Toward the Development and Validation of a Model of Environmental Citizenship of Young Adults

Abstract

Growing civic participation in environmental movements shows that societies are more engaging in behavior that can be described as environmental citizenship. This behavior, however, may not necessarily be driven by factual knowledge and accurate beliefs. In the present study, a simplified model of environmental citizenship proposed by Hawthorne and Alabaster is tested in a sample of Lithuanian young adults in order to develop a working model of environmental citizenship that could be later used in intervention designs as a change model. A convenience sample of 267 young adults from 18 to 40 years of age participated in the study by filling in a questionnaire that assessed various components that comprise a model of environmental citizenship. The results indicate that, with a few modifications to better fit the data, the model fit the data well and could explain approximately 50% of the variance of environmental citizenship. The study uncovered a gap between factual (concrete) environmental knowledge and environmental literacy (perceived competence), illustrating the need to address this divide in order to ensure evidence-based participation in environmental movements and environmental citizenship behavior.

1. Introduction

Environmental issues are at the forefront of global priorities because of both economic factors related to the use of limited resources and the environmental harm done by extracting and using these resources. The ever increasing environmental consciousness and civic participation related to environmentalism also leaves no doubt that societies see environmental issues as being of extreme importance [1]. In order to address environmental issues effectively, citizens need to act in order to shape local and global environmental policy through legal means and civic participation; this sort of social action is often called environmental citizenship [2].

Civic participation, however, needs to be based on a thorough and nuanced understanding of the issues in question if such social action is to lead to realistic and sustainable solutions. Thus, objective and factual knowledge on environmental issues plays a crucial part in sustainable behavior and environmental citizenship [3]. As a matter of fact, environmental knowledge is one of the components predicting environmental citizenship [4] and can also be considered as part of a larger construct of environmental citizenship [5,6].

This study is focused on young adults as they have the potential to be agents of change in environmental policy through their active and competent civic participation [7]. A recent qualitative study on the perceptions of education for environmental citizenship in Lithuania (where the present study took place) indicated that stakeholders representing both secondary and higher education, in addition to policy makers, feel as if there is not enough being done to help young people develop their environmental citizenship, but they are also hopeful regarding the trends in the country [8]. Thus, this study is a pilot of a working model of environmental citizenship that could be later used as a change model with variables that are targetable through interventions and as a descriptive benchmark for environmental citizenship in Lithuania.

It must be noted that in a recent conceptualization of environmental citizenship it is defined as “the active participation of citizens, in the private and public sphere, through individual and collective actions, toward solving current environmental problems and preventing the creation of new ones, parallel to developing a healthy relationship with nature” [2,9]. However, from this point of view, environmental citizenship is not a singular construct, but an umbrella term encompassing environmentally beneficial beliefs, values, attitudes, skills, competences, and behavior [6]. Thus, while in the present study we treat environmental citizenship as a singular dependent variable, the whole model that is tested can be said to be measuring environmental citizenship as a holistic system of interacting parts.

Environmental citizenship is a relatively new concept that has gained some interest from researchers and educators in recent years [6,10,11,12] and there are studies being conducted that operationalize environmental citizenship in a variety of ways, many of which include measures of pro-environmental behaviors and attitudes [5,13]. There is a lack, however, of a dedicated model that could be used to understand environmental citizenship as a process, i.e., as not only a measure of environmental citizenship, but also as a means of understanding how environmental citizenship develops. While several models have been proposed [2,4,5], they are more directed towards a conceptual understanding of the various components of environmental citizenship, not at how these components interact and shape relevant outcomes. This study addresses this gap in the literature.

The aim of the present study is to test an updated and simplified model of environmental citizenship proposed by Hawthorne and Alabaster [4]. The original model, published in 1999, was developed after an extensive investigation of the literature and was later tested empirically. However, the initial model can be too complex to be practically applicable in an intervention setting, as it includes components that are generally stable over time, such as personality traits or political beliefs [14]. After an extensive literature search, no cases where this model would have been applied as a structural model for investigating environmental citizenship were found, thus this appears to be the first attempt to apply the model in such a way.

The original model consists of more than 30 distinct variables, which can further be sub-divided into even more variables [4]. This makes an assessment of the aforementioned model quite complicated. This is because some of the variables require quite straightforward measurements (e.g., demographics), while other variables require long questionnaires (e.g., personality traits and other individual difference variables, environmental attitudes). Such an approach makes the model quite extensive, but practically difficult to use because of the length of the potential questionnaire for such a model. The problem further deepens if one intends to use the same questionnaire in a longitudinal context, because (even after assessing individual difference variables just once) the sheer length of the instrument can deter respondents from repeated participation in a study. If applied in an educational context, one would also expect a questionnaire that can comfortably be filled-in in no more than 45 min and without tiring the participants. Thus, a smaller model might be preferable, allowing for an easier assessment of environmental citizenship that does not take too much time and is not as emotionally and cognitively taxing.

For the present study, the model by Hawthorne and Alabaster [4] was reduced to its core components that are more easily targeted in an educational setting and through social interventions (Figure 1). The core components were chosen based on how many other variables seem to converge into them, while still retaining the general structure of the model and the relationships among the variables. This decision was made in hopes to provide an easy to apply framework for interventions, where interventions can focus on specific broad concepts that they can target, and any specificity would then stem from the intervention itself.

1.1. The Components of the Model

The components of the model tested in the present study (Figure 1) will be introduced below. In designing this research, several variables were considered as potential covariates or moderators, namely—political orientation [15] and the big five personality traits [16]. Political affiliation/interest and personality traits were included in the original model of environmental citizenship by Hawthorne and Alabaster [4], however, these variables tend to be more stable over time and are therefore not as useful as components of a structural model that consists of components that are subject to change, such as attitudes or knowledge. The aforementioned variables are used in an exploratory manner only and are used to investigate their potential role in the model for future research.

1.1.1. Need for Learning

The need for learning or the desire to learn about environmental issues encompasses one’s attitudes toward learning—whether one perceives it as a desirable and enjoyable activity or not. It also reflects one’s willingness to engage in various learning activities related to environmental issues. In general, a desire to learn is a pedagogical concept that describes one’s want and need to engage with learning materials and the need to acquire knowledge [17,18].

1.1.2. Abstract Environmental Knowledge

Abstract environmental knowledge represents one’s perceived knowledge of what actions are most environmentally friendly [4,19,20]. It must be stressed that the subjective nature of this construct means that one can perceive oneself to be knowledgeable regarding environmental issues, but this may or may not mean that one possesses factual knowledge regarding one’s actions.

1.1.3. Concrete Environmental Knowledge

Concrete environmental knowledge complements abstract environmental knowledge in the sense that concrete knowledge is based on the correct understanding of facts. It is the practical counterpart of the subjective construct of abstract knowledge. Concrete knowledge informs one’s real-life actions [4] (i.e., knowledge on what can and cannot be recycled may influence one’s consumer behavior) and can be practically useful in promoting pro-environmental behavior both in private and in public [3].

1.1.4. Environmental Awareness (Consciousness)

Environmental awareness represents one’s awareness of the consequences of one’s actions on the environment [20]. Environmental awareness is built on existing environmental knowledge and potentially shapes the attitudes one holds regarding the environment [4]. Similar constructs have been found to be an important part of models predicting pro-environmental behavior, such as the comprehensive action determination model [21,22].

1.1.5. Environmental Attitudes

Environmental attitudes have found their place as a prominent predictor of pro-environmental behavior and other related constructs in various contexts [23]. Perhaps the most widely used instrument to assess general environmental attitudes is the New Ecological Paradigm scale [23,24,25]. Environmental attitudes represent one’s subjective beliefs regarding the balance of nature and how it relates to the growth of human activity [25].

1.1.6. Environmental (Self-)Education

Environmental (self-)education can be viewed as a process of acquiring concrete and abstract knowledge regarding the environment, but also can contribute to the development of environmentally positive attitudes and environmental awareness, as the more one engages in learning about environmental issues, the more one understands why environmentally responsible behavior is necessary [4]. It must be noted that engaging in self-education or participating in educational activities may have varied results, depending on the quality of the education, thus this construct is inherently subjective as it reflects only engagement in educational activities and not their content or quality. Therefore, environmental (self-)education can be best understood as one’s willingness to acquire knowledge regarding environmental issues.

1.1.7. Environmental Literacy

Environmental literacy is the outcome of acquired environmental knowledge and environmental education [4]. It encompasses one’s understanding of the complex issues regarding the environment and an understanding of the possibilities of civic participation to be an agent for positive change in mitigating environmental issues. Environmental literacy can also be understood as ones competence to engage in environmental citizenship behavior [5].

1.1.8. Environmental Citizenship

As mentioned before, environmental citizenship can be understood as both a singular construct of pro-environmental behavior [4] and as an umbrella term encompassing all variables leading to one’s engagement in civic action publicly and in private to mitigate environmental issues [2,5]. In the tested model environmental citizenship is regarded as an outcome variable, operationalized through one’s willingness to engage in civic activities that would help in mitigating environmental issues [8].

1.1.9. Need for Action

Need for action or the desire to act largely stems from one’s personality [4] as it depends on one’s general approach toward social situations and the willingness to engage in various activities—all largely determined by one’s inherent traits [26,27,28,29]. In the current model, the need for action reflects one’s civic participation in pro-environmental movements.

1.2. Relationships between the Components of the Model

In their original research, Hawthorne and Alabaster [4] found support for their proposed model, however, more recent studies have both somewhat changed the conceptualization of environmental citizenship [2,6], and have included different variables in models assessing it [5,11,30]. The recently developed Environmental Citizenship Questionnaire [5], while a valuable tool in its own right, does not propose concrete structural relationships among the variables that comprise environmental citizenship, leaving the development of change models a task for future research. Other research on environmental citizenship has focused on assessing pro-environmental behavior, knowledge, and attitudes as an outcome without separately assessing environmental citizenship [13].

The lack of a unified approach in studying environmental citizenship has long been identified as a problem [4] and is only now being systematically addressed [5]. However, there is still a need to explore alternative ways of both conceptualizing and investigating environmental citizenship. The present study addresses this need.

No hypotheses are proposed in this research as it is exploratory in nature. The only expected outcome is that the directional relationships indicated in Figure 1 will be observed. Beyond the model of environmental citizenship by Hawthorne and Alabaster [4], the paths present in the model have individual studies supporting them. Environmental literacy has been conceptualized as an aggregate of environmental knowledge, as well as skills, attitudes, and behavior related to pro-environmental actions [31], and environmental knowledge has been shown to be linked to environmental awareness. Environmental education, while being linked to pro-environmental attitudes, behaviors, and perceptions [32,33,34] is also related to the desire to learn about environmental issues [35]. Furthermore, environmental attitudes and environmental activism also have been found to be strongly linked [36]; thus, the whole model tested in the present study is generally supported by the literature, albeit in parts. The model is regarded as subject to change based on data, as this is a pilot study to assess the functioning of a simplified model adapted from Hawthorne and Alabaster [4] in order to use the model in future research with a representative sample.

2. Materials and Methods

2.1. Sample Characteristics

The sample for this study was planned based on power analysis, with the intention to detect correlations of r = 0.3 or higher with the statistical power of 0.95. The minimum sample size for this would be 134 responses, but we set a safer limit of 150 observations. Participants were recruited through social media and snowball sampling as well as taking advantage of student mailing lists. A total of 293 individuals filled in the questionnaire. Because the study is aimed at young (emerging) adults, respondents who are 41 years of age or older were excluded from the analysis. This resulted in a final sample size of 267 observations, which exceeds the planned sample size.

Participant age ranged from 18 to 40 years of age, with a mean age of 24 years (SD = 5). Most participants were women (200 cases, 76% of total sample), four participants declined to indicate their sex. Approximately 48% of the participants indicated that they were employed at the time. Only seven participants (2.6% of the total sample) indicated that they were not studying at the time; almost half of the participants were studying in the field of social sciences (49.4%), the rest were studying in the fields of humanities (22.8%), natural sciences (15%), technology (7.2%), and agricultural science (3%).

2.2. Procedure

Data collection started on 24 October 2021 and ended on 12 November 2021. Potential participants were sent an invitation to fill in an online survey on environmental citizenship. This invitation was also disseminated through social media and encouraged potential participants to share the invitation with their friends.

Upon clicking the link to the survey, participants were given an introduction about the study and were informed that by starting to fill in the survey they are giving consent to use their data. The survey was completely anonymous. Participants filled in demographic questions first, followed by other measures.

2.3. Measures

The survey consisted only of these measures which are listed below. Participants did not complete any additional measures. Measures were presented to participants in the order they are listed below.

2.3.1. Demographic Variables

Participants were asked to indicate their biological sex, their age, their employment status, their field of study (if they are a student at the time), their net income and their monthly expenses. A variable of leftover income was computed by subtracting monthly expenses from monthly income.

2.3.2. Political Beliefs

Political beliefs [15] were assessed with 10 items rated on a 5-point scale (1—completely disagree, 5—completely agree); the measure assessed two dimensions (socially liberal vs. socially conservative and libertarian vs. statist) with 5 items per dimension. Higher scores on both dimensions represent a view that favors more individual freedom.

The internal consistency of the social dimension was poor (ω = 0.404) and was marginally acceptable on the economic dimension (ω = 0.560). This measure is used for exploratory purposes and is not tested in a model.

2.3.3. Personality Traits

The Big Five personality traits were assessed with the extra-short form of the BFI-2 inventory [16]. The measure assesses five personality traits: openness, conscientiousness, extraversion, agreeableness, and neuroticism. All traits were assessed with three items each. Items were rated on a 5-point scale (1—completely disagree, 5—completely agree). While the measure showed good model fit in a confirmatory factor analysis (CFI = 0.982, TLI = 0.977, SRMR = 0.060, RMSEA = 0.031, DWLS estimation was used), not all items loaded significantly into their factors.

The internal consistency was acceptable for conscientiousness (ω = 0.699), extraversion (ω = 0.669), neuroticism (ω = 0.784), and was poor for openness (ω = 0.348) and agreeableness (ω = 0.385). This measure is used for exploratory purposes and is not tested in a model.

2.3.4. Need for Learning

To assess the need for learning about environmental issues, a measure was developed based on the Attitude/Motivation Test Battery [37]. The measure consists of 10 items (e.g., “when learning about the environment, I would prefer to focus on information rather than playing games”, “learning about environmentally friendly behavior is interesting to me”). Items were rated on a 5-point scale (1—completely disagree, 5—completely agree). See Appendix A for full item list.

The scale showed good internal consistency (ω = 0.917).

2.3.5. Abstract Environmental Knowledge

Abstract environmental knowledge was assessed with a measure adapted from Kim and Stepchenkova [19] and Mohiuddin et al. [20]. The measure consists of 5 items (e.g., “I know how to recycle properly”) rated on a 5-point scale (1—completely disagree, 5—completely agree). See Appendix A for full item list.

The scale showed good internal consistency (ω = 0.782).

2.3.6. Concrete Environmental Knowledge

Concrete environmental knowledge was assessed with an objective knowledge test [38]. The original test consisted of 30 items, four of which were removed because of irrelevance, leaving 26 items with five response options, of which one is correct for each item. The total score for a participant was computed by adding up all correct responses. The scale in the form that was used in the present study has a minimum score of 0 and a maximum score of 26.

To assess internal consistency, the KR-20 coefficient was computed (KR-20 = 0.529). Bearing in mind that this is an objective knowledge test, this value can be considered acceptable.

2.3.7. Environmental Awareness (Consciousness)

Environmental awareness (also sometimes called environmental consciousness) was assessed with 5 items (e.g., “I understand the consequences of climate change”) all rated on a 5-point scale (1—completely disagree, 5—completely agree). The measure was adapted from Mohiuddin et al. [20]. See Appendix A for full item list.

The scale showed good internal consistency (ω = 0.828).

2.3.8. Environmental Attitudes

Environmental attitudes were assessed using the New Ecological Paradigm (NEP) scale [25]. The scale consists of 15 items all rated on a 5-point scale (1—completely disagree, 5—completely agree). Example item: “Humans are severely abusing the environment.”

The scale showed good internal consistency (ω = 0.795).

2.3.9. Environmental (Self-)Education

To assess environmental (self-)education, 8 items were designed for this research. Out of these items, 5 assess (self-)education during the past month and 3 items assess change in opinion regarding environmental topics during the past month. Based on an exploratory factor analysis, the items regarding (self-)education and change in opinion tended not to fall into one factor, thus only the first five items assessing (self-)education are used. The scale consists of items such as “During the past month I researched environmental questions at least a couple of times” and “During the past month I refreshed my knowledge regarding the environmental impact of various foods.” See Appendix A for full item list.

The scale showed good internal consistency (ω = 0.874).

2.3.10. Environmental Literacy

Environmental literacy was assessed with a measure developed by Hadjichambis and Paraskeva-Hadjichambi [5] (see Part B.2 of the ECQ). The measure consists of 11 items rated on a 5-point scale, assessing the degree to which one believes one knows how to contribute to the mitigation of environmental issues (1—I completely do not know how, 5—I fully know how). Example item: “I know how to contribute to the prevention of environmental problems.”

The scale showed good internal consistency (ω = 0.899).

2.3.11. Environmental Citizenship

Environmental citizenship was assessed with a measure developed by Hadjichambis and Paraskeva-Hadjichambi [5] (see Part C.9 of the ECQ). The scale consists of three items rated on a 5-point scale (1—very unlikely, 5—very likely). Example item: “I would try to change society in such a way that it becomes more environmentally friendly.”

The scale showed good internal consistency (ω = 0.756).

2.3.12. Need for Action

The need for action was assessed with a measure developed by Hadjichambis and Paraskeva-Hadjichambi [5] (see Part A.1 of the ECQ). The scale consists of 6 items rated on a 4-point scale, assessing one’s participation in pro-environmental movements and activities (1—I have done that in the past half year, 2—I have done that in the past year, 3—I have done that but more than a year ago, 4—I have never done that). Example item: “Have you participated in an environmental action group?”

The scale showed good internal consistency (ω = 0.824).

2.4. Analysis Strategy and Data Availability

All analyses were conducted with JAMOVI 2. Structural equation modelling was conducted using the robust variant of the diagonally weighted least-squares (DWLS) method because of its ability to accurately estimate models with non-normally distributed data and its general effectiveness with estimating effects in small samples [39,40,41,42,43]. Data and the pre-registration of the study are openly available (see data availability statement at the end of the article).

3. Results

Descriptive statistics and linear relationships among the variables tested in the model of environmental citizenship as well as demographic variables are presented in

Table 1. Descriptive statistics of demographic variables and variables included in the tested model.

	M (SD)	S	K	r
				1	2	3	4	5	6	7	8	9	10	11	12
1. Age	23.8 (4.85)	1.320	1.280	—
2. Income (in Euros)	547 (484)	1.470	2.92	0.568 ***	—
3. Expenses (in Euros)	189 (164)	1.560	4.760	0.385 ***	0.462 ***	—
4. Leftover income (in Euros)	356 (433)	1.610	4.160	0.493 ***	0.942 ***	0.135 *	—
5. Need for learning about environmental issues	3.12 (0.833)	−0.049	−0.380	0.104	0.002	0.053	−0.021	—
6. Abstract environmental knowledge	3.54 (0.668)	−0.347	0.129	0.054	0.025	0.078	−0.01	0.375 ***	—
7. Concrete environmental knowledge	19.5 (2.820)	−0.147	−0.563	0.163 **	0.020	0.078	−0.004	0.135 *	0.154 *	—
8. Environ-mental awareness	4.05 (0.657)	−0.697	1.890	−0.087	−0.123 *	−0.054	−0.112	0.303 ***	0.457 ***	0.103	—
9. Environmental attitudes (NEP)	3.82 (0.492)	−0.257	0.256	−0.112	−0.059	0.010	−0.067	0.322 ***	0.144 *	0.183 **	0.393 ***	—
10. Environmental (self-)education	2.83 (1.010)	−0.213	−0.649	0.132 *	0.044	0.095	0.005	0.668 ***	0.409 ***	0.189 **	0.229 ***	0.207 ***	—
11. Environmental literacy	3.05 (0.675)	−0.204	0.803	0.073	−0.008	0.027	−0.028	0.374 ***	0.591 ***	0.070	0.426 ***	0.083	0.404 ***	—
12. Environmental citizenship	3.46 (0.851)	−0.404	0.355	−0.023	−0.037	0.072	−0.071	0.567 ***	0.348 ***	0.119	0.384 ***	0.307 ***	0.489 ***	0.463 ***	—
13. Need for action	1.51 (0.607)	1.760	3.250	−0.011	−0.002	0.073	−0.037	0.143 *	0.161 **	−0.073	0.033	−0.041	0.190 **	0.106	0.201 ***

Note. * p < 0.05, ** p < 0.01, *** p < 0.001; S—Skewness, K—kurtosis.

. All variables that make up the model of environmental citizenship have acceptable skewness and kurtosis values for linear models. Many of the variables are strongly positively correlated, but none are correlated to the degree that collinearity would become an issue in constructing a linear model. Linear relationships between the variables of the model and political orientation as well as personality traits are presented separately in

Table 2. Linear relationships between political orientation, personality traits, and the variables included in the model of environmental citizenship.

	Social Political Dimension	Economic Political Dimension	Agreeableness	Conscientiousness	Extraversion	Openness	Neuroticism
Age	−0.194 **	−0.200 **	−0.013	0.150 *	0.162 **	0.021	−0.151 *
Income	−0.108	−0.097	0.117	0.222 ***	0.187 **	0.012	−0.127 *
Expenses	−0.083	0.010	0.070	0.168 **	0.151 *	0.072	−0.018
Leftover income	−0.081	−0.115	0.106	0.180 **	0.127 *	−0.024	−0.127 *
Need for learning about environmental issues	−0.020	0.021	0.215 ***	0.201 ***	0.243 ***	0.316 ***	−0.028
Abstract environmental knowledge	0.104	0.088	0.199 **	0.217 ***	0.155 *	0.261 ***	−0.166 **
Concrete environmental knowledge	−0.018	−0.371 ***	−0.129 *	0	−0.020	0.111	−0.093
Environ-mental awareness	0.048	0.038	0.147 *	0.161 **	0.086	0.156 *	0.006
Environmental attitudes (NEP)	0.082	−0.108	0.095	0.096	−0.001	0.071	0.133 *
Environmental (self-)education	0.004	−0.027	0.027	0.121 *	0.189 **	0.307 ***	−0.057
Environmental literacy	−0.068	0.056	0.066	0.175 **	0.256 ***	0.164 **	−0.128 *
Environmental citizenship	0.06	0.012	0.179 **	0.249 ***	0.329 ***	0.333 ***	−0.026
Need for action	−0.017	0.016	0.171 **	0.036	0.158 **	0.208 ***	−0.050

Note. * p < 0.05, ** p < 0.01, *** p < 0.001.

.

The economic dimension of the political compass had a significant negative correlation with concrete environmental knowledge, indicating that it is likely that individuals who prefer more economic freedom are less knowledgeable about environmental issues. All five personality traits had significant relationships with the various components of the model of environmental citizenship, suggesting that personality traits might be important for a more complete understanding of the development of environmental citizenship.

Since all variables that were planned to be included into a structural model have both acceptable internal consistency and appropriate skewness and kurtosis, a structural equation model of environmental citizenship was tested using the robust variant of the diagonally weighted least-squares estimation method (Figure 2).

It is generally accepted that in determining model fit, CFI and TLI values should be at least 0.950, while SRMR values should be 0.09 or lower and RMSEA values should be 0.06 or lower [44]. The initial model, displayed in Figure 1, showed insufficient fit to the data, indicating that the specified paths do not correspond to the actual relationships among the variables in the model (CFI = 0.690, TLI = 0.571, SRMR = 0.154, RMSEA = 0.178; χ²(26) = 245, p < 0.001). Thus, it was decided to investigate the modification indices of the tested model to search for suggested paths that would also be theoretically appropriate and reasonable.

Based on the inspection of the modification indices, it was decided to amend the model with three additional paths: from abstract environmental knowledge leading to environmental literacy, from environmental (self-)education leading to environmental literacy, and from the need for learning about environmental issues leading to environmental citizenship. Environmental literacy has often times been defined as consisting of environmental knowledge, as well as other components [31,45,46]; thus, the paths from knowledge toward literacy are consistent with the current literature. Additionally, an environmental citizen is one that always strives to learn about the current environmental issues and the ways one can help in mitigating them [2,6]; thus, the addition of the path leading from the need for learning about environmental issues toward environmental citizenship seems appropriate in the context of the developing theory around environmental citizenship and education for environmental citizenship. These paths, while not specified in the initial research by Hawthorne and Alabaster [4], seem theoretically appropriate and practically reasonable.

The updated model demonstrated excellent fit to the data (CFI = 0.978, TLI = 0.965, SRMR = 0.060, RMSEA = 0.051; χ²(23) = 38.7, p = 0.021). The results of the path-analytical model are presented in Figure 3.

Overall, the model is able to explain approximately 50% of the variance of environmental citizenship, which is a substantial amount for path-analytical models in which we do not control for internal consistency with the use of latent variables [47].

4. Discussion

Initially, the role of political orientation and personality traits was explored as possible future additions to the model. Both political views [36,48,49] and personality [28,29,50,51,52,53] have previously been explored in relation to pro-environmental behavior and environmental activism, showing that these constructs indeed may be valuable in creating a more complete understanding of environmentally relevant behavior. In the present study, both instruments had poor internal consistency, thus we are unable to draw any robust conclusions regarding their association with the components of the tested model of environmental citizenship. However, the significant correlations that were observed provide a basis to explore both political orientation and personality traits in future research on environmental citizenship, using more robust measures of the aforementioned constructs. Civic behavior is inherently political [15,54], it is inseparable from one’s political beliefs. It must be noted that both political beliefs and personality traits have at least some hereditary basis [14], making them important variables to consider if we are to understand the environmental citizenship of different individuals. In this sense, it becomes reasonable to adopt a person-oriented approach toward these variables [55,56,57], relating them with the model not as covariates, but as moderators for groups of differing individuals with relatively stable characteristics that cannot be targeted through interventions.

The original model of environmental citizenship that was tested did not fit the data well, but all expected relationships, except the path leading from environmental attitudes to the need for action, were significant. These results provide some support for the original model, but also encourage to search for alternative structures of it in order to better represent the relationships that were found in the data. Thus, based on the modification indices of the first model, a second iteration of the model was computed with three additional paths, which resulted in a model that fit the data perfectly.

Overall, the relationships fond in the model are mostly consistent with the previous literature [4], with some exceptions. While environmental literacy is thought to result from pro-environmental knowledge [31], in the updated model, concrete knowledge becomes an insignificant predictor of environmental literacy. Previous research also has found environmental attitudes to be linked with environmental activism [36], yet in the present study the path from environmental attitudes toward need for action remains insignificant in both the original and the updated models. Overall environmental citizenship is significantly predicted by the need for learning, environmental literacy, and the need for action, which combined can explain around 50% of the variance of the dependent variable.

Interestingly, environmental education contributes significantly to both abstract and concrete knowledge, but only abstract knowledge contributes significantly toward explaining environmental literacy. One would reasonably expect that factual knowledge would be a prominent predictor of literacy, being often thought as a component of it [31,58]. However, because environmental literacy, as well as abstract environmental knowledge are subjective self-reports in the present study, they likely do not represent actual competence and are rather the expressions of perceived competence [59].

The aforementioned finding illustrates a divide between one’s perceived competence regarding environmental issues and how to address them and their actual fact-based knowledge. People are prone to overestimate their competence, yet to have factual knowledge and perceived literacy regarding environmental issues uncorrelated is worrying in times of ever-growing interest in environmental activism.

A mismatch between factual knowledge and perceived competence regarding environmental issues has the potential to lead to undesirable social and environmental consequences [59], especially if activist groups strive for goals that are not evidence-driven. It thus seems that, at least in the present sample of young Lithuanian adults, environmental education should be a priority, focusing in particular to bridge their perceived competence with their actual competence. One way of doing so is through education for environmental citizenship, with a strong emphasis on evidence and factual knowledge.

A stronger basis of factual knowledge could be a positive force in environmental activism [7], as an informed citizen is harder to mislead and to co-opt for a movement that may have undisclosed intentions unrelated to its officially stated cause. An environmental citizen, by definition, is one who is both informed and empowered to act in ways that benefit the environment [2]. The data suggest that in the present sample only the latter is true. Such findings are a strong indication that quality education for environmental citizenship is needed [6,8].

Limitations and Future Directions

The convenience sample of the present study is a limiting factor. While the data provide insight to the structure of the tested model of environmental citizenship, women are over-represented in the sample and the snowball sampling method used in the present study may have resulted in a sample that is more homogeneous than would be optimal for making robust conclusions.

It must be noted that due to poor internal consistency, the reader should carefully interpret the correlations between measures of the political compass and other variables, as well as the correlations between openness and agreeableness and other variables. These variables were included in the analysis purely for exploratory and descriptive purposes. Future studies should strive to use more robust measures of pollical beliefs that reliably capture the nuanced view of each respondent. It may be prudent to include longer versions of personality trait measures as well in order to reliably assess participants’ traits, as in the present study the extra-short version of the BFI-2 was used.

This study is intended as an exploration of the structure of the model, and the size of the sample was intentionally kept low for practical reasons. The size of the sample and the way the sample was gathered, in addition to the fact that women are overrepresented in the sample, means that the results might not be as robust with other samples and are not necessarily representative. Future research should strive for larger samples that would allow at least partial use of latent variables when testing the model. Assessing the model with latent variables would provide additional insight into the true effect sizes of the relationships represented in the model. Additionally, researchers could look into whether the model functions the same for men and women.

While the data of the present study, in addition to previous research, suggest that political orientation and personality traits have potential value in studying environmental citizenship, future research should use more robust measures of these constructs to be able to use them as moderators. Such use of these variables also implies the need for a substantially larger sample that would allow multigroup structural equation modelling.

5. Conclusions

The updated model of environmental citizenship functions well and should be investigated further to provide more robust data on its functioning.

In the present sample, young adults’ concrete (factual) environmental knowledge is not related to their perceived environmental literacy, thus indicating that their environmental citizenship cannot be expressed to its full potential, as it requires one to have all the necessary information and competence to make correct decisions through civic participation.