ENVIRONMENTAL EDUCATION: FACTORS BEHIND CURRICULUM ADOPTION SARAH L. HOADLEY

Transcription

1 ENVIRONMENTAL EDUCATION: FACTORS BEHIND CURRICULUM ADOPTION By SARAH L. HOADLEY A thesis submitted in partial fulfillment of the requirements for the degree of MASTERS OF SOCIOLOGY WASHINGTON STATE UNIVERSITY Department of Sociology AUGUST 2007

2 To the Faculty of Washington State University: The members of the Committee appointed to examine the thesis of Sarah L. Hoadley find it satisfactory and recommend that it be accepted. Chair ii

3 ENVIRONMENTAL EDUCATION: FACTORS BEHIND CURRICULUM ADOPTION Abstract by Sarah L. Hoadley, M.A. Washington State University AUGUST 2007 Chair: Nella VanDyke Questions surrounding environmental issues and particularly global warming are a hot topic in the popular media. In this study, I predict factors that influence U.S. states incorporation of environmental curriculum utilizing state theoretical perspectives. While many scientists, academics and concerned citizens acknowledge the need for environmental curriculum, few researchers have explored the prevalence of its availability or the factors behind its adoption. The current research adds to literature on the state by focusing on education as a form of policy. It also expands the sociology of education literature by utilizing state theorizing when examining curriculum adoption and evaluating a subject not previously explored. My findings suggest that business interests, environmental NGOs, political actors and economic indicators are four factors affecting curriculum decisions of environmental education. These findings expand state theorizing and show support for business dominance, Marxist and state-centered theoretical approaches when examining the issue of environmental curriculum adoption. iii

4 TABLE OF CONTENTS ABSTRACT. iii INTRODUCTION 1 LITERATURE REVIEW. 3 BUSINESS DOMINANCE PERSPECTIVES.. 4 STATE-CENTERED PERSPECTIVES 8 DATA AND METHODS RESULTS. 17 DISCUSSION CONCLUSION. 24 TABLES TABLE 1, VARIABLE AND DATA SOURCE 25 TABLE 2, DESCRIPTIVE STATISTICS.. 26 TABLE 3, LOGISTIC REGRESSION TABLE 4, LOGISTIC REGRESSION 28 WORKS CITED. 29 APPENDIX CODING SHEET. 32 CODE BOOK 34 iv

5 Introduction Education acts as a socializing agent for individuals in society. It both teaches individuals how to interact and maintain themselves within the capitalist structure and legitimizes subject areas of most importance and need in our society. Environmental education is an important area of focus, because global society is currently witnessing an alarming number of destructive outcomes caused by human environmental degradation. For example, thousands of plant and animal species have been lost due in part to the human encroachment on the natural habitat. However, U.S. states vary if they incorporate environmental education. Although political scientists, sociologists, and education researchers look at how business and politics affect state policy, few researchers have focused on factors influencing the specific content of curriculum. I propose that education policy makers are influenced by economic interests as well as constituency groups. Specifically, my research seeks to answer the following questions: What factors contribute to the adoption of environmental education curriculum? Also, do corporate and political actors play a role in the specific content of curriculum? Political scientists and sociologists vary on the extent to which they think business interests influence state policy. Business dominance and Marxist theorists (Block 1987; Domhoff 1983, 1990; Mills 1956; Useem 1984) suggest that political actors will try to satisfy business interests while formulating social policy because their very position relies on business to maintain a healthy economy. Fred Block (1987:13) suggests that capitalists attempt to assure responsiveness by the state through various means including campaign contributions, lobbying activities, and favors to politicians and civil servants. State actors receive something from the capitalists and in turn provide capitalist friendly 1

6 policy. Utilizing Marxist and business dominance theories of the state when looking at education policy, I predict that states structure education curriculum favorable to the demands of corporate interests. Other state theorists disagree about the extent of corporate influence, instead suggesting that policy makers have a great deal of autonomy (Hooks 1993; Skocpol 1980). State managers rely on citizen support for elected seats in office and therefore these civil servants must to some degree cater to the demands of the constituency. Although the state actors may take corporate interests into account, these scholars suggest that community members and social movements also play a role influencing policy. Some education researchers echo this state-centered ideology of multiple actors influencing education decisions. For example, Richard Arum (2000:395) looks at the organization of the school community and contends that a broader view should be recognized to include state regulation, professional associations, and market institutions as institutional forces. This state-centered perspective provides evidence suggesting that multiple mechanisms affect school policy decisions. However, few education researchers investigate economic, political and community influences on specific curriculum topics. My research adds to this literature by exploring the conditions that contribute to the incorporation of environmental education in the state mandated public school curriculum. In order to broaden the understanding of both state and education research I examine why states have varying environmental education requirements and what factors influence the adoption of such curriculum. I do so by performing a cross-sectional content analysis of each state s public education curriculum content standards, coupled 2

7 with Census Bureau information, a list of environmental corporate criminals, information from the National Center for Educational Statistics and a listing of all environmental nongovernmental organizations headquarters in the U.S (referred to as NGO from here on). Literature Review State theorists and education scholars provide varying arguments regarding the degree to which corporations and the community influence education policy. Power elite and Marxist theorists argue that states provide education that is favorable to corporate interests (Block 1987; Domhoff 1983, 1990; Mills 1956; Useem 1984). From this perspective, environmental curriculum would be limited to a modest incorporation of environmental issues, because corporate polluters do not want states to encourage environmental activism among citizens. Consistent with this perspective, education researchers find evidence of corporate influence on both the organizational structure of schools and curriculum outcomes within the public education system (Apple 1996; Mitchell and Boyd 2001; Tyack 1974). State-centered theorists, however, focus on policy as a product of state actors attending to the demands of citizens and social movements, in addition to corporate interests (Skocpol 1980). In this case, states would adopt environmental curriculum if pressured by citizens and environmental NGOs. Similar to this perspective, education researchers consider education policy as a product of multiple interests, not just the sole interest of the corporate class. In the following pages I delve into these lines of research and outline the implications of utilizing a combination of state and state-centered perspectives to argue that 1.) corporate interests influence education policy and the specific content of curriculum 2.) states adoption of 3

8 environmental education will be determined by multiple factors 3.) states with excessive environmental degradation will be vested in adopting environmental education in order to improve the condition of the immediate environment. Business Dominance Perspectives State theorists disagree over the extent to which corporate interests influence the state. Power elite and Marxist theorists (Block 1987; Domhoff 1983, 1990; Mills 1956; Useem 1984) presume that the corporate class is highly influential in shaping state policy. Michael Useem (1984) looks at the organization of corporations and contends that corporate political activities represent a classwide principle of shared interest through the expansion of influence for informal lobbying, funding of candidates and access to governmental bodies. For power elite theorists, corporations now work on a united front. Domhoff (1990) similarly suggests that, due to the role of institutions and structures in capitalist society, industry has seen a dramatic rise in its influence on politics. My research extends this same theoretical approach by investigating the degree to which business, or in Domhoff s words the power elite, have influence on state curriculum content standards. Since manufacturing is a major cause of environmental degradation, it is in the interest of business or corporate polluters to limit the amount of information to the public concerning environmental matters. And while corporations cannot limit all access to such education, I propose that they do play a role in states incorporation of environmental curriculum into existing subject areas. In order for business to continue running profitably, corporate leaders work to influence policy which promotes capital gains because it benefits the business and state economic conditions. For example, if a manufacturing plant can produce more product with fewer 4

9 environmental regulations, then it would behoove business leaders to promote business interests over environmental concerns. Following this line of reasoning I expect states with more corporate polluters not to have environmental education requirements, because business interest would be focused strictly on economics. Environmental sociologists also address the concern that business will push for policy that represents capital gain. It is in the interest of business to promote corporate friendly policy because as capitalists, the goal is economic wealth. Such theorizing suggests that environmental conditions will reflect the extent of economic wealth achieved by each state (Alario and Freudenburg 2003; Dunlap and Mertig 1992; Fisher and Freudenburg 2004). If a state is successful in manufacturing, then the expectation is that environmental conditions would not be very favorable because of the sheer amount of pollution due to production. In line with environmental sociologists I suggest that business puts pressure on political leaders to adopt policy which is favorable to economic growth over environmental concerns. Marxist theorists (Block 1987; Gold, Lo, and Wright 1975; Poulantzas 1978) cast aside the notion of the one dimensional corporate class model, arguing that the state itself is set up to promote policy which addresses the long-term needs of further capital accumulation. Block (1987:8) suggests that the structure of the state acts in a way so the capacity of capitalism to rationalize itself is the outcome of a conflict among three sets of agents- the capitalist class, the managers of the state apparatus, and the working class. This perspective suggests that theoretical arguments limited to the direct influence of corporate class actors may be unrealistic. In many instances members of the class are either in direct competition with one another, which limits accumulation, or have limited capacity to look for long-term solutions and thus rely on short-term answers 5

10 for further gains. Thus, it is difficult for the corporate class as a unified body to rationalize reform activity. The corporate and working classes have influential power, but the structure of the state is such that it will act for further capital accumulation even if this is in direct opposition to corporate or citizen demands. Block (1987:22) states, Working-class pressures, for example, might lead to an expansion of educational resources available for the working class, but there is every likelihood that the content of the education will be geared to the needs of accumulation. This Marxist reasoning suggests the state structure has the power to determine policy and this will reflect decisions based on capital growth for the state. In line with such theorizing, the current research considers the possibility of the state structure rather than the corporate class as a defining link between political influence and education reform. I do so by measuring both political and economic indicators for each state. Why would corporate interests care about education curriculum? The corporate class cares about specific curriculum content simply because of the significant role that educational institutions perform for society. Education socializes youth and prepares them for the work environment. Michael Apple (1996) focuses on the political factors related to adopting a national curriculum. Apple (1996:25) concludes that unfortunately, rightist groups are indeed setting the political agenda in education. He refers to the rightist groups as a combination of business interests, the New Right and neoconservative intellectuals (27-28). Apple (1996:95) concludes the existing curriculum is never a neutral assemblage of knowledge it is always based on an assertion of cultural authority. In a capitalist society cultural authority represents the 6

11 perpetuation of economic accumulation. But this accumulation, in many instances, does not reflect environmental preservation. Apple (1996:96) suggests: The relationship between schooling and economic, political, and cultural power is not an afterthought. It is a constitutive part of the very being of schooling Understanding this requires that we see schools and the curricula, teaching, and evaluation that go on in them in ways that do not make invisible these connections between what we do as educators and the larger relations of power. Education researchers including Apple (1996) have demonstrated a corporate influence on education consistent with business dominance and Marxist theories of the state. For example, David Tyack (1974) utilizes a historical investigation of public urban schools and concludes that a reform took place in the early twentieth century which established a corporate organization of administration in public schools. The political actors involved in the reform were, according to Tyack (1974:137), leading business and professional men. A major function of this reform was to change public curriculum to fit the corporate class needs of a complacent and uniform work force. Michael Apple (1996:28) similarly contends that capitalists, aim at providing the educational conditions believed necessary for increasing international competitiveness, profit, and discipline. In addition, Mitchell and Boyd (2001) suggest that reading and mathematics curriculum within the U.S. has become more uniform in response to trends in globalization and external social and political processes. My research adds to the education literature theoretically by applying theories of the state when examining specific curriculum adoption. Connecting the education and political literature, I seek to understand the organizational influences that drive the curriculum content of environmental education. 7

12 Following Mitchell and Boyd s (2001) theoretical argument for reading and mathematics, I add to this body of literature by addressing this issue for environmental education. Few researchers examine the extent to which corporate interests shape the content of education curriculum. Based on the political, environmental and education research discussed thus far, I expect that states with higher numbers of corporate polluters and a business friendly atmosphere will not have environmental education. In line with this reasoning I expect states with Republican political control not to have environmental standards because generally speaking Republicans support pro-business activities. Business (and specifically manufacturing) is often a source of various forms of air and water pollution (Jorgenson 2006, forthcoming 2007). It behooves these organizations to keep this information out of public schools, because a negative response might mean a capital loss. The money that corporations generate also benefits the state economic structure, so even if it is the state structure and not business that is influencing curriculum, it would benefit all involved to go with decisions that promote capital. In addition to training workers, it might be that corporate interests want schools to restrict environmental education. Limiting citizen s knowledge of environmental issues limits the questioning of harmful practices and continues the steady flow of production. State-Centered Perspectives Other state theorists agree with Block s (1987) belief that state managers and the working class play a role in policy reform, but they disagree about the extent of influence the corporate class plays (Hooks 1993; Skocpol 1980). For state-centered theorists such as Skocpol (1980) state actors must also act on demands from community members and social movements rather than just the corporate class. Those in political power need to 8

13 provide legislation that acts in the interest of the constituency voting them into office. In certain circumstances, it behooves state organizations to implement anti-corporate or anti-working class policy because in the long run capital accumulation will still be attained even if a particular population experiences a negative short-term effect. This theorizing indicates that it is too limiting to simply see all state managers as agents who act solely for corporate interests. State agents are also influenced by various social movements and community members. If states are experiencing more environmental problems, then it is likely that community members and NGOs may try to influence state managers to incorporate more environmental education so positive action could be taken to help the states overall environmental conditions. More toxic waste could mean more environmental education because state actors are responding to community demands. Members of the community would care because they have a vested interest in places they lay stake to as home and they do not want to live in environmentally hazardous conditions. This ideology has been termed NIMBY, by the environmental justice scholars meaning, not in my back yard. An entire literature on Environmental Justice specifically looks at how communities respond to various kinds of environmental hazards (Jones and Rainey 2006; Petrie 2006; Stephenson 2005). Although community responses are not always successful at combating these negative outcomes, this area of literature does serve as an example of how important it is to consider measurements of community response when looking at all kinds of policy including education policy. I capture community interest by including measures of the number of environmental NGO headquarters for each state. A 9

14 state with several environmental organizations suggests a citizenry who are active, organized and environmentally conscious. Other education researchers echo the state-centered approach and investigate community factors associated with education policy. When discussing the concept of the school community, Arum (2000) points out that we should apply a broad definition of the school community. Amy Binder (2000, 2002) takes this approach looking at curriculum challenges of creationism and afrocentrism in three metropolis areas. Binder (2000, 2002) concludes, similar to what Skocpol (1980) suggested, that varying community responses to creationism and afrocentrism result in different levels of incorporation of curriculum. Likewise, social movement theory also predicts that social movement organizations will have an impact on policy (Amenta, Carruthers and Zylan 1992; Andrews 1997; Giugni, McAdam and Tilly 1999; McAdam 1999; McAdam and Su 2002). Taking this approach of applying a broad definition to the school community, I examine political, environmental and social factors that may be associated with curriculum reform. Data and Methods Utilizing individual states as the unit of analysis, I compiled a data set to investigate factors associated with the incorporation of environmental curriculum. The number of cases is 50 (curriculum data for Washington D.C. was not available throughout data collection due to web site construction, so I dropped this case from the analyses). The data collected are a cross-sectional content analysis of state curriculum content standards and data from the U.S. Census Bureau, a list of environmental 10

15 corporate criminals, information from the National Center for Educational Statistics and a listing of environmental NGO headquarters in the U.S., as I describe in more detail below (see Table 1 for variables and data sources, see Table 2 for descriptive statistics). In order to explore the relationship between state curriculum and contextual factors, I downloaded the curriculum content standards from each state and developed a coding sheet (see Appendix). The independent variables are measures of the political and environmental context including environmental performance, environmental mobilization, political, economic and school indicators. Dependent Variables I performed a cross-sectional content analysis on standards of curriculum for environmental education to construct the dependent variables measuring environmental education. I collected this information from each state s Department of Education web sites in the spring of 2006 from early January through late May. I then coded the curriculum content standards (see coding sheet Appendix). The first dependent variable identifies whether the state specifically labeled environmental education in the content standards. I operationalized this variable with a dichotomous measure coded 1 if the state labeled the curriculum as environmental, or 0 if not. I obtained this coding by performing a key word search on the home page looking for environmental education. This variable serves as my first indicator of the state s overall interest in environmental education. It was clear from the pilot sample that states discussed issues like conservation and interdependence in existing subject areas even if they did not label it specifically as environmental education. In fact, the data indicate only 17 of the 50 states (34%) labeled curriculum as environmental education. Therefore 11

16 I also evaluated each subject areas (science, social studies, and health) main contents to determine if any environmental education concepts existed within this curriculum that was not labeled. I defined a second dependent variable measuring environmental education by coding whether the state incorporates environmental education beginning in Kindergarten even if not specifically labeled. Scientists and education researchers have both addressed the importance of providing this type of curriculum from an early age (Hudson 2001; Orr 2002; Balsamo 2005). I operationalized this variable with a dichotomous measure coded 1 if a state had such curriculum beginning in Kindergarten, and 0 if not. Twenty-nine or 58% of the states require environmental education at this age even though they do not necessarily label the curriculum as environmental education. Such a measure represents strong environmental education, because it shows the state s willingness and understanding of environmental concerns and a need for action. Independent Variables To explore the various factors influencing education policy, I include five sets of independent variables. The independent variables include measures of the social, political and educational context that may influence environmental curriculum for each state. I choose 1994 as my data collection year, because the earliest year curriculum was revised was 1995 (Alabama, Colorado, and Montana) and I wanted my independent variables to be measured before this date. Due to the timing of Presidential elections, I measured the popular vote cast for President in Specifically, I focus on five sets of independent variables that center on- environmental performance, environmental mobilization, political, economic, and educational indicators (see Table 1 for variables and data sources). 12

17 The first set of variables looks specifically at environmental performance in each state. My first measure is the number of corporate polluters found in each state. The count of corporate polluters came from the corporate polluters list compiled by Russell Mokhiber, looking at the top 100 corporate criminals of the 1990s. This list includes 14 categories of crime, but for this project only environmental offenses were used, which resulted in an N=38. I tallied these 38 cases according to the number of offenses for each state ( The second variable is the number of hazardous waste sites in each state. This measure was constructed by dividing the number of hazardous waste sites listed on the National Priority List by the total acreage of state land. The third variable is the amount of toxic release in each state. This measure was constructed by evaluating how much toxic release was emitted divided by the total state acreage. I collected both the toxic release and hazardous waste site information from the Census Bureau Statistical Abstract (U.S. Bureau of the Census 1995, 1997). All of these variables, in addition to the percentage of employees concentrated in manufacturing discussed below help test my competing hypotheses. The first outcome might be that higher percentages of toxic release and hazardous waste sites might have a negative effect on curriculum, which would reflect a state response. This supports my business dominance and Marxist (first) hypothesis that business and the state structure influence education. If business dominance and Marxist theories of the state are correct and corporate interests do influence state policy then I would expect a high concentration of corporate polluters and toxic releases to coincide with not having environmental education. On the other hand, political actors might experience pressure from citizens and social movements to clean up the environmental conditions, which 13

18 supports my state-centered (second) hypothesis that constituency concerns influence education decisions. Political actors would feel compelled to act on citizen demands because they are voted into office by the constituency. The second set of variables capture environmental mobilization which illustrates the institutional factors associated with environmentalism. The environmental mobilization measure includes a count of how many environmental NGO headquarters are in each state, demonstrating community involvement and constituency concerns. The count of environmental NGO headquarters came from The Encyclopedia of Associations, vol The state-centered hypothesis declares: states with pollution and environmental organizations will have environmental education requirements. Constituents would be organized in states with environmental NGO headquarters and the political actors may feel pressure to fill the demands of the voters. The third set of independent variables focus on political representation because we would expect liberals to support pro-environmental curriculum more than conservatives. These measures are labeled political indicators and all four were measured dichotomously. The first variable equals 1 if the majority of the state voted for the Democratic candidate for president, and 0 if not. The second variable equals 1 if the state elected in a Democrat Governor, or 0 if not. The third variable equals 1 if the state had both a majority Democrat in the Senate and the House of Representatives, and 0 if not. The final variable equals 1 if the state had a mixture of Democrat and Republican representation in the Senate and the House of Representatives, and a 0 if not. All four variables were constructed from the Census Bureau Statistical Abstract (U.S. Bureau of the Census 1997). 1 Special thanks to Erik Johnson for sharing these data. 14

19 The fourth set of independent variables gives an idea of the type of reliance each state has for industry that may be causing environmental degradation. I include two measures of a state s wealth. The first is a ratio of the state s expenditures to the revenues and the second is a percentage of how many of the community members are living below the poverty line. The wealth of a state helps illustrate the potential for business to influence policy. If a state is relatively poor, it is more likely to rely on business to maintain a healthy economy, thus allowing more business influence on policy. In addition, I include a measure of the percentage of the work eligible population in the state working in manufacturing. Such a measure could have one of two effects. If this measure has a negative effect on environmental education, it may be because the citizenry rely on manufacturing for income so they influence policy to maintain business and state profit. If the size of the manufacturing sector has a positive effect on environmental education, it may be because the citizenry sees the pollution, feels the effects, and takes action to promote environmental regulations because they want a healthy environment to raise their families and enjoy life. All three measures of state industry reliance come from the Census Bureau Statistical Abstract (U.S. Bureau of the Census 1995). The final variable is the number of schools in a state turned into a dummy variable labeled as a 1 if a state has an above average number of schools (above average was 1,954 schools or above) and 0 if not (National Center for Educational Statistics 1994). This particular variable acts as a control variable for my analyses. I control for the number of schools in a state to focus on political and economic factors for curriculum adoption and comprehensiveness in relation to the size of the educational system. I used 15

20 a dichotomous measure because the number of schools was collinear with the number of corporate polluters. In addition to the number of schools as a control variable, the number of hazardous waste sites and toxic releases also act as control variables in my analyses. It was important for me to include such measures because these environmental indicators reflect if environmental pollution is an issue in a state. Although these three control variables are not theoretically driven, they do help to control for the environmental performance and educational institutional size for the analyses. I began my analyses by conducting diagnostics to make certain my data did not violate statistical assumptions. Findings indicated that some variables were correlated with population, so instead of including this as a control variable I transformed variables and divided by the total population of the state to include this control. Both the number of corporate polluters and the number of NGOs include such a control. In addition, I also transform several variables to control for state size by dividing by the total state acreage. Both the number of hazardous waste sites and toxic releases were transformed in this manner. Thus, I was able to control for both population and state size by transforming variables. Other collinearity problems were solved by transforming variables as described above. Logistic Regression To test my hypotheses, I conducted logistic regression on my two dependent variables, both of which are measured as dummy variables. The purpose of these analyses was to predict what factors account for the incorporation of environmental curriculum. The first model predicts how a state performs environmentally by evaluating if industry pollutes and to what extent. The second model illustrates the effects of 16

21 community presence. The third model evaluates the relevance of political parties. And the fourth and fifth models look at the multiple factors together. Results I began by investigating the first dependent variable, which measures whether a state has curriculum labeled as environmental or not. Model 1 of Table 3 examines the extent to which corporate polluters, hazardous waste sites and toxic releases influence the adoption of environmental education. The first theoretical assumption suggests that having corporate polluters will negatively influence the adoption of environmental education. The results indicate that having corporate polluters is significant (p<.10), but rather than it having a negative effect as business dominance and Marxist theories predict, having corporate polluters is positively associated with the inclusion of environmental education. Model 1 shows that each additional corporate polluter is associated with a 1.6 times greater likelihood that a state will adopt environmental education. This significance remains through Model 5, which shows little support for the business dominance and Marxist theorizing. This finding suggests that perhaps constituency as opposed to business concerns play a more significant role in curriculum adoption as state-centered theorists suggest. Model 2 tests the second theoretical assumption, which suggests that having environmental NGOs would have a positive influence on environmental education and reflects environmental mobilization. The results indicate a non-significant effect, suggesting little support for the state-centered approach that constituency concerns 17

22 influence the adoption of environmental standards. This finding is puzzling when considering the finding discussed above. Future research would evaluate this in more depth and consider alternative variables that could represent varying aspects of a constituency. Model 3 examines the influence of political indicators. The results show an interesting outcome, which suggests that local political actors play a significant role in education policy. Having a Democratic Governor has a significant positive effect on environmental curriculum. Including the other indicators (Models 4 & 5) increases the significance of the Democratic Governor to the p<.05 level. Having a Democratic Governor is associated with an 8.49 times greater likelihood that a state will adopt environmental education. Such results indicate that while it is important to have a Democratic Governor for a state to adopt environmental curriculum, it appears as though constituency groups are not a significant factor in whether a state has any explicitly distinguished form of environmental education. All three economic indicators in Model 5 also show non-significance. In addition, including the number of schools in Model 5 shows a non-significant effect. The above results identify the factors associated with states adopting environmental curriculum. Next I explore how the theoretical assumptions relate to the comprehensiveness of mandated environmental education (the dependent variable is coded as a 1 if the state mandates environmental material starting in Kindergarten whether labeled or not, and 0 if not). Model 1 of Table 4 examines the extent to which having corporate polluters, hazardous waste sites and toxic releases influence the comprehensiveness of environmental education beginning in Kindergarten. The first 18

23 theoretical assumption suggests that having corporate polluters negatively impacts environmental education. The results indicate that having corporate polluters is not significant until Model 5, when all other variables are included in the analyses (significant p<.10). These results are consistent with both business dominance and Marxist theorizing, which would predict a negative effect for this variable because if business is polluting then business interests would not want to promote environmental education. Having corporate polluters is associated with a times less likelihood that a state will include environmental education in existing subject areas beginning in Kindergarten whether labeled or not. Model 2 tests the second theoretical assumption which suggests that having environmental NGOs will positively influence comprehensive environmental education, and reflects environmental mobilization. The results indicate a significant effect (p<.05), lending support to the state-centered approach that constituency concerns influence comprehensive environmental education. Each additional environmental NGO headquarter in a state is associated with a 3.97 times greater likelihood of that state providing environmental education starting in Kindergarten whether labeled or not. This measure retains significance in the saturated model. Thus, having more environmental NGO headquarters in a state is significantly related to environmental education policies. Model 3 examines the influence of political indicators. Having both a Democratic Congress and Mixed Congress is significant at the p<.10 level. However, these variables become non-significant as other variables are added to the model. The Democratic Governor variable becomes significant in Model 4 first at the p<.10 level and then in Model 5 at the p<.05 level. Having a Democratic Governor is associated with a

24 times greater likelihood of that state providing environmental education than states with governors from other political parties. Such results are consistent with my expectation that having both environmental NGO headquarters and liberal political actors influences the comprehensiveness of environmental education. In addition, in Model 5 having a population below the poverty line is also significant at the p<.05 level. A state has a 1.47 greater likelihood of having environmental education in Kindergarten whether labeled or not, if some of the state population is below the poverty line. This is counter to my predictions, because I would expect a state with financial concerns to be focused on economics which generally favors corporate policy over environmental concerns. Finally, Model 5 shows that the effect of the number of schools in a state is nonsignificant. The number of schools in a state does not significantly affect the comprehensiveness of environmental education beginning in Kindergarten. Discussion My research adds to both state literature and sociology of education by evaluating the factors of environmental curriculum adoption and comprehensiveness. My findings suggest that environmental performance (especially if a state has corporate polluters), environmental mobilization, political actors, and an economic indicator influence a state s decision to adopt and include comprehensiveness of this education in existing subjects. More specifically and in line with specific aspects of business dominance, Marxist and state-centered perspectives, I find that business, the state structure, and constituency demands play a role in a state s adoption of and comprehensiveness of environmental education. Past state research focused on various forms of reform, but 20

25 never specifically education policy. Education researchers consider how state actors, corporate influence and constituency demands influence education reform, but seldom consider effects on specific curriculum. While the business dominance perspective solely focuses on business in relation to reform, the Marxist perspective also includes the state and the working class as influential powers. According to business dominance and Marxist theorizing, states with a conservative political representation and a larger number of corporate polluters are apt not to have environmental education. A political system dominated by conservatives would suggest a business friendly environment to advance the capitalist free market economic structure. I find mixed support for both the business dominance and Marxist perspectives. Table 3 indicates a positive effect for environmental education which suggests the opposite of what business dominance and Marxist perspectives propose. On the other hand Table 4 indicates a negative effect and is suggesting a relationship between business and policy consistent with business dominance and Marxist theorizing. One implication of this finding is that it appears as though business does have some sort of influence on the comprehensiveness of environmental content standards, but not for the adoption of such curriculum. State actors must perform a balancing act and meet the demands of citizens and NGOs as the state-centered perspective suggests. But state actors must also meet the demands of corporations as the business dominance and Marxist perspectives suggest. The current research fills a gap in both the state and education literature, by focusing specifically on environmental education. A possible weakness in my analyses is the lack of ability to look at the specific actors involved in curriculum reform. It is possible that corporate influence has additional relationships to education not currently 21

26 measured. As an example, perhaps a relationship exists between corporate polluters and a state s Governor. It could be that corporate polluters serve on committees that have influence on state politics. Such an interaction would suggest that corporate influence would show up in the political indicators rather than in corporate measures. A possible way to get at such a relationship in future research is to incorporate network analysis to investigate how the corporate class is involved in each state s reform. To further the understanding of constituency demands on state actors for education decisions, I utilize the second theoretical perspective incorporating statecentered reasoning. According to this perspective, states with a Democratic political representation, more environmental NGO headquarters, and those experiencing environmental pollution will be more prone to the incorporation of environmental standards. If citizens are aware of environmental pollution, more pressure will be placed on political actors to incorporate environmental education. My findings show mixed results. While constituency concerns appear to be significant for the second dependent variable looking at the comprehensiveness of environmental education, they are not significant when looking at the first dependent variable based on adoption of environmental education. Perhaps it is easier for NGOs to have influence on specific education projects rather than having an effect on the adoption of curriculum. In many states NGOs provide information and resources for specific community projects and these include an environmental focus. Perhaps these mixed finding are a result of constituency concerns being addressed through programming in the schools rather than petitioning for curriculum adoption. 22

27 State curriculum standards are influenced by political actors and social movements. I find such a relationship by looking at a new area of curriculum. Utilizing state theory shows that state actors affect education decisions. As current environmental conditions continue to degrade, understanding the influential forces that contribute to environmental education will be important at the local, state and national level. Some education scholars have gone as far as speculating that how we plan for environmental education now will have substantial effects on the environmental state in the future (Hudson 2001; Orr 2002). While the current research incorporates both environmental performance and mobilization indicators, it does so while also recognizing that additional variables might be useful in explaining further the factors behind environmental education curriculum adoption and comprehensiveness. With incorporation of supplementary indicators, such as environmental standards, the current research could be expanded upon. For example, it might be helpful in future research to include other environmental mobilization indicators like a time series investigation of a state s environmental legislation. Extending the analysis to include a time series dimension would allow for tracking the individual state s circumstances surrounding environmental legislation in relation to environmental education. As a result of the current findings, I suggest that evaluating the incorporation of environmental education in public schools is aided by combining aspects of business dominance, Marxist and state-centered theorizing. However, to better understand the corporate and community influence on education, future research could benefit from looking at specific actors involved in curriculum reform. 23

28 Conclusion The results indicate that political actors, specifically a state s Governor, environmental NGO headquarters, corporate polluters and the economic situation of a state s population have an influence on environmental curriculum and comprehensiveness. The analyses reveal mixed findings, which indicate that further empirical testing is needed surrounding this subject matter. While my study explores a new area of research, to improve and expand the project I would like to look at a time series analysis following the states that have curriculum labeled as environmental education. Understanding the conditions of a state prior to and after the inception of environmental education could provide new insight into the social actors and organizations that influence education decisions. States environmental education standards are important for all in the global society to understand because the environment will continue to be a topic of debate and concern in the years to come. From a local level, empowering future citizenry with environmental knowledge is necessary information which can be incorporated into existing subject areas such as science and social studies. To reduce air pollution, provide clean and safe water supplies and encourage rather than destroy biological diversity, students must be given the tools and hands on experience to understand real world results. 24

29 TABLE 1: Variable and Data Source Category Variable Data Source Dependent Environmental Education labeled Department of Education websites Environmental Education in K Department of Education websites Independent Environmental Performance Indicators Number of corporate polluters Corporate Crime Reporter by Russell Mokhiber Number of hazardous waste sites 1995 Census Statistical Abstract Amount of toxic releases 1997 Census Statistical Abstract Environmental Mobilization Indicators Number of non-governmental organizations (NGOs) The Encyclopedia of Associations Political Indicators Majority popular vote for President 1997 Census Statistical Abstract Majority popular vote for Governor 1997 Census Statistical Abstract Control of House and Senate 1997 Census Statistical Abstract Economic Indicators Ratio of expenditures to revenue 1996 Census Statistical Abstract Percent below poverty line Percent of population in manufacturing 1997 Census Statistical Abstract 1995 Census Statistical Abstract School Indicators Number of schools in state National Center for Educational Statistics 25

30 TABLE 2: Descriptive Statistics Variables Range Mean/Freq Dependent Environmental Education labeled 0-1 0= 66% 1= 34% Environmental Education in K 0-1 0= 42% 1= 58% Independent Number of corporate polluters (1.39) Number of hazardous waste sites (26.40) Toxic releases (1,000 pounds) ,061 39,339 (43,730) Number of environmental NGO headquarters (10.25) Majority Democratic vote for President ' = 36% 1= 64% Democratic Governor 0-1 0= 64% 1=36% Democractic House and Senate 0-1 0= 86% 1= 14% Mixed House and Senate 0-1 0= 36% 1= 64% Ratio of expenditures to revenues (0.59) Percent below poverty line (3.93) Number of manufacturing employees 10-1, (368) Number of schools in state- dummy = 64% 1= 36% Note: Numbers in parentheses are standard deviation for non-dichotomous measures. N=50 26

31 TABLE 3: Logistic Regression of Environmental Education on Selected Independent Variables, 1994 Model 1 Model 2 Model 3 Model 4 Model 5 Top Corporate Polluters 0.474* 0.548** 0.569* (0.253) (0.283) (0.330) [1.606] [1.729] [1.767] Number of Hazardous Waste Sites- intesity ( ) ( ) ( ) [6.060] [2.020] [1.080] Number of Toxic Releases-intensity (0.975) (1.287) (1.500) [0.632] [0.386] [0.697] Number of Environmental NGO Headquarters (0.616) (0.850) (1.283) [2.044] [2.493] [1.406] Majority Vote for Democrat President (0.811) (1.059) (1.060) [1.486] [2.869] [2.253] Democrat Governor 1.155* 1.802** 2.138** (0.651) (0.808) (0.959) [3.175] [6.067] [8.487] Democrat Congress (1.381) (1.604) (1.898) [0.139] [0.108] [0.203] Mixed Congress * (0.899) (1.206) (1.213) [0.404] [0.131] [0.183] Ratio of Expenditures to Revenues (7.170) [5.753] Percent Below Poverty Line (0.146) [0.960] Percent Population in Manufacturing (9.830) [0.195] Number of Schools-dummy (1.287) [2.439] Constant ** (0.450) (0.393) (0.665) (0.767) (7.141) Pseudo R N=50 Notes: Standard errors are in parentheses; Odd ratios are in brackets. ***p<.01 **p<.05 *p<.10 27

32 TABLE 4: Logistic Regression of Environmental Education in Kindergarten on Selected Independent Variables, 1994 Model 1 Model 2 Model 3 Model 4 Model 5 Top Corporate Polluters * (0.212) (0.315) (0.428) [1.032] [0.740] [0.477] Number of Hazardous Waste Sites-intesity (76.349) (82.259) ( ) [1.020] [1.250] [5.130] Number of Toxic Releases-intensity (0.953) (1.063) (1.734) [2.103] [1.253] [11.360] Number of Environmental NGO Headquarters 1.378** 2.277** 3.513** (0.669) (0.983) (1.493) [3.967] [9.749] [33.533] Majority Vote for Democrat President * * (0.870) (1.000) (1.217) [0.282] [0.199] [0.109] Democrat Governor * 2.057** (0.663) (0.807) (1.036) [2.829] [4.276] [7.824] Democrat Congress 2.288* (1.250) (1.362) (1.614) [9.858] [8.112] [1.361] Mixed Congress 1.835* (1.000) (1.119) (1.386) [6.264] [5.510] [7.313] Ratio of Expenditures to Revenues (8.665) [ ] Percent Below Poverty Line 0.385** (0.159) [1.470] Percent Population in Manufacturing (9.348) [68.702] Number of Schools-dummy (1.361) [0.685] Constant (0.418) (0.354) (0.698) (0.783) (8.199) Pseudo R N=50 Notes: Standard errors are in parentheses; Odd ratios are in brackets. ***p<.01 **p<.05 *p<.10 28