WILLINGNESS TO PROTECT THE ENVIRONMENT

Reed Geertsen and Darren Hendrickson, Utah State University

This research analyzes the effects of various indicators of social position, environmental quality, and generational change on public willingness to finance local environmental protection. The study variables were assessed using identical measures in two statewide surveys of nonmetropolitan Utah conducted in 1973 and 1991. Data were analyzed through generational-aging-age comparisons and multiple regression models. The results of analyses revealed a negative aging effect from 1973 to 1991; however, no generational effects were detected for younger-, middle-, or older-age cohorts. Several indicators of social position and perceived environmental quality were found to be statistically associated with one's willingness to finance local environmental protection after controlling for the effects of study year. The implications of these findings are discussed with respect to the growing body of research on public attitudes toward the environment.

How do social conditions influence public willingness to protect the natural environment? Although several research studies have examined people's perceptions of environmental problems, many of these studies have yielded varied and contradictory outcomes. Some see environmentalism as a social movement that peaked in the 1970s. Others place more emphasis on environmental attitudes as a reflection of one's location in the life cycle or position in the social structure. At the same time many are speculating that Americans are less willing to protect the environment today than they were two decades ago? The major focus of the present study concerns longitudinal comparisons of data collected in 1973 and 1991 using identical measures of study variables in two statewide surveys of nonmetropolitan Utah. It was undertaken to determine the relative effects of generational change and various indicators of social position as well as perceived environmental quality on public willingness to finance local environmental protection.

The Utah population provides a unique social-cultural group for assessing environmental attitudes and social position. A predominant theme in the literature is that Judeo-Christian religions assume a mastery-over-nature perspective (Attfield, 1991; White, 1973). This ideology is rooted in the idea that humans can control the natural environmental. Empirical research has found that non-Judeo-Christians are more likely to be concerned about environmental issues than Judeo-Christians (Hand & Van Liere, 1984; Kanagy & Willits, 1993). Furthermore, Hand and Van Liere (1984) assert that the Mormon Church with headquarters in Utah is one of a few "conservative" denominations that shows a particularly strong inclination toward a mastery-over-nature perspective. This suggests possible differences between Mormons and non-Mormons in their support for environmental action.

The years 1973 and 1991 are useful points in time for investigating possible changes in public willingness to finance environmental action. Many studies in the literature point to generational differences in environmental concern. In writing about the U.S. environmental movement, Albrecht (1976, p. 154) concluded that "... environmentally related attitudes and behavior evident in the United States in the late 1960s and early 1970s border on the incredible." Erksine (1972) reported trend data indicating that Americans were seriously concerned about the state of the natural environment, perhaps at a level never before experienced. Other researchers have referred to the 1970s as the environmental decade (Dunlap, 1989; Mitchell, 1989).

Historically, the late 1960s and early 1970s exhibited a dramatic rise in concern for environmental issues. This nationwide increase in environmental awareness appears to be the result of several factors, including greater media attention to air and water pollution, widespread media coverage of the Santa Barbara oil spill in 1969, and the celebration of the first Earth Day in 1970 (Albrecht, 1976; Dunlap & Mertig, 1991). As Americans became increasingly concerned about environmental issues, both federal and state governments responded by creating a variety of legislation aimed at protecting the natural environment. Most notable at the federal level was the passage of the National Environmental Policy Act in 1969 and the creation of the Environmental Protection Agency in 1970 (Dunlap, 1991; Rosenbaum, 1990). The Utah State Legislature, responding to rising environmentalism, also created land-use legislation (Halliday, 1973). By the early 1970s, environmental issues had become an important concern of the American public.

During the mid to late 1970s, environmental awareness and activism began to decrease. The energy crisis of the 1970s seems to have diverted public attention away from environmental issues (Dunlap, 1991). At the same time there was a dramatic change in the nation's political leadership. A Democratic presidency, which appeared to be concerned about environmental issues, gave way to the less environmentally-sensitive Republican presidency. With this shift in political power came widespread efforts to reduce environmental regulation and a sharp decline in the creation of new environmental laws (Dunlap, 1991; Sale, 1993). Eventually, the Reagan neglect of the environment created a backlash with an apparent re-commitment to environmental issues among the American public. Dunlap (1989) speculates that the mid-1980s saw an increase in concern for environmental issues because many felt they could not trust the Reagan administration to sufficiently protect the environment.

By the late 1980s and the early 1990s, some investigators claim that the American public recaptured some of its earlier dedication to environmental issues. For example, Rosenbaum (1990, p. 25) has stated that "the strength of public support for environmental protection in the late 1980s, measured by public opinion polls, appears robust, widespread, and sometimes unprecedented." Similarly, Dunlap (1991) suggests that in recent years there has been an increase in the levels of public perceptions of environmental problems. Nevertheless, the pervasiveness of environmentalism in the early 1990s remains unclear.

Previous research has reported a connection between environmental concern and various indicators of social location such as education, political affiliation, and age. Initially, research identified the environmental movement with the middle and upper classes. Then, two separate studies of environmental attitudes narrowed the class influence to education (Buttel & Flinn, 1976; Costantini & Hanf, 1972). In a reexamination of the relationship between social class and environmental beliefs, Buttel and Flinn (1978, p. 445) concluded that education accounts for "virtually all of that part of the variance in environmental attitudes accounted for by social class indicators." More recent research has further directed attention toward the positive relationship between education and environmental concern, downplaying the role of social class. Jackson (1983) reported a positive association between educational attainment and willingness to pay for environmental protection. Mohai and Twight (1987) argued that education is second only to age in consistently predicting environmental attitudes. On the other hand, Howell and Laska (1992), analyzing longitudinal data collected at three time points during the 1980s, concluded that educational attainment was the most important predictor of environmentalism, even surpassing age.

Other early work in environmental sociology established a connection between political orientation and environmental attitudes (Dunlap, 1975; Tognacci, Weigel, Wideen & Vernon, 1972). However, some contradictory findings have been reported. Dillman and Christenson (1972) did not find a significant relationship between political party membership and environmental attitudes. Conversely, Buttel and Flinn (1976) have argued that political ideology, not political affiliation, provides a more meaningful predictor of environmental attitudes. The distinction between conservative and liberal orientations has been singled out by Buttel and Johnson (1978), while Samdahl and Robertson (1989) have tried to identify various categories of liberalism as potential determinants of environmental concern. In a review of the literature dealing with environmental concern, Van Liere and Dunlap (1980) hold that regardless of the measure of politics employed, the results are always found to be in the expected direction. That is to say that Republicans and conservatives tend to display lower levels of pro-environmental attitudes than Democrats and liberals.

Two studies have special relevance for the present research. The first reviewed all available studies prior to 1980 that addressed the relationship between environmental concern and five sociodemographic variables: age, social class, residence, political ideology, and sex. In this review, Van Liere and Dunlap (1980, p. 192) expressed confidence "...in concluding that younger, well-educated, and politically liberal persons tend to be more concerned about environmental quality than their older, less educated, and politically conservative counterparts." The second study was published in 1992 and represents a follow-up to the earlier work. In this study Jones and Dunlap (1992) tried to assess whether the social bases of environmental concern had changed over time. They concluded that the correlates of environmental concern had remained pretty much the same from 1973 to 1990.

Some studies of environmental concern included assessments of the perceived seriousness of various forms of pollution (cf. Jones & Dunlap, 1992; Van Liere and Dunlap, 1980). They found higher levels of concern among those who had noticed visible signs of air pollution or had become worried about contaminated water supplies. These studies suggest a probable link between perceptions of pollution and willingness to finance environmental protection. For example, a person who feels that air and water quality are deteriorating logically would seem to have added reason to support higher taxes for taking action to address environmental problems. In the medical literature, the perceived seriousness of a health problem has been found to be in important precondition for taking a health action to keep from being personally affected by the problem (Cockerham, 1997). The same outcome might be expected from a perception of environmental problems, particularly since air and water quality have important implications for health.

Finally, a number of studies have tried to determine and explain the relationship between age and a person's attitude toward the natural environment. Buttel (1979) was one of the first to report a negative relationship between age and pro-environmental attitudes in a midwestern state. Mohai and Twight (1987) replicated Buttel's study using national-level data and likewise concluded that age was a more consistent predictor of environmental attitudes than other positional variables. Other studies have reported similar findings (Honnold, 1981; Jones & Dunlap, 1992; Van Liere & Dunlap, 1980). Less well documented is the relationship between age and a person's willingness to finance environmental protection (Jackson, 1983). The questions of age and early socialization experiences and how they may operate in the formation of attitudes and/or public opinion has given rise to two opposing theoretical points of view. The present study was undertaken, in part, to determine which of these opposing points of view is supported by data collected in 1973 and 1991 in nonmetropolitan Utah.

The first of the two theoretical approaches examined in this study has been labeled the age-difference approach. The younger members of society are often perceived as agents of social change. Likewise, young adults tend to be viewed as more radical in their political orientation. Ryder (1965) contends that young adults are "old enough to participate," but "...not old enough to have become committed...." This lack of role obligation grants young adults the freedom to express diverse points of view, participate in social movements, and perceive numerous social issues as problematic. Essentially, young adults have less to lose when participating in "anti-establishment" social movements. On the other hand, older societal members become more rooted in family obligations and role relationships and thereby become more committed to the existing social structure and perpetuation of the status quo (Glenn, 1977; Hornback, 1974). Malkis and Grasmick (1977) were the first to suggest a connection between youthful detachment and pro-environmental attitudes including support for increased environmental spending. Key to this approach is the idea that the young are detached from many societal obligations because of their age. In summary, age-differences theory predicts a negative relationship between age and willingness to pay more taxes to protect the environment regardless of the particular historical circumstances affecting different age cohorts.

The second theoretical orientation examined in this study is cohort-differences or generational-change theory. A cohort is defined as people who have experienced similar life events or share something in common (Weeks, 1989). Unlike age-differences theory and its focus on detachment, cohort-difference theory focuses on socialization as an important dynamic for change. Cohort-differences theory argues that cohorts have differing perceptions because of the unique collective experiences occurring at particular stages in the life-cycle. Various theorists have argued that events occurring during a cohort's "formative" years leave lasting impressions that help shape attitudes and perceptions (Glenn, 1980; Mannheim, 1952; Ryder, 1965; Schuman & Rieger 1992).

Kanagy, Humphrey & Firebaugh (1994) maintain that part of the negative correlation between age and environmentalism in the 1970s is due to events such as the first Earth Day and media coverage of the Santa Barbara oil spill. They argue that exposure to various high-profile environmental issues among this cohort left a socialization imprint that should last throughout their life cycle. In other words, this particular cohort should continue to have higher levels of environmental concern as it ages and replaces older cohorts; and this cohort replacement should help diffuse pro-environmental attitudes throughout the population. Their findings suggest that cohort replacement may have contributed to the growth of environmentalism during the 1980s. In summary , cohort-differences theory predicts that middle-aged Americans in the 1990s will show stronger support for protecting the environment than their middle-aged counterparts did in the 1970s. Likewise, younger adults in the 1990s should show stronger support than their cohort counterparts from the previous generation due to socialization imprints at an early, impressionable age (under 18).

The 18-year span between the two study years in our data (1973-1991) provides a unique opportunity to test these two different interpretations of attitude formation and life-cycle vs. generational change.

The precepts of the two theoretical orientations addressed in this study and the broader review of the literature were used to construct several hypotheses. According to age-differences theory, younger persons in both study years should be more willing to finance local environmental action than older persons. The reason for this is fewer attachments to society. According to cohort-differences or generational-change theory, respondents in 1991 should be more willing to finance local environmental action than respondents in 1973. The reasons for this are younger-age socialization and two levels of cohort replacement. First, the 18-36 cohort of 1973 has replaced the 37-54 cohort by 1991. Second, the under age 18 cohort of 1973 has replaced the 18-36 cohort by 1991. Both of these cohorts should have socialization imprints from the environmentalism of the 1970s. Events such as the first Earth Day celebration, the Santa Barbara oil spill, and rising concern over pollution occurred during their younger, more impressionable years, and thus should produce greater sensitivity to the need to protect the natural environment.

Other hypotheses derived from the literature predicted a greater willingness to finance environmental action among the better educated, those identified with the Democratic party, non-Mormons, and those who believe air and water pollution are becoming more serious problems.

The data for this study were obtained from two state-wide surveys of nonmetropolitan Utah residents. The first survey was conducted in 1973 as part of a land use study of Utah and included 941 respondents. The selection of respondents followed two-stage area probability sampling procedures. Interviews with respondents were conducted in their respective homes. The second data set, collected in 1991, was part of a research project concerned with changes in the state of Utah. The sample size was 318. The interviews for the 1991 survey were conducted via telephone. Households in the 1991 survey were selected through a computer-generated, random-digit dialing process. Both samples had favorable completion rates in excess of 80 percent.

A variety of state and local issues was included in both surveys. Both the 1973 and the 1991 interviews were carried out by trained interviewers. Possible interviewer bias and errors were minimized through the utilization of a standardized questionnaire instrument. The gender of the person to be interviewed in each household was predetermined according to household composition. If the household was designated as male, an adult female was interviewed only in instances where no adult male lived at that residence. In cases of multiple adults of the designated gender, interviews were conducted with the adult with the most recent birthday.

A comparison of the two samples revealed that nonmetropolitan Utah had become somewhat younger, better educated, less Democrat, and less Mormon in 1991 that it was in 1973. For example, nearly 21 percent of the sample had completed a B.S. degree in 1991 compared with a little more than 13 percent in 1973. Likewise, affiliation with the Democratic Party had dropped from about 27 percent in 1973 to approximately 17 percent in 1991.

Public willingness to finance local environmental protection was measured by the following question: "Would you be willing to pay more taxes to protect the natural environment if you knew the money would be spent in your local area? Responses were then coded as: 1 = "No," 2 = "Probably," and 3 = "Definitely."

Since the time span between the two data collections was eighteen years, age was collapsed into 18-year cohorts. These categories were used to assess generational change, with each category representing a cohort as indicated in cohort-differences theory. Age was collapsed into three categories: 1 = "18-36," 2 = "37-54," and 3 = "55 and over." Education was assessed by asking respondents to indicate the highest degree or grade that they had completed. Responses on education were coded into four categories: 1 = "Less than high school," 2 = "High school graduate," 3 = "Some college," and 4 = "College graduate or beyond." Political affiliation was coded as: 1 = "Other" and 2 = "Democrat." Religious preference was dichotomized into: 1 = "Non-Mormon" and 2 = "Mormon."

A combined measure of perceived air and water pollution was developed from the following two questions: (1) Do you think air pollution is a serious problem, some problem, or no problem for the state of Utah? (2) Do you think water pollution is a serious problem, some problem, or no problem for the state of Utah? Responses to the two questions were coded as: 0 = "No problem," 1 = "Some problem," and 2 = "Serious problem." The responses were then summed to give a perceived pollution score with a range of 0-4.

The generational-aging-age analysis provides data on three separate effects: age-difference effects, aging effects, and generational effects. This methodology was employed to empirically investigate the age-differences and cohort-differences hypotheses, as well as their corresponding theoretical orientations. The methodology is somewhat similar to an aging-period-cohort analysis developed by Palmore (1978). Some modifications were made to more precisely test the research hypotheses in this study. To simplify the analysis, willingness to pay more taxes was dichotomized into "definitely" versus "other."

Figure 1 summarizes how each of the effects was calculated. As indicted by Palmore (1978), each of the age groups should cover a time span identical to the time between study years. Since this study compares samples gathered 18 years apart (1973-1991), age categories were broken down into 18-year intervals (18-36, 37-54, and 55+). Age-difference effects provide test results for age-differences theory. These effects were computed by subtracting the score for the youngest cohort in a study year from the middle-aged cohort in that same study year (B-A and E-D). (See next page for Figure 1.)

Aging effects show cohort replacement changes over time; that is, how perceptions have changed as people in the same cohort have aged. By way of illustration, the youngest cohort in 1973 (18-36) was compared with the middle-aged cohort in 1991 (37-54) since the youngest cohort had aged some 18 years by 1991 (E-A in Figure 1). Generational effects were computed by taking the value for an age-group in 1973 and subtracting it from an equivalent age-group's value in 1991. For the young cohorts, this is represented in Figure 1 as D-A. This is the cohort- replacement effect for the under 18 generation in 1973. This group should have experienced the greatest socialization imprints from the environmentalism of the 1970s. A Pearson's chi-square was calculated for each of the effects to estimate how far each difference exceeded chance variation.

Table 1 reports the findings of the generational-aging-age analysis. With respect to age-differences theory, there are significant differences between the youngest and the middle-aged cohorts. In 1973, there is a difference of -19.2 percentage points between these two cohorts. The youngest cohort is much more willing to pay higher taxes to protect the environment than persons in the next oldest cohort. This is statistically significant at the .001 level. The age difference for 1991 also shows a sizable difference (-11.2%) in the proportion of respondents who are willing to finance environmental protection, but the difference lacks statistical significance. The findings from Table 1 suggest that younger persons are more willing to finance environmental protection due to age-related rather than time-specific factors.

The aging effects between the young cohort in 1973 and the middle cohort in 1991 show a -12.4 percent difference in willingness to definitely finance environmental protection. This percentage difference corresponds with a statistically significant chi-square (4.42); however, the difference is negative rather than positive as predicted by cohort replacement. This finding suggests that as younger people age their level of environmental concern is not carried with them into the next generation, but instead decreases. Likewise, Table 1 doesn't show any significant generational effects, especially for the two youngest cohorts where positive changes were hypothesized using the cohort-replacement logic of cohort-differences theory.

Multiple Regression

The second stage in the statistical analysis involved the computation of zero-order correlation coefficients and multiple-regression models for each study year. Multiple regression was employed because it provides a comparative assessment of the independent effects of the study variables. It has a number of assumptions that must be met for an ideal analysis (Lewis-Beck, 1980). One of those assumptions is that variables are continuous, an assumption that is rarely met in the social sciences. As a result, any findings from regression analysis must be interpreted with caution.

Table 2 displays the zero-order correlation coefficients (Pearson's r) and standardized regression coefficients (beta) for the multiple regression analysis from the 1973 sample. Beta values show the independent effects of the study variables, while the Pearson's-r coefficients show the overall variable effects. All of four indicators of social position have a statistically significant effect on willingness to finance local environmental protection for the 1973 sample. The age variable has a statistically significant beta (-.146). Also, the relationship between age and environmental protection is in the direction predicted by the age hypothesis. The education variable has a statistically significant beta (.096). Likewise, political affiliation is statistically significant (beta = .090) and in the expected direction, meaning Democrats are more willing to fund environmental protection than other political groups. Within the multiple regression model, religion is also significant (beta = -.070). It should be noted that the corresponding betas, while statistically significant, are small and indicate only partial association. The overall multiple regression model for 1973 is statistically significant (F = 11.513) but only explains about 5 percent of the variance in willingness to pay more taxes to protect the natural environment. (See next page for Table 2.)

Table 2 also displays the results from the multiple regression between the independent variables and willingness to finance local environmental protection for the 1991 sample. Age remains statistically significant with a beta = -.153. The relationship continues to be in a negative direction, supporting both the age-differences theory and the age hypothesis. Religion appears to have a stronger effect in 1991 than in 1973. The 1991 beta coefficient of -.176 for religion is substantially higher than the -.070 religion coefficient in 1973. The negative beta and Pearson's-r coefficients support the religion hypothesis. This hypothesis predicted that non-Mormons would be more willing to finance environmental protection than their Mormon counterparts. In the 1991 sample, both education and political affiliation show no significant effects. The regression model explains seven percent of the variance and has a significant F-value (5.875).

The final stage in the statistical analysis provided information on the direct effects of variables when added sequentially as groups of variables. In this hierarchical regression, variables were grouped logically based on the literature review and theoretical orientations. The independent variables were grouped into four models and the results are reported in Table 3. Model 1 shows the effects of age, education, and political affiliation on willingness to finance environmental protection. Consistent with previous findings in the literature, the betas for age, education, and political affiliation are all statistically significant. The negative beta score for age (-.164) indicates that younger persons are more willing to finance environmental protection, a finding that lends support to age-differences theory.

Model 2 adds religion to the variables included in Model 1. This variable was added hierarchically to see how the effects of religion compare with the effects of variables examined in previous studies. Age, education, and political affiliation continue to have statistically significant, albeit weak, independent effects that cannot be explained by religion. The variable religion also shows a statistically significant effect on willingness to finance environmental protection that cannot be explained by age, education, or political affiliation. The negative beta value (-.100) for religion means that Mormons are less willing to finance environmental protection than non-Mormons, in line with the religion hypothesis.

Perceptions of air/water pollution are added in Model 3. This variable has a particularly strong and independent statistical effect (beta = .192) on willingness to pay more taxes to protect the local environment. Model 3 increases the variance explained in one's willingness to finance environmental protection to 9 percent. This change is statistically significant. This appears to be much more important than the various indicators of social position examined in the first two regression models.

Model 4 provides a further test of the overall generational-change hypothesis by adding study year to the regression equation. As indicated in Table 3, all of the independent variables included in the previous three models retain their significance, whereas study year has no effect on one's willingness to finance environmental protection. The variance explained by Model 4 remains unchanged at 9 percent. These findings cast further doubt on the cohort-differences or generational-change hypothesis as it relates to the financing of environmental action in nonmetropolitan Utah.

Two contrasting theoretical points of view were examined in this study: the age-differences approach versus the cohort-differences approach. The results of the generational-aging-age analysis for the environmental protection variable showed significant support for the age hypothesis but not for the generational-change hypothesis. In this regard, youth was found to be a significant predictor of willingness to protect the natural environmental, but generational change was not. Therefore, the generational-change hypothesis must be rejected. The relationship between age and environmental concern uncovered in this study is consistent with much of the literature. In conclusion, the age-differences approach, with its emphasis on stratification and detachment, was found to be the most plausible orientation based on findings for nonmetropolitan Utah.

Similar to other research, an individual's willingness to finance environmental protection in nonmetropolitan Utah is influenced significantly by education and political affiliation. This result is consistent with the findings of other studies in other states, such as the work by Van Liere and Dunlap (1980) and Jones and Dunlap's (1992) research. As predicted, those with more education expressed a greater willingness to fund environmental protection. Likewise, political affiliation was related to financing environmental protection in the manner predicted by the political- affiliation hypothesis.

The relatively small amount of variance explained by the age, education, and political- affiliation regression models implies that these indicators of social position may not be the most important factors influencing one's willingness to finance environmental protection. It is important to note that other studies have found similarly small amounts of explained variance using roughly the same variables (Jones & Dunlap, 1992; Van Liere & Dunlap, 1980). These relatively weak effects of many of the social position variables underscore the need for broadening the investigation to include other possible sources of influence similar to what we have attempted to do in the present study.

Our study shows that religion is a significant determinant of one's willingness to fund environmental protection in nonmetropolitan Utah. Specifically, non-Mormons expressed more willingness to finance environmental protection. This finding supports Hand and Van Liere's (1984) mastery-over-nature perspective. It may also be partly a problem of economics. Those persons belonging to the Mormon religion have various religious economic obligations, including what Mormons call "tithing." All faithful Mormons are expected to donate 10 percent of their annual income to support church programs (McConkie, 1966). This income burden among Mormons may lead to a reluctance to consider parting with even more income for government programs, such as environmental protection.

Even more pronounced is the finding that one's perception of selected environmental problems has a greater independent effect on support for environmental protection than any of the indicators of social position examined in this study. Clearly, a significant part of the willingness to pay more taxes to protect the local environment is the belief that air and/or water pollution have become a more serious problem. This is consistent with research on health behavior which indicates a strong connection between perceived seriousness of health problems and willingness to take action to prevent these problems (Cockerham, 1997). There may be some similarity between the type of action investigated in this study and what medical sociologists call "health action" since both deal with the willingness of individuals to take actions designed to prevent future problems. More research should be undertaken to investigare this possible link.

One of the questions addressed in the literature concerns the changing relationships between variables over time. For example, does education have the same influence on various environmental issues today that it had in the 1970s era of environmental activism? To answer this question with respect to financing environmental protection, possible relationships between study variables were examined for both study years. When the sample for 1973 was examined, all of the indicators of position were significantly correlated with the dependent variable. The relationships of age, education, political affiliation, and religion with willingness to finance environmental protection were in the directions predicted by the various social-position hypotheses. In 1991, stronger relationships with age and religion were observed, while education and political affiliation dropped into the range of chance variation. Part of this lack of significance can be attributed to the smaller sample size in 1991. However, it may also indicate some possible changes in relationships between demographic variables and willingness to finance environmental protection in the 1990s as evidenced by the relatively low beta coefficients.

The most striking finding of this study is that support for environmental protection is basically the same in the 1990s that it was in the 1970s. One possible explanation is that mountain states like Utah were less affected by Earth Day, the coastal oil spills, and other forms of environmental activism so characteristic of the 1970s. Our findings underscore the need for additional longitudinal assessments in other parts of the country to verify the lack of support reported in this study for cohort-differences theory and the supposition of generational change and cohort diffusion in relation to protecting the natural environment through increased public spending.

Albrecht, S. L. (1976). Legacy of the Environmental Movement. Environment and Behavior, 8, 147-169.

Attfield, R. (1991). The Ethics of Environmental Concern. Athens, GA: University of Georgia Press.

Buttel, F. H. (1979). Age and Environmental Concern: a Multivariate Analysis. Youth and Society, 10, 237-255.

Butell, F. H. & Flinn, W. F. (1976). Environmental Politics: Structuring of Partisan and Ideological Cleaves in Mass Environmental Attitudes. Sociological Quarterly, 17, 477-490.

Butell, F. H. & Flinn, W. F. (1978). Social Class and Mass Environmental Beliefs: a Reconsideration. Environment and Behavior, 10, 433-450.

Buttel, F. H., & Johnson, D. E. (1978). Dimensions of Environmental Concern: Factor Structure, Correlates, and Implications for Research. Journal of Environmental Education, 9, 49-64.

Cockerham, W. F. (1997). Medical sociology. Englewood Cliffs, NJ: Prentice-Hall.

Constantini, E., & Hanf, K. (1972). Environmental Concern and Lake Tahoe: a Study of Elite Perceptions, Backgrounds, and Attitudes." Environment and Behavior, 4, 209-239.

Dillman, D. A., & Christenson, J. A. (1972). The Public Value for Pollution Control. In W. R. Burch, N. H. Cheek, & L. Taylor (Eds.), Social behavior, natural resources, and the environment (pp. 237-256). New York: Harper Row.

Dunlap, R. E. (1975). The Impact of Political Orientation on Environmental Attitudes and Actions. Environment and Behavior, 7, 428-454.

Dunlap, R. E. (1989). Public Opinion and Environmental Policy. In J. P. Lester (Ed.), Environmental politics and policy: Theories and evidence (pp. 87-134). Durham, NC: Duke University Press.

Dunlap, R. E. (1991). Trends in Public Opinion Toward Environmental Issues: 1965-1990. Society and Natural Resources, 4, 285-312.

Dunlap, R. E., & Mertig, A. G. (1991). The Evolution of the U.S. Environmental Movement from 1970-1990: an Overview. Society and Natural Resources, 4, 209-218.

Erksine, H. (1972). The Polls: Pollution and its Costs. Public Opinion Quarterly, 36, 120-135.

Glenn, N. D. (1977). Cohort analysis. Beverly Hills, CA: Sage.

Glenn, N. D. (1980). Values, Attitudes, and Beliefs. In O. G. Brim Jr., and J. Kagan (Eds.), Constancy and Change in Human Development (pp. 596-640). Cambridge, MA: Harvard University Press.

Halliday, R. S. (1973, November 11). State Land Use Bill Gains Okay. Salt Lake Tribune, p. B-1.

Hand, C. M., & Van Liere, K. D. (1984). Religion, Mastery over Nature, and Environmental Concern. Social Forces, 63, 555-570.

Honnold, J. A. (1981). Predictors of Public Environmental Concern in the 1970s. In D. E. Mann (Ed.), Environmental Policy Formation (pp. 63-75). Lexington, MA: Lexington Books.

Honnold, J. A. (1984). Age and Environmental Concern: Some Specification of Effects. Journal of Environmental Education, 16, 4-9.

Hornback, K. E. (1974). Orbits of Opinion: The Role of Age in the Environmental Movement's Attentive Public, 1968-1972. Unpublished doctoral dissertation, Michigan State University, Lansing.

Howell, S. E., & Laska, S. B. (1992). The Changing Face of the Environmental Coalition: a Research Note. Environment and Behavior, 24, 134-144.

Jackson, J. E. (1983). Measuring the Demand for Environmental Quality with Survey Data. Journal of Politics, 45, 335-350.

Jones, R. E., & Dunlap, R. E. (1992). The Social Bases of Environmental Concern: Have They Changed over Time? Rural Sociology, 57, 28-47.

Kanagy, C. L., Humphrey, C. R., & Firebaugh, G. (1994). Surging Environmentalism: Changing Public Opinion or Changing Publics? Social Science Quarterly, 75, 804-819.

Kanagy, C. L., & Willits, F. K. (1993). A Greening of Religion: Some Evidence from a Pennsylvania Sample. Social Science Quarterly, 74, 674-683.

Lewis-Beck, M. S. (1980). Applied Regression: An introduction. Newbury Park, CA: Sage.

Malkis, A., & Grasmick, H. G. (1977). Support for the ideology of the environmental ovement: Tests of alternative hypotheses. Western Sociological Review, 8, 25-47.

Mannheim, K. (1952). Essays on the Sociology of Knowledge. New York: Oxford University Press.

McConkie, B. R. (1966). Mormon Doctrine. Salt Lake City, UT: Bookcraft.

Mitchell, R. C. (1989). From Conservation to Environmental Movement: The Development of Modern Environmental Lobbies. In M. J. Lacey (Ed.), Government and Environmental Politics (pp. 81-113). Washington DC: Congressional Quarterly.

Mohai, P. & Twight, B. W. (1987). Age and Environmentalism: an Elaboration of the Buttel Model Using National Survey Evidence. Social Science Quarterly, 68, 798-815.

Palmore, E. (1978). When Can Age, Period, and Cohort Be Separated? Social Forces, 57, 282-295.

Rosenbaum, W. A. (1990). Environmental Politics and Policy. Washington, DC: CQ Press.

Ryder, N. B. (1965). The Cohort as a Concept in the Study of Social Change. American Sociological Review, 30, 843-861.

Sale, K. (1993). The Green revolution: The American environmental movement, 1962-1992. New York: Hill-Wang.

Samdahl, D., & Robertson, R. (1989). Social Determinants of Environmental Concern: Specification and Test of the Model. Environment and Behavior, 21, 57-81.

Schuman H., & Rieger, C. (1992). Historical Analogies, Generational Effects, and Attitudes Toward War. American Sociological Review, 57, 315-326.

Tognacci, L., Weigel R. H., Wideen, M. F., & Vernon, D. (1972). Environmental Quality: How Universal Is Public Concern? Environment and Behavior, 4, 73-86.

Van Liere, K. D., & Dunlap, R. E. (1980). The Social Bases of Environmental Concern: a Review of Hypotheses, Explanations and Empirical Evidence. Public Opinion Quarterly, 44, 181-197.

Weeks, J. R. (1989). Population. Belmont, CA: Wadsworth.

White, L. (1973). The Historical Roots of Our Ecological Crisis. In I. G. Barbour (Ed.), Western man and environmental ethics (pp. 18-30). Reading, MA: Addison-Wesley.