Gender Differences in Environmental Concern

Abstract

Research on environmental concern has consistently found that women have modestly stronger pro-environmental values, beliefs, and attitudes than do men. Scholars have proposed and examined several explanations and have found that only a few hypotheses receive somewhat consistent empirical support, including the institutional trust hypothesis. Given that recent research suggests that men and women have equivalent levels of trust in social institutions, we chose to revisit the institutional trust hypothesis. We use a structural equation modeling technique on General Social Survey data from 2000 and 2010. In both years, we found that women report greater pro-environmental views and concern about environmental problems than do men. Yet, we found only minimal gender differences in institutional trust and no evidence that institutional trust mediates the relationship between gender and environmental concern. Our study does not support the institutional trust hypothesis. We end by identifying potential implications of our findings and suggestions for future research.

Keywords

gender environmental concern institutional trust structural equation modeling

Introduction

To explain a consistently found phenomenon—wherein women tend to have modestly stronger pro-environmental values, beliefs, and attitudes than do men (e.g., Dietz, Kalof, & Stern, 2002; Greenbaum, 1995; McCright, 2010)—environmental concern scholars have examined a range of hypotheses and found only a few having received somewhat consistent empirical support (e.g., Blocker & Eckberg, 1997; Davidson & Freudenburg, 1996; McCright, 2010; Xiao & Hong, 2010). One such hypothesis is the institutional support hypothesis, which combines insights from gender socialization theory and the risk analysis literature. According to this hypothesis, the modest gender differences in environmental concern are likely because women tend to have less trust and confidence in key social institutions such as government and science than do men, which in turn increases their concern for the environment when compared with men’s levels.

The institutional trust hypothesis has escaped researchers’ attention in recent years, and we believe it is time to revisit it for two reasons. First, women’s weaker trust in social institutions relative to men has been attributed to broader unequal gender power relations and the fact that social institutions are male dominated (e.g., Gustafson, 1998). Empirical studies prior to the mid-1990s, as summarized by Davidson and Freudenburg (1996), tend to find weaker trust in social institutions among women than men, which presumably led to women’s stronger concern for the environment. However, recent research on gender and institutional trust suggests that this relationship may have substantially changed since the 1990s. Indeed, recent empirical studies find that gender differences in institutional trust have largely disappeared (e.g., Carman, 2011; Goold, Fessler, & Moyer, 2006; Kim, 2010; Schyns & Koop, 2010), directly questioning the validity of the institutional trust hypothesis.

Second, as Freudenburg (1993, p. 914) argues, the negative impact of institutional trust on environmental concern may stem from a phenomenon termed recreancy. This concept refers to a paradoxical situation in which advancements of modern science and technology and detailed division of labor in our society have not only permitted an unprecedented level of prosperity but also have created such a level of dependency on social institutions that we are all vulnerable to potential “institutional failures.” Yet, this recreancy argument may be more applicable to concern about specific technological risks, such as nuclear waste and industrial pollutants, than to environmental concern more generally. Indeed, recent studies (e.g., Adaman et al., 2011; Xiao, 2011) find that, depending on how institutional trust or environmental concern is measured, institutional trust may have a positive association with pro-environmental beliefs and attitudes instead of a negative one.

In this study, we use the 2000 and 2010 General Social Survey (GSS) data sets to examine in depth the institutional trust hypothesis for explaining gender differences in environmental concern. These two GSS data sets provide many survey questions we use to form multi-item measures of institutional trust and different dimensions of environmental concern, as well as a standard battery of sociodemographic variables. In the next section, we first briefly review research on gender differences in environmental concern, emphasizing the prevailing hypotheses as identified by Blocker and Eckberg (1997) and Davidson and Freudenburg (1996). We then focus our attention on the institutional trust hypothesis.

Gender Differences in Environmental Concern

In the literature, environmental concern is typically conceptualized as a complex, multifaceted construct that refers to the extent to which people are concerned with environmental problems and support environmental protection—which includes expressing a willingness to pay or sacrifice (Dunlap & Jones, 2002). According to Dunlap and Jones (2002), environmental concern contains many interrelated facets that can be conceptualized as forming two general components. One component refers to general environmental beliefs and worldviews, and the other refers to concerns about more specific environmental problems. In the literature on gender differences in environmental concern, scholars have examined a variety of facets of both components of environmental concern.

While early work on environmental concern in the 1960s and 1970s tended to find little evidence of a gender difference (e.g., Van Liere & Dunlap, 1980), research in the past few decades consistently finds that women report stronger pro-environmental values, beliefs, and attitudes than do men (e.g., Davidson & Freudenburg, 1996; Dietz et al., 2002; Greenbaum, 1995; McStay & Dunlap, 1983; Mohai, 1992; Zelezny, Chua, & Aldrich, 2000). Such a gender difference is particularly clear in those studies examining concern about environmental problems that pose health and safety risks to participants and their families (e.g., exposure to toxic waste and proximity to a nuclear waste storage facility). Theorizing in this area draws upon insights from sociological theories of gender that emphasize gender differences in the socialization process and/or social roles and status (Blocker & Eckberg, 1997; Davidson & Freudenburg, 1996).

Gender socialization theory (e.g., Chodorow, 1978; Gilligan, 1982) argues that boys and girls acquire different values and learn different social expectations from society’s dominant culture through early childhood socialization. For instance, in the United States, boys learn to become competitive, independent, and unemotional, and that they will be expected to economically provide for their family after growing up. Girls learn to become compassionate, cooperative, and empathetic, and that they will be expected to assume the role of a nurturing caregiver when they grow up. A simple argument is that such differences in early child socialization become a source of gender differences in environmental concern. A direct application of such an argument is the “Concerns About Health and Safety” hypothesis (Blocker & Eckberg, 1997) or “Safety Concerns Hypothesis” (Davidson & Freudenburg, 1996), which explains why studies dealing with worry about specific health- and safety-related environmental problems tend to find the greatest gender differences (e.g., Freudenburg & Davidson, 2007; Greenbaum, 1995; Marshall, 2004; Mohai, 1992).

Scholars often extend the gender socialization argument to consider such factors that may mediate the effect of gender on environmental concern: scientific/environmental knowledge, religious beliefs and religiosity, basic value orientations, and trust in social institutions. Briefly, gender differences in environmental concern are theorized as at least partially the consequences of gender differences in these mediating factors. Empirical research provides uneven support for these hypotheses.

There is largely no evidence for the suggested mediating effects of knowledge or religious beliefs. Even though there is somewhat consistent evidence that men have greater scientific and technical knowledge than women (e.g., Arcury, Scollay, & Johnson, 1987; Blocker & Eckberg, 1997; Hayes, 2001; Xiao & Hong, 2010), research tends to find weak or even positive association between knowledge and environmental concern (e.g., Blocker & Eckberg, 1997; Davidson & Freudenburg, 1996; Hayes, 2001; Xiao & Hong, 2010). Similarly, while empirical research does tend to find weak but consistent evidence that women’s religious beliefs are stronger than are men’s, there is much less evidence of a consistent relationship between religious beliefs and environmental concern (e.g., McCright, 2010). The arguments for the mediating effects of value orientations (Stern, Dietz, & Kalof, 1993) and trust in social institutions, or the “institutional trust hypothesis,” have more consistent empirical support (Davidson & Freudenburg, 1996).

In addition to these hypotheses derived from gender socialization theory, scholars also have proposed several explanations that emphasize gender differences in social roles and statuses. In the literature, researchers have focused on three such social roles: employment status, parenthood, and homemaker status. Empirical findings regarding all three social roles seemed largely inconclusive in the past (e.g., Blocker & Eckberg, 1997; Davidson & Freudenburg, 1996). Furthermore, recent research tends to find virtually no effects of employment, homemaker status, and parenthood on environmental concern (e.g., Freudenburg & Davidson, 2007; McCright, 2010; Xiao & Hong, 2010; also see Blocker & Eckberg, 1997 and Davidson & Freudenburg, 1996, for more details about these hypotheses). Several scholars thus argue that gender differences in environmental concern are independent of the social roles and statuses that men and women differentially occupy (McCright, 2010; Mohai, 1997).

In sum, among these explanations of gender differences in environmental concern, three arguments seem more promising than others: the safety concerns, value orientations, and institutional trust hypotheses. In the next section, we focus on the institutional trust hypothesis for a few reasons we elaborate below.

Gender, Institutional Trust, and Environmental Concern

The institutional trust hypothesis essentially proposes that institutional trust mediates the relationship between gender and environmental concern. More specifically, this hypothesis posits that (a) men have higher levels of trust in institutions, such as science and technology and government, than do women; (b) levels of trust and confidence in such institutions correlate negatively with environmental concern; and (c) institutional trust mediates gender’s effect on environmental concern at least partially, if not entirely. Early research as summarized by Davidson and Freudenburg (1996) and Blocker and Eckberg (1997) provided fairly consistent support for this hypothesis. However, more recent work seems to challenge this hypothesis.

Social scientists continue to examine gender differences in levels of trust in institutions. Several recent studies find that men report greater trust in various institutions than do women (e.g., Cole, Kincaid, & Rodriguez, 2004; Goldfinch, Gauld, & Herbison, 2009; Van der Meer, 2010). However, more numerous are studies that find no such gender differences in institutional trust (e.g., Carman, 2011; Goold et al., 2006; Hero & Tolbert, 2004; Kim, 2010; Lawless, 2004; Schyns & Koop, 2010; Slomczynshi & Janicka, 2009). In addition, a few studies even find that women have stronger confidence and trust in institutions than do men (e.g., Fuse & Hanada, 2009; Kelleher & Wolak, 2007). If women and men are becoming more similar in their trust and confidence in institutions than they were in the past, then the explanation of gender differences in environmental concern proposed by the institutional trust hypothesis would no longer be valid.

Regarding the inverse relationship between institutional trust and environmental concern, the risk perception literature continues to find that higher levels of institutional trust are indeed associated with lower levels of risk perceptions (e.g., Bronfman, Vazquec, Gutierrez, & Cifuentes, 2008; Norgaard, 2007; Poortinga, Cox, & Pidgeon, 2008; Siegrist, Cvetkovich, & Roth, 2000). However, a few recent studies suggest that trust in management authorities and the government tends to increase people’s willingness to pay for various environmental measures such as higher taxes (e.g., Jones, Panagiotidou, Spilanis, Evangelinos, & Dimitrakopoulos, 2011), higher prices for natural resources (e.g., Speelman, Farolfi, Frija, D’Hases, & D’Haese, 2010), and carbon dioxide emission reductions (e.g., Adaman et al., 2011). This signals that the relationship between institutional trust and environmental concern may depend upon how the latter is operationalized: for example, perceived risks of environmental problems versus willingness to pay for environmental protection.

Another recent study by Xiao (2011), which examines how public perception of science and technology is correlated with environmental concern, focuses specifically on the operationalization of both constructs. Xiao identifies three different dimensions of perception of science and technology: (a) belief in the promise of science and technology, (b) reservations about science and technology, and (c) support for science and technology. There is no gender difference in the first and the third dimensions, but women have stronger reservations about science and technology than do men. Furthermore, belief in the promise of science and technology is negatively related to endorsement of an ecological worldview and concern for global environmental problems. Yet, support for science and technology is positively associated with these two dimensions of environmental concern. Such findings further suggest that the effect of institutional trust on environmental concern may vary depending on which specific dimensions of both constructs are examined.

In sum, we believe that this recent research reveals a clear need to revisit the institutional trust hypothesis. We also note that the conceptualization and operationalization of institutional trust has been relatively inconsistent in the environmental concern literature. As summarized above, some researchers refer to public trust in social institutions in general (e.g., Davidson & Freudenburg, 1996), while others tend to focus on trust in more specific social institutions such as government, management authorities (such as those in charge of nuclear facilities and regulations), and science and technology (e.g., Blocker & Eckberg, 1997). In this study, following the work of Bord and O’Connor (1992) and Freudenburg (1993), we chose to utilize measures of trust in several key social institutions: business, government, and science. Operationalization details are provided later in the “Method” section.

Hypotheses and Analytical Strategy

As stated above, the institutional trust hypothesis of gender differences in environmental concern is actually a group of specific hypotheses. Following the arguments in the literature as summarized by Davidson and Freudenburg (1996) and Blocker and Eckberg (1997), we hypothesize the following:

Hypothesis 1: Women express greater environmental concern than do men, after controlling for other sociodemographic variables.

Hypothesis 2: Women have lesser trust in social institutions than do men, controlling for other sociodemographic variables.

Hypothesis 3: Institutional trust is negatively correlated with environmental concern.

Hypothesis 4: Institutional trust at least partially mediates the effect of gender on environmental concern.

To adequately examine the mediating effects of institutional trust on the gender–environmental concern relationship, we use the structural equation modeling (SEM) technique to construct and test a mediating effect model. In this model, gender and a standard battery of sociodemographic variables are exogenous variables and have direct effects on measures of institutional trust and environmental concern. Institutional trust not only receives direct effects from the exogenous variables but also has direct effects on environmental concern. This model thus enables the examination of the indirect effects of gender on environmental concern via institutional trust. Figure 1 displays this conceptual path diagram.

Figure 1.

Conceptual path diagram of two structural equation models in the study.

Method

Data sets and Variables

In this study, we utilize two data sets derived from the 2000 and 2010 GSS. Both surveys contain an environment module that provides items to form measures of environmental concern and institutional trust. Due to a split-sample design, in 2000 and 2010, only about a third of the total sample answered both sets of questions on environmental concern and institutional trust, resulting in final sample sizes of 857 (out of 2,817) in 2000 and 763 (out of 2,044) in 2010. Fortunately, these subsamples were split randomly and thus still retain their representativeness, albeit with weaker statistical power due to smaller sample sizes.

Measuring institutional trust

Each year since 1973, GSS has included a set of 13 items asking about respondents’ confidence in a group of various social institutions. Cook and Gronke (2001, 2005) suggest that these items form two dimensions of institutional trust, one group of “institutions of opposition” including the press, TV, and organized labor, and another group of “institutions of order” that includes the remaining 10 institutions. However, for the purpose of this study and following the conceptualization and operationalization commonly used in the literature of environmental concern (e.g., Bord and O’Connor, 1992; Freudenburg, 1993), we chose five items to measure confidence in three institutions. Two items, “CONBUS” and “CONFINAN,” are combined to measure “trust in business” (i.e., major companies, banks, and financial institutions). Another two items, “CONFED” and “CONLEGIS,” form a measure of “trust in government” (i.e., the executive branch of federal government and Congress. Finally, a single item, “CONSCI,” measures “trust in science.” Table 1 contains their exact wording.

Table 1.

Original General Social Survey Variable Names and Coding for Variables in the Study.

Variable	Original GSS name	Description	Coding in the study
Confidence in social institutions items: “I am going to name some institutions in this country. As far as the people running these institutions are concerned, would you say you have a great deal of confidence, only some confidence, or hardly any confidence at all in them?”
Trust in business	CONBUS	Major companies banks and financial institutions	1 = not at all
	CONFINAN
Trust in government	CONFEDCONLEGIS	Executive branch of federal government Congress	2 = only some
Trust in science	CONSCI	Scientific community	3 = a great deal
General pro-environmental beliefs items: “How much do you agree or disagree with each of these statements?”
Pro-environmental views	SCIGRN	Modern science will solve environmental problems.	1 = strongly agree
	GRNECON	We worry too much about the future of the environment, and not enough about prices and jobs today.	2 = agree
	GRNPROG	People worry too much about human progress harming the environment.	3 = neither
	IMPGRN	There are more important things to do in life than protect the environment.	4 = disagree
	GRNEXAGG	Many of the claims about environmental threats are exaggerated.	5 = strongly disagree
Concern about specific environmental problems items: “In general, do you think that [ . . . ] is dangerous to the environment?”
Concern about problems	CARSGEN	Air pollution by cars	1 = not at all dangerous
	INDUSGEN	Air pollution by industry	2 = not dangerous
	CHEMGEN	Pesticides and chemicals used in farming	3 = somewhat dangerous
	WATERGEN	Pollution of rivers, lakes, and streams	4 = dangerous
	GENEGEN	Modifying the genes of certain crops	5 = very dangerous
Sociodemographic items
Female	SEX	Sex	0 = male, 1 = female
Employed	WRKSTAT	Labor force status	0 = unemployed, 1 = employed
Parenthood	BABIES, PRETEEN, TEENS	Number of children under 18 at Home	0 = no child under 18 at home, 1 = child below 18 years at home
Age	AGE	Age	actual age in years
Non-White	RACE	Race	0 = White, 1 = non-White
Education	EDUC	Years of education	actual years of formal schooling
Income	CONCINC	Household income in constant dollars	actual amount in dollars
Political ideology	POLVIEWS	Thinks of self as conservative or liberal	1 = extremely conservative to 7 = extremely liberal
Party affiliation	PARTYID	Political party affiliation	1 = strong Republican to 7 = strong Democrat
City size	SIZE	Population size where respondent lives	actual amount in 1,000s

Note. In the final structural equation models, “trust in business” and “trust in government” are both simple additive indexes that combine the two items in each group. The variables “Pro-Environmental Views” and “Concern About Problems” are latent variables measured by their corresponding GSS items. All items were recoded so that higher scores mean greater institutional trust, stronger pro-environmental views, and greater concern about environmental problems, respectively. All sociodemographic variables are observed variables.

Measuring environmental concern

The environment module in both years provides several items that can be used to form measures of environmental concern. We follow Dunlap and Jones’s (2002) lead in conceptualizing environmental concern with two main components: one that contains general pro-environmental beliefs such as the perceived importance of the environment and support for environmental protection and another that contains more specific views about substantive environmental problems.

We ran a series of preliminary Confirmatory Factor Analysis (CFA) models and formed two distinctive measures of environmental concern, which parallel the two dimensions above. See Table 1 for details. One measure, which we termed pro-environmental views, consists of five items tapping general pro-environmental beliefs. The other measure, which we termed concern about problems, includes five items assessing the perceived dangerousness of specific environmental problems. Factor loadings of these items range from 0.27 to 0.80, which show adequate to very good reliability. For both sets of items, greater values represent greater environmental concern.

Furthermore, as in the literature (e.g., Corral-Verdugo, Bechtel, & Fraijo-Sing, 2003; Stern, Dietz, Abel, Guagnano, & Kalof, 1999; Xiao, 2011), we hypothesize that more general pro-environmental beliefs causally precede beliefs and attitudes about more specific environmental problems. Thus, in the subsequent SEM models, our pro-environmental views measure has a direct effect on our concern about problems measure.

Sociodemographic and political variables

The primary independent variable in this study is gender (female = 1). Besides gender, GSS includes several sociodemographic and political variables that we use as controls in our models. These variables include employment status (employed = 1), parenthood (child below 18 years at home = 1), age (in years), race (non-White = 1), education (years of school completed), household income (in constant dollars), political ideology (1 = extremely conservative to 7 = extremely liberal), political party affiliation (1 = strong Republican to 7 = strong Democrat), and the population size of where the respondent lives (in 1,000s). See Table 1 for more details.

Results and Discussion

Using data from both the 2000 and 2010 GSS, we estimated two identical structural equation models, using the path diagram illustrated in Figure 1, one for each year. In these models, gender, the three measures of institutional trust, and all control variables are observed exogenous variables. “Pro-Environmental Views” and “Concern About Problems” are latent variables, each measured by its corresponding indicators as shown above. Thus, our SEM models integrate the two aforementioned CFA measurement models into the structural models.

The results of our SEM analyses are reported in Tables 2 to 4. Tables 2 and 3 display the standardized direct effects of our predictor variables on the institutional trust measures, pro-environmental views, and concern about problems measures in 2000 and 2010. Table 4 displays the standardized indirect and total effects for all the variables in the models for 2000 and 2010. Both SEM models have good fits: goodness-of-fit index (GFI) = 0.96-0.97; adjusted goodness-of-fit index (AGFI) = 0.92-0.93; incremental fit index (IFI) = 0.94; comparative fit index (CFI) = 0.94; root mean squared error of approximation (RMSEA) = 0.04.

Table 2.

Standardized Regression Coefficients (Direct Effects)—Results of SEM Examining the Mediating Effects of Institutional Trust in 2000.

Predictors	Trust in business	Trust in government	Trust in science	Pro-environmental views	Concern about problems
Female	0.00	0.09*	−0.03	0.08*	0.04
Employed	−0.08*	−0.06	−0.04	−0.01	0.02
Parenthood	−0.05	−0.05	−0.05	−0.03	−0.02
Age	−0.11*	−0.13*	−0.06	−0.15*	0.05
Non-White	−0.08*	0.07	−0.06	−0.15*	0.09*
Education	0.08*	−0.02	0.12*	0.13*	−0.01
Income	0.09*	0.03	0.09*	0.07	−0.10*
Party affiliation (Republicans to Democrats)	−0.03	0.04	0.03	0.16*	−0.02
Political ideology (more liberal)	−0.12*	0.03	0.00	0.18*	0.00
City size	0.07*	0.14*	0.05	−0.05	0.05
Trust in business	—	—	—	−0.07	−0.06
Trust in government	—	—	—	−0.03	0.01
Trust in science	—	—	—	0.11*	−0.06
Pro-environmental views	—	—	—	—	0.52*
R² coefficient	.07	.06	.06	.17	.28

Note. SEM = structural equation modeling.

p < .05.

Table 3.

Standardized Regression Coefficients (Direct Effects)—Results of SEM Examining the Mediating Effects of Institutional Trust in 2010.

Predictors	Trust in business	Trust in government	Trust in science	Pro-environmental views	Concern about problems
Female	0.01	−0.03	−0.10*	0.13*	0.06
Employed	0.01	0.02	−0.02	0.13*	0.00
Parenthood	0.02	0.06	0.00	0.06	−0.01
Age	−0.07	−0.10*	0.02	0.01	−0.04
Non-White	0.10*	0.14*	−0.03	−0.19*	0.08*
Education	−0.02	−0.06	0.15*	0.15*	−0.12*
Income	0.02	0.01	0.09*	0.06	−0.12*
Party affiliation (Republicans to Democrats)	−0.04	0.22*	0.07	0.22*	0.05
Political ideology (more liberal)	−0.14*	0.09*	0.06	0.23*	0.04
City size	0.02	0.00	−0.04	0.05	−0.01
Trust in business	—	—	—	−0.16*	−0.13*
Trust in government	—	—	—	0.11*	0.05
Trust in science	—	—	—	0.21*	0.04
Pro-environmental views	—	—	—	—	0.47*
R² coefficient	.06	.15	.04	.39	.33

Note. SEM = structural equation modeling.

p < .05.

Table 4.

Standardized Indirect and Total Effects—Results of an SEM Examining the Mediating Effects of Institutional Trust in 2000 and 2010.

	Pro-environmental views				Concern about problems
	Indirect		Total		Indirect		Total
Predictors	2000	2010	2000	2010	2000	2010	2000	2010
Female	−.01	−.02	.07*	.11*	.04 (.00)	.05*(.02)	.08*	.11*
Employed	.00	.00	−.01	.13*	.00	.06*	.02	.06
Parenthood	.00	.00	−.04	.07	−.01	.03	−.04	.02
Age	.01	.00	−.15*	.02	−.07	.01	−.02	−.02
Non-White	.00	−.01	−.15*	−.19*	−.07	−.10*	.02	−.02
Education	.02	.03*	.15*	.18*	.06	.09*	.06	−.03
Income	.00	.02	.07*	.08	.03	.04	−.07	−.09*
Party affiliation (Republicans to Democrats)	.01	.05*	.17*	.26*	.09*	.14*	.08	.20*
Political ideology (more liberal)	.00	.04*	.18*	.28*	.10*	.16*	.09*	.19*
City size	.00	−.01	−.05	.04	−.03	.01	.02	.00
Trust in business	—	—	−.07*	−.16*	−.04	−.08*	−.10*	−.21*
Trust in government	—	—	−.03	.11*	−.02	.05	.00	.10*
Trust in science	—	—	.11*	.21*	.06	.10*	.00	.14*
Pro-environmental views	—	—	—	—	—	—	.52*	.47*

Note. SEM = structural equation modeling. Significance tests for indirect effects are based on bootstrap standard errors (see Hayes, 2009, for more details). For gender’s indirect effects on concern about problems, we report both the total indirect effects and the partial indirect effects in parentheses that exclude the contribution by pro-environmental views.

p < .05.

We examine the evidence bearing on each of our four institutional trust hypotheses in turn. Hypothesis 1 posits that women have greater environmental concern than do men, after controlling for other sociodemographic and political variables. For this hypothesis, we examined the total effect of gender on both measures of environmental concern. Table 4 shows that, in both 2000 and 2010, gender has a statistically significant positive total effect on both pro-environmental views and concern about problems. These results largely confirm the common finding in recent literature that women tend to be modestly more concerned with the environment than are men, regardless of the type of measures (e.g., Dietz et al., 2002; Greenbaum, 1995; McCright, 2010). Hypothesis 1 thus enjoys consistent support.

We now turn to Hypothesis 2, which posits a lower level of institutional trust among women compared with men. Tables 2 and 3 report the direct effects of gender on the three institutional trust measures. Overall, we found that women and men largely have equal trust in social institutions with only two exceptions (out of six total comparisons). In 2000, women expressed slightly stronger trust in government (the executive branch of federal government and Congress) than did men, which is surprising as the literature suggests that the opposite would be the case. In 2010, women had less trust in science than men, which is consistent with the literature. For both statistically significant gender differences, the size of effect is small. Thus, Hypothesis 2 is not supported in 2000 and only weakly supported in 2010, consistent with the results of recent studies discussed earlier (e.g., Carman, 2011; Goold et al., 2006; Hero &Tolbert, 2004; Kim, 2010; Lawless, 2004; Schyns & Koop, 2010; Slomczynshi & Janicka, 2009).

Hypothesis 3 posits a negative correlation between institutional trust and environmental concern, when higher scores represent greater trust and concern, respectively. We examined the total effects of institutional trust on pro-environmental views and concern about problems in Table 4. The results in this table are quite inconsistent. The relationship between institutional trust and environmental concern varies substantially depending on the year, the measure of institutional trust, and the facet of environmental concern.

First, the 2000 and 2010 results show that trust in business relates negatively to both facets of environmental concern, as expected. However, since we found no gender difference in trust in business in either 2000 or 2010, this finding does not support the institutional trust hypothesis. Second, neither trust in government nor trust in science has any statistically significant effect on either measure of environmental concern in 2000. Thus, Hypothesis 3 is only partially supported in 2000. However, results from the 2010 model show that trust in government and trust in science actually relates to higher pro-environmental views and concern about environmental problems. This finding contradicts the prediction of Hypothesis 3, but it is consistent with the recent institutional trust studies we mentioned above. Therefore, Hypothesis 3 is not supported in 2010.

Finally, Hypothesis 4 posits that institutional trust at least partially mediates the effect of gender on environmental concern. Here, we examined the standardized indirect effect of gender on both measures of environmental concern in Table 4. In both 2000 and 2010, the results show only nonsignificant indirect effects of gender. Clearly, institutional trust does not mediate the relationship between gender and environmental concern. Thus, Hypothesis 4 is not supported.

Combining the above findings together, we have a clear conclusion. Briefly, our results do not support the institutional trust hypothesis. There is consistent evidence for modest gender differences in environmental concern. However, such gender differences are independent of variation in institutional trust. That is, there is no consistent evidence for the hypothesized gender difference in institutional trust, and there is clearly no evidence for the mediating effect of institutional trust on the relationship between gender and environmental concern. Combined with the results of other studies (e.g., Carman, 2011; Fuse & Hanada, 2009; Kelleher & Wolak, 2007; Schyns & Koop, 2010), our findings suggest that the institutional trust hypothesis is not a valid explanation of gender differences in environmental concern.

Besides gender, other sociodemographic and political variables perform largely as expected in predicting environmental concern. Overall for general views of the environment, education, race, political ideology, and party affiliation are consistent and robust predictors across the 2 years of data. More highly educated, White, politically liberal, and Democratic respondents report more general pro-environmental beliefs than do their respective counterparts. As for concern about specific environmental problems, only income, political ideology, and party affiliation are consistent and robust predictors across the 2 years of data. Lower income, politically liberal, and Democratic respondents report greater concern about specific environmental problems than do their respective counterparts. The standardized total effects of political ideology and party affiliation on pro-environmental views and concern about problems are consistently larger in 2010 than they are in 2000; this is consistent with the political polarization trend on climate change that McCright and Dunlap (2011) find using Gallup data over the same period (see also Dunlap & McCright, 2008). Not surprisingly, our general pro-environmental views measure is the most influential predictor of concern about specific environmental problems. The performance of these predictors suggests that our latent measures of environmental concern have considerable construct validity.

While we did not set out to test the social roles hypotheses for explaining gender differences in environmental concern, our results nevertheless bear upon two of them. Neither employment status nor parenthood has much significant effect on environmental concern in our study. Only in 2010 does employment status have a statistically significant effect on general pro-environmental views. Employed respondents report stronger pro-environmental beliefs than do their unemployed counterparts. It logically follows that employment status and parenthood cannot mediate much of the gender differences in environmental concern found in this study. Thus, our results are consistent with recent research that finds no support for the social roles hypotheses (e.g., McCright, 2010).

Conclusion

In this study, we conducted an in-depth examination of the institutional trust hypothesis that has been offered as an explanation of gender differences in environmental concern. While this hypothesis garnered some empirical support before the mid-1990s (e.g., Blocker & Eckberg, 1997; Davidson & Freudenburg, 1996), little research has examined it since—especially with analytical techniques that properly model the hypothesized mediating effect of institutional trust on the relationship between gender and environmental concern. Given that recent research suggests that men and women have equivalent levels of trust in social institutions (e.g., Carman, 2011; Goold et al., 2006; Kim, 2010; Schyns & Koop, 2010), we chose to revisit this institutional trust hypothesis.

Using 2 years of GSS data from 2000 and 2010, we used the SEM technique to formally examine the direct, indirect, and total effects of gender and institutional trust on environmental concern. With 2 years of GSS data allowing identical measures of institutional trust, general pro-environmental views, and concern about specific environmental problems, we were able to replicate our measurement models and the overall structural models across 2 different years—enhancing our confidence in the our findings.

Consistent with the literature since the 1990s (e.g., Dietz et al., 2002; Greenbaum, 1995; McCright, 2010), we found modest gender differences in two measures of environmental concern—general pro-environmental views and specific concerns about environmental problems—in both 2000 and 2010. Women consistently reported more pro-environmental views and greater levels of concern about specific environmental problems than did men. This holds even when controlling for a set of relevant sociodemographic and political variables that earlier research has found to be correlated with environmental concern.

Yet, also consistent with recent research (e.g., Cook & Gronke, 2005; Gronke & Cook, 2007; Price & Romantan, 2004), we found a little or no gender difference in institutional trust in either year. There is no gender difference in trust in business in either 2000 or 2010. Women actually showed slightly more trust in government than did men in 2000, but then such difference disappeared in 2010. Only when we examined trust in science did we find lesser trust among women than men, and such difference is small in size and only statistically significant in 2010.

In addition, our results confirm findings in more recent research that trust in specific institutions, namely, government and science, actually enhance pro-environmental views and concerns about environmental problem. This finding suggests that we need to move beyond any conceptualization of a singular relationship between institutional trust and environmental concern. Instead, it is necessary to consider that trust in different social institutions has a distinctive association with concern for the environment.

Furthermore, we found no evidence for the mediating effect of institutional trust on the relationship between gender and environmental concern. The modest gender differences in environmental concern are independent of variation in institutional trust. Combined with the finding of a diminishing effect of trust in social institutions on environmental concern, we believe that the institutional trust hypothesis no longer has the empirical support that it once enjoyed prior to mid-1990s.

One implication of these results is that differential trust in major social institutions by gender is not likely the cause of the gender differences in concern about major environmental problems in recent years, such as worry about the risks of anthropogenic climate change, the Deepwater Horizon oil spill in the Gulf of Mexico, or hydraulic fracturing for natural gas in the Marcellus Shale. Communication or risk management strategies aimed at increasing public trust in major social institutions will not likely result in substantially reduced environmental concern. Furthermore, such strategies will not likely reduce the gendered difference in concern for these major environmental problems.

So, why do we consistently observe modest gender differences in environmental concern? As we stated at the outset, recent research finds that such gender differences are independent of the social roles and statuses that men and women differentially occupy (McCright, 2010; Mohai, 1997). Our study supports this conclusion. That leaves us with a few remaining hypotheses derived from gender socialization theory, which have received somewhat consistent empirical support: the health and safety concerns, risk perception, and value orientations hypotheses (e.g., Bord & O’Connor, 1997; Dietz et al., 2002; McCright, 2010).

We urge scholars to find ways to simultaneously test these key socialization hypotheses. To do so will require that scholars design new survey instruments to model latent factor measures of these key variables (e.g., risk perceptions, value orientations, and environmental concern). In these new surveys, scholars also should use refined measures of gender. Continued use of survey items measuring gender as demographically male or female is less than optimal. We argue for the use of single-item or multi-item indicators of gender identity, which measure individuals’ masculinity and femininity along a continuum, such as Bem’s (1993) Sex Role Inventory. Finally, this future work should use CFA and SEM, where possible, due to their superior ability for modeling complex variables as latent factors and estimating theory-driven structural models that are crucial for the examination of mediating effects.

Footnotes

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Chenyang Xiao, PhD, is an assistant professor of sociology at American University. His research focuses on public beliefs of and attitudes toward environmental problems such as global warming as well as general environmental values and worldviews, often with an international scope.

Aaron M. McCright is associate professor of sociology in Lyman Briggs College, the Department of Sociology, and the Environmental Science and Policy Program at Michigan State University. His research investigates how interrelationships among scientific developments, political processes, and social dynamics influence society’s capacity for recognizing and dealing with environmental degradation and technological risks.

References

Adaman

Karah

Kumbaroglu

Ozkaynak

Zenginobuz

(2011). What determines urban household’s willingness to pay for CO₂ emission reductions in Turkey: A contingent valuation survey. Energy Policy, 39, 689-698.

Arcury

T. A.

Scollay

S. J.

Johnson

T. P.

(1987). Sex differences in environmental concern and knowledge. Sex Roles, 16, 463-472.

Batson

C. D.

Schoenrade

Ventis

W. L.

(1993). Religion and the individual. New York, NY: Oxford University Press.

Bem

S. L.

(1993). The lenses of gender. New Haven, CT: Yale University Press.

Blocker

T. J.

Eckberg

D. L.

(1997). Gender and environmentalism. Social Science Quarterly, 78, 841-858.

Bord

R. J.

O’Connor

R. E.

(1992). Determinants of Risk Perceptions of a Hazardous Waste Cite. Risk Analysis, 12(3), 411-416.

Bord

R. J.

O’Connor

R. E.

(1997). The gender gap in environmental attitudes. Social Science Quarterly, 78, 830-840.

Bronfman

Vazquec

E. L.

Gutierrez

V. V.

Cifuentes

L. A.

(2008). Trust, acceptance, and knowledge of technological and environmental hazards in Chile. Journal of Risk Research, 11, 755-773.

Carman

(2011). The process if the reality: Perceptions of procedural fairness and participatory democracy. Political Studies, 58, 731-751.

10.

Chodorow

N. J.

(1978). The reproduction of mothering. Berkeley: University of California Press.

11.

Cole

R. L.

Kincaid

Rodriguez

(2004). Public opinion on federalism and federal political culture in Canada, Mexico, and the United States, 2004. Publius: Journal of Federalism, 31, 201-225.

12.

Cook

T. E.

Gronke

(2001). Dimensions of trust: Is public confidence in the media distinct from other institutions? Paper presented at the Annual Meeting of the Midwest Political Science Association. Retrieved from http://people.reed.edu/~gronkep/docs/MIDW2001.pdf

13.

Cook

T. E.

Gronke

(2005). The skeptical American: Revisiting the meanings of trust in government and confidence in institutions. Journal of Politics, 67, 784-803.

14.

Corral-Verdugo

Bechtel

R. B.

Fraijo-Sing

(2003). Environmental beliefs and water conservation: An empirical study. Journal of Environmental Psychology, 23, 247-257.

15.

Davidson

D. J.

Freudenburg

W. R.

(1996). Gender and environmental risk concerns. Environment and Behavior, 28, 302-339.

16.

Dietz

Kalof

Stern

P. C.

(2002). Gender, values, and environmentalism. Social Science Quarterly, 83, 353-364.

17.

Dunlap

R. E.

Jones

R. E.

(2002). Environmental concern: Conceptual and measurement issues. In Dunlap

R. E.

Michelson

(Eds.), Handbook of environmental sociology (pp. 482-524). Westport, CT: Greenwood Press.

18.

Dunlap

R. E.

McCright

A. M.

(2008). A widening gap: Republican and democratic views on climate change. Environment, 50, 26-35.

19.

Freudenburg

W.R.

(1993). Risk and Recreancy: Weber, the Division of Labor, and the Rationality of Risk Perceptions. Social Forces, 71(4), 909-932.

20.

Freudenburg

W. R.

Davidson

D. J.

(2007). Nuclear families and nuclear risks. Rural Sociology, 72, 215-243.

21.

Fuse

Hanada

(2009). Political distrust and gendered attitudes: The Japanese state and women. Sex Roles, 60, 843-858.

22.

Gilligan

(1982). In a different voice. Cambridge, MA: Harvard University Press.

23.

Goldfinch

Gauld

Herbison

(2009). The participation divide? Political participation, trust in government, and e-government in Australia and New Zealand. Australian Journal of Public Administration, 68, 333-350.

24.

Goold

S. D.

Fessler

Moyer

C. A.

(2006). A measure of trust in insurers. Health Services Research, 41, 58-78.

25.

Greenbaum

(1995). Taking stock of two decades of research on the social bases of environmental concern. In Mehta

M. D.

Ouellet

(Eds.), Environmental sociology (pp. 125-152). North York, Ontario, Canada: Captus Press.

26.

Gronke

Cook

T. E.

(2007). Disdaining the media: The American public’s changing attitudes toward the news. Political Communication, 24, 259-281.

27.

Gustafson

P. E.

(1998). Gender differences in risk perception: Theoretical and methodological perspectives. Risk Analysis, 18, 805-811.

28.

Hayes

A. F.

(2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76, 408-420.

29.

Hayes

B. C.

(2001). Gender, scientific knowledge, and attitudes toward the environment. Political Research Quarterly, 54, 657-671.

30.

Hero

R. E.

Tolbert

C. J.

(2004). Minority voices and citizen attitudes about government responsiveness in the American states: Do social and institutional context matter? British Journal of Political Science, 34, 109-121.

31.

Jones

Panagiotidou

Spilanis

Evangelinos

K. I.

Dimitrakopoulos

P. G.

(2011). Visitors’ perceptions on the management of an important nesting site for loggerhead sea turtle (Caretta caretta L.): The case of Rethymno coastal area in Greece. Ocean & Coastal Management, 54, 577-584.

32.

Kelleher

C. A.

Wolak

(2007). Explaining public confidence in the branches of state government. Political Research Quarterly, 60, 707-721.

33.

Kim

(2010). Public trust in government in Japan and South Korea: Does the rise of critical citizens matter? Public Administration Review, 70, 801-810.

34.

Lawless

J. L.

(2004). Politics of presence? Congresswomen and symbolic representation. Political Research Quarterly, 57, 81-99.

35.

Marshall

B. K.

(2004). Gender, race, and perceived environmental risk. Sociological Spectrum, 24, 453-478.

36.

McCright

A. M.

(2010). The effects of gender on climate change knowledge and concern in the American public. Population and Environment, 32, 66-87.

37.

McCright

A. M.

Dunlap

R. E.

(2011). The politicization of climate change and polarization in the American public’s views of global warming, 2001-2010. Sociological Quarterly, 52, 155-194.

38.

McStay

J. R.

Dunlap

R. E.

(1983). Male-female differences in concern for the environmental quality. International Journal of Women’s Studies, 6, 291-301.

39.

Mohai

(1992). Men, women, and the environment. Society & Natural Resources, 5, 1-19.

40.

Mohai

(1997). Gender differences in the perceptions of most important environmental problems. Race, Gender & Class, 5, 153-169.

41.

Norgaard

K. M.

(2007). The politics of invasive weed management: Gender, race, and risk perception in rural California. Rural Sociology, 72, 450-477.

42.

Poortinga

Cox

Pidgeon

N. F.

(2008). The perceived health risks of indoor radon gas and overhead powerlines: A comparative multilevel approach. Risk Analysis, 28, 235-248.

43.

Price

Romantan

(2004). Confidence in institutions before, during, and after indecision 2000. Journal of Politics, 66, 939-956.

44.

Schyns

Koop

(2010). Political distrust and social capital in Europe and the USA. Social Indicators Research, 96, 145-167.

45.

Siegrist

Cvetkovich

Roth

(2000). Salient value similarity, social trust, and risk/benefit perception. Risk Analysis, 20, 353-362.

46.

Slomczynshi

K. M.

Janicka

(2009). Structural determinants of trust in public institutions, cross-national differentiation. International Journal of Sociology, 39, 8-29.

47.

Speelman

Farolfi

Frija

D’Hases

D’Haese

(2010). The impact of the water rights system on smallholder irrigators’ willingness to pay for water in Limpopo Province, South Africa. Environment and Development Economics, 15, 465-483.

48.

Stern

P. C.

Dietz

Abel

Guagnano

G. A.

Kalof

(1999). A value-belief-norm theory of support for social movements: The case of environmentalism. Research in Human Ecology, 6, 81-97.

49.

Stern

P. C.

Dietz

Kalof

(1993). Value orientations, gender, and environmental concern. Environment and Behavior, 25, 322-348.

50.

Van der Meer

(2010). In what we trust? A multi-level study into trust in parliament as an evaluation of state characteristics. International Review of Administrative Sciences, 76, 517-536.

51.

Van Liere

K. D.

Dunlap

R. E.

(1980). The social bases of environmental concern. Public Opinion Quarterly, 44, 181-197.

52.

Xiao

(2011). Public attitudes toward science and technology and concern for the environment: Testing a model of indirect feedback effects. Environment and Behavior. Advance online publication. doi:10.1177/0013916511414875.

53.

Xiao

Hong

(2010). Gender differences in environmental behaviors in China. Population and Environment, 32, 88-104.

54.

Zelezny

L. C.

Chua

Aldrich

(2000). Elaborating on gender differences in environmentalism. Journal of Social Issues, 56, 443-457.