Foreign aid, grassroots activism, and the strength of applied community studies in aid-receiving countries: the case of community psychology

Introduction

Local capacity building for the sustainability of international development initiatives has been a major focus of global aid donations. Today, there are numerous aid donors including global multilateral organizations such as the United Nations and the World Bank Group, as well as coalitions of countries such as the Organization for Economic Co-operation and Development (OECD) and the European Commission's Directorate-General for International Cooperation and Development (DG DEVCO). In recent years, these organizations have tried to shift focus from top-down aid initiatives to more on the ground, grassroots interventions.

In this study, we consider the impact of foreign aid from OECD countries on the existence and strength, not of income or wealth, but of the field of community psychology (CP) in 67 countries receiving that foreign aid. CP's emphasis on applied research and practice focused on analyzing and solving social, political, health, educational, and other problems at the local level, make it a critical professional field in the countries that need such training and expertise most and where it may have the greatest impact (Hanitio & Perkins, 2017). While we focus on CP in this paper, there are numerous other applied community-based research fields that could be impacted by similar mechanisms. The Global Development of Applied Community Studies dataset used in this article currently includes the fields of CP, community sociology, community development, community social work, interdisciplinary community studies, applied/development anthropology, development economics, public health, urban/regional planning/development, public administration/policy studies, community/popular education, and liberation theology/religious studies. For this study, we consider the case study of CP.

This paper presents a predictive model of factors that influence the development of CP as a professional and academic field, based on such indicators as professional organizations or conferences, graduate and undergraduate courses or programs, and publications. Controlling for societal income inequality (GINI), economic productivity (GDP), educational infrastructure, civil liberties, and history of grassroots nonviolent political action, we will explore the relationship between the amount of money received from OECD countries—in the form of Official Development Assistance (ODA)—and the existence and level of development of CP across 67 aid-receiving countries. The interaction effect between aid and political activism will contribute to our understanding of how foreign aid can have unintended negative consequences at both the local and national levels. Aid must be used to improve human and community wellness, not by supplanting local expertise and human capital, but by developing and supporting them.

This study examines important concerns and conflicts between international aid, social development, political activism, and allied social science fields such as CP. It is critical to understand the macrosystemic relationships that strengthen or weaken CP and discover the reasons why the field has yet to develop in the lowest-income, greatest-need countries (Hanitio & Perkins, 2017), beyond the simple and obvious factor of educational resources. This research considers the impact of foreign aid by controlling for higher educational infrastructure and by focusing on the interaction of aid with another key factor: the history and extent of grassroots activism in the country. There has been no prior empirical study on the moderating effect of foreign aid on the relationship between social-political activism and the strength of CP.

The Global Growth of Community Psychology

The field of CP has grown internationally fairly steadily since the late 1970s (with roots earlier than that). The term CP was first adopted in the U.S. in 1965 and by the 1970s it began taking root in different forms in Canada, Europe, Latin America, a few countries in Asia, Australia, and New Zealand. By the 1990s, journals, conferences, and academic programs devoted explicitly to the field began to emerge in multiple countries on every continent (Perkins, 2009; Reich, et al. 2007). Montero (1996) compared the development of CP in Latin America and the United States with two vastly different social and political contexts. However, both involved an epistemological crisis whereby positivism was found lacking as a response to complex social problems. Thus, a widespread yearning for a more just epistemology in psychology and the need to reduce dependence on centralized control through local capacity building of researchers, professionals, and community members have been important drivers of the global growth of CP.

Yet that does not explain the considerable variation in such growth or the conditions that promote or inhibit it. In the first quantitative or mixed-methods study of its kind, Hanitio and Perkins (2017) showed that a history of nonviolent action predicted the strength of both CP and community development in 91 countries globally, and that, unlike community development, CP was stronger in “more” developed countries (based on the Human Development Index). They also explored brief qualitative case studies contrasting Chile, which has a well-developed field of CP but a smaller community development field, and Ghana, with its well-developed community development field but little CP.

This paper responds to and builds on Hanitio and Perkins (2017) by focusing specifically on countries that receive aid from OECD nations and on the relationship of that aid to the growth of CP, which is vitally important both for those higher-poverty, less-resourced countries and for the future of CP. Further, it likely depends on different factors than those in donor countries, due to the history of colonization which left many former colonies stripped of financial resources and host to systems of education that were implemented by and for the colonizers. Those countries that were once colonizers map almost perfectly onto those now considered donor countries, and those that receive aid are mostly former colonies (Rist, 2014). Thus, a separate analysis of aid-receiving countries will allow us to consider the unique and ongoing impact of foreign aid between former colonizer and colonized countries on the growth of CP.

Dependency and Globalization

Dependency theory is one lens used to examine the relationship between countries in the Global North and South. Dependency analysts argue that countries in the Global North create international structures which privilege themselves, and cause formerly colonized countries in the Global South to remain tied to the economies of the former colonizing countries in the Global North (Dos Santos, 1970). Cardoso (1977) notes two major types of dependency analyses: one leading to constant underdevelopment vs. one which re-creates the capitalist state inside aid-receiving countries such that an elite is formed internally, mirroring that of the colonial relationship, causing a 'double exploitation' of dominated groups. Each conceptualization of dependency may be valid, but apply to different countries with different implications for education and human capital development in each country.

Some dependency theory ideas continue in the contemporary literature on globalization. Globalization is defined by Sites (2000) as international integration based on strengthened economic, political, communication, and migratory connections and reduced barriers between countries. Globalization has been a millenium-long trend, but has become of particular interest in recent decades. Some scholars view the role of Global North countries as creating dependency and exploiting the Global South through institutions such as the World Bank and International Monetary Fund (IMF) (Stiglitz, 2002). Others suggest that globalization has positive impacts on most countries, generally decreases global inequality, and contributes to development in aid-receiving countries (Milanovic, 2013).

Foreign Aid and Capacity Development

Foreign aid may serve the geopolitical aims of OECD countries in their desire to influence the economic, defense, and alignment policies of aid-receiving countries. But regarding development goals, there is ample evidence supporting dependency theory and the ineffectiveness of aid, even in IMF Working Papers (Masud & Yontcheva, 2005). Most studies focus on the effects of aid on GDP and other macroeconomic indicators, and rarely find that aid has a significant impact on those. Instead we focus on the effect of aid on capacity development in one field: CP. What are the potential mechanisms for such an effect? The stated goals of foreign aid are not to support the recipient countries continually, but rather to help build countries to the point at which they are able to support themselves sustainably (Riddell, 2008; Rist, 2014). According to the OECD (2006), capacity development refers to the “process whereby people, organizations, and society as a whole unleash, strengthen, create, adapt, and maintain capacity over time” (p. 9).

Aid specifically for education has been considered important for human capital formation (i.e., capacity development; Shultz, 1961). Since the 2000s, higher education became the largest recipient of aid. Despite this trend, most studies of aid effectiveness focus on outcomes at the basic and secondary levels of education (Riddell & Niño-Zarazua, 2016).

Capacity development is now considered a major focus of donors (Riddell, 2008; Riddell & Niño-Zarazúa, 2016) and is an integral part of ensuring sustainability of projects. In 2005, OECD countries, including those in the European Union, signed the Paris Declaration in which they dedicated themselves to five principles for increasing aid effectiveness: ownership by recipient countries; donor alignment with recipient objectives; harmonization of process; measuring results; and mutual accountability between donors and recipients. DG DEVCO, one of the largest global aid donors, also points to this agreement as an important historical moment for the organization. They note that the agreement represents a move towards greater 'coordination, harmonisation and transparency' (European Commission, 2019). This paper adds to the literature by considering the relationship of aid to capacity development in the field of CP. Training in CP builds capacity specifically for the type of local social, health, wellness, and empowerment projects that aid donors aim to support.

Bräutigam and Knack (2004) found a strong statistical relationship between high aid levels and deterioration in governance in a sample of African countries. Controlling for GDP and violence in these countries, aid was still found to have an independent and negative effect. They conclude that the implementation, not the amount, of aid is the likely cause of this deterioration and propose that aid could be used for developing institutions and human resource capacity if administered in a way that emphasizes local ownership and direction.

In a case study of development professionals and projects in Cambodia, donors and local officials both agreed that capacity development was “by far the most important aim of technical assistance” (Godfrey et al. 2002, p. 361). However, the projects were still owned and directed by donor agencies in their “identification, design and implementation, to the detriment of capacity development” (p. 369). This resulted in very few projects driven by local demand, and institutions built around the expectations of continual aid. These authors recommend not withdrawing aid, but rather transferring ownership of projects and programs to locals and paying more attention to increasing local capacity in a sustainable way (Godfrey et al. 2002).

Despite some limited successes, capacity development efforts have been largely ineffective (World Bank, 2005). Riddell (2008) concludes that aid would be able to produce capacity development if it were not donor driven nor focused mainly on economic development, the outcome of interest to most donors. The focus should be more on the process of capacity development rather than the elusive goal of economic growth (Riddell & Niño-Zarazúa, 2016).

In sum, foreign aid's impact on donor's primary goal of economic growth is negligible, but its effect on capacity development—at both the individual and institutional levels—should be more positive, but is often counterproductive. These findings support dependency theory in that a connection between the core and periphery causes a further negative impact on periphery countries. However, it is important to emphasize that none of the above researchers suggest that aid should be withdrawn, rather they suggest that it be re-focused.

Effects of Nonviolent Grassroots Activism on Development and Capacity

The staff and volunteers of grassroots community organizations understand, not only local needs, but the dynamics of institutions, conflict, empowerment, collaboration, values, and tradition (Finsterbusch & Van Wicklin, 1987; Uvin & Miller, 1996). As government-led, top-down interventions hamper the development of civil society, Kellogg (2012) urges Western donors to prioritize outreach to increase the number of local nonviolent grassroots organizations doing meaningful advocacy work in their respective communities. Despite the challenges, supporting grassroots activism benefits sustainable development.

Perhaps more surprising, a history of grassroots activism may also benefit a country's development capacity in the form of trained human and intellectual capital. Reich et al. (2007) describe grassroots social movements as a notable contributor to the growth of CP. By improving the quality of life, allowing local voices to be heard, and encouraging community-wide participation, social movements have acted as a precursor to the development of CP (Reich et al., 2007). Hanitio and Perkins (2017) created a model to test that across 91 countries and found that nonviolent grassroots activism at the country level significantly predicts the development of both CP and community development, controlling for civil liberties, social and economic development, and population size. Hanitio and Perkins (2017) recognized, however, that top-down development interventions may also influence the growth of applied community studies and called for research comparing the influence of foreign aid and grassroots activism.

Foreign Aid and Grassroots Organizations

The impact of grassroots organizations is often limited when decisions are influenced more by outsiders than indigenous knowledge of the local situation (Uvin & Miller, 1996). This sharp contrast between top-down and bottom-up approaches to community development is even more problematic when the initiative is driven by foreign aid, no matter how well intentioned. When discussing the deterioration of governance in Africa, Bräutigam and Knack (2004) describe the perverse incentives created by foreign aid that discourages grassroots participation. Similarly, Makoba (2002) calls attention to the ways aid from large donors leads to large development projects which undermine the independence and effectiveness of local grassroots groups and precipitate inequality, inefficiency, corruption, and ignorance of local conditions. Local communities must be in control to meet their own needs.

Grassroots organizations' goals are constrained by excessive upward accountability requirements created by aid programs, which exacerbates pressures and discontent with such programs expressed by the local community (Dixon & McGregor, 2011). For example, donors require reports and proposals written in English, which limits available grassroots partners to English-speaking groups and restricts their capacity to reach the rather lower-skilled indigenous population (Dixon & McGregor, 2011). This specific case exemplifies the negative moderating effect foreign aid can have on the viability and impact of grassroots organizations. What is unknown, however, is the effect of aid on local professional and research disciplines, such as CP, that could assist development efforts and local problem-solving. Our study fills that gap and will test whether grassroots activism plays as important a role in the existence and strength of CP in less developed countries as it does in wealthier countries (Hanitio & Perkins, 2017), and what the role of foreign aid is in that relationship.

Hypotheses

The existing literature sheds light on the negative link between foreign aid and the strength of CP, quality of governance, economy, political stability, and community development. As an extension of the theoretical model developed by Hanitio and Perkins (2017), we propose an updated theoretical framework (Figure 1) to predict the strength of CP in a given country based on (a) the strength of existing grassroots activism, (b) the amount of foreign aid received (Official Development Assistance per capita, which may have a direct and/or a moderating effect on the development of CP), and (c) controlling for the independent influences of GDP per capita, income inequality, educational infrastructure, and civil liberties. The country-level characteristics, which include the control variables, represent the economic, political, social, and educational situation of a given country. The predictive strengths of the antecedent variables on the strength of CP become more accurate with these country-level characteristics fixed.

As a test of this model, we propose the following four hypotheses:

Hypothesis 1: Development aid and the strength of CP will be negatively correlated, as suggested by dependency theory.

Hypothesis 2: Nonviolent grassroots activism and CP will be positively correlated (Hanitio & Perkins, 2017) in this subsample of aid-receiving countries.

Hypothesis 3: Both development aid and nonviolent grassroots activism will significantly explain the variance in the strength of CP over and above the influence of other country-level control variables: GDP per capita, educational infrastructure, income inequality, and civil liberties.

Hypothesis 4: The relationship between nonviolent grassroots activism and the strength of CP will be moderated by development aid. In other words, when development aid is low, nonviolent action and CP will be significantly and positively related, but when development aid is high, this relationship will be nonsignificant.

Method

Global Development of Applied Community Studies (GDACS) Project

The GDACS dataset assesses 12 community-focused applied research disciplines (CP, community sociology, community development, community social work, development anthropology, development economics, public health, urban/regional planning/geography, public administration/policy studies, community/popular adult education, liberation theology/faith-based community development, and interdisciplinary community studies) in each of 104 selected countries, constituting 94.3% of the world population. The sampling process for selecting countries for the dataset occurred in two stages: the first 30 countries were known to have established work or training in CP; next, all countries exceeding 10 million in population were selected (excluding North Korea for which accurate information is unobtainable), and finally smaller countries with which the research team had particular familiarity were added; for more details, see Hanitio and Perkins (2017).

Sample

For the present analysis, which focus specifically on countries that receive foreign aid, we selected 67 countries from the GDACS dataset which were listed by the OECD as recipients of Official Development Aid and Country Programmable Aid. Of the 73 countries in the GDACS dataset that receive aid, six were excluded from this analysis. Palestine was missing data for foreign aid, GDP, and civil liberties. Syria was excluded due to the recent history of war in the region and having atypical and changing foreign aid profiles and sources. Barbados, Cuba, and Somalia were excluded because they do not have GINI ratings for recent years. Chad was not listed in the Webometrics dataset, and so was excluded because it did not have a rating for educational infrastructure.

The sample for study consisted of 27 countries from Africa, 20 from Asia, three from the Caribbean, five from Central America, one from Europe, one from North America and 10 from South America. The smallest GDP per capita in the sample for the year 2015 was $365 while the largest was $28,056. The smallest population was just under 3 million, while the largest was 1.4 billion. The Appendix contains a list of the 67 countries along with their values for each of the variables used.

Variables and Measures

Outcome variable: The strength of community psychology. In the Global Development of Community Studies project, the development of each field is coded and recoded, based primarily on web searches and secondarily on evidence in the research literature, by multiple trained researchers after establishing an acceptable level of inter-coder agreement. Each field received a score on a ten-point scale, where the points are established by the following rubric (Hanitio & Perkins, 2017): one point for a formal professional organization or conference; one point for any undergraduate courses; one point for any graduate courses; one point for any undergraduate programs; two points for one graduate program or three points for multiple graduate programs (as graduate-level training was deemed more important and relevant than undergraduate programs for professional preparation); one point for less than five publications in the subject, or two for more than five publications but no journal, or three if the country has a CP journal. The measure had sufficient variance (SD = 3.0), while internal consistency was not expected or necessary for this additive scale based on the presence or absence of various resources, not psychological factors. Thus, our dependent variable was the strength of CP on the 0-10 scale as measured for each of 67 aid-receiving countries. Further rationale and limitations of this scale can be found in the Discussion section below.

Predictor variables: Official development assistance, grassroots activism

OECD official development assistance. ODA is defined as “government aid designed to promote the economic development and welfare of developing countries... aid includes grants, 'soft' loans (where the grant element is at least 25% of the total) and the provision of technical assistance” (OECD website). ODA is administered to countries that are designated to be 'developing' by the OECD, and it is provided by those 'developed' countries that are members of the OECD. The aim of ODA includes the alleviation of poverty in the aid-receiving countries in the long term. To track development goals, the OECD is a part of the Global Partnership for Effective Development Cooperation, which has a list of indicators geared towards measuring and creating a more effective development co-operation between donors and recipients of aid. The indicators are based on the following core principles of effective development co-operation principles: ownership, focus on results, inclusive development partnerships and transparency, and mutual accountability among development partners. While ODA does not represent all development aid received by a country, it represents a significant portion (UNDP, 2011). There are other sources of development aid that are not reflected by this measure, such as money from large nonprofit organizations that work directly with the government, as well as money from South-South cooperation (OECD, 2019).

Nonviolent grassroots activism. A country-level indicator of historical grassroots political and social activism is taken from the Global Nonviolent Action Database (GNAD; Swarthmore College, 2015). The GNAD records nonviolent action campaigns by country and provides a rating of the success of these actions. The database contains over 1,000 nonviolent action cases from more than 200 countries. As in Hanitio and Perkins (2017), we used a product score consisting of the frequency of nonviolent actions in a country's history multiplied by the average success of these movements in order to create a score that reflected both the frequency and impact of these actions. A base-10 log transformation was used to adjust for positive skewness in the distribution. For more information, see Hanitio and Perkins (2017). This measure includes events of nonviolent action that in some cases occurred before the first signs of CP as a discipline. Thus, this measure is meant to serve as a proxy for a culture of nonviolent activism rather than representing the direct impact of these instances of nonviolent action that took place during the development of the field of CP.

Control variables

GDP per capita. The 2015 Gross Domestic Product (GDP) reflects the economic productivity of a country in a given year. The GDP for each country was divided by that country's population, creating a standardized variable: GDP per capita. Countries with lower GDPs tend to receive more development aid, thus controlling for GDP is important to separate the impacts of a country having low resources from the impacts of development aid. However, as stated above, GDP was correlated with all the control variables and the development aid variables, and as such we explored both models with and without GDP to observe differences in the effects of its inclusion.

Educational infrastructure. Hanitio and Perkins (2017) used the U.N. Human Development Index, which combines national income, health and education indicators. In order to separate the effects of income and education, we decided to use GDP (above) for income and create a new higher educational infrastructure variable, which is represented by data from the Webometrics project, based at the Cybermetrics Lab at the Spanish National Research Council (CSIC). Our measure is simply the number of universities represented in each country in the 2016 Webometrics database, divided by the population of that country. This measure of educational infrastructure is particularly relevant to this project as the development of all academic fields is likely influenced by the prevalence of institutions of higher education. Thus, controlling for the number of universities with web presence in each country should reduce the chances that the results are reflective of general educational infrastructure rather than specifically the fields of applied community studies of interest.

Income inequality. The GINI coefficient is a widely accepted measure of the degree of income inequality in a country. The higher the GINI, the greater the income inequality in that country. Inequality may be a driver of creation and expansion of applied community studies fields as greater inequality is related to increased social problems (Wilkinson & Pickett, 2009) which are the raison d'être of community studies; plus the charitable and tax-based resources from wealthier citizens may support community studies institutions (that are geared toward mitigating or reducing the negative effects of inequality).

Civil liberties. Civil liberties ratings were adapted from Freedom House's annual survey of experts on political rights and civil liberties in 210 countries and territories conducted in 2014 and reported in 2015. The civil liberties measure contains 15 indicators in four categories: Freedom of Expression and Belief, Associational and Organizational Rights, Rule of Law, and Personal Autonomy and Individual Rights (Freedom House, 2015). We reverse-coded the scale so that 1 represents lowest civil liberties and 7 represents highest civil liberties.

Approach to Data Analysis

All analyses were conducted at the country level. ODA per capita and non-violent action were mean-centered for all analyses so that we could test the effect of the interaction of those two variables.

Correlations. First, a correlation table (Table 1) was constructed to understand the simple relationships between the variables (Hypothesis 1 and 2).

Hierarchical linear regressions. Two hierarchical linear regressions (Table 2) were conducted to determine the relationship of development aid to the strength of CP over and above the predictive effects of other sets of control variables (Hypothesis 3). The main predictive variable of interest in this paper is the measure of development aid, Official Development Assistance, from OECD nations for the year 2014, the most complete year in the dataset for the countries of interest. GDP per capita (“GDP”) was used as a control variable for one of the models, but was not used for the other model. The main hypothesis is that development aid is predictive of the strength of community studies over and above the control variables. GDP is significantly correlated with many of the predictor variables and as such, to test the effects of multicollinearity in the model, it was excluded from one set of equations. Those countries with the lowest GDPs receive more aid as they have most need. Thus, for one of the models, GDP was included as a control variable to determine whether or not development aid predicted the strength of community studies above and beyond the influence of GDP.

Interaction of development aid and nonviolent action. Two more hierarchical linear regressions (Table 3) were conducted including the interaction between development aid and non-violent action (Hypothesis 4). The hierarchical regression tests the significance of the moderating effect. To better understand the moderating effect of foreign aid, we plotted a graph of the simple slopes of the nonviolent action-strength of CP linkage. The graph represents how the relationship between nonviolent action and the strength of CP changes as the aid levels change within a country. Again, given GDP's multicollinearity, separate models tested the effect of including GDP per capita in the model.

Steps of the hierarchical regressions. The order of the steps was determined by grouping structural economic and infrastructure variables first, then political cultural variables of civic freedoms and historical grassroots activity, which together provide control variables for the influence of foreign aid. That predictor variable of interest was added last in each model in order to determine its predictive power over and above the other variables (see Tables 1-4).

Results

Correlations

Table 1 displays the simple bivariate Pearson correlations for all factors in the model: GDP per capita, educational infrastructure per capita, national inequality (GINI), civil liberties, nonviolent action, country programmable aid per capita, official development assistance per capita, with the outcome variable: the strength of CP.

As predicted, GDP per capita (r = .428, p < .001), inequality (r = .436, p < .001), civil liberties (r = .450, p < .001), and nonviolent action (r = .459, p < .001) were positively correlated with the strength of CP. As hypothesized, Official Development Assistance had a negative correlation with CP (r = -.332, p < .01).

Hypothesis 1 was supported, ODA is negatively correlated with the strength of CP. Hypothesis 2 was also supported, nonviolent action is positively correlated with the strength of CP.

Hierarchical Regressions

All four of the regression models significantly predicted the strength of CP. Altogether, the factors in model 1 (see Table 2) predicted 39.5% of the variance in strength of CP (p < .0001); in model 2 (see Table 2) the factors predicted 38.8% of the variance in the strength of CP (p < .0001).

Comparing models 1 and 2, we can see that GDP has a suppression effect on ODA per capita. In model 1, which includes GDP, the addition of ODA to the model does not produce a significant change in the predictive power of the model. However, in model 2, which does not include GDP, ODA predicted 4.6% of the variance in the strength of community psychology (p < .05). Similar to a finding by Hanitio and Perkins (2017), nonviolent grassroots action also predicted a significant amount of the variance in the strength of community studies in all models. This provides support for Hypothesis 3.

Assessing the Moderating Effects of Development Aid

To test the possibility of a significant moderator effect, both models were run again with the interaction of foreign aid x nonviolent action entered in the final step (see Table 3, Models 3 and 4). The moderator effect is present when the interaction term between the predictor and moderator is significant.

These models that included the interaction effect significantly predicted the variance of the strength of CP. Model 3 predicted 42.7% of the variance (p < .0001) and model 4 predicted 43.1% of the variance (p < .0001).

In model 3, which included GDP per capita as a control variable, the control variables accounted for 32.4% of the variance in the first step. In the second step, civil liberties and nonviolent action accounted for a significant addition of 9.8% of the variance. In the third step, ODA accounted for a non-significant addition of 2.7% of the variance. In the final step, the ODA-nonviolent action interaction accounted for a significant addition of 3.8%. The interaction of foreign aid and nonviolent action is significant (β = -.259, p < .05). Model 4 replicated model 3, except that GDP was excluded. Comparing models 3 and 4, GDP is shown to again have a suppression effect on ODA. In model 4, ODA significantly predicts 4.6% of the variance in the strength of CP.

In support of Hypothesis 4, foreign aid was found to significantly moderate the relationship between nonviolent grassroots action and the strength of CP. Overall, these results indicate that foreign aid has a direct influence on the strength of CP beyond what can be accounted for by nonviolent action, and also moderates the relation between nonviolent action and the strength of CP.

To further examine the interaction effects that emerged, we plotted the simple slopes of the nonviolent action-strength of CP linkage at 1 SD below the mean and 1 SD above the mean of foreign aid. We also tested whether each slope was statistically significant. As shown in Figure 2, the results matched the predicted pattern: the nonviolent action-strength of CP linkage exists in the low aid condition (simple slope = 3.25, p = .0019), but was not found to be significant in the high aid condition (simple slope = -1.14, p = .49). Thus, Hypothesis 4 was fully supported. Specifically, when foreign aid levels are low, those countries with high grassroots activism levels developed CP much more than those with low grassroots activism levels. However, when the foreign aid levels are high, there are no differences between a country with high nonviolent grassroots presence and a country with low nonviolent grassroots presence.

Discussion

The results indicate support for Hypothesis 1, Hypothesis 2, mixed support for Hypothesis 3 and support for Hypothesis 4. Development aid was negatively related to the strength of CP. Mirroring findings from Hanitio and Perkins (2017), nonviolent action had a positive relationship with the strength of CP in the aid-receiving countries in our sample.

Development aid was also shown to be a significant factor in predicting the strength of CP in three of four multivariate models tested (models 2, 3 and 4). The exceptional case (Model 1) involved predicting CP when GDP per capita was included as a control variable. This appears to be due to multicollinearity between GDP per capita and (less) foreign aid received per capita; along with the positive effect of GDP on strength of CP, which is consistent with Hanitio and Perkins' (2017) broader finding that Human Development Index is a strong predictor of CP across 91countries (including wealthier OECD nations). Thus GDP per capita appears to have a suppression effect on the influence of development aid in the models predicting CP.

The results from this study showed a significant, positive relationship between nonviolent activism and the strength of CP, and a significant, negative relationship between foreign aid and the strength of CP. More importantly, foreign aid was found to moderate the relationship between nonviolent action and the strength of CP as specified in Hypothesis 4. Hypotheses 1 and 2 support and confirm the findings of Hanitio and Perkins (2017). Hypothesis 4 extends those findings and investigates the interaction between the primary predictor variables of interest in both studies.

The results of the bivariate correlations and multiple regression models support dependency theory in that foreign development aid received is “negatively” related to the strength of applied community studies fields. While development aid aims to build human capital in the aid-receiving countries, these results indicate that development aid may, perhaps unintentionally, supplant or discourage at least a certain, critical form of human capital—i.e., “indigenous” applied community research training and professional resources. This relationship is further complicated when we consider the interaction between aid and nonviolent action. When aid is high, the link between nonviolent action and strength of CP becomes non-significant. When aid is low, however, nonviolent action is strongly related to the strength of CP.

The results support and extend for aid receiving countries another key finding of Hanitio and Perkins (2017)—that a history of non-violent political activism is a significant positive predictor of the strength of CP. This might have provided some hope that even if dependency or other pressures inhibit community studies, grassroots activity can be a catalyst for their emergence and growth. However, given the interaction effect, we observe that nonviolent action only has this relationship in countries with lower levels of aid.

An interesting finding which is different from the Hanitio and Perkins (2017) is that in all of the models in this study income inequality was significantly positively related to the strength of CP. Hanitio and Perkins (2017) included both aid receiving and donor countries. Thus, it is possible that income inequality matters in this case specifically because we focus here on the aid receiving countries.

These results lead to some important implications for CP and grassroots development. The findings confirm and extend previous studies, which found a relationship between nonviolent grassroots activism and the strength of CP (Hanitio & Perkins, 2017; Reich et al., 2007). The present results are also consistent with other studies that examined the effectiveness of grassroots action in a local community context (Finsterbusch & Van Wicklin, 1987; Kellogg, 2012; Uvin & Miller, 1996). From these studies and our theoretical model, we can speculate that countries with high levels of grassroots activity may have greater needs for CP training and resources, and thus, expand the practice of CP compared to countries with a relative lack of demand for those studies.

Implications and Future Research

Overall, our results indicate that even when controlling for educational infrastructure, GDP per capita, income inequality, civil liberties, and nonviolent activism, development aid still has a negative and significant relationship with the strength of CP, except when controlling for GDP per capita due to its shared variance with that field. While some dependency theorists might advocate reducing or even ending development aid, we are suggesting instead that aid officials reconsider the mechanisms of aid so that the dual goals of community and human capital development can be better realized.

Further qualitative research is needed to determine the ways in which development aid could be better employed in order to support human capital in aid receiving countries, and the ways in which development aid is currently managed and how that impacts applied community studies fields. There are many possibilities. Foreign aid could have major impacts on the development of academic applied community studies in a few major ways: (1) whether aid is used directly to financially support tertiary education including in these fields and/or (2) whether the projects and organizations that are awarded development aid sustainably employ enough local expertise (researchers and students) that there is a demand for professional training.

Given the negative relationship between foreign development aid and the strength of CP, as well as previous literature on this relationship:

• It is possible that development aid is being used for development projects that do not involve local researchers or local students of applied community studies fields, and thus there is not a demand for this professional training created by development aid.-
• It is possible that development projects do not hire local experts at the same frequency, nor pay them the same as foreign counterparts and so there is little demand to pursue these fields as persons in aid-receiving countries.
• It is possible that development aid is granted largely to international non-profits rather than local community organizations, and so the community development expertise is entrusted to foreigners who work outside of the country and gain credentials outside of the country, thus not creating a demand for these fields.
• It is possible that the local academy and development projects are disconnected for other reasons, such as a lack of historical support for academic disciplines most likely to develop applied community studies.
• It is possible that the presence of more development projects means there are more spaces outside of the academy for applied community researchers to learn, and so human capital may be created but not in academic institutions nor leading to academic publication. And it is possible that each of the above reasons apply, but perhaps differently in different countries.
• Finally, it seems likely that countries receiving less or no aid are forced to develop and use their own indigenous community planning, research and evaluation resources due to the lack of external supply of those.

We recommend further research on these possibilities to more clearly understand the negative relationship between development aid and the strength of community studies fields. Only then can we know what both aid providers and receiving countries can do to support the growth of those fields. As we noted, some international development organization policies attempt to address some of the issues mentioned above, particularly around the issue of involving local organizations and local talent in development interventions. For example, the European Structural and Investment Funds support human capital development, including at the graduate and professional levels, through its European Social Fund. Although its focus is on lower-income regions of Europe, the same approach could be taken with development aid to the even poorer countries we studied. In addition, the European Commission's Directorate-General for International Cooperation and Development (DG DEVCO) has adopted the following key principles of development effectiveness: country ownership, transparency and accountability, focus on results, and inclusive development partnerships. The DG DEVCO has so far produced: a Joint Programming initiative to improve partnerships between the DG DEVCO and local organizations; a Public-Private engagement strategy to increase collaboration between local private sector actors and the DG DEVCO; and the European Union Results Framework to increase the transparency of aid funded projects (European Commission, 2019). These steps signify that there is progress towards an implementation of aid that also strengthens the local community. However, questions still remain about the level of involvement of local participants in the development processes. Further research would help to clarify how and whether local academics are involved in these processes, as well as the mechanisms by which these initiatives impact the development of the field of CP.

Our results suggest that development aid is likely a stronger predictor of the strength of CP than nonviolent action. Countries that received high levels of aid saw less development in CP, regardless of the strength of nonviolent action. Yet, among countries that received low levels of aid, those with more grassroots activity were more likely to have an established CP discipline than those with less grassroots activity. However, the results in support of the moderating role of foreign aid on the relationship between nonviolent action and CP suggest that there is likely no standard, universal answer to increasing the strength of CP. The moderating effect we found calls into question the problems of the current aid system. We realize that there is a gap between the donors' goals and local outcomes and suggest strategies that focus on local ownership and find an effective way to tie the foreign resources with the needs and goals of local grassroots organizations.

In addition, we offer a practical implication for large donors, CP experts, grassroots organizations, and developing countries. We emphasize the importance of understanding the antecedents of CP before anything else. With this basic understanding of the underlying relationships, donors and aid recipients can work toward creating a collaborative environment with an emphasis on local ownership that will expand CP and potentially other applied community studies disciplines. Large donors should recognize the power of nonviolent grassroots action as a predictor of the strength of CP, which may be a crucial source of human capital development and professional training. Partnerships with community groups and utilization of local grassroots leaders may even be a possible requirement for large development projects.

Limitations and Strengths of the Study

The GDACS study uses coded ratings as proxies for the development of applied community studies fields, which means we are limited in that we can report only the information found on the internet and only consider certain indicators—such as programs, courses, publications, conferences and organizations—of formalized, academic fields of applied community studies. This is perhaps the most important point to acknowledge: our results and thus conclusions do not necessarily apply to community research conducted by organizational staff or community volunteers who have not been university-trained or do not publish or present at conferences. As important as informal training of nonprofessional community researchers is, it would be difficult to determine what effect development aid may have on those fields. We must also note that the 0-10 estimates of field strength in a given country are admittedly rough. The least sensitive part of the strength of field scale, however, is at the high end (e.g., the U.S., U.K., Canada, Italy, and Spain all having the same maximum value of 10 for CP, despite their differences in number of programs, journals, conferences, and CP students, faculty and professionals). No country analyzed here had the resources to receive a maximum strength of CP value of 10. In fact, there were many more countries for which we could find no evidence of CP, which is a much smaller field than the others included in the GDACS project. A final limitation is the fact that countries were coded over a five-year period from 2013 to 2018. We mitigated this by adding values as we learned of new conferences, courses, or publications in a given country. A validation study of the strength of discipline scale is currently in progress.

There are also several strengths of the study to note. This is the first study that attempts to look at the global development of disciplines focused on addressing problems at the local community level, and includes most of the largest countries that receive foreign aid. The dataset uses standard and generally reliable social and economic indicators and large-sample international surveys aggregated to the national level along with an effort to quantify the strength of each discipline in each country, which allows us to consider the empirical relationship between those macro indicators and elusive estimates of the development of applied academic fields. Additionally, the focus on academic research makes countries more comparable because we can hold a set of common assumptions even with slight differences in definition. The focus on professional training also shows how fields have become institutionalized, mostly in public universities, and so implies that these structures will have longevity, as opposed to grassroots community research which is valued, but harder to measure and may not be institutionalized and so less supported and unstable.

Conclusions

At the country level, development aid is negatively correlated, and nonviolent action is positively correlated with the strength of CP in a sample of countries that receive foreign aid. Additionally, development aid explains the variance in the strength of CP over and above other country level variables when GDP per capita, which is strongly positively related to CP, is not included as a control variable. Finally, development aid has a moderating effect on nonviolent action, whereby in countries with lower aid there is a stronger positive relationship between nonviolent action and the strength of CP. We suggest that the discrepancy between the capacity building aims of development aid and the strength of these fields might be tied to the way that aid and local ownership relate at the country level. Further research, particularly qualitative research, should be done to determine the mechanisms that drive these statistical relationships. This inquiry adds to our understanding of the global growth of CP, and further inquiry into the growth and distribution of other fields would help us better understand the ways that applied community studies might be leveraged in development efforts in aid receiving countries.