve. Finally, racial and ethnic differences in giving persist: Asian men, White women, and Asian women are less likely to donate than their counterparts, yet Hispanic men donate more than non-Hispanic men.
Our study replicates the existing studies in the literature on gender difference in giving using a nationally representative dataset and two measures of donation: total donations in dollars and a share of income. Estimates confirm the robustness of the existing findings and provide additional findings in which being a woman, being married, and having a higher income are positively related to the amount given, but not necessary with giving as a share of income. Our study can also helps donors, researchers, fundraisers, managers in non-profit organizations, and policymakers understand behavioral differences in giving and formulate appropriate strategies to maximize charitable bequests.
TABLE 1
Selected Characteristics of Individual Respondents
Selected Variables
All Sample Men Women
No. % of total No. % of total No. % of total Age 18-25 2333 0.099 1140 0.103 1193 0.097 26-45 10821 0.461 5072 0.457 5749 0.466 46-65 10300 0.439 4896 0.441 5404 0.438 Total 23454 1.000 11108 1.000 12346 1.000 Race White 19055 0.812 9315 0.839 9740 0.789 Black 2888 0.123 948 0.085 1940 0.157
Asian or Pacific Islander 1184 0.050 704 0.063 480 0.039
Multi-Race 327 0.014 141 0.013 186 0.015
Total 23454 1.000 11108 1.000 12346 1.000
Ethnicity Hispanic 3186 0.136 1551 0.140 1635 0.132
Non-Hispanic 20268 0.864 9557 0.860 10711 0.868
Total 23454 1.000 11108 1.000 12346 1.000
Education Less than HS Diploma 2930 0.125 1363 0.123 1567 0.127 At least High School Diploma 11029 0.470 5066 0.456 5963 0.483 Associate degree 2470 0.105 1045 0.094 1425 0.115 Bachelor's degree 4528 0.193 2306 0.208 2222 0.180 Advanced degree 2497 0.106 1328 0.120 1169 0.095 Total 23454 1.000 11108 1.000 12346 1.000 Marital Married 12996 0.554 6692 0.602 6304 0.511 Status Unmarried 10458 0.446 4416 0.398 6042 0.489 Total 23454 1.000 11108 1.000 12346 1.000
Annual Less than $25,000 5190 0.221 2085 0.188 3105 0.251
Income $25,000-49,999 6178 0.263 2750 0.248 3428 0.278
$50,000-74,999 4485 0.191 2208 0.199 2277 0.184 $75,000-99,999 3035 0.129 1594 0.144 1441 0.117 More than $100,000 4566 0.195 2471 0.222 2095 0.170
Total 23454 1.000 11108 1.000 12346 1.000
Housing House Owned 15212 0.649 7331 0.660 7881 0.638
Tenure House Rented 8242 0.351 3777 0.340 4465 0.362
Vehicle Own at least one vehicle 20430 0.871 9828 0.885 10602 0.859 Ownership Do not own a vehicle 3024 0.129 1280 0.115 1744 0.141
Total 23454 1.000 11108 1.000 12346 1.000
Metropolitan Metropolitan Living 22171 0.945 10580 0.952 11591 0.939 Living Non-metropolitan Living 1283 0.055 528 0.048 755 0.061
Total 23454 1.000 11108 1.000 12346 1.000 Region Northeast 4339 0.186 2004 0.181 2335 0.190 Midwest 5363 0.229 2600 0.235 2763 0.225 South 8252 0.353 3672 0.331 4580 0.372 West 5425 0.231 2802 0.253 2623 0.213 Total 23379** 1.000 11078** 1.000 12301** 1.000 * 75 respondents (30 of men and 45 of women) did not report their region.
TABLE 2
Total Donations by All Sample, by Gender, and by Marital Status: 2006
All Sample Men Women Married Unmarried
Total Monthly Donations Per Person (in Dollars) Total Observations 23454 11108 12346 12996 10458 Donor Observations 10859 5125 5734 7083 3776 % of Total 0.463 0.461 0.464 0.545 0.361 Mean 294.81 290.53 298.65 308.13 269.83 Standard Deviation 237.97 224.56 249.30 248.38 214.91 Min 140 140 140 140 140 Max 3860 2890 3860 3700 3860 Total Monthly Donations Per Person (as a Share of Annual
Income) Total Observations 23453 11108 12345 12995 10458 Donor Observations 10858 5125 5733 7082 3776 % of Total 0.463 0.461 0.464 0.545 0.361 Mean 0.014 0.017 0.011 0.010 0.022 Standard Deviation 0.213 0.301 0.069 0.227 0.183 Min 0.0003 0.0003 0.0003 0.0003 0.0004 Max 18.2 18.2 3.2 18.2 7.1
TABLE 3
Total Monthly Donations (in Dollars) by All Sample, by Gender, and by Marital Status: Tobit Estimation Results
All Sample Men Women Married Unmarried
(n=23,454) (n=11,108) (n=12,346) (n=12,996) (n=10,458) Independent Variables Estimated Coefficient Marginal Effect Estimated Coefficient Marginal Effect Estimated Coefficient Marginal Effect Estimated Coefficient Marginal Effect Estimated Coefficient Marginal Effect Intercept -302.38*** - -283.04*** - -278.29*** - -289.45*** - -275.48*** -(28.753) - (44.909) - (38.345) - (45.311) - (42.884) -Age: 26-45 20.70* 7.02* -0.074 -0.025 38.04** 12.88** 42.29** 16.27** 5.936 1.711 (11.051) (3.753) (15.405) (5.230) (15.788) (5.359) (20.052) (7.723) (13.883) (4.005) Age: 46-65 121.33*** 41.66*** 75.90*** 25.98*** 156.64*** 53.90*** 131.10*** 50.78*** 127.02*** 37.45*** (11.210) (3.903) (15.780) (5.181) (15.966) (5.600) (20.280) (7.924) (14.253) (4.301) Gender: Female 33.28*** 11.26*** - - - - 8.085 3.108 64.40*** 18.41*** (5.822) (1.963) - - - - (7.505) (2.886) (9.465) (2.685) Race: White -71.12*** -25.00*** -43.450 -15.131 -93.56*** -33.05*** 15.039 5.731 -173.74*** -54.14*** (19.868) (7.261) (34.203) (12.216) (26.345) (9.754) (31.971) (12.077) (28.180) (9.560) Race: Black -36.26* -12.01* -16.859 -5.655 -58.87** -19.27** 22.739 8.879 -115.47*** -31.35*** (21.281) (6.895) (36.693) (12.160) (27.962) (8.873) (34.380) (13.637) (29.919) (7.670) Race: Asian -127.65*** -39.38*** -129.79*** -40.05*** -109.11*** -34.05*** -90.58** -32.62*** -142.95*** -37.07*** (23.168) (6.504) (37.374) (10.463) (32.637) (9.396) (35.287) (11.882) (34.864) (8.124) Ethnicity: Hispanic 10.640 3.627 27.68** 9.56** -7.387 -2.486 -23.72** -8.99** 56.76*** 16.95*** (9.032) (3.098) (12.572) (4.418) (12.964) (4.343) (11.558) (4.323) (14.550) (4.498)
Education: At least high school diploma 65.49*** 22.26*** 51.05*** 17.40*** 77.96*** 26.42*** 61.87*** 23.94*** 65.31*** 18.81***
(9.893) (3.374) (14.128) (4.827) (13.866) (4.710) (13.060) (5.084) (15.349) (4.417)
Education: Associate degree 95.06*** 34.28*** 75.68*** 27.10*** 110.44*** 39.99*** 85.77*** 34.84*** 104.93*** 32.44***
Education: Bachelor's degree 137.25*** 49.91*** 110.58*** 39.73*** 162.23*** 59.72*** 111.22*** 45.13*** 170.64*** 54.03***
(11.654) (4.533) (16.568) (6.301) (16.390) (6.549) (15.171) (6.487) (18.370) (6.366)
Education: Advanced degree 175.01*** 66.56*** 151.23*** 56.77*** 194.43*** 74.76*** 161.50*** 68.65*** 185.97*** 61.09***
(13.066) (5.548) (18.381) (7.625) (18.669) (8.126) (16.799) (7.853) (21.275) (7.931)
Marital status: Married 77.63*** 26.12*** 111.38*** 37.13*** 50.79*** 17.15*** - - -
-(6.486) (2.164) (9.139) (3.073) (9.296) (3.136) - - - -Annual income: $25,000-49,999 40.32*** 3.87*** 44.89*** 15.54*** 41.99*** 14.41*** 71.95*** 28.62*** 7.108 2.052 (8.909) (3.117) (13.423) (4.765) (12.009) (4.183) (14.723) (6.055) (11.827) (3.421) Annual income: $50,000-74,999 62.72*** 21.94*** 67.06*** 23.58*** 69.93*** 24.50*** 81.51*** 32.53*** 41.62*** 12.28*** (10.006) (3.621) (14.698) (5.391) (13.903) (5.050) (15.237) (6.309) (14.710) (4.448) Annual income: $75,000-99,999 98.96*** 35.63*** 97.92*** 35.32*** 112.00*** 40.59*** 114.58*** 46.85*** 92.38*** 28.44*** (11.320) (4.336) (16.101) (6.229) (16.087) (6.242) (16.071) (6.974) (18.660) (6.127)
Annual income: More than $100,000 154.09*** 56.52*** 164.67*** 60.62*** 157.96*** 58.17*** 171.50*** 70.06*** 159.94*** 51.69***
(11.029) (4.383) (15.788) (6.413) (15.701) (6.282) (15.630) (6.774) (18.777) (6.773)
Housing Tenure: House Owned 57.91*** 19.34*** 42.44*** 14.25*** 67.34*** 22.43*** 50.77*** 19.03*** 56.18*** 16.26***
(7.169) (2.358) (10.221) (3.417) (10.055) (3.296) (10.389) (3.795) (10.039) (2.915)
Vehicle Ownership: Own at least one vehicle 85.37*** 27.46*** 104.06*** 33.02*** 70.50*** 22.88*** 93.45*** 33.57*** 84.45*** 23.42***
(9.777) (2.980) (14.604) (4.338) (13.293) (4.136) (17.200) (5.761) (12.277) (3.268)
Other Control Variables YES - YES - YES - YES - YES
-Log likelihood -87045.813 - -40891.436 - -46093.956 - -55877.100 - -31082.320
TABLE 4
Total Donations (as a Share of Annual Income) by All Sample, by Gender, and by Marital Status: Tobit Model Estimation Results
All Sample Men Women Married Unmarried
(n=23,453) (n=11,108) (n=12,345) (n=12,995) (n=10,458) Independent Variables Estimated Coefficient Marginal Effect Estimated Coefficient Marginal Effect Estimated Coefficient Marginal Effect Estimated Coefficient Marginal Effect Estimated Coefficient Marginal Effect Intercept -0.194*** - -0.267*** - -0.045*** - -0.189*** - -0.151*** -(0.019) - (0.041) - (0.008) - (0.031) - (0.024) -Age: 26-45 -0.007 -0.002 -0.023 -0.006 -0.002 -0.001 0.017 0.005 -0.022*** -0.005*** (0.007) (0.002) (0.014) (0.004) (0.003) (0.001) (0.014) (0.004) (0.007) (0.002) Age: 46-65 0.029*** 0.008*** 0.016 0.004 0.012*** 0.003*** 0.050*** 0.014*** 0.017** 0.004** (0.007) (0.002) (0.014) (0.004) (0.003) (0.001) (0.014) (0.004) (0.008) (0.002) Gender: Female 0.003 0.001 - - - - -0.006 -0.002 0.014*** 0.003*** (0.004) (0.001) - - - - (0.005) (0.001) (0.005) (0.001) Race: White -0.028** -0.008* -0.023 -0.006 -0.014** -0.004** -0.009 -0.002 -0.046*** -0.011** (0.014) (0.004) (0.031) (0.008) (0.006) (0.002) (0.022) (0.006) (0.017) (0.005) Race: Black -0.010 -0.003 -0.005 -0.001 -0.004 -0.001 0.000 0.000 -0.023 -0.005 (0.015) (0.004) (0.034) (0.009) (0.006) (0.002) (0.023) (0.007) (0.018) (0.004) Race: Asian -0.045*** -0.011*** -0.061* -0.015* -0.011 -0.003* -0.049** -0.013** -0.032 -0.007 (0.016) (0.004) (0.034) (0.008) (0.007) (0.002) (0.024) (0.006) (0.020) (0.004) Ethnicity: Hispanic 0.010* 0.003* 0.026** 0.007** 0.001 0.000 -0.002 0.000 0.020** 0.005** (0.006) (0.002) (0.012) (0.003) (0.003) (0.001) (0.008) (0.002) (0.008) (0.002)
Education: At least high school diploma 0.039*** 0.010*** 0.049*** 0.013*** 0.014*** 0.004*** 0.040*** 0.011*** 0.035*** 0.008***
(0.006) (0.002) (0.013) (0.003) (0.003) (0.001) (0.009) (0.003) (0.008) (0.002)
Education: Associate degree 0.049*** 0.013*** 0.062*** 0.017*** 0.018*** 0.005*** 0.048*** 0.014*** 0.048*** 0.012***
Education: Bachelor's degree 0.070*** 0.019*** 0.095*** 0.026*** 0.023*** 0.007*** 0.065*** 0.019*** 0.070*** 0.018***
(0.007) (0.002) (0.015) (0.004) (0.003) (0.001) (0.010) (0.003) (0.010) (0.003)
Education: Advanced degree 0.074*** 0.021*** 0.102*** 0.029*** 0.026*** 0.008*** 0.070*** 0.021*** 0.074*** 0.020***
(0.008) (0.003) (0.017) (0.005) (0.004) (0.001) (0.011) (0.004) (0.012) (0.003)
Marital status: Married 0.034*** 0.009*** 0.070*** 0.018*** 0.005*** 0.001*** - - -
-(0.004) (0.001) (0.008) (0.002) (0.002) (0.001) - - - -Annual income: $25,000-49,999 -0.013** -0.003** -0.026** -0.007** -0.009*** -0.002*** -0.020** -0.006** -0.012** -0.003** (0.006) (0.001) (0.012) (0.003) (0.002) (0.001) (0.010) (0.003) (0.006) (0.001) Annual income: $50,000-74,999 -0.012* -0.003* -0.018 -0.005 -0.008*** -0.002*** -0.024** -0.006** -0.007 -0.002 (0.006) (0.002) (0.013) (0.003) (0.003) (0.001) (0.010) (0.003) (0.008) (0.002) Annual income: $75,000-99,999 -0.002 0.000 -0.010 -0.003 -0.004 -0.001 -0.015 -0.004 0.010 0.003 (0.007) (0.002) (0.015) (0.004) (0.003) (0.001) (0.011) (0.003) (0.010) (0.002)
Annual income: More than $100,000 0.013* 0.003* 0.017 0.005 0.000 0.000 0.000 0.000 0.031*** 0.008***
(0.007) (0.002) (0.014) (0.004) (0.003) (0.001) (0.011) (0.003) (0.010) (0.003)
Housing Tenure: House Owned 0.028*** 0.007*** 0.034*** 0.009*** 0.010*** 0.003*** 0.027*** 0.008*** 0.023*** 0.005***
(0.005) (0.001) (0.009) (0.002) (0.002) (0.001) (0.007) (0.002) (0.005) (0.001)
Vehicle Ownership: Own at least one vehicle 0.033*** 0.008*** 0.055*** 0.014*** 0.010*** 0.003*** 0.040*** 0.011*** 0.025*** 0.006***
(0.006) (0.002) (0.013) (0.003) (0.003) (0.001) (0.012) (0.003) (0.007) (0.002)
Other Control Variables YES - YES - YES - YES - YES
-Log likelihood -4579.595 - -3898.570 - 3705.303 - -2602.200 - -1888.505
REFERENCES
Anderson, V. L. “Gender Differences in Altruism among Holocaust Rescues.” Journal of Social
Behavior and Personality, 8, 1993, 43-58.
Andreoni, J. and Vesterlund, L. “Which is The Fair Sex? Gender Differences in Altruism.”
Quarterly Journal of Economics, 116(1), 2001, 293-312.
Andreoni, J., Brown, E. and Rischall, I. “Charitable Giving by Married Couples: Who Decides and Why Does it Matter?” Journal of Human Resources, 38(1), 2003, 111-133.
Becker, G. “Human Capital.” 1964. New York: Columbia Press.
Belfield, C. R. and Beney, A. P. “What Determines Alumni Generosity? Evidence for the UK.”
Educational Economics, 8(1), 2000, 65-80.
Bolton, G. E. and Katok, E. “An Experimental Test for Gender Differences in Beneficent Behavior.” Economic Letters, 48(3-4), 1995, 287-292.
Bryant, W. K., Jeon-Slaughter, H., Kang, H. and Tax, A. “Participating in Philanthropic Activities: Donating Money and Time.” Journal of Consumer Policy, 26(1), 2003, 43-73. Coleman, J. S. “Social Capital in the Creation of Human Capital.” American Journal of
Sociology, 94(Supplement), 1988, S95-S120.
Conley, D. “The Racial Wealth Gap: Origins and Implications for Philanthropy in the African American Community.” Nonprofit and Voluntary Sector Quarterly, 29(4), 2000, 530-540. Eckel, C. C. and Grossman, P. J. “Are Women Less Selfish Than Men? Evidence from Dictator
Experiments.” The Economic Journal, 108(448), 1998, 726-735.
Eckstain, S. “Community As Gift-Giving: Collectivistic Roots of Volunteerism.” American
Sociological Review, 66(6), 2001, 829-851.
Farr, J. Social Capital: A Conceptual History. Political Theory, 32(1), 2004, 6-33.
Giving USA Foundation. “Giving USA 2007: The Annual Report on Philanthropy for the Year 2006.” 2007, Brown, M. S. Ed. New York: AAFRC Trust for Philanthropy.
Giving USA Foundation “Giving USA 2008: The Annual Report on Philanthropy for the Year 2007.” 2008. New York: AAFRC Trust for Philanthropy.
Giving USA Foundation “Giving USA 2009: The Annual Report on Philanthropy for the Year 2008.” 2009. New York: AAFRC Trust for Philanthropy.
Hall, H. “Gender Differences in Giving: Going, Going, Gone?” New Directions for
Philanthropic Fundraising, 2004(43), 2004, 71-81.
Janoski, T, Musick, M. and Wilson, J. “Being Volunteered? The Impact of Social Participation and Pro-Social Attitude on Volunteering.” Sociological Forum, 13(3), (1998). 495-519. Kaplan, A. E. and Hayes, M. J. “What We Know about Women as Donors.” New Directions for
Philanthropic Fundraising, 1993(2), 1993, 5-20.
Lammers, J. C. “Attitudes, Motives, and Demographic Predictors of Volunteer Commitment and Service Duration.” Journal of Social Service Research, 14(3/4), 1991, 125-140.
List, J. A. “Introduction to Field Experiment in Economics with Application to the Economics of Charity.” Experimental Economics, 11(3), 2008, 203-212.
List, J. A. “Young, Selfish and Male: Field Evidence of Social Preferences.” The Economic
Journal, 114(492), 2004, 121-149.
McDonald J. F. and Moffitt, R. A. “The Use of Tobit Analysis.” The Review of Economics and
Statistics, 62(2), 1980, 318-321.
Marital Status on Giving and Volunteering in Indiana.” Nonprofit and Voluntary Sector
Quarterly, 35(4), 2006, 65-587.
Musick, M. A., Wilson, J. and Bynum, W. B. Jr. “Race and Formal Volunteering: The Differential Effects of Class and Religion.” Social Forces, 78(4), 2000, 1539-1571.
NSGVP “Gender Difference in Giving and Volunteering.” 2000. National Survey of Giving, Volunteering and Participating.
Okunade, A. A., Wunnava, P. V. and Walsh, R. Jr. “Charitable Giving of Alumni: Micro-Data Evidence from A Large Public University.” American Journal of Economics and Sociology, 53(1), 1994, 73-84.
Piper, G. and Schnepf, S. V. “Gender Difference in Charitable Giving in Great Britain.”
Voluntas, 19(2), 2008), 103-124.
Putman, R. D. “Bowling Alone: America’s Declining Social Capital.” Journal of Democracy, 6(1), 1995, 65-78.
Spence, L. J., Schmidpeter, R. and Habisch, A. “Assessing Social Capital: Small and Medium Sized Enterprises in Germany and the U.K.” Journal of Business and Ethics, 47, 2003, 17-29.
Tobin, J. “Estimation of Relationships for Limited Dependent Variables.” Econometrica, 26(1), 1958, 24-36.
