International Journal of Educational Advancement

, Volume 7, Issue 3, pp 229–242

Who Decides in Giving to Education? A Study of Charitable Giving by Married Couples

Special Issue Paper

DOI: 10.1057/palgrave.ijea.2150063

Cite this article as:
Rooney, P., Brown, E. & Mesch, D. Int J Educ Adv (2007) 7: 229. doi:10.1057/palgrave.ijea.2150063

Abstract

Using data from the Center on Philanthropy Panel Study and the Panel Study on Income Dynamics, we analyzed whether husbands or wives were more likely to determine whether and how much money to donate to educational institutions. Among donor households, we are able to examine what socio-economic-demographic factors explain differences in whether men or women are more likely to decide to give to educational institutions. We also compare the “who decides” question for education to who decides about giving overall. We find that after controlling for other factors in multiple regression analyses, the educational attainment of both spouses is positively associated with increased giving to education, as is the number of children living at home, family income, and wealth (excluding home values). The age of the husband does not matter but there is a positive association with the age of the wife and amounts given to education. Men have little or no influence on the decision to give to education at all or the amounts donated to education. Conversely, women decision-makers are more likely to have a positive effect on both the likelihood of giving to education and the amounts given to education.

Keywords

education givingcharitable givinggiving married

Who Decides in Giving to Education? A Study of Charitable Giving by Married Couples

Giving to charity is an important topic that has been studied across multiple academic disciplines. The extant research has looked at the individual characteristics of donors as well as those in households—examining how gender, education, income, age, race, and marital status affect charitable giving behavior (e.g., Belfield and Beney, 2000; Conley, 2000; Andreoni et al., 2003; Bryant et al., 2003; Rooney et al., 2005; Mesch et al., 2006). Very little research exists, however, as to the charitable giving decision-making process itself. “Focusing solely on individuals may leave important information out of the picture” since other household members may influence “both whether and how much to give for a particular cause” (Burgoyne et al., 2005, p. 384). Knowing who in a household decides on charitable giving, how decisions are reached as to how much and where charitable contributions should be made, and what factors affect this decision making “would provide vital data for researchers who seek accurate trends in giving” (Burgoyne et al., 2005, p. 384).

Furthermore, this type of information would be useful to those in the fundraising profession. Individuals in the fundraising field are increasingly becoming aware of the importance of understanding patterns of giving behavior and the decision-making process as they strive to improve upon targeting, soliciting, and cultivating different donors. Understanding who decides within couples and how that might differ across subsectors—as well as how to approach couples so as to maximize effectiveness, minimize mistakes, and increase participation levels—is critical for those in the fundraising profession.

Although there is considerable literature on how couples make decisions and influence household expenditures (i.e., Manser and Brown, 1980; McElroy and Horney, 1981; Pahl, 1983, 1995; Thomas, 1990; Lundberg and Pollak, 1993, 1996; Volger, 2005; Browning and Chiappori, 1998; Gray, 1998; Phipps and Burton, 1998; Tichenor, 1999; Volger et al., 2006), Andreoni et al. (2003) study is one of the first to apply these principles to charitable decision-making. Their study examined intra-household decision-making and found evidence that bargaining, predominantly favoring husbands, characterizes how household charitable decisions are made. Overall, the results of their study found that single men and women exhibit different tendencies toward giving; single women were more likely than single men to give across all charities except one category. Among married people, women were more likely than men to give to all but two categories of charities and women spread their giving dollars more thinly across several categories, while men had a greater tendency to concentrate their giving. When decisions were made jointly, however, husbands had more influence over their wives in deciding on charitable giving. Furthermore, they found education and income to be the primary determinants of control over charitable resources—being the primary earner strengthens one's bargaining power in marriage as does the husband's education relative to the wife's.

Purpose of the Study

The purpose of this paper is to build upon the Andreoni et al. (2003) study by examining one particular category of charities—education. Research on giving to higher education is very limited (Liu, 2006). Yet, as institutions of higher education face budget cuts from state and federal sources, private funding to educational institutions is becoming increasingly important in order to sustain operating budgets. As such, colleges and universities are looking for ways to diversify their revenue streams and supplement funding through private giving (Liu, 2006). With the increasing competitiveness in the market, combined with rising educational costs and reductions in student financial aid, college and university alumni are ever more essential in providing financial support to their institutions (Briechle, 2003; Liu, 2006; Weerts and Ronca, 2007). Research on giving to higher education generally has focused on predicting the factors and characteristics of alumni who give (e.g., Taylor and Martin, 1995; Baade and Sundberg, 1996; Okunade and Berl, 1997; Belfield and Beney, 2000; Clotfelter, 2003; Weerts and Ronca, 2007), the institutional and macroeconomic factors in explaining variations in giving (e.g., Briechle, 2003; Gunsalus, 2004; Liu, 2006), alumni motivation for giving (e.g., Diamond and Kashyap, 1997; Weerts and Ronca, 2007), fundraising practices in higher education (e.g., Harrison, 1995; Harrison et al., 1995), and the determinants of donor revenue (e.g., Cunningham and Cochi-Ficano, 2001; see review of literature in Liu, 2006).

Virtually no research, however, has been conducted as to how decisions among couples are made to educational institutions. The Andreoni et al. (2003) study found that when the woman is the decision-maker, she is significantly more likely to give to education than is the husband or a jointly deciding couple. Burgoyne et al. (2005) found that the choice of charity is often a reflection of personal choice, reflecting individuals’ interests and concerns in articulating his or her identity in the act of giving. Although their study revealed that participants who were actively involved in organized religion tended to support charities favored by their place of worship, they did not examine choice of charity as a function of who decides.

Our study addresses some of the gaps in this literature. We investigate the following research questions: (1) What are the patterns of household giving to education and can we predict whether or not a donor will give to education? (2) How does giving to education vary across income and wealth? (3) How does giving to education vary according to who decides? (4) Is there a difference between giving to education versus secular giving and religious giving? (5) Does the educational attainment of donors and prospective donors affect whether or not they give at all—and if so how much?

This paper addresses an important gap in the research. By using the Center on Philanthropy Panel Study data about giving and volunteering behaviors paired with data from the overall Panel Study on Income Dynamics (PSID), we can estimate whether there are differences in the amounts given to education (as well as to religious and secular nonprofits overall) when the husband or wife decides separately, jointly, or when one of the spouses is the primary decision-maker. We look at these differences at the simple mean level as well as after controlling for income, wealth, educational attainment, number of children, and age in a regression model. Furthermore, we examine whether educational attainment is important in determining the decision-making process among couples. That is, are couples more likely to decide jointly when they both have a college degree (Andreoni et al., 2003), or are there other determinants that are important but were not available in prior research? We also estimate whether or not the size of the gift is affected by who decides—again both at the mean level, but also after controlling for other variables statistically. Finally, we look at the relationships between giving to education and secular giving and religious giving. Are these differences primarily a function of educational attainment? If so are they more driven by that of the husband or wife?

Economic Models of Household Decision Making

Generally, there are three broad styles of collective models found in the economics literature, each predicting different ways in which marriage partners reach expenditure decisions (e.g., Pahl, 1983, 1995; Phipps and Burton, 1998; Brown, 2005; Volger, 2005). Under an income pooling model, a married couple can make decisions as if it were a single individual economic unit, where the couple “thinks as one.” Alternatively, one member may have dictatorial control over resources; however, the outcome is that the household “thinks as one.” In either case, the household pools their resources. In a second model, cooperative bargaining, couples function as two separate economic units and can choose to allocate their money by pooling some of their income or by keeping their money separate. This model adopts the point of view that there is some conflict between the couple's preferred spending patterns. Each partner, however, also cares about the well-being of the other and wants the other to reach the highest level of well-being consistent with his or her own level of satisfaction. In this case, the bargaining power in the household is a critical issue in how to divide the surplus produced by the marriage (Brown, 2005). In the third model, non-cooperative behavior, partners also function as two separate economic units in which a partner could not cooperate or cease to consult one another on their spending choices. In this case, control over spending depends strongly on how much income each partner brings to the marriage.

Charitable Giving Decision Making in Households

Are these economic models of household decision making consistent with couples’ decision making in giving to charity? The Andreoni et al. (2003) study indicates that there is a parallel between the way in which charitable decisions are made and other household financial decisions. Other researchers have found similar results. Using focus groups with individuals who were currently married or living with a partner, Burgoyne et al. (2005) found that charitable giving decisions within the family were treated very much like other uses of money—“if both partners agree that they want to give to certain causes on a regular basis, then this gets discussed and built into the normal household outgoings” (p. 395). Planned, larger gifts to charity tended to be joint decisions as part of the regular household budget, but smaller gifts tended to be individual decisions. These decisions were moderated by the level of household income and, to some extent, older children—but only if they had input into other aspects of family spending. In a study using data from the Giving in the Netherlands Panel Study, Wiepking and Bekkers (2006) found that the majority of decisions on charitable giving were made jointly—over 80 percent deciding as one economic unit. Their results support other research in that educational attainment is a significant predictor as to who decides and the presence of children in the home and home ownership have a positive effect on charitable decision making. In general, research has found that, in terms of decision making over charitable giving, the greater the share of household resources and the more the education, the greater the influence a partner has over the other (i.e., Pahl, 1983; Volger, 2005; Andreoni et al., 2003; Wiepking and Bekkers, 2006).

Since the Andreoni et al. (2003) study, several recent trends have emerged regarding charitable giving by married couples (Brown, 2005). First, married women's influence over charitable giving is growing—as married women's earnings increase, they take a more active role in decision making (Brown, 2005). Data analyzed from the PSID found that the most commonly reported mode of decision making is collaboration, reported by three quarters of the sample and, for the 10 percent of couples who reported that only one member of the couple decides, wives were more than twice as likely as husbands to be the decision maker (Brown, 2005). These research results have significant implications for educational nonprofits and charities. Particularly in higher education, where women have surpassed men in enrollment and have become more financially independent (Briechle, 2003), it is critical to understand the factors that influence giving to our colleges and universities. “This economic evolution creates an opportunity for fundraising professionals to elicit financial support from women in an effort to advance the mission of their specific institution” (Briechle, 2003, p. 20). Second, recent research has found that households in which the male partner decides on charitable gifts are the largest donors, whereas households with female decision makers are the smallest donors (Wiepking and Bekkers, 2006). Third, demographics of the donors can affect decision making in households. For example recent research shows that couples with a strict Protestant religious denomination have a higher probability of having the male partner decide on charitable giving (Wiepking and Bekkers, 2006); the presence of children in the household increases the likelihood of giving (Banks and Tanner, 1997); and couples may change their household financial decisions during their lifecycle (Pahl, 1995).

Methodology

Sample

The PSID is the largest (n=∼8,000 households) and longest-running (started in 1968) panel study in the world. The PSID is implemented by the University of Michigan's Institute for Survey Research. Each household is asked many of the same questions and a few different questions in every wave. The sample includes a nationally representative sample and an over-sample of low income and minority households. This analysis only utilizes the nationally representative sample within the PSID. The PSID interviews the same households each time and has consistently had year-to-year continuation rates in excess of 95 percent. Furthermore, as children age and are emancipated, they become a new data point and hence behaviors can be tracked not only over time within a household but also across generations both within and across families.

Starting in 2001, the Center on Philanthropy at Indiana University developed and sponsored a philanthropy module within the PSID. Respondents are asked about their giving and volunteering to various subsectors. The questions include appropriate prompts for what types of donations to include and which ones to exclude as well as what types of charities are included in each section (and which ones are not). In addition, starting in 2003, the Center on Philanthropy at Indiana University developed and sponsored questions about religious attendance and “who decides” within couples about philanthropic giving.

Estimation techniques

Given the fact that giving is truncated at zero (i.e., one cannot give a gift less than $0), we have to take into account this truncation bias. We use probits to estimate the marginal impact of each of the independent variables on the probability of the households being a donor to education in this national survey. Similarly, we use tobits to estimate the incremental effects of the independent variables on the amounts donors give to education (or religion, and/or secular causes). Given that the vast majority of households decide about giving to education jointly, we used this category as the reference category and all comparisons of giving to education by other decision-makers is compared to this joint category. We also test whether or not the results are sensitive to type of regression (e.g., logit or probit) and/or the functional form (e.g., absolute dollars or a log–log model). Finally, we test whether or not our results are specific to education or can be generalized to include giving to religious organizations and/or all secular nonprofits.

Dependent variables

Donor to Education=1 if donated to education at all; zero otherwise.

Amount Donated to Education=dollar amount donated to education.

Amount Given to Secular=dollar amount donated to all secular causes.

Amount Given to Religion=dollar amount donated to all religious charities.

Amount Given in Total=dollar amount donated to all charities.

Log of Amount Given to Education (plus10)=natural log of the dollar amount donated to education plus $10 (to permit inclusion of the log of households that donated zero dollars).

Log of Amount of Total Given (plus10)=natural log of the dollar amount donated to all charities plus $10 (to permit inclusion of the log of households that donated zero dollars).

Independent variables

Educational Attainment of Husband=number of years of education of husband.

Educational Attainment of Wife=number of years of education of wife.

Number of Children=number of children under 18 still living at home.

Family Income in 2003=total family income measured in dollars.

Wealth in 2003, excluding home=total family wealth measured in dollars, but not including homes.

Age of Husband (in years)

Age of Wife (in years)

Male Decides Alone=1 if the male spouse decides about giving to education alone; zero otherwise.

Female Decides Alone=1 if the female spouse decides about giving to education alone; zero otherwise.

Male Mainly Decides=1 if the male spouse mainly decides about giving to education; zero otherwise.

Female Mainly Decides=1 if the female spouse mainly decides about giving to education; zero otherwise.

Decides Separately=1 if the spouses each decides separately about giving to education; zero otherwise.

Results

Descriptive statistics

In Table 1, we see that 18.8 percent of the sample gives to education at all and they give an average gift of $89 to education, $607 to secular causes overall, and $1,112 to religious causes. The average educational attainment of the husbands is 13.5 years and that of the wives is 12.9 years. The average household has almost one child (0.9 kids). The average family income is $81,857 with a wealth (excluding homes) of $223,624. The average man is 45.3 years old and the average woman is 43.8 years old. Only 8.8 percent of men decide alone whether to give to education versus 16.4 percent of women. Men are the main decision makers in only 1.6 percent of households versus 3 percent of women. Couples decide these things separately in 9 percent of the cases. In the vast majority of the cases (e.g., well over half), the decision to give to education is made jointly.
Table 1

Descriptive statistics

 

n

Mean

Std. Dev.

Min

Max

Giving to education: percentage

3,278

0.188

0.391

0

1

Amount given to education

3,278

88.589

739.562

0

30,000

Amount given to secular

3,278

606.968

1750.146

0

36,625

Amount given to religion

3,278

1111.772

3178.8

0

1,10,000

Log of amount given to education (plus10)

3,278

2.812

1.203

2.303

10.309

Log of amount of total given (plus10)

3,278

5.744

2.297

2.303

11.608

Amount given in total

3,278

1630.15

3665.535

0

1,10,000

Educational attainment of husband

3,086

13.489

2.406

1

17

Educational attainment of wife

3,018

12.921

3.465

0

17

Number of children (under 18) still at home

3,278

0.875

1.081

0

6

Family income in 2003

3,278

81856.69

106640.5

−19,000

36,60,650

Wealth in 2003, excluding home

3,278

223623.9

10,49,770

−3,08,000

3.53E+07

Age of husband

3,278

45.29

14.97

18

94

Age of wife

3,148

43.75

14.213

16

90

Male decides alone

3,278

0.088

0.283

0

1

Female decides alone

3,278

0.1635

0.37

0

1

Male mainly decides

3,278

0.016

0.125

0

1

Female mainly decides

3,278

0.03

0.17

0

1

Decides separately

2,446

0.093

0.291

0

1

In Table 2, we see that the mean levels of giving by both single men and women are quite low at all income levels less than $100,000, and for both genders. Single men and women earning over $100,000 are the only ones giving more than the mean levels, but these high-income households give over twice the mean levels to education. Among couples in which the decision was made “mostly by the male,” giving to education is clustered almost exclusively in two income groups: those earning $30,000–$50,000 gave an average of $233 to education and those earning more than $100,000 gave an average of $583 to education (overall mean=$300). Among couples in which the decision was made “mostly by the female,” giving to education grows in a nearly exponential manner with income: those earning $30,000–$50,000 gave an average of $68 to education, but there is a dip to $32 for those earning $50,000–$75,000; those earning $75,000 to $100,000 gave $544; and those earning more than $100,000 gave an average of $1,327 to education (overall mean=$599).
Table 2

Mean giving to education by income and decision-making authority

Income

Male

Female

Mostly male

Mostly female

Joint

Separate

All couples

n

$10,000 or less

17

    

25

3

72

10,001–30,000

4

17

  

21

4

8

378

30,001–50,000

9

13

233

68

65

51

30

692

50,001–75,000

26

26

23

32

81

53

44

815

75,001–100,000

16

48

 

544

61

69

59

529

over 100,000

113

184

583

1327

228

247

252

792

all incomes

46

58

300

599

112

131

89

 

n

283

536

52

98

1244

248

  

Interestingly, among the two sets of couples who made the decision either jointly or separately, giving to education grows with income and at similar rates and with nearly the same amounts for the overall means ($112 and $131, respectively) and those earning over $100,000 ($228 and $247, respectively). Among all couples, very few decisions to give were made mostly by men (3.2 percent) or mostly by women (6 percent), and only 15 percent were made separately; a full three-fourths of the couples (75.8 percent) made the decision to give to education jointly.

Regression results

Factors affecting the probability of being a donor to education (at all)

Our results are remarkably robust whether using a probit or a logit and hence we will treat them interchangeably. Not surprisingly, the educational attainment levels of both men and women have a positive and significant effect on the probability of whether or not a household is a donor to education (see Table 3). As families have more children, they are more likely to be donors to education. Income has a positive effect on the likelihood of being a donor, but surprisingly, wealth (excluding home) does not matter. Husband's age does not matter, but increases in the age of the wife are positively associated with the likelihood of being an education donor.
Table 3

Probit and Logit Models for Giving to Education

dep vbl: whether give to education or not

dep vbl: whether give to education or not

Probit model

Logit model

Variable

Coefficient

Variable

Coefficient

Constant

−4.659655***(0.3048421)

Constant

−7.998869***(0.5420015)

Education: husband

0.1015595***(0.0169497)

Education: husband

0.177443***(0.0298271)

Education: wife

0.1135076***(0.0178478)

Education: wife

0.1948701***(0.0309489)

Number of children (at home)

0.1758364***(0.032155)

Number of children (at home)

0.2953025***(0.0548437)

Family income

2.04e-06***(5.22e-07)

Family income

3.37e-06***(8.66e-07)

Wealth (excluding home)

3.09e-08(3.09e-08)

Wealth (excluding home)

4.75e-08(5.21e-08)

Age: husband

−0.0085396(0.0071226)

Age: husband

−0.0137757(0.0124317)

Age: wife

0.0199115***(0.0074709)

Age: wife

0.0333796**(0.0130706)

Male decides alone

−0.1761117(0.1111884)

Male decides alone

−0.2961216(0.1957657)

Female decides alone

0.1838358**(0.0801451)

Female decides alone

0.3063296**(0.139306)

Male mainly decides

0.3036928(0.2023556)

Male mainly decides

0.5271218(0.3308657)

Female mainly decides

0.4093173***(0.1518307)

Female mainly decides

0.6816128***(0.2556255)

Couple decides separately

0.2828981***(0.1069419)

Couple decides separately

0.4673717***(0.1807798)

n

2191.00

n

2191.00

Pseudo R2

0.1293

Pseudo R2

0.1288

Standard errors are in parentheses

*p<0.1, **p<0.05, ***p<0.01

Who decides whether to give to education at all?

In cases is which the man decides exclusively or mostly, there is no significant effect on the probability of being a donor. Conversely, in cases in which the female decides either entirely or mostly, the household is much more likely to give to education. Perhaps surprisingly, couples that decide about giving separately are significantly more likely (holding everything else constant) to give something to education than households that decide jointly.

Factors affecting the amounts donated to education

Our results are remarkably robust whether using raw dollar amounts or a log–log model and hence we will characterize them as one. We find that both the husband's and the wife's educational attainment is positively associated with the amounts donated to education, as is the number of children (under 18 at home), as well as income and wealth. Like in the probits, the age of the husband does not affect the amounts given to education but the age of the wife is positively associated with increased giving to education (Table 4).
Table 4

Tobit Models for Dollars Given to Education

dep vbl: $ Given to education

dep vbl: ln(Amount Given to Education+10)

Tobit model

Tobit model

Variable

Coefficient

Variable

Coefficient

constant

−8469.308***(587.9593)

Constant

−14.51934***(1.098383)

Education: husband

193.0368***(31.17848)

Education: husband

0.4253772***(0.056697)

Education: wife

192.1625***(32.50938)

Education: wife

0.4061529***(0.0591086)

Number of children (at home)

258.5459***(58.51521)

Number of children (at home)

0.5908191***(0.1060711)

Family income

0.0012182***(0.0003888)

Family income

1.98e-06***(7.36e-07)

Wealth (excluding home)

0.000142***(0.0000368)

Wealth (excluding home)

1.49e-07**(6.96e-08)

Age: husband

−21.88957*(13.08924)

Age: husband

−0.0357524(0.0235865)

Age: wife

41.98737***(13.72314)

Age: wife

0.0762938***(0.0247447)

Male decides alone

−372.484*(206.7125)

Male decides alone

−0.6073378*(0.3681019)

Female decides alone

223.0657(146.4282)

Female decides alone

0.4961071*(0.2648613)

Male mainly decides

443.0009(347.8523)

Male mainly decides

0.978251(0.6357363)

Female mainly decides

1252.741***(254.0462)

Female mainly decides

1.356679***(0.4819305)

Couple decides separately

332.1213*(190.3044)

Couple decides separately

0.9294566***(0.3454599)

n

2191.00

n

2191.00

Pseudo R2

0.0276

Pseudo R2

0.0763

Standard errors are in parentheses

*p<0.1, **p<0.05, ***p<0.01

Who decides how much to donate to education?

Decisions made exclusively by men or women are not significant at traditional levels of significance, as are decisions made mainly by men. Decisions made mostly by women or separately by couples, however, are associated with increased levels of giving to education, holding all other factors constant.

Factors affect amounts donated to religious and secular charities

In order to compare the results about giving to education and their context validity, we ran similar regressions for amounts donated to religious organizations and secular nonprofits. We find that both the husband's and the wife's levels of educational attainment are positively associated with increased giving amounts to both secular and religious nonprofits, but their ages do not matter. The number of children is positively associated with religious giving but not secular giving. As expected, income and wealth are positively associated with increased amounts of both religious and secular giving (Table 5, 6).
Table 5

Tobit Model for Dollars Given to Secular Causes

dependent variable: secular giving amount

Tobit model

Variable

Coefficient

constant

−3582.988***(401.09)

Education: husband

124.4078***(23.01768)

Education: wife

105.525***(24.99597)

Number of children (at home)

10.37826(46.49681)

Family income

0.0027993***(0.000385)

Wealth (excluding home)

0.0002743***(0.000037)

Age: husband

1.228991(9.761254)

Age: wife

11.45585(10.20093)

Male decides alone

−271.3346*(147.9234)

Female decides alone

10.90736(112.931)

Male mainly decides

188.5785(308.615)

Female mainly decides

753.2153***(227.4447)

Couple decides separately

538.6329***(158.3239)

n

2191.00

Pseudo R2

0.0105

Standard errors are in parentheses

*p<0.1, **p<0.05, ***p<0.01

Table 6

Tobit Model for Dollars Given to Religion

dependent variable: religious giving amount

Tobit model

Variable

Coefficient

constant

−5813.181***(926.66)

Education: husband

95.54966*(52.07481)

Education: wife

145.766**(57.01673)

Number of children (at home)

469.851***(107.3384)

Family income

−0.0005359(0.0008686)

Wealth (excluding home)

0.0006117***(0.0000837)

Age: husband

28.75533(22.80453)

Age: wife

37.81845(23.85502)

Male decides alone

−1414.971***(340.4461)

Female decides alone

−1737.95***(262.2431)

Male mainly decides

−75.03716(690.5132)

Female mainly decides

−891.2774*(538.6378)

Couple decides separately

−1441.37***(374.847)

n

2191.00

Pseudo R2

0.0069

Standard errors are in parentheses

*p<0.1, **p<0.05, ***p<0.01

Who decides how much is donated to religious and secular charities?

For men and women deciding alone, there is a no association with the amounts given to secular charities but both groups experience a negative association for religious giving. In cases in which the male is mainly responsible, there is no relationship with the amounts given to religion or for secular giving amounts. When females are the main decision-makers, there is a positive effect on secular giving and no effect on religious giving, holding other factors constant. When couples decide the amounts to give separately, there is a positive association with secular giving and a negative association with religious giving amounts, holding everything else constant.

Discussion

Let us return to our research questions (original numbering listed parenthetically) and see how we would answer them given more complete information. (1) What are the patterns of household giving to education and can we predict whether or not a donor will give to education? (2) Do income and wealth play a special role in giving to education? (5) Does the educational attainment of donors and prospective donors affect whether or not they give at all—and if so how much? The educational attainment of both spouses is positively associated with both the likelihood of being a donor to education and the amounts donated to education, as are the number of children living at home and income. Wealth has a mixed effect: it does not seem to matter in the decision to give to education but it does affect how much is given to education. Perhaps surprisingly, the husband's age does not matter but the wife's age is positively associated with increased likelihood to give and amounts given to education.

(3) How does giving to education vary according to who decides? (4) Is there a difference between giving to education versus secular giving and religious giving? Clearly, who the decision-maker is in the question of “who decides” in giving to education matters. While most of these decisions are taken jointly, we do see some significant differences in the cases in which men and women are the main decision makers or they are made entirely separately. Bad news for men: after controlling for other factors, men are not likely to have a significant effect in the decision to give to education at all nor the dollar amounts given. On the other hand, women, whether deciding on their own or having the main influence, are more likely than not to have a significant effect on the likelihood of being education donors and consequently on the amounts donated to education. These patterns are not identical for religious giving and total secular giving but they are pretty consistent. While women may suffer from a pay gap in the labor market, men seem to suffer from an influence gap in the philanthropic giving decision-making sphere.

Implications for Practitioners

The findings from our study indicate the importance of women as donors to education. Our study finds a clear and consistent pattern of women as donors that is significantly different from that of men. This is consistent with other research showing that different donor groups have different motivations and preferences for giving (i.e., Briechle, 2003; Liu, 2006). As the trend continues, in which women are becoming more affluent and moving into the ranks of middle and upper classes at an increasing rate, fundraisers would be well advised to pay attention to the giving preferences of women.

Copyright information

© Palgrave Macmillan Ltd 2007

Authors and Affiliations

  1. 1.Center on Philanthropy at Indiana University IUPUISuite 301 IndianapolisUSA