Why So Many Measures of Nonprofit Financial Performance? Analyzing and Improving the Use of Financial Measures in Nonprofit Research

Abstract

Financial measures provide an empirical basis from which nonprofit researchers and practicing managers can approximate organizational capacity, financial health, and performance. These measures are used in nonprofit research to predict organizational activities and funding opportunities. Yet, little empirical evidence exists to tell us what these measures assess and whether they capture underlying concepts in the way we assume. Using Internal Revenue Service (IRS) Form 990 data, this article explores the following research question: Can accounting measures be organized into theoretically intuitive and empirically defensible constructs? To answer this question, a literature review of nonprofit financial health studies and textbooks was conducted, and dimension reduction techniques were employed. The findings suggest that the answer to the research question is not as simple as expected, and we should exercise more caution in how we use financial measures in nonprofit research.

Keywords

nonprofit financial performance accounting ratios financial measures

Introduction

Scholars generally agree that nonprofit financial performance is evaluated through a careful examination of multiple factors, such as liquidity, solvency, margin, and profitability. These constructs illustrate how much cash a nonprofit has on hand, how much debt the nonprofit has accrued, how efficient the nonprofit is in the use of its resources, and how stable the nonprofit is over time. This financial knowledge empowers nonprofit managers to create operating budgets, monitor organizational finances, measure progress toward predetermined financial targets, and establish financial reserves sufficient to sustain future operations.

To capture liquidity, solvency, margin, and profitability, nonprofit managers and researchers use various indicators (e.g., days of cash on hand as a liquidity indicator), and nonprofit financial management textbooks recommend multiple measures for each construct. For example, Weikart, Chen, and Sermier (2013) offer three liquidity measures: the current ratio, the working capital ratio, and the quick ratio. McLaughlin (2009) and Coe (2011) offer similar liquidity measures and add to the list days’ receivables and days of cash on hand, respectively. Zietlow, Hankin, and Seidner (2007) expand the list of liquidity measures and add the cash ratio, the cash reserve ratio, and the asset ratio. According to Weikart et al. (2013), “Each of these ratios can be used to determine the extent to which a nonprofit need not worry about its cash flow” (p. 136). The proliferation of measures raises the questions of why there are so many measures offered to capture the same construct, and whether for purposes of practice or analysis each liquidity measure is as suitable as the next.

The basis for the numerous and varied measures is not fully rationalized in the literature, but the implicit assumption in offering multiple measures is that these constructs (liquidity, solvency, margin, and profitability), and financial performance more generally, are complex and cannot be captured by single measures. McLaughlin (2009) hints at the need for multiple measures when he states that the current ratio “is actually a rather crude measure” (p. 76). He goes on to say that “groups needing a more fine-tuned measure of liquidity” (McLaughlin, 2009, p. 76) should try the quick ratio. Zietlow et al. (2007) elaborate on the purpose for multiple liquidity measures when they propose that each ratio “gives a slightly different perspective on the spendable funds of the organization” (p. 213).

If each ratio provides a different perspective, then multiple measures should be evaluated in concert with one another to garner a full picture of each construct. However, this nuance is lost on nonprofit researchers who consistently fail to use these measures in conjunction with one another, and instead employ the measures singularly and interchangeably, implicitly assuming that one measure of liquidity is as good as the next. Two problems arise from this practice. First, construct validity is undermined when complex constructs, such as liquidity, solvency, margin, and profitability, are reduced to a single measure. If these constructs are multidimensional, then choosing a single measure as an indicator presents only a partial picture. Second, because scholars choose, calculate, and employ these measures differently, nonprofit researchers are failing to test similar constructs, and thus are not contributing to cumulative research.

This article examines empirically whether financial measures converge and to what degree these measures are indicators of theoretically intuitive underlying constructs. I present the findings from these analyses in two parts: In the first part of the article I briefly discuss nonprofit financial performance; map our conceptual understanding of liquidity, solvency, margin, and profitability; and offer an examination of the measures used in the academic literature to capture these constructs. Then, I investigate the dimensionality of the constructs and present the results of dimension reduction analyses performed on several measures. The findings demonstrate the disjunction between the conception of these constructs in the literature and the methods nonprofit researchers employ to capture them empirically. Because of this counterintuitive finding, I turn to another analysis and consider the field more generally.

Thus, in the second part of the article I shift the focus from the four constructs discussed earlier to a broader analysis of financial measures used in nonprofit research. After a brief summary of the literature that uncovers 70 financial measures, I present the results of several statistical tests aimed at deriving a data-driven classification of the measures. The findings illustrate the uniqueness of these measures and call attention to the need for nonprofit researchers to exert greater discrimination in the selection and calculation of financial measures.

Accounting Concepts and Corresponding Measures in Studies of Financial Health

Nonprofit financial management textbooks suggest monitoring various measures considered indicators of liquidity, solvency, margin, and profitability, to assess nonprofit financial performance (Figure 1). This conception is mirrored in nonprofit research, where multiple studies apply financial ratios as representative indicators of the four accounting constructs. However, while the textbooks imply that the financial measures are individual and complementary, the academic literature habitually treats the measures as redundant and interchangeable. This part of the article reviews nonprofit financial health studies to identify the most commonly cited indicators of the four constructs, then examines the relationship among the indicators, and between the indicators and underlying accounting concepts.

Figure 1.

Financial performance: Constructs and sample of related indicators.

Tuckman and Chang (1991) developed measures for identifying vulnerable nonprofits and in so doing began a line of inquiry that focused attention on the importance of nonprofit financial health. Subsequent studies (Greenlee & Trussel, 2000; Hager, 2001; Hodge & Piccolo, 2005; Keating, Fischer, Gordon, & Greenlee, 2005; Trussel, 2002) built on their work and incorporated elements of models used in for-profit bankruptcy prediction literature (Altman, 1968; Ohlson, 1980) to produce probabilities of nonprofit financial vulnerability. The list of predictors generally includes three categories of variables—accounting variables (e.g., net income divided by total assets), revenue variables (e.g., revenue diversification), and efficiency variables (e.g., administrative expenses divided by total expenses)—with accounting variables representing a large majority.

As illustrated above, accounting variables are intended to represent broader concepts, such as solvency and profitability. Like the measures total assets and total expenses used to capture organizational size in the literature (Ashley & Faulk, 2010; McGinnis Johnson, 2013), accounting measures such as net income divided by total revenue and working capital divided by total assets are employed to capture accounting constructs (Keating et al., 2005). Four common accounting constructs are frequently referenced in the literature: liquidity, solvency, margin, and profitability. For each construct, common financial measures are assumed to measure the construct.

Nonprofit Assets

Unlike for-profit organizations, nonprofit organizations are prohibited due to the nondistribution constraint from allocating profits to officers of the organization or stakeholders, and any profit accumulation must remain within the nonprofit (Hansmann, 1980). In the nonprofit sector, “Net assets do not represent cash balances of the organization; rather, net assets represent a claim of ownership on assets owned by the organization. Net assets represent those assets reinvested within the NPO rather than used up” (Calabrese, 2011a, p. 3).

Generally Accepted Accounting Principles (GAAP) require nonprofits to classify total net assets into three categories: unrestricted net assets, temporarily restricted net assets, and permanently restricted net assets (Calabrese, 2011a; Financial Accounting Standards Board [FASB], 1993). However, with the exception of recent studies on endowments and capital structures (Bowman, 2011a; Bowman, Tuckman, & Young, 2012; Calabrese, 2011a, 2011b), net assets have been lumped together as a singular category. Unrestricted net assets more closely approximate the actual amount of equity a nonprofit organization may spend in times of financial distress or use as collateral for borrowing (Calabrese, 2011b). According to Calabrese (2011a), existing theory suggests that net assets are desirable not only for long-term financing (labeled solvency here) but also for short-term financing (labeled liquidity here). The reasoning is because total net assets represent the nonprofit organization’s equity regardless of donor-imposed restrictions, whereas unrestricted net assets represent the cash balances that can be used by nonprofit managers to reinvest in the organization to overcome short-term financial shocks.

Solvency

Solvency is measured in nonprofit financial vulnerability literature in three ways: total net assets divided by total revenue (Tuckman & Chang, 1991), total net assets divided by total assets (Bowman, 2011a; Keating et al., 2005), and total assets less total liabilities (analogous to Keating et al.’s, 2005, insolvency risk variable). Given the consistency with which these measures are labeled solvency in nonprofit research, I test whether they capture an underlying solvency construct.

Liquidity

Liquidity “consists of cash or financial resources without donor restrictions, which can be efficiently converted into cash quickly” (Bowman, 2011b, p. 179) and is the cash available to continue the organization’s operations in the short run. Two common indicators are used to assess liquidity. The first is working capital divided by total assets (Greenlee & Tuckman, 2007; Keating et al., 2005); working capital refers to the difference between current assets and both current liabilities and temporarily restricted net assets. The second is months of spending (Bowman, 2011a, 2011b), which is “the number of months an organization could survive after losing all current income and maintaining its spending on operations at a constant level” (Bowman, 2011b, p. 179). Months of spending incorporates unrestricted net assets in the numerator, so that only immediately available funds without donor restriction are considered. I test these liquidity measures to see if they represent an underlying liquidity construct.

Profitability

Profitability shows how much the organization nets after accounting for expenses and generally conveys the long-term sustainability of the nonprofit organization. Two variables representing profitability are common to accounting-based nonprofit research: net income divided by total assets (Bowman, 2002, 2011a; Keating et al., 2005) and revenues less expenses (Bowman et al., 2012; Keating et al., 2005). I examine whether these financial measures, consistent with the assumptions that they represent a firm’s profitability, converge and capture an underlying profitability construct in the data.

Margin

Margin refers to the efficiency of earnings and represents a nonprofit’s short-term sustainability. The commonly used measure for margin is net income divided by total revenue (Chabotar, 1989; Coe, 2011; Greenlee & Trussel, 2000; Greenlee & Tuckman, 2007; Hager, 2001; Hodge & Piccolo, 2005; Keating & Frumkin, 2001; Keating et al., 2005; Ryan & Irvine, 2012; Trussel, Greenlee, & Brady, 2002; Tuckman & Chang, 1991). Bowman (2011a) proposes the measure “markup” and contends that it is superior to margin for use in the nonprofit sector. Markup is “an organization’s annual surplus expressed as a percentage of spending on operations” (Bowman, 2011b, p. 179). Analogous to the assumptions for the solvency, liquidity, and profitability constructs, I test whether the margin measures converge to represent an underlying margin construct.

The habitual use of individual measures to represent constructs such as liquidity, solvency, margin, and profitability in the nonprofit management literature implies that these constructs are unidimensional. Based on this literature review, I test the underlying dimensions represented by these measures. If the unidimensionality assumption is warranted, then the nine measures should coalesce to form the four accounting constructs discussed above.

Data

Data for this research are the digitized data files from the National Center for Charitable Statistics (NCCS), which contain Internal Revenue Service (IRS) Form 990 information for all filing 501(c)(3) organizations from 1998 to 2003.¹ The data were cleaned following the recommendations of Bowman et al. (2012) and Calabrese (2011b). The steps taken and the proportion of the sample meeting each criterion are enumerated in Table 1. To overcome dependencies in the data (i.e., organizations with multiple years of 990 data in the digitized data files), only 2003 data are selected, resulting in a sample size of 100,788.

Table 1.

Proportion of Sample Meeting Conditions for Selection.

Sample selection	n	% yes
Total sample size	1,388,480	100
Public charity	1,376,137	99.1
Individual return	1,385,597	99.8
Long form	1,104,666	79.6
Accrual accounting	699,408	50.4
Statement of Financial Accounting Standards 117	1,084,112	78.1
Active: Total expenses > depreciation + interest paid	1,378,447	99.3
Active: Total assets > 0	1,332,137	95.9
No obvious data errors (negative liabilities, negative expenses, no program expenses)	1,055,536	76.0
Federal Information Processing Standards code	1,383,598	99.6
U.S. states and District of Columbia	1,387,288	99.9
All	549,688	39.6
Fiscal Year 2003 subset	100,788	—

Dimension Reduction—Nine Common Accounting Measures

If liquidity, solvency, margin, and profitability are unidimensional constructs, then the nine accounting measures should each load exclusively on the one dimension for which the literature proposes that it is an indicator. In this section, several factor analyses are performed to examine the relationship of these variables to one another and to discern the underlying dimensions they represent.

Factor Analysis

An exploratory factor analysis (EFA), using principal component extraction and a Varimax rotation, was conducted on the cleaned data set consisting of the accounting measures to determine how well the proposed constructs fit the unidimensional assumption. The proposed constructs and the nine accounting measures are summarized in the first column in Table 2. Inferring from the use of these measures in the academic literature, we should expect to see the nine measures load unidimensionally on their respective components: The first three measures in Table 2 should load on Component 1, the next two measures on Component 2, the following two measures on Component 3, and the final two measures on Component 4.

Table 2.

Component Matrix—All Subsectors.

Constructs and corresponding accounting measures	Component
Constructs and corresponding accounting measures	1	2	3	4	5
Solvency
Total net assets / total revenue	−.945	.002	.004	.001	.000
Total net assets / total assets	−.001	.000	.004	.744	.017
Total assets − total liabilities	−.007	.000	.841	−.002	−.001
Profitability
Net income / total assets	.000	.000	.000	.002	.997
Total revenue − total expenses	.010	.001	.841	.003	.001
Liquidity
Working capital / total assets	.001	.000	−.003	.744	−.015
Months of spending	−.002	.893	.000	.000	.000
Margin
Net income / total revenue	.945	.002	.007	.001	.000
Markup	.001	.893	.000	.000	.000

Note. Extraction method: principal component analysis. Rotation method: Varimax with Kaiser Normalization.

The results of the principal component analysis in Table 2 diverge from this expectation, however. For the nine measures, the principal component analysis returned five components with an eigenvalue greater than 1.0 (the conventional criterion for determining dimensionality), rather than the expected four components. Furthermore, the variables do not load as the literature suggests. The results demonstrate that none of the measures converge to form underlying dimensions as assumed. For example, the first three items in Table 2 should load together (as stated above), yet they load on three separate components.

The items loading on Component 1 both have total revenues in the denominator, so it is possible revenues are driving this dimension. But one would expect income and assets to load similarly, and net income divided by total revenue loads positively on Component 1, while total net assets divided by total revenue loads negatively. Similar to Component 1, Component 4 items both have total assets in the denominator, perhaps overwhelming the long-term focus of total net assets and the short-term focus of working capital. Component 3 items, total assets minus total liabilities and total revenue minus total expenses, are both long-term measures; and Component 2 items, months of spending and markup, are both short-term measures. It is unclear why net income divided by total assets loads independently on Component 5 and does not load with the other measures that have total assets in the denominator (Component 4).

Subsector Analysis

Given the likelihood that subsectors with typically large endowments (e.g., higher education, hospitals) fundamentally differ from those subsectors with smaller endowments (e.g., human services, religion), a principal component analysis was conducted on each subsector.² Even after the subsector effect was removed, the variables continued to load similarly—which is to say not as the literature suggests. The factor analytic results conducted on each subsector mirror those in Table 2. While the results across subsectors show some small variation, subsector has a negligible effect on the dimensionality among the accounting ratios.

Index Construction

To test the feasibility of creating an index for each of the accounting constructs (liquidity, solvency, margin, and profitability), a reliability analysis was conducted. The items composing each construct had extremely low Cronbach’s alpha scores and intraclass correlations (all scores were at or near 0), implying that the indicators proposed in the literature should not be combined to create indexes.

Correlation Analysis

To examine the relationships among the nine measures further, I conducted correlation analysis and tested for multicollinearity. The correlation matrix shows moderate correlations among the variables with only one correlation above .7 and many at or near 0 (Table 3). These modest correlations suggest that each of the financial ratios is measuring different things, despite similarities in their measurement (i.e., same numerator or denominator).³ This finding challenges the literature-driven conception regarding the interchangeability of measures.

Table 3.

Correlation Matrix—All Subsectors.

		1	2	3	4	5	6	7	8	9
1	Total net assets / total revenue	1.000
2	Total net assets / total assets	.001	1.000
3	Total assets − total liabilities	.005	.002	1.000
4	Net income / total assets	.000	.004	.000	1.000
5	Total revenue − total expenses	−.003	.002	.413	.001	1.000
6	Working capital / total assets	.000	.108	−.002	.000	.001	1.000
7	Months of spending	.004	.000	.000	.000	.000	.000	1.000
8	Net income / total revenue	−.786	.000	.000	.000	.014	.001	.000	1.000
9	Markup	.000	.000	.000	.000	.001	.000	.595	.001	1.000

Implications

By using single measures to capture particular underlying constructs, nonprofit researchers are portraying accounting constructs unidimensionally. However, these analyses demonstrate that the assumptions regarding the unidimensionality of the four accounting constructs lack empirical support. It is likely that these measures failed to converge into neatly prescribed constructs because the relationship among these measures is complex and the constructs they represent are multidimensional. This interpretation would support the guidance offered in nonprofit financial management textbooks that indicators should be evaluated in conjunction with one another.

In sum, these findings suggest that nonprofit researchers should exert more caution in the discussion and application of these accounting constructs. Construct validity is undermined when researchers employ measures singularly to represent multidimensional concepts. These findings also raise broader concerns regarding the use of financial measures in nonprofit research. To explore the implications stemming from these findings, I broaden the scope of the analysis and conduct additional empirical examination. Hence, the second part of this article extends the analysis from the four accounting constructs to examine the measurement and use of financial measures in nonprofit literature more generally. Results from the second part of this article demonstrate that in addition to viewing accounting constructs multidimensionally, researchers should avoid using financial measures interchangeably.

Financial Measures in Nonprofit Literature

Researchers employ financial measures to stratify their sample (Dumont, 2013), measure financial performance (Newton, 2013) and organizational performance (Brown, 2005), and predict organizational activities and funding opportunities. Researchers have used financial measures to predict grantee selection (McGinnis Johnson, 2013), grant amount (Ashley & Faulk, 2010), advocacy activities (Garrow & Hasenfeld, 2014), innovations (Jaskyte, 2013), charitable donations (Tinkelman & Mankaney, 2007; Trussel & Parsons, 2008), individual donations through social media (W. Saxton, 2013), fundraising performance (Scherhag & Boenigk, 2013), executive compensation (Sedatole, Swaney, Yetman, & Yetman, 2013), governance quality (Newton, 2013), internal accountability (Ryan & Irvine, 2012), adoption of web-based accountability practices (G. Saxton & Guo, 2011), compliance with financial reporting standards (Verbruggen, Chistiaens, & Milis, 2011), and the propensity and intensity of nonprofit collaboration with local government (McIndoe, 2013).

Financial ratios are used as quantitative surrogates for hard to measure concepts. For instance, organizational size is a common predictor or control variable in the literature, and measures of organizational size are used as proxies for capacity. However, few authors explain why they choose to operationalize size as they do. The studies cited above variably estimate organizational size as total revenue, total expenses, or total assets—with some authors adding log or natural log transformation. Researchers make two assumptions by using organizational size as a proxy for capacity and employing various measures to estimate organizational size: First, organizational size empirically captures capacity (financial or otherwise), and second, various measures of organizational size are empirically equivalent.

The use of financial measures to represent concepts is prevalent, but little attention is given to what these measures capture and the relationship among them. The result is a body of nonprofit literature that is overrun with financial measures operationalized in a variety of ways that are intended to represent a few concepts. This section reviews the use of financial ratios in the nonprofit finance literature and seeks to organize these ratios into theoretically intuitive and empirically defensible categories.

Literature Review

The findings from the factor analysis conducted on the nine commonly used accounting measures in the first part of this article expose two problems with the traditional conception of accounting concepts and by extension nonprofit researchers’ and managers’ use of financial measures. First, the assumption that accounting measures are singular indicators of underlying concepts is accepted a priori knowledge, but this assumption is not empirically supported. The second problem is the vagueness with which the literature has defined these accounting constructs and the imprecision with which various accounting measures have been employed as proxies for these constructs. This imprecision has led to studies that refer to solvency or profitability singularly, while measuring them in a multitude of ways. For example, solvency is variably measured as current assets divided by current liabilities (Zietlow, 2012), change in unrestricted net assets, change in total net assets, net operating revenue (Bowman, 2011b), total debt service divided by total revenues (Finkler, Purtell, Calabrese, & Smith, 2013), and total liabilities divided by total assets (Keating et al., 2005; Ohlson, 1980; Weikart et al., 2013).

In the following paragraphs, I present a review of financial measures in nonprofit literature and look for areas of convergence among the measures that may lead to a better conceptual understanding of what these measures represent. A comprehensive literature review of dozens of nonprofit financial health articles and six nonprofit financial management textbooks published over the last two decades resulted in the identification of 154 financial measures. Once the 154 measures were standardized (i.e., put in a common language), 70 unique measures remained.

An example of standardization is illustrated in the calculation of the measure, days cash on hand. Some authors only incorporated cash in the numerator, whereas others included cash equivalents such as marketable securities. Similarly, while some formulas used total expenses in the denominator, others looked at operating expenses and accounted for depreciation. Thus, the formula for days cash on hand was standardized as follows:

\frac{cash + cash equivalents}{(total operating expenses - depreciation) / 365} .

On the basis of the literature review, I attempted to organize the 70 measures into theoretically intuitive constructs. Relying predominantly on the authors’ descriptions, I placed the measures into categories. For example, working capital divided by total assets was classified as a liquidity ratio, net income divided by total revenue was designated a margin ratio, and so on. However, it quickly became evident that in addition to several measures falling into each theoretical construct, some of the same measures were variously attributed to different constructs. In other words, the analysis not only suggested multiple solvency measures as expected, but it also showed that some of the same measures were labeled solvency by one author and liquidity by another. For instance, Zietlow (2012) categorizes the measure current assets divided by current liabilities as solvency, while many others (Coe, 2011; Greenlee & Tuckman, 2007; Keating & Frumkin, 2001; McLaughlin, 2009; Ryan & Irvine, 2012; Weikart et al., 2013; Zietlow et al., 2007) label the measure liquidity.

Two additional examples illustrate the point further. First, net income divided by total revenue is variably referred to as margin (Coe, 2011; Hager, 2001; Hodge & Piccolo, 2005; Keating et al., 2005; Trussel et al., 2002; Tuckman & Chang, 1991), operating margin (Greenlee & Trussel, 2000), net operating results (Chabotar, 1989), stability (Trussel & Parsons, 2008), sustainability (Ryan & Irvine, 2012), and profitability (Keating & Frumkin, 2001; McLaughlin, 2009; Zeller, Stanko, & Cleverley, 1996). Second, total liabilities divided by total assets is variably referred to as solvency (Keating et al., 2005; Ohlson, 1980), long-term solvency (Weikart et al., 2013), equity (Hodge & Piccolo, 2005; Trussel, 2002; Trussel et al., 2002), flexibility (Carroll & Stater, 2009), leverage (Bacon, 1992; Bowman, 2002; Calabrese, 2011b; Keating & Frumkin, 2001), debt (Ryan & Irvine, 2012), and uncategorized (Ashley & Faulk, 2010; Greenlee & Tuckman, 2007; Zietlow et al., 2007).

This variability, both in the measurement of the financial measures and in the theoretical constructs the measures are purported to represent, illustrates the need for clarification. Hence, the rest of this article is dedicated to pursuing conceptual clarity. In the following section, the 70 financial measures identified in the literature review are explored via factor analysis.

Dimension Reduction—All Accounting Measures

Given the shortcomings of the traditional conceptualization of financial measures, this section applies inductive reasoning to explore the viability of establishing underlying constructs from a purely data-driven perspective. Multiple exploratory factor analyses were performed to uncover latent constructs, and once again, the results affirm the disjunction between our conceptual understanding of the accounting measures and empirical data.

Factor Analysis—All 70 Variables

A factor analysis with principal component extraction and a Varimax rotation was performed on all 70 variables (listed in the Appendix). The output yielded 22 components with an eigenvalue greater than 1.0. The number of components precludes any meaningful interpretation; therefore, attempts were made to reduce the number of components using decision criteria other than the standard Kaiser Criterion. An examination of the scree plot suggested, conservatively, 22 components, and if one were bold in interpreting the “elbow,” 21 components. Cattell’s scree plot test introduces researcher subjectivity; because it did not improve interpretation in this instance, additional criteria were explored. Several studies suggest that the Kaiser criterion is among the least accurate methods for selecting the number of components to retain, with studies suggesting that the Kaiser criterion consistently overestimates the number of major components (Costello & Osborne, 2005; Lance, Butts, & Michels, 2006; Rummel, 1970; Velicer & Jackson, 1990). These same studies universally agree that Horn’s parallel analysis (PA) criterion is much more promising and yields more accurate numbers of components. PA is a Monte Carlo–based simulation method that selects components after comparing the factor analyzed data with a matrix of data sets with random numbers representing the same number of cases and variables (Garson, 2012). The components greater than randomly generated components, based on a specified threshold, are chosen. To reduce the number of components and ensure selectivity in the approximation of components, 100 parallel data sets were run with a 99% significance level. Despite this rigorous analysis, the PA criterion yielded 21 components.

The high number of components coupled with extensive cross-loadings, despite a Varimax rotation (which, as an orthogonal rotation, should increase interpretability and reduce cross-loadings by restricting correlation between the components), frustrates interpretation. In an effort to yield more interpretable and theoretically intuitive results, three decision criteria were applied to the data to reduce the number of components. The criteria were applied cumulatively. Those steps are detailed in the following sections.

Decision Criterion 1—Correlation

Factor analysis uncovers the latent constructs or dimensions underlying a set of variables by reducing the attribute space among them. Therefore, while not mathematically required, moderate to high correlations among the variables are assumed; otherwise, the analysis will yield as many components as variables. This assumption is why many researchers require correlations greater than .30 to include a variable in the analysis (Garson, 2012).

The first decision criterion applied to the data was the requirement that each variable must have a correlation of .30 or greater with at least one other variable in the analysis. If a variable did not have a correlation greater than or equal to .30, it was removed. This decision criterion led to the removal of seven variables.

A principal component analysis with a Varimax rotation was performed on the remaining 63 variables, and once again the results were not interpretable. The Kaiser criterion yielded 21 components, the scree plot suggested 20 components, and a PA (on 100 parallel data sets with a 99% significance level) resulted in 20 components.

Decision Criterion 2—Communality

A variable’s communality, h²,

is the squared multiple correlation for the indicator variable as a dependent variable using the factors as predictors. Communality measures the percent of variance in a given variable explained by all the factors jointly and may be interpreted as the ‘reliability’ of the indicator. (Garson, 2012, p. 36)

Therefore, variables with a low communality (i.e., a small percent of the variable’s variance is jointly explained by the components) can drive up the number of components.

The next step then was to remove the measures with a low communality. Considerable discretion exists regarding what constitutes a low communality, with some authors suggesting .25 (Garson, 2012) and others suggesting .40 (Rummel, 1970). Given the large number of measures and the desire to substantively reduce the number of components to a manageable (and interpretable) number, two cutoffs were employed. First, I removed measures with communality less than .50. A review of the data showed that all variables had communality greater than or equal to .50, hence a second more rigorous cutoff of .70 was implemented, which led to the removal of eight measures.

A principal component analysis with a Varimax rotation was performed on the remaining 55 variables. As in the foregoing analyses, the results defied interpretation. The Kaiser criterion yielded 18 components, and the scree plot and PA each recommended 17 components.

Decision Criterion 3—Multicollinearity

While factor analysis relies on moderate to high intercorrelations to reduce variables to fewer underlying constructs (dimensions), multicollinearity “increases the standard error of factor loadings, making them less reliable and thereby making more difficult the process of inferring labels for factors” (Garson, 2012, p. 54). Therefore, some researchers advocate the elimination of collinear variables (Garson, 2012).

The decision criteria were applied cumulatively. Thus, the 55 remaining variables were regressed on asset disruption (a common dependent variable in financial vulnerability studies) and collinearity statistics were generated. Of the 55 variables, 30 suggested evidence of multicollinearity. Therefore, additional variables would need to be removed. Removing all 30 measures with evidence of multicollinearity would decimate the sample, so criteria were applied to remove multicollinearity while retaining as many measures as possible.

This step involved identifying all correlations greater than or equal to .90 and observing the explanatory power of each variable in predicting asset disruption (employing explanatory power as a criterion was deemed preferable to the indiscriminate removal of variables), then removing those variables with high multicollinearity and low explanatory power. This process was conducted until the collinearity statistics showed all remaining variables were within the acceptable ranges (tolerance > .20 and variance inflation factor [VIF] < 5; Garson, 2014). This process resulted in the removal of 17 additional variables.

The final factor analysis was performed on the remaining 38 measures. Despite employing all three decision criteria, the results yielded no improvement in interpretability. With “only” 38 measures, the number of recommended components remained high, and cross-loadings were common for the majority of the measures. Table 4 provides a summary of the results from the exploratory factor analyses performed after each decision criterion was applied.

Table 4.

Number of Components by Decision Criterion.

	Number of components
	All measures	Correlation ≥ .30	Communality ≥ .70	No multicollinearity
Number of variables	70	63	55	38
Method for determining components
Kaiser criterion	22	21	18	16
Scree plot	21	20	17	14
Parallel analysis	21	20	17	15

Implications

This research set out to explore whether financial measures can be organized into theoretically intuitive and empirically defensible constructs. The findings from the first part of this article suggest that accounting constructs are multidimensional, and nonprofit researchers should consider using multiple measures to capture them. The findings from the second part of this article demonstrate the difficulty in organizing empirically dissimilar measures into fewer underlying dimensions, and call attention to the hazards for research and practice of employing financial measures interchangeably.

Discussion

Nonprofit scholars agree that liquidity, solvency, margin, and profitability are central to financial performance and can be observed through several representative financial measures. However, the literature is inconsistent with regard to what the measures assess and is unclear on which measures should be employed to capture each construct. Although nonprofit financial management textbooks appear to suggest that accounting constructs are multidimensional, these exhortations lack empirical support. Hence, nonprofit managers and researchers are left to choose the appropriate measures, and without clear guidance, they may choose on the basis of available data, expediency, or model fit. The result of this variability is a collection of studies susceptible to attributing more generalizability and salience to their findings than perhaps warranted.

The conceptual and empirical disjuncture demonstrated in this article yields three potential interpretations. First, theory is ahead of data. It is possible that our assumptions regarding the representativeness of financial measures are correct, but our measurement capacity and available data are falling short. Form 990 data have several documented weaknesses, including completeness and accuracy of the data, representativeness of the sample to the population of nonprofits, and misclassification of revenues and expenses. Furthermore, Form 990 data do not distinguish between restricted and unrestricted cash balances, or specify the purposes for which revenue is raised (e.g., operating vs. capital campaign). Therefore, organizations could appear liquid, when in reality their funds are restricted, or organizations could appear to be profitable when in fact they are deficit spending.⁴

Second, financial measures may defy classification in nonprofit data. It is conceivable that theoretical coherence is spurious because the measures are representative of different concepts of interest, and pursuing a theoretically intuitive and empirically defensible taxonomy of accounting measures is fruitless, even counterproductive.

The third and most likely interpretation from the results is that the dimension reduction techniques failed to reduce measures to neat constructs because the constructs are multidimensional. Correlations among the variables did not uncover conceptually meaningful relationships, but that does not mean the measures fail as indicators. Consider organizational performance, which can be evaluated on the basis of programmatic effectiveness (e.g., number of clients served; Carman, 2009) or organizational efficiency (e.g., level of input required to generate outputs; Tinkelman & Donabedian, 2007). There is no indication that these measures would load on the same factor, yet both are indicators of organizational performance.⁵

Conclusion

Despite the prevalence of financial measures in nonprofit literature, scant empirical research has assessed the dimensionality of these measures in nonprofit data. This gap in the literature is surprising given the possible implications. If we cannot empirically substantiate the widely held assumptions regarding financial measures and what they represent, then we must question our approach to research and our prescriptions for nonprofit practitioners. Are nonprofit studies that make use of financial measures building on one another or are they, by using different financial measures to capture the same concepts, testing different things?

This study did not set out to determine the “correct” liquidity, solvency, margin, and profitability measures. Rather, this article intended to examine the use of financial measures in nonprofit research. This analysis takes an important first step in demonstrating that financial measures are not interchangeable and accounting constructs are multidimensional. Future studies should build on the work started here and reevaluate the conceptual link between constructs and their proposed financial measures.

The implications from these findings extend well beyond the relatively small pool of researchers engaged in nonprofit financial research. Nonprofit scholars as well as practicing managers focusing on volunteerism, cross-sector collaboration, marketing, and numerous other areas of nonprofit inquiry employ financial measures and trust that the measures represent an organization’s liquidity, solvency, margin, and profitability. This research illustrates the potential hazards associated with employing empirically dissimilar financial measures interchangeably and using single measures to represent important concepts.

Hence, I conclude with three recommendations. First, nonprofit researchers should focus more attention on construct validity by making financial measures the element of interest, not concepts such as solvency, margin, profitability, or liquidity. More caution should be paid to uniformity in the application of accounting ratios in the literature. Otherwise, researchers are failing to engage in the same conversation, and we may be failing to build a foundation of knowledge. Second, where a construct is the element of interest, researchers should consider employing multiple measures for capturing it. One method would be to identify interrelated measures via dimension reduction or another appropriate technique and combine them into indices or component scores. Employing multiple measures could present a more multifaceted and nuanced picture. Finally, researchers should conduct sensitivity and robustness tests using different ratios for underlying financial concepts to improve validity. Regardless of the indicator selected, this analysis calls for more attention to the choice of financial measures, as well as the interpretation and consequences of that choice.

Footnotes

Appendix

Accounting Measures—Standardized Formulas and 2003 IRS Form 990 Calculation.

Description of formula in the literature	Formula calculation	Construct referenced in the literature	Form 990 calculation
Days’ receivables	(Accounts receivable × 365) / total operating revenue	Liquidity	(Line 47cB × 365) / Line 1d + 2 + 3 + 11
Equity ratio	(Total net assets − endowment assets) / (total assets − endowment assets)	Long-term capacity	(Line 73B − 55aB − 55bB - 56B) / (Line 59B − 55aB − 55bB - 56B)
Markup	100% × (Change in unrestricted net assets + depreciation / total operating expenses)	Short-term sustainability	100% × [(Line 67B − 67A) + Line 42A / Line 44A − 42A]
Liquidity	12 × (Current assets − [current liabilities + temporarily restricted net assets]) / (total assets − depreciation)	Liquidity	12 × [((Line 45B + 46B + 47cB + 48cB + 49B + 52B + 53B) − ((Line 60B + 61B) + Line 68B)) / Line 59B − 42A]
Months of spending	12 × (Current assets − [current liabilities + temporarily restricted net assets]) / (total expenses − depreciation)	Liquidity	12 × [((Line 45B + 46B + 47cB + 48cB + 49B + 52B + 53B) − ((Line 60B + 61B) + Line 68B)) / Line 17 − 42A]
Months of spending	12 × (Unrestricted net assets − [equity in property, plant, and equipment − tax exempt debt and mortgages]) / (total operating expenses)	Short-term capacity	12 × [Line 67B − (Line 55cB + 57cB − 64aB − 64bB) / Line 44A − 42A]
Average age of plant	Accumulated depreciation/depreciation	Fixed asset age	Line 55b / Line 42A
Days’ cash on hand	Cash + cash equivalents / ([total operating expenses – depreciation] / 365)	Liquidity, working capital efficiency	Line 45B + (Line 46B + 54B) / ((Line 44A − 42A) / 365)
Cash ratio	Cash + cash equivalents / current liabilities	Liquidity	Line 45B + (Line 46B + 54B) / (Line 60B + 61B)
Modified cash ratio	Cash + cash equivalents / total assets	Liquidity	Line 45B + (Line 46B + 54B) / Line 59B
Cash reserve ratio	Cash + cash equivalents / total expenses	Liquidity	Line 45B + (Line 46B + 54B) / Line 17
Defense interval	Cash + cash equivalents + accounts receivable / average daily operating expenses	—	Line 45B + (Line 46B + 54B) + Line 47cB / ((Line 44A − 42A) / 365)
Quick ratio	Cash + cash equivalents + accounts receivable / current liabilities	Liquidity	Line 45B + (Line 46B + 54B) + Line 47cB / (Line 60B + … + 64aB)
Total surplus	Change in total net assets	Surplus, solvency	Line 73B − 73A
Return on assets	Change in total net assets / total assets	Profitability	(Line 73B − 73A) / Line 59B
Return on net assets, growth rate in equity	Change in total net assets / total net assets	Return on equity	Line 73B − 73A / Line 73B
Net income ratio	Change in total net assets / total operating revenue	—	Line 73B − 73A / Line 1d + 2 + 3 + 11
Unrestricted surplus	Change in unrestricted net assets	Surplus, solvency	Line 67B − 67A
Profitability	Change in unrestricted net assets / total assets	Profitability	(Line 67B − 67A) / Line 59B
Profit margin	Change in unrestricted net assets / total revenue	—	(Line 67B − 67A) / Line 12
Liquidity	Current assets − (current liabilities + temporarily restricted net assets) / total assets	Liquidity	((Line 45B + 46B + 47cB + 48cB + 49B + 52B + 53B) − ((Line 60B + 61B) + Line 68B)) / Line 59B
Net working capital	Current assets − current liabilities	Liquidity	(Line 45B + 46B + 47cB + 48cB + 49B + 52B + 53B) − (Line 60B + 61B)
—	Current assets − current liabilities / total assets	—	((Line 45B + 46B + 47cB + 48cB + 49B + 52B + 53B) − (Line 60B + 61B)) / Line 59B
Primary reserve ratio	Current assets − current liabilities / total expenses	—	((Line 45B + 46B + 47cB + 48cB + 49B + 52B + 53B) − (Line 60B + 61B)) / Line 17
Quick ratio	Current assets − inventories / current liabilities	Liquidity	((Line 45B + 46B + 47cB + 48cB + 49B + 52B + 53B) − Line 52B) / (Line 60B + 61B)
Current ratio	Current assets / current liabilities	Liquidity, working capital efficiency	(Line 45B + 46B + 47cB + 48cB + 49B + 52B + 53B) / (Line 60B + 61B)
Asset ratio	Current assets / total assets	Liquidity	(Line 45B + 46B + 47cB + 48cB + 49B + 52B + 53B) / Line 59B
Average payment period	Current liabilities / ([total operating expenses − depreciation] / 365)	Fixed asset age	(Line 60B + 61B) / ((Line 44A − 42A) / 365)
—	Current liabilities / current assets	Liquidity	(Line 60B + 61B) / (Line 45B + 46B + 47cB + 48cB + 49B + 52B + 53B)
Depreciation rate	Depreciation / gross fixed assets	Fixed asset age	Line 42A / Line 57a
Asset productivity	Earnings before interest / total assets	—	Line 12 − 4 − 5 / Line 59B
Fixed asset ratio, financial distress costs	Gross fixed assets / total assets	—	Line 57a / Line 59B
Endowment	Prospectively endowed: If liquid investments (i.e., securities, not buildings) are greater than or equal to total expenses than 1, otherwise 0	Surplus	If (Line 54B + Line 56B) ≥ Line 17 then 1, otherwise 0
Return on investments	Investment income / invested financial assets	Profitability	Line 4 + 5 + 6c + 7 + 8d / Line 45B + 46B + 54B + 55cB + 56B
Surplus margin, net surplus	Net income	Surplus	Line 18
Return on assets	Net income / total assets	Profitability, long-term sustainability	Line 18 / Line 59B
Savings indicator, savings ratio	Net income / total expenses	—	Line 18 / Line 17
Return on equity	Net income / total net assets	Equity	Line 18 / Line 73B
Operating margin ratio	Net income / total operating revenue	Margin	Line 18 / (Line 1d + 2 + 3 + 11)
Total margin, net operating results, surplus margin	Net income / total revenue	Profitability, stability, margin, sustainability	Line 18 / Line 12
Cash flow to total debt	Net income + depreciation / current liabilities + mortgages	—	Line 18 + Line 42A / (Line 60B + 61B) + Line 64bB
Cash flow to total debt	Net income + depreciation / total liabilities	Capital	Line 18 + 42A / Line 66B
Times interest earned	Net income + interest expense / interest expense	Debt coverage	Line 18 + Line 41A / Line 41A
Nonoperating gains	Net nonoperating gains / total revenue	Liquidity	(Line 4 + 5 + 6c + 7 + 8d + 9c + 10c) / Line 12
Operating surplus	Net operating income	Solvency	(Line 1d + 2 + 3 + 11) − (Line 44A − 42A)
Debt coverage ratio	Net operating income / tax exempt debt + mortgages	—	(Line 1d + 2 + 3 + 11) − (Line 44A − 42A) / Line 64aB + 64bB
Return on assets	Net operating income / total assets − endowment assets	Long-term sustainability	(Line 1d + 2 + 3 + 11) − (Line 44A − 42A) / (Line 59B − 55cB − 56B)
Operating margin	Net operating income / total operating revenue	Profitability	(Line 1d + 2 + 3 + 11) − (Line 44A − 42A) / (Line 1d + 2 + 3 + 11)
Net operating ratio	Net operating income / total revenue	—	(Line 1d + 2 + 3 + 11) − (Line 44A − 42A) / Line 12
Operating margin	Net operating income + depreciation / total revenue	Profitability	(Line 1d + 2 + 3 + 11) − (Line 44A − 42A) + Line 42A / Line 12
Debt ratio	Tax exempt debt + mortgages / total assets	—	Line 64aB + 64bB / Line 59B
Debt burden ratio	Tax exempt debt + mortgages / total expenses	—	Line 64aB + 64bB / Line 17
Debt-to-equity ratio	Tax exempt debt + mortgages / total net assets	—	Line 64aB + 64bB / Line 73B
Debt service ratio	Tax exempt debt + mortgages / total operating revenue	Debt structure	Line 64aB + 64bB / Line 1d + 2 + 3 + 11
Debt service burden	Tax exempt debt + mortgages / total revenue	Solvency	Line 64aB + 64bB / Line 12
Asset growth	Total assets (boy) / total assets (boy)	—	Line 59B / Line 59A
Fixed asset financing	Total liabilities / net fixed assets	Capital structure	Line 66B / Line 57cB
Debt ratio, leverage, gearing ratio	Total liabilities / total assets	Flexibility, solvency, equity, debt	Line 66B / Line 59B
Debt-to-equity ratio	Total liabilities / total net assets	Capital structure, long-term solvency, long-term risk	Line 66B / Line 73B
Equity balance	Total net assets	Equity	Line 73B
Liquidity funds indicator	Total net assets − permanently restricted net assets − equity in property, plant, and equipment / total expenses	—	Line 73B − Line 69B − (Line 55cB + 57cB − 64aB − 64bB) / Line 17
Equity financing	Total net assets / total assets	Capital structure, profitability, long-term capacity	Line 73B / Line 59B
Net asset reserve ratio	Total net assets / total expenses	—	Line 73B / Line 17
Adequacy of equity, total margin, equity ratio	Total net assets / total revenue	Stability, flexibility, profitability, equity	Line 73B / Line 12
Growth potential	Total revenue − (change in total assets + net income)	—	Line 12 − ((Line 59B − 59A) + Line 18)
Current asset turnover	Total revenue / current assets	Working capital efficiency	Line 12 / (Line 45B + … + 53B)
Fixed asset turnover	Total revenue / net fixed assets	Fixed asset efficiency	Line 12 / Line 57cB
Total asset turnover, capital-turnover ratio, return ratio	Total revenue / total assets	Fixed asset efficiency	Line 12 / Line 59B
—	Total revenue / total expenses	Fiscal performance	Line 12 / Line 17
Leverage ratio	Unrestricted net assets / total liabilities	—	Line 67B / Line 66B

Note. The table presents the 70 formulas identified during the literature review. In addition to the formula calculation and Form 990 calculation, commonly employed descriptors of the formulas and their corresponding constructs are included. The descriptions and constructs are culled from the literature and are not this author’s interpretation. “—” represent formulas employed in the literature, based on this author’s review, without a description or reference to a construct.

Acknowledgements

I would like to thank the anonymous reviewers for their helpful comments on earlier drafts of this article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biography

Christopher R. Prentice is an assistant professor in the Department of Public and International Affairs at the University of North Carolina Wilmington, where he teaches public and nonprofit management. His research focuses on government–nonprofit relations and nonprofit finance.

References

Abramson

(1995). Sources of data on nonprofit finance. Nonprofit Management & Leadership, 5, 443-451.

Altman

(1968). Financial ratios, discriminant analysis, and the prediction of corporate bankruptcy. Journal of Finance, 23, 589-609.

Ashley

Faulk

(2010). Nonprofit competition in the grants marketplace: Exploring the relationship between nonprofit financial ratios and grant amount. Nonprofit Management & Leadership, 21, 43-57.

Bacon

(1992). Do capital structure theories apply to nonprofit hospitals? Journal of the Midwest Finance Association, 31, 86-90.

Bowman

(2002). The uniqueness of nonprofit finance and the decision to borrow. Nonprofit Management & Leadership, 12, 293-311.

Bowman

(2011a). Financial capacity and sustainability of ordinary nonprofits. Nonprofit Management & Leadership, 22, 37-51.

Bowman

(2011b). Finance fundamentals: Building capacity and sustainability. Hoboken, NJ: John Wiley.

Bowman

Tuckman

Young

(2012). Issues in nonprofit finance research: Surplus, endowment, and endowment portfolios. Nonprofit and Voluntary Sector Quarterly, 41, 560-579.

Brown

(2005). Exploring the association between board and organizational performance in nonprofit organizations. Nonprofit Management & Leadership, 15, 317-339.

10.

Calabrese

(2011a). The accumulation of nonprofit profits: A dynamic analysis. Nonprofit and Voluntary Sector Quarterly, 41, 300-324. doi:10.1177/0899764011404080

11.

Calabrese

(2011b). Testing competing capital structure theories of nonprofit organizations. Public Budgeting & Finance, 31, 119-143.

12.

Carman

J. G.

(2009). The accountability movement: What’s wrong with this theory of change? Nonprofit and Voluntary Sector Quarterly, 39, 256-274. doi:10.1177/0899764008330622

13.

Carroll

Stater

(2009). Revenue diversification in nonprofit organizations: Does it lead to financial stability? Journal of Public Administration Research and Theory, 19, 947-966.

14.

Chabotar

(1989). Ratio analysis comes to nonprofits. The Journal of Higher Education, 60, 188-208.

15.

Coe

(2011). Nonprofit financial management: A practical guide. Hoboken, NJ: John Wiley.

16.

Costello

Osborne

(2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment Research & Evaluation, 10(7), 1-9.

17.

Dumont

(2013). Nonprofit virtual accountability: An index and its application. Nonprofit and Voluntary Sector Quarterly, 42, 1049-1067.

18.

Financial Accounting Standards Board. (1993). Statement of Financial Accounting Standards No. 117: Financial statements of not-for-profits. Norwalk, CT: Author.

19.

Finkler

Purtell

Calabrese

Smith

(2013). Financial management for public, health, and not-for-profit organizations (4th ed.). Upper Saddle River, NJ: Pearson.

20.

Froelich

Knoepfle

(1996). Internal revenue service 990 data: Factor or fiction? Nonprofit and Voluntary Sector Quarterly, 25, 40-52.

21.

Froelich

Knoepfle

Pollak

(2000). Financial measures in nonprofit organization research: Comparing IRS 990 return and audited financial statement data. Nonprofit and Voluntary Sector Quarterly, 29, 232-254.

22.

Garrow

Hasenfeld

(2014). Institutional logics, moral frames, and advocacy: Explaining the purpose of advocacy among nonprofit human service organizations. Nonprofit and Voluntary Sector Quarterly, 43, 80-98.

23.

Garson

G. D.

(2012). Blue Book Series: Factor analysis. Asheboro, NC: Statistical Associates.

24.

Garson

G. D.

(2014). Blue Book Series: Multiple regression. Asheboro, NC: Statistical Associates.

25.

Greenlee

Trussel

(2000). Estimating the financial vulnerability of charitable organizations. Nonprofit Management & Leadership, 11, 199-210.

26.

Greenlee

Tuckman

(2007). Financial health. In Young

(Ed.), Financing nonprofits (pp. 315-335). Lanham, MD: Alta Mira Press of Rowman and Littlefield.

27.

Hager

(2001). Financial vulnerability among arts organizations: A test of Tuckman-Chang measures. Nonprofit and Voluntary Sector Quarterly, 30, 376-392.

28.

Hansmann

(1980). The role of nonprofit enterprise. Yale Law Journal, 89, 835-901.

29.

Hodge

Piccolo

(2005). Funding source, board involvement techniques, and financial vulnerability in nonprofits. Nonprofit Management & Leadership, 16, 171-190.

30.

Jaskyte

(2013). Does size really matter? Organizational size and innovations in nonprofit organizations. Nonprofit Management & Leadership, 24, 229-247.

31.

Keating

Fischer

Gordon

Greenlee

(2005). Faculty Research Working Paper Series: Assessing financial vulnerability in the nonprofit sector (Paper No. 27). Hauser Center for Nonprofit Organizations, Harvard University.

32.

Keating

Frumkin

(2001). How to assess nonprofit financial performance (NASAA Training: 28-57). Retrieved from http://www.nasaa-arts.org/Learning-Services/Past-Meetings/Reading-5-Understanding-Financial-Statements.pdf

33.

Lance

Butts

Michels

(2006). The sources of four commonly reported cutoff criteria: What did they really say? Organizational Research Methods, 9, 202-220.

34.

McGinnis Johnson

(2013). Necessary but not sufficient: The impact of community input on grantee selection. Administration & Society. Retrieved from http://aas.sagepub.com/content/early/2013/11/06/0095399713509241.full.pdf+html

35.

McIndoe

(2013). Reinforcing the safety net: Explaining the propensity for and intensity of nonprofit-local government collaboration. State and Local Government Review, 45, 283-295.

36.

McLaughlin

(2009). Streetsmart: Financial basics for nonprofit managers. Hoboken, NJ: John Wiley.

37.

Newton

(2013, September). Finance Seminar Series: Managerial incentives, financial performance, and governance quality: Evidence from nonprofit organizations. Oklahoma University. Retrieved from https://housing.ou.edu/dam/price/Finance/CFS/paper/pdf/Newton_Paper.pdf

38.

Ohlson

(1980). Financial ratios and the probabilistic prediction of bankruptcy. Journal of Accounting Research, 19, 109-131.

39.

Rummel

R. J.

(1970). Applied factor analysis. Evanston, IL: Northwestern University Press.

40.

Ryan

Irvine

(2012). Not-for-profit ratios for financial resilience and internal accountability: A study of Australian international aid organizations. Australian Accounting Review, 22, 177-194.

41.

Saxton

Guo

(2011). Accountability online: Understanding the web-based accountability practices of nonprofit organizations. Nonprofit and Voluntary Sector Quarterly, 40, 270-295.

42.

Saxton

(2013). The social network effect: The determinants of giving through social media. Nonprofit and Voluntary Sector Quarterly, 43, 850-868.

43.

Scherhag

Boenigk

(2013). Different or equal treatment? Donor priority strategy and fundraising performance assessed by a propensity score matching study. Nonprofit Management & Leadership, 23, 443-472.

44.

Sedatole

Swaney

Yetman

(2013, June 17). Accounting-based performance metrics and executive compensation in nonprofit organizations (Accounting Seminar). University of Miami. Retrieved from http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2151326

45.

Tinkelman

Donabedian

(2007). Street lamps, alleys, ratio analysis and nonprofit organizations. Nonprofit Management & Leadership, 18, 5-18.

46.

Tinkelman

Mankaney

(2007). When is administrative efficiency associated with charitable donations? Nonprofit and Voluntary Sector Quarterly, 36, 41-64.

47.

Trussel

(2002). Revisiting the prediction of financial vulnerability. Nonprofit Management & Leadership, 13, 17-31.

48.

Trussel

Greenlee

Brady

(2002, June). Predicting financial vulnerability in charitable organizations. The CPA Journal, 66-69.

49.

Trussel

Parsons

(2008). Financial reporting factors affecting donations to charitable organizations. Advances in Accounting, 23, 263-285.

50.

Tuckman

Chang

(1991). A methodology for measuring the financial vulnerability of charitable nonprofit organizations. Nonprofit and Voluntary Sector Quarterly, 20, 445-460.

51.

Velicer

Jackson

(1990). Component analysis versus common factor-analysis: Some further observations. Multivariate Behavioral Research, 25, 97-114.

52.

Verbruggen

Chistiaens

Milis

(2011). Can resource dependence and coercive isomorphism explain nonprofit organizations’ compliance with reporting standards? Nonprofit and Voluntary Sector Quarterly, 40, 5-32.

53.

Weikart

Chen

Sermier

(2013). Budgeting & financial management for nonprofit organizations: Using money to drive mission success. Los Angeles, CA: CQ Press.

54.

Zeller

Stanko

Cleverley

(1996). A revised classification pattern of hospital financial ratios. Journal of Accounting and Public Policy, 15, 161-182.

55.

Zietlow

(2012, September 26). A financial health index for achieving nonprofit financial sustainability. Retrieved from http://ssrn.com/abstract=2049022

56.

Zietlow

Hankin

Seidner

(2007). Financial management for nonprofit organizations: Policies and practices. Hoboken, NJ: John Wiley.