Philanthropy Among Young People: Empirical Methods Research Paper

Introduction

America is a nation bursting with civic spirit, generosity, philanthropy, and volunteerism. This remains true even as the country struggles through the third year of fallout from the collapse of the sub-prime mortgage market and, many economists warn, the U.S.A. is already in frank recession (see, e.g., Linde Group, 2008; IMF). And yet, it is difficult to overlook a torrent of donations that reached a record $306 billion last year (Grammage, 2008) dispensed by over a million charities, large and small (Stone, Association of Fundraising Professionals, 2008).

Institutions, foundations, and the extremely wealthy grab headlines, as do universities eager to modernize facilities with tax-deductible windfalls. Little is heard, however, about volunteerism and outright philanthropy from university students themselves and the youth in general, despite hoary pronouncements about the idealism of the young. This paper, therefore, aims to examine the extent of empirical research about philanthropic behavior among the young and redress what gaps there may be.

Literature Review

Philanthropy in America

The reported figure of $306 billion includes grants, scholarships, employee matching gifts, and other amounts reported as “grants and contributions paid during the year” on the 990-PF tax form. The ten leading private fund sources, by market value of assets and amounts disbursed are:

These well-known foundations may make the headlines with the occasional large donation but, says the Boston College Center on Wealth and Philanthropy, it is the mass of high-net-worth individuals that adds up. Nearly three-fourths of the record donations last year came from individuals. A somewhat dated Gallup poll reveals that nearly 90% of Americans donated money to some charity over a given year (Carroll, 2006) or at least liked the prestige of donating enough to make such a claim. At the top end, just 8 percent of U.S. households account for fully half of all giving in America (Schervish, 2008).

As to volunteerism, an excellent conceptual framework is to “…act in recognition of a need, with an attitude of social responsibility, and without concern for monetary profit, going beyond one’s basic obligations.” (Ellis, qtd. in Jordon, 1994). Concretely, the current study shall be consistent with prior research in defining volunteer participation as including:

1. Those outside the scope of a college course or service group.
2. Carried out under the aegis of a service or church group that may not be affiliated with the school.
3. Fundraising, participation in church-sponsored activities, building houses for Habitat for Humanity, environment-rescue projects, or succoring the homeless.

The research will, however, exclude participation in structured service-learning activities since philanthropic intent is the core of the study.

The modern-day catalyst seems to have been a 1960 address by then-candidate Sen. John F. Kennedy at the University of Michigan; his address to the 10,000 students then assembled is widely credited with reviving a spirit of community service and civic involvement that had been thought lost after the era of pioneering and settling the American West had passed (Berson, 1995). Certainly, the clarion call of the young Kennedy to serve America and promote peace by volunteering for duty in developing countries (and the equally ringing declaration in his 1962 inaugural address, “Ask not what your country can do for you, but what you can do for your country”), inspired more than 195,000 Americans in the decades that followed (Peace Corps, 2008) to help the civil rights movement of the time and volunteer for overseas duty with the Peace Corps. Within the nation’s borders, at last count, more than half of adult Americans acknowledged plans to volunteer around twenty billion hours over a given year in charitable activities (Rose, 2004).

By the end of the century, two landmark laws had been passed: the National and Community Service Act of 1990 and the National and Community Service Act of 1993. But civic spirit had waned. Participation rates in volunteer programs declined, especially among the young and the better-educated, those who had at least attended or actually finished college. Ehrlich (2000) opines that this is paradoxical since college should shape an openness towards other communities and cultures, thereby providing fertile ground for civic engagement. This is consistent with the fact that 19 in 20 Peace Corps volunteers today have at least an undergraduate degree (Peace Corps, 2008).

Still, there is evidence of high rates of volunteerism by college students, albeit self-reported. Service-learning components were introduced into universities and colleges as early as the sixties. Such programs are more prevalent than ever, providing both students and faculty avenues for community and civic volunteerism. The Cooperative Research Program of the Higher Education Research Institute reported at the start of the decade that four in five college freshmen could claim to have done volunteer work in a prior 12-month period and more than half did community service as part of classroom requirements (Pritchard, 2001). This is up substantially from the 43% of freshmen reported by the American Freshman Survey in 1988 as having volunteered a minimum of an hour weekly and the two-thirds of all undergraduates ever involved in volunteer work reported in the Undergraduate Survey of 1993.

Personal Characteristics and Other Correlates of Philanthropic Behavior

In good times or bad, not-for-profit organizations are clearly obligated to target solicitations very well so as to keep overhead to the minimum. Personal contact works when approaching foundations and very high net-worth individuals, of course. Mass mailings and direct-response media have to be exquisitely fine-tuned because individual donors are the biggest source of donations nationally (Hatfield, 2008). accounting for no less than 75% to 76% of dollar inflows (Sargeant, Ford, and Hudson, 2007; Stone, 2008). Hence, it is crucial to understand donors’ characteristics so charitable organizations can target those most likely to give, thereby saving time and money in the process as well.

There is evidence that college education is a sound predictor of charitable giving. Fulkerson (1995) reports, for instance, that forty-seven percent of individual donors in 1995 were college graduates. Brown (2001) also points to a correlation between higher education and an increase in charitable giving. The explanation for these two findings may lie in the fact that those who choose to obtain a college degree may be more invested in their communities, to begin with, and learn about the value of charitable organizations in the school. Secondly, college degrees are related to either better means to start with or subsequently accelerate one’s rise through the hierarchy of an organization and the social order of the surrounding community. Therefore, those with degrees are more likely to be asked for charitable donations and are better able to afford philanthropy.

Income considerations aside, volunteer projects provoke “purer” motivations such as Pierson (2002) declares, the self-satisfaction one gain from simply helping others, the idealism of wanting to correct social inequities, and even somewhat compulsory involvement in church apostolates.

As to the socio-demographic profile of student volunteers, the evidence is sparse and mixed. A twenty-year-old survey among California State University undergraduates suggested that incidence of volunteerism varied by gender (a 36:27 female-to-male ratio), was more popular among mainstream whites and Native Indians than the Asians and Filipinos who are important minorities in the state, and seemed for some reason more attractive to students 30 years and older (41% participation) than to the majority who were 29 or younger (less than 30%, O’Brien, 1993). More recently, however, a wider-ranging meta-analysis of varied studies by Levine and Curton (1998) revealed no substantive differences in volunteerism by age, gender, ethnicity, or college type attended.

Sociological Theory

Going to college at all, Pascarella and Terenzini (1991) report in a perhaps self-evident conclusion, is a function of family income and occupational status. On the other hand, Blau and Duncan (1967) constructed psychological and sociological models proposing that the antecedents of higher educational attainment are:

• Parental education (besides the aforementioned occupation and income;
• The quality of achievement in high school;
• Personal aspirations;
• Support from teachers and parents;
• Mental ability of the student.

There is some evidence that the inclination for volunteerism propensity is inversely related to being materialistic. Several retrospective studies show that non-volunteer students scored higher on self-description scales about the importance of “being well-off financially” as a goal in life (Astin, 1991; O’Brien, 1993; Astin and Sax, 1998; Vogelgesang and Astin, 2000). With the benefit of hindsight some nine years after graduation, however, it seems that volunteerism linked rather strongly with going on to graduate school, higher-income post-graduation, and being confident about socializing with varied ethnic groups or income strata.

Giving Among the Young

There is no question that age is no hindrance to volunteer work. In fact, the idealism so characteristic of the young is absolutely essential for volunteerism, whatever the circumstances under which it is supplied.

One concedes that the young in their twenties and thirties, just starting out on their careers or on the way up, do not yet possess the disposable income to make sizeable donations. Following the tradition of their parents and elders – fully 89% of Americans gave at one time or another to the “March of Dimes”, the drive to eradicate polio — the “Millennials” and “Gen X’ers” start giving early, do so more consistently, support new causes and are keen to be involved long after sending in their check (Stone, 2008). As a result, one observes such phenomena as United Away airing an appeal on the Super Bowl for a $5 donation via cellular phone SMS, others using ballpark electronic scoreboards or live appeals during R & B concerts, charities using Google Earth to show youthful donors the progress of village water supply projects in various countries, Boston’s Museum of Fine Arts putting up a FaceBook page and inviting fans, the Chicago Symphony Orchestra and the Ann Arbor (MI) Symphony Orchestra posting on YouTube to reach out to young benefactors, and Oxfam employing email blasts to reach 400,000 potential benefactors overnight. All these evidently lower the participation threshold for the young and even students. In the reverse direction, online donation payments, 6% of American families already seem comfortable giving over the Internet (Ibid.) and this is fine for “entry-level” donations.

As to causes supported, Giving USA reports that those in their 20s and 30s have a distinct preference for ecological causes and reaching out internationally although the lion’s share of giving by any age cohort does go to local/community concerns. As an example of the latter, college students and young workers seem to like giving time as literacy coaches in elementary schools around the neighborhood.

If new technology has provided ways for the young to surmount low disposable incomes, is it a question then of doubtful intent and lack of the impulse to give? In fact, a pilot experiment by Wineburg (1985) suggests that schoolchildren in seventh through ninth grades were inclined to give and that the propensity for giving was inversely proportional to age. However, a question can be raised about the reliability of this finding owing to the fact that the test subjects had been “sensitized” by the Jewish tradition of bringing tzedakah (or charity), symbolizing righteousness, for collection in class.

Closer to the age cohort that this study will cover, Nelson (1999) quantified the results of a telephone poll conducted in Los Angeles county. In 1999, for some reason, donation prevalence fell to two-thirds of families, lower than the national average. Even worse, the donation incidence among those aged 18 to 24, the leading beneficiary, fell by half to less than 25% where religious organizations were concerned, and that for non-religious beneficiaries declined from 12% to 10%.

The Information Gap

A search of the literature reveals plentiful cues on factors that influence postsecondary achievement generally and, to an extent, career choices. In the other direction – testing the relationship between career choice (implied by academic major selected) and propensity for philanthropy – there is a decided dearth of empirical research.

Research Question

Given the findings that there are certain occupational and other personal differences among those more actively involved in donating time and money, the central research question this study addresses is whether such career inclinations manifest earlier. Is philanthropic behavior also strongly associated with the choice of certain fields of study in college?

Accordingly, the study hypotheses should be stated as follows:

• H₀, the null hypothesis: There is no difference in philanthropic behavior across fields of academic study.
• H_a, the alternative hypothesis: Certain fields of academic study are associated with a greater incidence of philanthropic behavior.

An empirical and cross-sectional study should therefore address the following research objectives:

1. Test the assumption that choice of academic major is a fully autonomous choice of college students and may therefore correlate highly with other personal inclinations such as giving behavior.
2. Assess the extent of financial and time resources college students have available for philanthropy.
3. Quantify past, present, and family philanthropy as antecedents for future giving.
4. Investigate whether there are meaningful differences in philanthropic behavior across specific majors or academic departments.

Key Concepts and Variables

The dependent variable (DV) in this study shall be philanthropic behavior, defined as donating either time, money, or both to the needy, for disaster relief and for other charitable causes. This shall specifically exclude time and money allotted to course projects and co-curricular activities under the aegis of the University.

The influencing or independent variable (IV) shall be academic major or field of interest operationalized as the home department of Junior- and Senior-year undergraduates, as well as graduate students, at the time fieldwork, is to be carried out.

The study will also gather data gender, on prior giving behavior, philanthropic intentions after graduation, track record of family giving, social class, disposable income, and available free time (expressed as units carried during the term). In analysis, it is likely that such intervening or explanatory variables may have to be held constant in order to test more fully the relationship between the IV and DV. Alternatively, data analysis may demonstrate that one or several counts as important IVs.

Other variables such as age and ethnicity may not even be gathered at all since there is no theoretical basis for believing that these influence the DV in some way.

Sample/Participants

Given the absence of scientific research in the area, this study will have the greatest theoretical utility if it were to be generalized to all college upperclassmen and graduate students around the country. On a pragmatic level, however, one should consider the current plan as embodying a pilot project that may well yield satisfying results and can then be replicated in other campuses.

While the population of interest consists of all university students who have already made a choice of major (which would suggest simple random probability sampling), the relevant stipulation is that the sample frame for the study consists of all 73 bachelor’s degree programs and 32 graduate programs extant at UNCW (University of North Carolina Wilmington, n.d.). In analysis, it is likely that past and planned philanthropic behavior will be compared across:

1. Ten leading programs by student population.
2. Four major academic clusters consisting of the Arts, Science, Business, and the Schools of Nursing and Education combined (on the rationale that the latter two are service-oriented professions). In consultation with the study adviser, the Interdisciplinary majors and graduate students shall be grouped into one of these clusters.

The above specifications apply chiefly to ensuring a representative sample for the central study instrument, a structured questionnaire (see “Data Collection” and “Study Instrument” below). In the first stage of the study, when focus group discussions shall be employed, only pure random or convenience sampling within each of the above clusters is necessary.

The quantitative survey will employ stratified sampling because the researcher opts to divide the universe of UNCW who have chosen their academic majors four to ten mutually exclusive and exhaustive subsets. Within each stratum, the study will require simple random selection, an operation that is independent of the sampling done in every other stratum.

Such an approach enhances the validity of the study since one is effectively sampling across subsets of the independent variable. Moreover, each student with a major can be assigned to one and only one stratum and no one in the population is “unassignable”.

Other strengths that stratified sampling lend this study have to do with:

• Operational efficiency because the information that allows the researcher to classify each upperclassman and graduate student is readily available from the University database.
• Improved statistical efficiency because stratified samples boast smaller sampling errors than pure random samples (Cochran, 1963).

The reliability of the study is also enhanced by the fact that stratified sampling effectively forces the study to be representative by virtue of culling adequate numbers of students across categories of the IV and since the database exists, controlling by representativeness by weighting the results back to their true proportions in the UNCW population.

Weighting will be necessary because the researcher proposes minimum net (of non-response or spoiled questionnaires) sub-sample sizes to ensure adequate numbers for data analysis:

The weighting scheme, to take place prior to data analysis, will be based on the true distribution that is available in the University database.

Ethical Considerations

The study will comply fully with UNCW guidelines for social science research on human respondents by ensuring that:

• All participation will be voluntary. Students must sign a separate consent form when recruited for the focus groups and at the end of the structured questionnaire proper.
• No personal information will be collected other than those necessary to classify the student for philanthropic activity, academic major, social class, age, gender, and religion. In particular, no names, addresses, or contact information will be obtained other than what is necessary for post-fieldwork quality control and back-checking of vague/illegible answers. After this necessary activity, such names and contact information will be destroyed, never to be encoded for data processing.
• The researcher submits to the supervision of the UNCW Institutional Review Board under the guidelines of undertaking a scientific investigation and being designed to contribute to generalizable knowledge. Further, the proponent undertakes to submit the full research protocol, consent forms, FGD discussion guide, and survey questionnaire proper for review.
• Neither pilot interviews nor any fieldwork shall commence without written notice of approval from the IRB chair or full board.
• The research will involve no minors nor expose any respondent to physical, moral, or emotional risk. Neither will any responses provided result in grounds for academic discipline and sanction.

Data Collection

This study will employ a two-stage, qualitative/quantitative data collection plan.

The first stage will consist of four focus group discussions (FGD’s), one for each of the academic strata defined under “Sample/Participants” above. This will require merely convenience sampling in order to assemble panels of eight to 12 participants per group. Taking on the role of FGD moderator, the researcher will run sessions of about an hour each in a suitably private setting such as a student apartment, a lounge in student housing, or even outdoors as the weather permits. Verbatim feedback will be recorded on audio or videotape so as to subsequently permit content analysis in context.

The researcher relies on the synergy of group dynamics in the FGD to provoke a greater range of responses than might otherwise be gathered from individual interviews with a like number of respondents. Since philanthropy is laudable and embodies a social good, exposure of personal attitudes and giving behavior will likely not be inhibited by the public nature of a focus group.

Diligent content analysis of the first-stage exploratory research will in all likelihood give rise to additional questions or unanticipated answer categories for inclusion in the second-stage survey questionnaire.

The administration of this questionnaire comprises the final element of the stratified sampling plan previously described. The choice of survey method is dictated by three important considerations:

1. Efficiency – Undertake the group administration format because study length will be unconscionably long if the researcher had to conduct hundreds of face-to-face interviews.
2. Reliability and Validity – A supervised self-administered survey optimizes response rate better than mail, leave-behind, “take-one” dispensers, online, or email surveys. Both reliability and validity are seriously hampered when large numbers of target respondents justify non-participation with the attitude that philanthropy has no place in their lives or they are embarrassed to admit that they have no wish to give of their time or treasure.
3. Random selection – can be satisfied by selecting one to three classes at random from the department lists of courses required of Juniors, Seniors, or graduate students and requesting lecturers for about ten minutes of class time to explain the purpose of the study and be available for questions while the entire class fills out the survey.

The survey form shall be a structured, self-administered questionnaire that can be accomplished in a few minutes by students of average intelligence and alertness. It shall comprise sections for personal information, philanthropic behavior, and rationale for the academic course being undertaken (see “Study Instrument” below for item outline).

Survey/Study Instrument

Other than the fact that the questioning format in FGD’s is almost always open-ended and that the structured questionnaire is mostly close-ended or aided, item wording for the DV philanthropic behavior will be similar. The other two types of information to be collected, academic major and personal data, will be recorded in similar forms in both cases.

Independent variable Academic Major

1. Year level (PUT CHECKLIST)
2. What degree are you pursuing? You can tick off more than one if applicable. (CHECKLIST)
3. Under what department? (CHECKLIST)
4. What attracted you to enroll for this degree? (OPEN-END)
5. Other than your own conviction, were there any other influences on your choice of major? (YES/NO)
6. What influences were these? (OPEN-ENDED)

Personal Information (“Please check off or fill in the answer blanks below to tell us something about yourself. Please be assured that this study will merely group all the findings. Your answers can never be traced back to you.”)

1. Gender (TICK BOXES)
2. Age (CHECKLIST OF AGE RANGES)
3. Religion/Church attended (OPEN-ENDED)
4. Social class (CHECKLIST)
5. Recipient of a government grant, student loan, or other financial assistance (YES/NO)
6. Personal savings (in dollars)?

Philanthropic Behavior

Have you ever volunteered and spent at least half a day of your time helping…(MULTIPLE ANSWER: the needy, disaster relief, community clean-up, feeding program, child or elderly care, fund drive, and similar philanthropic activities? IF NON-VOLUNTEER, SKIP TO DONATIONS SECTION)

1. How old were you when you first started to volunteer for such activities? (CHECKLIST OF AGE RANGES)
2. Over the past 12 months, about how many hours in all did you volunteer for work with the needy, disaster relief, etc.?
3. Will you continue to do so after college? (YES/NO)
4. Why/Why not? (OPEN-ENDED)
5. Compared to what you did in the previous 12 months, do you anticipate volunteering the same, more or fewer hours after graduating?
6. By how many hours do you expect to lengthen or shorten your volunteer involvement?
7. (IF CONTINUE TO VOLUNTEER) Which recipients, clubs, drives, or activities will continue to get your volunteered time?
8. Have you ever given cash pledges or money of your own to help…(MULTIPLE ANSWER: the needy, disaster relief, community clean-up, feeding program, a civic club or non-profit organization, for a child or elderly care, fund drive, and similar philanthropic activities?)
9. How old were you when you first started to donate money? (CHECKLIST OF AGE RANGES)
10. Over the past 12 months, how much in total did you donate? (OPEN-END)
11. Will you continue to do so after college? (YES/NO)
12. Why/Why not? (OPEN-ENDED)
13. Compared to how much you donated in the previous 12 months, do you anticipate giving more or fewer funds of your own after graduating? (FORCED CHOICE CHECKLIST, INCLUDING “No change”)
14. By how much do you expect to increase, or cut back on, your donations each year?
15. (IF CONTINUE TO DONATE) Which recipients, clubs, drives, or activities will continue to deserve your personal donations?

Consent Form

Note that there is no answer option for dual time-money philanthropy. This will be taken care of by recording in the data processing.

Analytical Approach

Subsequent to data entry, two additional variables will be generated by first running a descriptive analysis of time (in hours/days) and money (in dollars) donated recently. The distribution curves, means, standard deviations, and quartile ranges will then be scrutinized to derive two new interval variables that will classify the students into non-donor, low-philanthropy, average, and high in reported philanthropic behavior as to time or money.

The second stage will be cross-tabulation of the IV academic major and all other personal information against the DV’s. Several cross-tabulation runs will be required to test outcomes for the dependent variables when these are expressed as:

• Nominal variables (“donor” or not).
• Separate runs for donating money, volunteering time, and students who do both.
• Degree of philanthropic behavior: Low, average, or high.
• The interval variable is “opinion about philanthropy being worthwhile”.

To this point, data analysis will have depended on frequency counts and modes for the nominal variables; mean, median, and variance for the ordinal items; and mean and variance for the interval scales.

Tests of significance will then proceed. One can, a priori, choose α = 0.05 as the decision rule, meaning a 5% chance of being wrong when rejecting the null hypotheses. If any of the indicated statistical tests indicate that the probability of occurrence of the observed result due to chance or sampling error is less than 5%, this gives us confidence for accepting the alternate hypotheses: choice of academic major or cluster indeed has a bearing on philanthropic behavior.

As to specific tests we shall employ:

1. The chi-square in the case of contingency tables displaying the frequency distribution of, for example, the nominal variables of donation incidence across the independent sub-samples of academic major.
2. The Z test for two means when scrutinizing means of the interval variables on attitude scales or the ratio variables measuring time and money donated.

References

Association of Fundraising Professionals. (2008). Recognizes the importance of giving in tough economic times. PR Newswire.

Astin, A. (1991). Student involvement in community service: Institutional commitment and the campus compact. Los Angeles: Higher Education Research Institute, UCLA.

Astin, A. & Sax, L. (1998). How undergraduates are affected by service participation. Journal of College Student Development, 39, 251-63.

Berson, J. S. (1995). Win/win/win with a service-learning program. 2008. Web.

Blau, P. & Duncan, O. (1967). The American occupational structure. New York: Free Press.

Carroll, J. (2006). Americans are more likely to donate money, not time, to charities. The Gallup Poll Briefing, 46-47.

Cochran, W. G. (1963). Sampling techniques. 2^nd ed. New York: John Wiley & sons.

Ehrlich, T. (2000). Measuring up: The state-by-state report card for higher education. Civic Engagement. Washington, DC: The National Center for Public Policy and Higher Education.

Foundation Center (2008). Top funders: Top US foundations. Web.

Gammage, J. (2008, October 2). Charities brace for lean times in 2009. McClatchy – Tribune Business News.

International Monetary Fund (2008, October) World economic outlook: Financial stress, downturns, and recoveries. Washington, DC.

Jordon, K. (1994). The relationship of service-learning and college student development. Blacksburg, VA: Virginia Polytechnic Institute and State University. Levine & Curton (1998).

The Linde Group (2008) General economic environment. Web.

Nelson, S. S. (1999). Study finds fewer donors to charity, aid: Givers in Los Angeles County tend to be older, survey reports. Officials fear ‘the beginning of a worrisome trend. :[Home Edition]. Los Angeles Times, p. B, 1:5.

O’Brien, E. (1993). Outside the classroom: Students as employees, volunteers and interns. Research Briefs, 4(1) 1-12.

Pascarella, E. & Terenzini, P. (1991). How college affects students: Findings and insights from twenty years of research. San Francisco: Jossey-Bass.

Peace Corps (2008). Fact sheet 2009. Web.

Peace Corps (n.d.). History/Decades of service. Web.

Pierson, C.T. (2002) Volunteerism in college: Impact on cognitive outcomes, learning orientations and educational aspirations. (Doctoral dissertation, University of Iowa, 2001).

Pritchard, I. (2001). Raising standards in community service learning. About Campus, 18-24.

Rose, D. (2004). The American philanthropic tradition. Executive Speeches, 18(4), 1-4.

Schervish, P.G. (2008). Despite downturn, there’s hope for U.S. charitable giving, Boston College researchers tell national conference of fundraisers. PR Newswire.

Stone, A. (2008). The new face of giving: With a high-tech boost, even minors have major impact. USA TODAY, p. E.1.

University of North Carolina Wilmington. (n.d.). About UNCW. 2008. Web.

Vogelgesang, L. & Astin, A. (2000) Comparing the effects of community service and service learning. Michigan Journal of Community Service Learning, 7, 25-34.

Wineburg, S. S. (1985) Factors affecting philanthropic behavior of Jewish adolescents. Journal of Social Psychology, 131(3) 345-354.