8 Case Studies on Both Descriptive Statistics and Inferential Statistics: Social Justice Nonprofit Organizations with Hypothesis Testing Examples
Learning Outcomes
-
- Introduction to Nonprofit Organizations
- Data Sets: Average Base Pay for Nonprofit CEOs
(from the Bayer Center for Nonprofit Management at Robert Morris University’s Rockwell School of Business) - Statistical Analyses
- 4a. Introduction
- 4b. Descriptive Statistics
- 4c. Inferential Statistics
- Hypothesis Testing Problem #1: To assess whether CEO salaries differ across different funding types.
- Hypothesis Testing Problem #2: To assess whether Program Manager salaries differ across different funding types.
- Hypothesis Testing Problem #3: To assess whether the availability of professional development benefits or no professional development benefits (binary) for CEOs differ across different funding types.
- Hypothesis Testing Problem #4: To assess whether the availability of professional development benefits or no professional development benefits (binary) for Non-Executives differ across different funding types.
Descriptive and Inferential Statistics:
Social Justice Nonprofits
Case #4: Social Justice Nonprofit Organizations
In the world of nonprofit organizations, we often hear the word social justice.
A significant number of people who believe in and work for social justice are employed in the nonprofit sector: an industry that requires organizations to compete for government and foundation funding.
While there is no single definition of a nonprofit organization, they are considered an integral part of society that bolsters economic prosperity, environmental integrity and social justice.[1] Nonprofits usually possess six characteristics. They are[2]:
- Formal organizations dedicated to pursuing mission-oriented goals through the collective actions of citizens.
- Legal entities organized and operated for a collective, public, or social benefit, and are not entities that operate as businesses aiming to generate a profit for their owners.
- Subject to the non-distribution constraint: any revenues that exceed expenses must be committed to the organization’s purpose, and not taken by private parties.
- Self-governing organizations that are governed by a board of directors, a group of volunteers that is legally responsible for making sure the organization remains true to its mission, safeguards its assets, and operates in the public interest.
- Dependent upon volunteers to support their activities. Smaller nonprofits may have no paid staff and are completely organized and managed by volunteers.
- Focused on meeting community needs through the provision of their services, goods and resources. They assist the community to drive economic development, the arts, cultural awareness, education, health, and spirituality — in virtually every sector of society.
Nonprofit organizations are accountable to the donors, founders, volunteers, program recipients, and the public community. Nonprofit organizations also are known simply as “nonprofits.”
For nonprofits to receive donated funds, they must register as a 501(c)(3) organization with the Internal Revenue Service. Recognition as a 501(c) (3) allows nonprofit organizations to access grant money from foundations, corporations, and the government without paying income tax. Nonprofit organizations depend upon a diverse set of funding streams to sustain their operations, such as, through grants and income from donations.
Nonprofit organizations are held to high moral standards because their social and community values are associated with their missions. Honesty, integrity, transparency, confidentiality, and equity are values that are typically expressed in a charitable nonprofit’s code of ethics–but there may be other values that are important to a nonprofit.
Some nonprofits, such as Social Justice Nonprofits may emphasize a justice-based system–a moral perspective that emphasizes fairness and equality. A Social Justice Nonprofit focuses squarely on communities most impacted by oppressive systems in the United States. They often receive the least amount of funding but are closest to the solutions. Such nonprofits have a long history of involvement with social movements, and by incorporating the language of social justice and social change into their programming, they send a clear signal to funders that they value those themes. By incorporating social justice discourse into their programming, these nonprofits strive to become institutional entrepreneurs, pushing the broader philanthropic community to consider equity as a funding priority.
A number of small and large foundations have announced new focus areas designed to address the equity issue. They are changing their grantmaking priorities, program areas, and hiring practices, and are pooling resources to advance one of the most vital issues of our time. Foundations recognize that investing in equity means investing in capacity, infrastructure, and leadership development, not just programs.
For the Boston Foundation[3]equity is at the center of its mission.[4]
“If we are to advance equity and close the gaps caused by this region’s greatest disparities, we must tackle the individual, systems-level and root-level causes of inequity. To do this, our work with nonprofits must expand beyond grantmaking to other dimensions of partnership that allow us to most effectively tackle these complex and challenging issues. In addition to making grants, we partner with nonprofits and civic organizations to advocate, convene, raise funds, conduct research, expand thinking, share stories and resources, and co-create strategic efforts that build a more equitable Greater Boston.”
The National Council of Nonprofits stands with the Boston Foundation and others for equity and justice and in denouncing racism, intolerance, and exclusion. Embedded within the W. K. Kellogg Foundation is a commitment to advancing racial equity and racial healing, to developing leaders and to engaging communities in solving their own problems. As stated in its report, The Business Case for Racial Equity[5], “Furthering the success of populations of color will not only serve an important social justice goal, but it will also be a major driver of our collective social and economic well-being.”
The Boston Foundation’s Social Justice Ecology (SJE) Framework[6] supports the conditions for social justice to thrive in the Greater Boston area. It particularly affirms that commitment to equity is a key element of nonprofit effectiveness. Using race as an example, such commitment is described as “the condition where one’s race identity has no influence on how one fares in society. Working towards a race equity culture involves the ongoing elimination of policies, practices, attitudes, and cultural messages that reinforce differential outcomes by race inside and outside of an organization.”
The SJE Framework was conceived by the Boston Foundation in response to three challenges currently facing nonprofits:
- The Racial Leadership Gap: Over 85% of Greater Boston nonprofit staff and board leaders are White. This has not changed over 20 years although the demographics in the Greater Boston area have changed.
- Historically marginalized people are excluded from shaping decisions that affect their lives. Leadership of nonprofits and foundations often are not even in conversation with the people in the communities they are working to impact. Plus, BIPOC (Black, Indigenous, People of Color) are not adequately represented in the seats of power and leadership within the nonprofit sector.
- Nonprofits are not investing in the operational and leadership capacities needed to deliver on their missions in an effective and sustainable manner.
The Boston Foundation views effective, connected, and representative leadership as a critical prerequisite for equity to thrive as a focus for the nonprofit organizational culture. Leaders of social justice nonprofits balance many roles: holding the vision and strategy of the organization and at the same time managing staff, fundraising, and other day-to-day activities. They use their leadership skills to create bold programs that make significant change both locally and nationally. Their day-to-day focus on equity helps to build a strong social justice ethos into the nonprofit sector, strengthen the role of nonprofit organizations in the United States as sites of democratic practice, and promote nonprofit groups as partners in building a movement for progressive social change.
According to Kim and Kunreuther (2019) [7]of the Building Movement Project (BMP), “While most [Executive] Directors have a strong sense of how to be social justice leaders, there is a lack of integration, and balance, of both transformative leadership skills and concrete management skills.”(pg. 2) When the Executive Director is a person of color, BMP’s 2022 report[8], Trading Glass Ceilings for Glass Cliffs, finds that they are faced with other challenges as well. “Ascending to an executive position does not end a leader’s struggle with racism, and sometimes increases those struggles.” The report, which shines a spotlight on the experiences and challenges facing nonprofit Executives/CEOs of color, recommends that “Executive leaders of color need support, not more training.” (page 5) Further, “The number of leaders of color in Massachusetts nonprofit organizations is growing, but POC-led organizations are still undervalued and underfunded compared with white-led peers.”[9]
The W. K. Kellogg Foundation invests in developing and strengthening leadership networks that equip and enable leaders committed to racial equity to advance the health, education and economic well-being of children, families and communities. The WKKF Leadership Network with the Center for Creative Leadership is an innovative fellowship for local leaders to connect, grow and lead transformational change toward a more equitable society. The 18-month Fellowship brings together 80 inspiring leaders from across the United States and sovereign Tribes. The program offers hands-on leadership development, personalized coaching, peer networking, and practical experience. As a result, Fellows gain critical knowledge and skills for navigating today’s challenges and advancing racial equity and racial healing, community engagement and collective leadership.
Case Study #4: Statistical Analyses
(Descriptive and Inferential Statistics)
4a. Introduction
In April 2023, the networks of the National Council of Nonprofits conducted a nationwide survey to secure the latest, comprehensive information about the nonprofit workforce. Almost three out of four respondents (72.2%) said salary competition affects their ability to recruit and retain employees, followed by budget constraints/insufficient funds (66.3%). Additional causes for nonprofit workforce shortages reported by nonprofits were stress and burnout (50.2%) and challenges caused by government grants and contracts (20.6%).
Survey results concluded that, “When nonprofits cannot hire enough employees to provide vital services, the public suffers. Data from this survey and others show that along with increased demands for services, there are longer waiting lists, reduced services, and sometimes elimination of services. When any of those happen, the ripple effects cannot be ignored: communities lose access to food, shelter, mental health care, and other vital services on which people depend.”[10]
The primary theoretical explanation of lower wages in the nonprofit sector is that the difference may be considered as a labor donation by the employee to the nonprofit. Nonprofit workers attracted to the mission of their organization are willing to accept lower wages because they see their forgone wages as a donation toward a cause they support (Handy et al. 2007).[11] Although theoretical explanations for negative wage differentials are more persuasive, they are found to be negative or positive depending on the funding type of the nonprofit studied or employment status, such as CEO vs. program managers.
Case Study #4 explores how the Chief Executive Officer (CEO) and Program Manager’s salaries differ by a nonprofit’s funding type. Data from the Bayer Center for Nonprofit Management at Robert Morris University’s Rockwell School of Business were examined: the 2021 Wage and Benefit Survey of Southwestern Pennsylvania Nonprofit Organizations.[12] The survey reports the compensation and benefit practices in effect on October 1, 2020, as reported by 185 nonprofit organizations in Southwestern Pennsylvania.
The sample includes nonprofit organizations in southwestern Pennsylvania with varied funding sources, budget sizes, and subsector purposes from 2011 through 2021. The nonprofits volunteered to participate versus being selected through sampling methods.
A primary and critical goal of nonprofit organizations is to attract, recruit, and retain well-qualified professional and support staff. Competitive compensation, attractive benefit packages, and equitable policies support this goal.
The following presents both descriptive and inferential statistics on the sample.
4b. Descriptive Statistics
The CEO leads the participating nonprofit organizations and reports to the Board of Directors. CEO salaries of nonprofits were segmented into one of five funding types: Contributions from Individuals, Contributions from Foundations and Trusts, Government, Program Service Fees, and Sales/Investments.
Of the 185 nonprofits, the annual operating expenses of participating nonprofits range from under $50,000 to over $100,000,000 per year. The median annual operating expenses among survey participants is $1,200,000. Table 1 groups organizations based on their annual operating expenses.
Annual Expense Groups | # of Organizations | Average Annual Expenses | Average Base Pay for CEO |
Less than $500,000 | 50 | $288,378 | $72,607 |
$500,000-$999,999 | 29 | $676,066 | $90,698 |
$1,000,000-$2,499,999 | 50 | $1,583,795 | $115,629 |
$2,500,000-$9,999,999 | 36 | $4,875,014 | $146,238 |
$10,000,000 and More | 20 | $29,400,667 | $217,102 |
Total | 185 | $4,739,073 | $122,822 |
As shown, the financial size of the nonprofit affects the pay for executive-level management. Based on the sample data in Table 1, there is a positive association between the average base pay for CEOs and the average annual expenses of the respective nonprofit. As one increases, the other increases as well.
Gender and CEO Compensation: On average, male Executive Directors/CEOs earn significantly higher pay than do females. As shown in Table 1, the average annual base pay for all Executive Directors/CEOs in the sample is $122,822. For men, the average Executive Director/CEO pay is $129,218 per year; for women, the average Executive Director/CEO pay is $118,767 per year. There is not sufficient data to report average base pay for Non-Binary Executive Directors/CEOs.
Mission/Field of Service and CEO Compensation: In the nonprofit sector, the mission of the organization is its greatest concern. Salamon (1992)[13] suggests there are six characteristics of the nonprofit sector: 1. formally constituted; 2. private, nongovernmental; 3. not created to generate profit for those in charge of the organization; 4. self-governing; 5. voluntary in some way; and 6. designed to serve the public purpose.
In general, nonprofits tend to focus more on social outcomes related to the mission over financial metrics. As shown in Table 2, the mission or field of service of a nonprofit organization can affect CEO compensation.
Field of Service/Mission | # of Organizations | Median Salary | Average Base Pay for CEO |
Basic Material Need |
13 |
$120,000 | $134,167 |
Counseling-Behavioral Health & Wellness |
5 |
$142,376 | $136,982 |
Culture/Arts |
18 |
$90,034 | $110,321 |
Economic/Neighborhood Development |
10 |
$89,500 | $92,071 |
Education and Child Care Services |
13 |
$115,529 | $135,978 |
Employment & Economic Opportunity |
6 |
$86,705 | $89,046 |
Environment/Animal Welfare |
17 |
$86,000 | $119,874 |
Foundation/Philanthropy |
6 |
$117,456 | $138,397 |
Health and Health Education | 8 | $86,500 | $109,582 |
Legal/Advocacy | 6 | $125,000 | $134,945 |
Management/Technical Assistance | 7 | $113,000 | $125,991 |
Social Support | 31 | $115,000 | $135,474 |
Youth/Recreation | 5 | $85,000 | $87,255 |
Total | 145 |
Primary Source of Funding and CEO Compensation: Table 3 shows how CEO compensation may differ based on the nonprofit’s primary source of funding.
Primary Source of Funding | # of Organizations | Median Salary | Average Base Pay for CEO |
Contributions from Individuals |
19 |
$100,000 |
$113,485 |
Contributions from Foundations or Trusts |
53 |
$ 99,000 |
$104,929 |
Government |
43 |
$115,529 |
$132,192 |
Program Services Fees |
25 |
$138,500 |
$137,062 |
Revenue from Sales & Investments |
9 |
$147,420 |
$169,385 |
Total |
149 |
In this section, descriptive statistics were used to help describe, show or summarize data in a meaningful way such that, for example, patterns might emerge from the data. They are simply a way to describe our data. The next section will use inferential statistics to allow us to use these samples to generalize about the populations from which the samples were drawn. It is, therefore, important that the sample accurately represents the population. The process of achieving this is called sampling. Inferential statistics arise out of the fact that sampling naturally incurs sampling error and thus a sample is not expected to perfectly represent the population. The methods of inferential statistics are the estimation of parameter(s) and testing of statistical hypotheses.
4c. Inferential Statistics
The following statistical analyses performs hypotheses testing on four problems:
Hypothesis Testing Problem #1: To assess whether CEO salaries differ across different funding types.
Hypothesis Testing Problem #2: To assess whether Program Manager salaries differ across different funding types.
Hypothesis Testing Problem #3: To assess whether the availability of professional development benefits or no professional development benefits (binary) for CEOs differ across different funding types.
Hypothesis Testing Problem #4: To assess whether the availability of professional development benefits or no professional development benefits (binary) for Non-Executives differ across different funding types.
Social Justice Nonprofits Hypothesis Testing #1
The Problem: To assess whether CEO salaries differ across different funding types.
Hypotheses #1:
Null Hypothesis H0: There are no significant differences in CEO salaries across different funding types.
Alternative Hypothesis Ha: There are significant differences in CEO salaries across different funding types.
The hypothesis tests if CEO salaries differ across different funding types. CEO salaries were divided into five groups (Group 1-Foundation; Group 2-Government; Group 3-Individuals; Group 4-Program Fees, and Group 5-Sales).
Since the Levene’s Statistic is significant (<.001), the equal variance assumption was not met. However, ANOVA results showed significant differences between the groups (F(4,888)=12.5, <.001). Therefore, the null hypothesis is rejected.
To check to see where group differences specifically occurred, post-hoc comparisons were assessed using Dunnett’s T3. The test indicated that the mean CEO salary for Group 1-Foundations (M=$100,476.77, SD=$33,013.52) was significantly different from the mean CEO salary for Group 2-Government (M=$117,800.31, SD=$40,157.27); the mean CEO salary for Group 1-Foundations (M=$100,476.77, SD=$33,013.52) was significantly different from the mean CEO salary for Group 4-Program Fees (M=$120,226.28, SD=$48,108.25); the mean CEO salary for Group 1-Foundations (M=$100,476.77, SD=$33,013.52) was significantly different from the mean CEO salary for Group 5-Sales (M=$139,970.86, SD=$58,911.84); and the mean CEO salary for Group 3-Individuals (M=$106,900.50, SD=$46,120.04) was significantly different from the mean CEO salary for Group 5-Sales (M=$139,970.86, SD=$58,911.84). The mean differences were significant at the .05 level. However, no significant differences were detected between Group 1-Foundation and Group 3-Individuals; Group 2-Government and Group 3-Individuals; Group 2-Government and Group 4-Program Fees; Group 2-Government and Group 5-Sales; Group 3-Individuals and Group 4-Program Fees; and Group 4-Program Fees and Group 5-Sales.
Levine Statistic | df1 | df2 | Sig. | F, Sig. | ||
All Combined CEO Salaries | Based on Mean | 14.699 | 4 | 888 | <.001 | 12.500,<.001 |
Funding Types | Mean Difference | Sig. | 95% Confidence Interval Lower Bound |
95% Confidence Interval Upper Bound |
Foundation vs. Government | -$17,323.54 | <.001 |
-$26,128.65 |
-$8,518.44 |
Foundation vs. Program Fees | -$19,749.51 | <.001 |
-$32,327.21 |
-$7,171.81 |
Foundation vs. Sales | -$39,494.09 | <.001 |
-$64,601.79 |
-$14,386.39 |
Individuals vs. Sales | -$33,070.36 | .007 |
-$59,850.41 |
-$6,290.30 |
Hypothesis | Regression Weights | Beta Coefficient | R2 | F | p-value | Hypothesis Supported |
Ha | Funding Types→ CEO Salaries | $ 6262.06 | .029 | 26.872 | <.001 | Yes |
Social Justice Nonprofits Hypothesis Testing #2
The Problem: To assess whether Program Manager salaries differ across different funding types.
Hypotheses #2:
Null Hypothesis H0: There are no significant differences in Program Manager salaries across different funding types.
Alternative Hypothesis Ha: There are significant differences in Program Manager salaries across different funding types.
The hypothesis tests if Program Manager salaries differ across different funding types. Program Manager salaries were divided into five groups (Group 1-Foundation; Group 2-Government; Group 3-Individuals; Group 4-Program Fees; and Group 5-Sales ).
Since the Levene’s Statistic is significant (.014), the equal variance assumption was not met. However, ANOVA results showed significant differences between the groups (F(4,888)=4.354, <.002). Therefore, the null hypothesis is rejected.
To check to see where group differences specifically occurred, post-hoc comparisons were assessed using Dunnett’s T3. The test indicated that the mean Program Manager salaries for Group 1-Foundations (M=$45,561.13, SD=$9,530.95) was significantly different from the mean Program Manager salaries for Group 3-Individuals (M=$41,663.43, SD=$8,727.38); the mean Program Manager salaries for Group 2-Government (M=$45,483.98, SD=$10,437.31) was significantly different from the mean Program Manager salaries for Group 3-Individuals (M=$41,663.43, SD=$8,727.38); and the mean Program Manager salaries for Group 3-Individuals (M=$41,663.43, SD=$8,727.38) was significantly different from the mean Program Manager salaries for Group 5-Sales (M=$45,726.94, SD=$8,179.95). The mean differences were significant at the .05 level.
However, no significant differences were detected between Group 1-Foundation and Group 2-Government; Group 1-Foundation and Group 4-Program Fees; Group 1-Foundation and Group 5-Sales; Group 2-Government and Group 4-Program Fees; Group 2-Government and Group 5-Sales; Group 3-Individuals and Group 4-Program Fees; and Group 4-Program Fees and Group 5-Sales.
Levine Statistic | df1 | df2 | Sig. | F, Sig. | ||
All Combined CEO Salaries | Based on Mean | 3.132 | 4 | 888 | .014 | .4.354, .002 |
Funding Types | Mean Difference | Sig. | 95% Confidence Interval Lower Bound |
95% Confidence Interval Upper Bound |
Foundation vs. Individuals |
$ 3,897.70 |
.002 |
$ 1,015.48 |
$6,779.92 |
Government vs. Individuals |
$ 3,820.55 |
<.001 |
$ 1,113.68 |
$6,527.41 |
Individuals vs. Sales |
-$4,063.51 |
.043 |
-$8,057.61 |
-$ 69.40 |
In addition, a bivariate regression was performed to test if funding types carries a significant impact on Program Manager salaries. The dependent variable, All Combined Program Manager Salaries, was regressed on the predict or variable, Funding Types, to test the Hypothesis Ha. Funding Types did not significantly predict All Combined Program Manager Salaries, F (1, 891) = 2.590, p <.108, which indicates that Funding Types does not play a significant role in shaping Program Manager Salaries (b=-$437.41, p=.108). These results clearly indicate the negative effect of the independent variable , Funding Types, on Program Manager Salaries. Also, the R2 depicts that the model only explains .3% of the variance in Program Manager Salaries. Table 2.2 shows the summary of the findings.
Hypothesis | Regression Weights |
Beta Coefficient |
R2 |
F |
p-value |
Hypothesis Supported |
Ha | Funding Types→ Program Manager Salaries | -$437.41 | .003 | 2.590 | .108 |
No |
Social Justice Nonprofits Hypothesis Testing #3
The Problem: To assess whether the availability of professional development benefits or no benefits (binary) for CEOs differs across different funding types.
Hypotheses #3:
Null Hypothesis H0: There are no significant differences in available professional development benefits vs. no professional development benefits (binary) for CEOs across different funding types.
Alternative Hypothesis Ha: There are significant differences in the availability of professional development benefits vs. no professional development benefits (binary) for CEOs across different funding types.
The hypothesis tests if the presence of professional development benefits vs. no professional development benefits for CEOs differ across different funding types. The availability of CEOs’ professional development benefits was divided into two groups: No Benefits (Group 0) and Benefits (Group 1).
Since the Levene’s Statistic is significant (<.001), the equal variance assumption was not met. However, ANOVA results showed significant differences between the groups (F(4,554)=5.144, <.001). Therefore, the null hypothesis is rejected.
To check to see where group differences specifically occurred, post-hoc comparisons were assessed using Dunnett’s T3. The test indicated that the availability of professional development benefits vs. no benefits was significantly different for Group 1-Foundations (M=.84, SD=.371) and Group 5-Sales (M=.56, SD=.502), and Group 4-Program Fees (M=.84, SD=.422) and Group 5-Sales (M=.56, SD=.502). The mean differences were significant at the .05 level.
However, no significant differences were detected between Group 1-Foundation and Group 2-Government; Group 1-Foundation and Group 3-Individuals; Group 1-Foundation and Group 4-Program Fees; Group 2-Government and Group 3-Individuals; Group 2-Government and Group 4-Program Fees; Group 2-Government and Group 5-Sales; Group 3-Individuals and Group 4-Program Fees; and Group 3-Individuals and Group 5-Sales.
Levine Statistic | df1 | df2 | Sig. | F, Sig. | ||
Benefits vs. Non-Benefits | Based on Mean | 14.838 | 4 | 554 | <.001 | 5.144,<.001 |
Funding Types | Mean Difference | Sig. | 95% Confidence Interval Lower Bound |
95% Confidence Interval |
Foundation vs. Sales | .273 | .026 | .02 | .52 |
Program Fees vs. Sales | .274 | .030 |
.02 |
.53 |
In addition, a bivariate regression was performed to test if Funding Types carries a significant impact on the presence of professional development benefits vs. no professional development benefits for CEOs. The dependent variable, Benefits CEO Binary, was regressed on the predicting variable, Funding Types, to test the Hypothesis Ha. Funding Types significantly predicted the availability of professional development benefits vs. no benefits for CEOs, F (1, 557) = 5.141, p =.024, which indicates that Funding Types can play a significant role in determining the availability of professional development benefits vs. non-benefits for CEOs (b=-.032, p=.024). These results clearly direct the positive effect of the independent variable, Funding Types. However, the R2 depicts that the model only explains .9% of the variance in available professional development benefits vs. no professional development benefits for CEOs.Table 3.2 shows the summary of the findings.
Hypothesis |
Regression Weights |
Beta Coefficient |
R2 |
F |
p-value |
Hypothesis Supported |
Ha | Funding Types→ Professional Development Benefits vs. No Benefits for CEOs |
-.032 |
.009 |
5.141 |
.024 |
Yes |
Social Justice Nonprofits Hypothesis Testing #4
The Problem: To assess whether available professional development benefits or no professional development benefits (binary) for Non-Executives differ across different funding types in 2021.
Hypotheses:
Null Hypothesis H0: There are no significant differences in available professional development benefits vs. no professional development benefits (binary) for Non-Executives across different funding types in 2021.
Alternative Hypothesis Ha: There are significant differences in available professional development benefits vs. no professional development benefits (binary) for Non-Executives across different funding types in 2021.
The hypothesis tests whether available professional development benefits vs. no professional development benefits for Non-Executives differ across different funding types. Non-Executive professional development benefits were divided into two groups: Benefits (Group 1) and No Benefits (Group 2).
Since the Levene’s Statistic is significant (.000), the equal variance assumption was not met. However, ANOVA results showed significant differences between the groups (F(4,9402)=869,.000). Therefore, the null hypothesis is rejected.
To check to see where group differences specifically occurred, post-hoc comparisons were assessed using Dunnett’s T3. The test indicated that the availability of professional development benefits vs. no benefits was significantly different for a majority of the groups: Group 1-Foundation (M=.78, SD=.413) and Group 2-Government (M=.92, SD=.274); Group 1-Foundation (M=.78, SD=.413) and Group 3-Individuals (M=.94, SD=.244); Group 1-Foundation (M=.78, SD=.413) and Group 4-Program Fee (M=.40, SD=.490); Group 1-Foundation (M=.78, SD=.413) and Group 5-Sales (M=.52, SD=.50); Group 2-Government (M=.92, SD=.274) and Group 4-Program Fee (M=.40, SD=.490); Group 2-Government (M=.92, SD=.274) and Group 5-Sales (M=.52, SD=.50); Group 3-Individuals (M=.94, SD=.244) and Group 4-Program Fee (M=.40, SD=.490) Group 3-Individuals (M=.94, SD=.244) and Group 5-Sales (M=.52, SD=.50); and Group 4-Program Fee (M=.40, SD=.490) and Group 5-Sales (M=.52, SD=.50). The mean differences were significant at the .05 level.
However, no significant differences were detected between Group 2-Government and Group 3-Individuals.
Levine Statistic | df1 | df2 | Sig. | F, Sig. | ||
Benefits vs. Non-Benefits | Based on Mean |
1655.84 |
4 |
9402 |
.000 |
869, .000 |
Funding Types | Mean Difference | Sig. | 95% Confidence Interval Lower Bound |
95% Confidence Interval |
Foundations vs. Government | -.137 | <.001 | -.18 | -.10 |
Foundations vs. Individuals | -.155 | .000 | -.20 | -.11 |
Foundations vs. Program Fee | .380 | <.001 | .33 | .43 |
Foundations vs. Sales | .258 | .000 | .20 | .32 |
Government vs. Program Fee | .516 | .000 | .49 | .55 |
Government vs. Sales | .395 | <.001 | .34 | .45 |
Individuals vs. Program Fee | .534 | .000 | .50 | .57 |
Individuals vs. Sales | .413 | <.001 | .36 | .47 |
Program Fee vs. Sales | -.121 | .000 | -.18 | -.06 |
In addition, a bivariate regression was performed to test if funding types carries a significant impact on the availability of professional development benefits vs. no benefits for Non-Executives. The dependent variable, Benefits Non-Executives Binary, was regressed on the predictor variable, Funding Types, to test the Hypothesis Ha. Funding Types significantly predicted the availability of professional development benefits vs. no professional development benefits for Non-Executives, F (1, 9405) = 1851.31, p =.000, which indicates that Funding Types can play a significant role in determining the availability of professional development benefits vs. no professional development benefits for Non-Executives (b=-.130, p=.000). These results clearly direct the positive effect of the independent variable, Funding Types. However, the R2 depicts that the model only explains 16.4% of the variance in the availability of professional development benefits vs. no professional development benefits for Non-Executives. Table 4.2 shows the summary of the findings.
Hypothesis |
Regression Weights |
Beta Coefficient |
R2 |
F |
p-value |
Hypothesis Supported |
Ha | Funding Types→ Professional Development Benefits vs. No Professional Development Benefits for Non-Executives |
-.150 |
.164 |
1851.312 |
.000 |
Yes |
This concludes our journey to becoming equity-minded through the lens of statistics. I hope that during your journey, you enjoyed periodic pauses and had mindful quietude as you reflected on what you learned. There is so much to learn about statistics itself and how it blends so well into understanding social justice and equity issues.
In closing, I want to thank you for your steadfastness and courage to travel this journey with me. I am always available for questions and comments. You can email me at yanthony @ framingham.edu. I would love to hear from you!
- McDonald RE, Weerawardena J, Madhavaram S, Mort GS (2015). From “Virtuous” to “Pragmatic” Pursuit of Social Mission: A Sustainability-Based Typology of Nonprofit Organizations and Corresponding Strategies. Management Research Review, September 21. ↵
- Hammack DC (2002). Nonprofit Organizations in American history: Research Opportunities and Sources. American Behavioral Scientist, 45 (11), 16381674. ↵
- Founded in 1915, the Boston Foundation is one of the largest community foundations in the nation, with net assets of some $1.7 billion. The Foundation is a partner in philanthropy, with nearly 1,000 separate charitable funds established by donors either for the general benefit of the community or for special purposes. ↵
- Boston Foundation (2024). Quote by Orlando Watkins, Vice President and Chief Program Officer. Retrieved from https://www.tbf.org/nonprofits on July 2, 2024. ↵
- Turner A (2018). The Business Case for Racial Equity: A Strategy for Growth. W. K. Kellogg Foundation: East Battle Creek, Michigan ↵
- Boston Foundation (2024). Retrieved from https://www.tbf.org/what-we-do/strategic-focus-areas/social-justice-ecology?q&sortBy=date&sortOrder=desc&page=1on July 2, 2024. ↵
- Kim H & Kunreuther F (2019). Vision for Change: A New Wave of Social Justice Leadership. The Building Movement Project. ↵
- Building Movement Project (2022). Trading Glass Ceilings for Glass Cliffs: A Race to Lead Report on Non-Profit Executives of Color. ↵
- Building Movement Project (2020). The Burden of Bias in the Bay State: The Non-Profit Racial Leadership Gap in Massachusetts. ↵
- National Council of Nonprofits (2024). The Nonprofit Workforce Shortage Crisis. Retrieved https://www.councilofnonprofits.org/nonprofit-workforce-shortage-crisis on July 17, 2024. ↵
- Handy F, Mook L, Ginieniewicz J, & Quarter J (2007). The Moral High Ground: Differentials Among Executive Directors of Canadian Nonprofits. Retrieved from http://repository.upenn.edu/spp_papers/101 on July 17, 2024. ↵
- The Bayer Center provides free downloadable archival reports from 2008 through 2021, each including 150 to 185 participating non-profits representing over 11,000 employees bi-annually. ↵
- Salamon LM (1992). America’s Nonprofit Sector: A Primer. New York: The Foundation Center. ↵