From Words to Deeds: How Do Knowledge, Effectiveness, and Personal Relevance Link Environmental Concern and Buying Behavior?

ABSTRACT

The gap between environmental attitudes and behaviors has been on research agendas for a while. Despite the enormous efforts of all concerned parties to increase consumer engagement in environmental issues, the levels of individual environmental concern are still higher than actual green purchasing. Considering the shortcomings in theory and practice, the purpose of this paper is to examine the link between environmental concern and environmental buying behavior by introducing three mediating variables, namely consumer environmental knowledge, perceived consumer effectiveness, and perceived personal relevance. The hypotheses were tested on a sample of 319 consumers using structural equation modeling. The results show that environmental concern predicts environmental buying behavior. Environmental concern also influences consumer knowledge, perceived consumer effectiveness, and perceived personal relevance, while environmental buying behavior is affected by knowledge and effectiveness. The model testing confirmed a partially mediated model. The findings offer several avenues for public policy makers, academics, and socially responsible companies that find the environment important.

KEYWORDS:

• Environmental concern
• environmental buying behavior
• consumer environmental knowledge
• perceived consumer effectiveness
• perceived personal relevance
• mediation

Introduction

With ever-growing consumption jeopardizing achievement of the Nations (2021) sustainable development goals, it seems that consideration of the environment in consumption is more important than ever. Previous studies on consumer behavior suggest that consumers with a higher level of environmental concern are more likely to engage in ecologically conscious consumer behavior; however, raising only environmental concern has been found to be insufficient to achieve meaningful change (Tam & Chan, 2017) with numerous examples showing how consumer demand has not resulted in widespread adoption of environmentally-friendly products (Wymer & Polonsky, 2015). Although the idea that there is an attitude-behavior gap has been circulating among academics and practitioners for some time (Auger & Devinney, 2007; Carrigan & Attalla, 2001; Jayaratne et al., 2015; Prothero et al., 2011; Roberts & Bacon, 1997; Vermeir & Verbeke, 2008), White et al.’s (2019) recent framework for encouraging sustainable consumer behavior demonstrates that there is still room for improvement, considering that the discrepancy between what consumers say and do poses a challenge for marketers in various domains even today (White et al., 2019). Furthermore, Newton et al. (2015, p. 1974) argue that “an unresolved issue in the literature is whether the relationship between environmental concern and environmental purchase intentions is conditional upon the existence of additional constructs.” Analogously, Gifford and Nilsson (2014) encouraged researchers to focus more on moderating or mediating effects when determining pro-environmental behavior.

We respond to this notion and propose to examine the mechanism that underlies the relationship between environmental concern and buying behavior through three mediating variables: consumer environmental knowledge, perceived consumer effectiveness, and perceived personal relevance. In this way, our contribution to the literature is multi-fold. By introducing knowledge, personal relevance, and effectiveness as mediators, we offer a novel set of constructs to explain the link between environmental concern and behavior. In line with this reasoning, we propose environmental concern is not only the determinant of behavior, which has been the focal relationship in previous research, but also of other constructs, such as knowledge, relevance, and effectiveness, which were rarely explained by environmental concern before (see Gifford & Nilsson, 2014; Joshi & Rahman, 2015). Considering the study of Millennials by Heo and Muralidharan (2019), whose main focus was on the relationship between environmental knowledge and pro-environmental behavior with environmental concern and perceived consumer effectiveness as mediators, we propose a different constellation of constructs. Thus, we aim to demonstrate an alternative and more refined perspective, which suggests the connection between environmental concern and buying behavior is not so straightforward, as it is possible that the predictive ability of environmental concern spans across additional constructs. Moreover, in the proposed set, we include a previously under-examined construct in the pro-environmental consumer behavior literature, i.e. perceived personal relevance, which enables us to capture the importance consumers place on environmental issues and respond to previous studies which identified a need to include more variables into the interrelationships between environmental concern, knowledge, and behavior (Di Martino et al., 2019; Pagiaslis & Krontalis, 2014). Personal relevance is important as environmentally-friendly products generally provide benefits for the individual, in addition to short- or long-term societal or environmental benefits for others (Davari et al., 2017). Finally, the proposed model is tested on a general consumer sample in a central European country, i.e. Slovenia, where environmentalism is gaining recognition (Golob et al., 2017). Our sample choice enables us to extend the findings of previous pro-environmental studies, which focused on young consumers (e.g., Heo & Muralidharan, 2019; Di Martino et al., 2019; Perera et al., 2018). Moreover, Tam and Chan (2017) found that the link between environmental concern and behavior differs across societies, so introducing a novel cultural setting could represent an incremental contribution as well.

In this manner, the purpose of this paper is to examine consumer environmental buying behavior, starting with environmental concern as the main factor in explaining this behavior. This relationship is further explored by proposing novel mediating roles for environmental knowledge and perceived consumer effectiveness, but also by adding another potentially important mediating factor, i.e. perceived personal relevance, in the context of Slovenian consumers, where the issue of sustainable development is gaining a great deal of attention.

Theoretical background and hypotheses development

While an increasing number of studies imply that the awareness of consumers as to the effects and consequences of their behavior on the environment is rising (Trudel, 2019; White et al., 2019), researchers are still puzzled by the complex relationship among consumer environmental concern, intention, and behavior (strong concern and intention to behave in an environmentally friendly manner, but an evident lack of actual environmentally friendly actions in consumer behavior). Novel research suggests that increased environmental concern and anti-consumption attitudes do not necessarily result in less consumption or a better impact on the environment (e.g., Kropfeld et al., 2018). It is becoming obvious that in order to achieve better behavioral results, a deeper understanding of the factors interfering in the relationship between environmental concern and behavior is still required.

Due to this apparent intention-behavior gap, researchers are making significant efforts to better understand the factors that might enhance the translation of environmental concern and intentions into genuine behavior, and to structure the existing knowledge by offering comprehensive integrative reviews. For example, Gifford and Nilsson (2014) named 18 different personal and social factors that influence the relationship between environmental concern and behavior. These factors vary from demographics (age, gender) to deep consumer characteristics (values, a sense of control, and responsibility), and further research thereon in moderating and mediating roles was suggested to provide a better understanding of their influence. Furthermore, the effects of demographics (income and education) on sustainable behavior and consumption were confirmed by Milfont and Markowitz (2016), while Diamantopoulos et al. (2003) showed the effectiveness of socio-demographic characteristics for differentiating between consumer knowledge and concern, while these characteristics were not shown to be significant for behavior. Joshi and Rahman (2015) researched the importance of factors influencing consumer perceptions, purchase intentions, and actual behavior (purchases) regarding green products. They found two major groups of relevant factors that determine consumer environmental behavior: individual factors (variables specifically related to the decision-maker, such as emotions, environmental concern, perceived consumer effectiveness or the extent to which one’s behavior can make a difference, personal values (environmental, social, ethical) and norms), and situational factors (product price, eco-labeling and certification, product attributes and quality, product availability, subjective norms/social norms and reference groups, brand image). Additionally, research so far has also investigated actions that can help bridge the gap between consumer environmental concern (as an attitude) and consumer behavior. Steg and Vlek (2009) reviewed and emphasized interventions as an effective way of fostering behavioral change related to the environment, while White et al. (2019) named social influence, habit formation, individual self, feelings and cognition, and tangibility as relevant psychological processes of consumer engagement in environmentally friendly behavior.

In recent times, scholars started to delve into more complex relationships among environmentally-charged constructs with novel studies testing various concepts in their efforts to explain environmental concern, intentions, and behavior. Researchers searched for additional constructs related to ethics to extend existing relationships in the Theory of planned behavior (Rex et al., 2015). Furthermore, habit and self-identity were reported as mediators in relation to environmental intentions (Gkargkavouzi et al., 2019), while environmental knowledge moderated the relationships between different lifestyles and green product purchasing (Sheng et al., 2019). Additionally, the relationship between consumer green intention and behavior was shown to be moderated by purchase situation (Grimmer et al., 2016). Previous studies empirically confirmed that the link between environmental concern and behavior was mediated by self-direction (Zibenberg et al., 2018), learned helplessness (Landry et al., 2018), religiosity (Bhuian et al., 2018), and culture and its characteristics (Morren & Grinstein, 2016; Tam & Chan, 2017).

The choice of constructs in our study was guided by the research of Brochado et al. (2017), who found that the individual or psychographic factors were more relevant than demographic variables. For example, environmental concern was found to be one of the factors with the strongest impact on environmental behavior in a recent study by Hosta and Zabkar (2020). What is more, Kang et al. (2013) led us to the trio of individual factors, i.e. perceived consumer effectiveness, environmental knowledge, and personal relevance. The importance of perceived consumer effectiveness, together with green product availability, was established in intention-behavior gap research (Nguyen et al., 2017), while Wei et al. (2018) recognized the mediating role of perceived consumer effectiveness on the relationship between environmental concern and willingness to pay more for environmentally friendly products. Perceived consumer effectiveness along with consumer knowledge was found to be a relevant factor of consumer pro-environmental self-identity in a recent study examining the mediating effects of pro-environmental self-identity on sustainable buying and curtailment (Dermody et al., 2018). Finally, Heo and Muralidharan (2019) proved the relationships between environmental knowledge, environmental concern, and perceived consumer effectiveness. These constructs were observed in a slightly different order than in some earlier research (e.g., Schahn & Holzer, 1990). Based on these outlines, below we present the main constructs and related hypotheses of our study.

Environmental concern

Environmental concern has been defined as having positive attitudes about sustainability, representing the level at which people are attentive to environmental problems and possible solutions (Bickart & Ruth, 2012). It captures different degrees of individuals’ consciousness with regard to environmental issues and their efforts to resolve them (Dunlap & Jones, 2002). The concept of environmental concern is very often listed as an imperative in understanding green consumer behavior (Heo & Muralidharan, 2019).

Prior studies suggest that consumers’ concern for environmental issues has a strong impact on purchasing environmentally friendly products (e.g., Heo & Muralidharan, 2019; Milfont & Markowitz, 2016; Mishal et al., 2017; Morren & Grinstein, 2016; Paul et al., 2016; H. Wang et al., 2019; White et al., 2019). It has been shown that environmental concern has a positive effect on other sustainable actions such as adopting renewable energy or a fictitious green energy brand (Bang et al., 2000; Hartmann & Apaolaza-Ibanez, 2010). Other empirical studies have also proved the importance of this factor in the process of environmentally conscious consumer behavior (Brochado et al., 2017; Kautish & Sharma, 2019). Follows and Jobber (2000) indicate that when consumers are purchasing green products they balance between environmental concern and the attributes of the product. Higher levels of concern make consumers consider the environmental features of a product (H. S. Kim & Damhorst, 1998). In line with these outlines, we hypothesize:

H1: Environmental concern has a positive effect on environmental buying behavior.

Environmental knowledge

According to Fryxell and Lo (2003, p. 48), environmental knowledge can be defined as “general knowledge of facts, concepts, and relationships concerning the natural environment and its major ecosystems.” The literature on consumer environmental practices often postulates that environmental concern is a general belief construct that functions as a predecessor of a variety of more specific constructs, such as environmental knowledge (Pagiaslis & Krontalis, 2014).

Following Zaichkowsky (1985), it is possible that consumers with a high level of involvement in an environmental issue are more willing to search for information. This may lead to increasing levels of knowledge regarding a particular environmental topic. The positive link between concern and knowledge has been proposed in previous studies in other settings (Bang et al., 2000; Marcketti & Shelley, 2009), although empirical support was mixed. In a recent study of green consumer behavior, Pagiaslis and Krontalis (2014) found that environmental concern has a positive effect on knowledge. Accordingly, we expect that environmentally concerned consumers who are involved in this issue are consequently more likely to be knowledgeable about the environment and eco-friendly products.

H2a: Environmental concern has a positive influence on environmental knowledge.

Consumer environmental buying behavior is often a reflection of consumers’ environmental knowledge, which is presented as one of the most influential factors affecting desired behaviors (Joshi & Rahman, 2015; Nguyen et al., 2017). As an important determinant, knowledge can contribute to permanent changes in attitudes and behavior (Bator & Cialdini, 2000). It can be explained by the fact that higher levels of environmental knowledge mean that consumers have more information, which guides them to buy more green products and, in this way, behave more sustainably (N. Lee et al., 2012; Mahesh & Ganapathi, 2012).

In exploring consumer environmental buying behavior, Goh and Balaji (2016) reported that consumers with higher levels of environmental knowledge were more likely to buy environmentally friendly products. In contrast, a recent study by He et al. (2019) did not show that environmental knowledge has a significant relationship with environmentally conscious consumer behavior. In general, many studies reveal that environmental knowledge positively influences both intention and actual pro-environmental consumption behavior (Mostafa, 2007; Pagiaslis & Krontalis, 2014; P. Wang et al., 2014). Based on the prevalent previous findings, we propose the following relationship between environmental knowledge and buying behavior:

H2b: Environmental knowledge has a positive influence on environmental buying behavior.

Perceived consumer effectiveness

Studying the gap between environmental concern and buying behavior, Roberts (1996) suggests that a critical factor in interpreting this relationship is perceived consumer effectiveness. Perceived consumer effectiveness is defined as a “domain-specific belief that the efforts of an individual can make a difference in the solution to a problem” (P. S. Ellen et al., 1991, p. 103). In the environmental setting, perceived consumer effectiveness can be explained as the belief that each individual can contribute to the reduction of environmental problems by the right choice of products (Roberts, 1996). The main issue is that many consumers believe the environmental impact of their product choices would be negligible (Wymer & Polonsky, 2015). They are not aware that their small everyday practices can lead to positive impacts on sustainable change (Jayaratne et al., 2015).

Roberts (1996) reported that when people feel that they can be very effective in saving the environment through a specific action, they will show more concern for the environment by performing that action. In developing our hypothesis, we rely on Y. J. Lee et al. (2014), who discovered that altruistic value, which is concerned with interest in others (e.g., human beings, the biosphere), positively influences perceived consumer effectiveness. Analogously, we can assume that consumers who are interested in environmental issues are predisposed to solve these problems. Consequently, they will have a positive view of their contribution to problem resolution (Y. J. Lee et al., 2014). Consumers who are more concerned about the environment may also believe that they can do something to save it. Following this reasoning, we hypothesize:

H3a: Environmental concern has a positive influence on perceived consumer effectiveness.

P. S. Ellen et al. (1991) argue that perceived environmental effectiveness is a different factor than environmental concern or attitudes and has an influence on environmentally conscious behavior. Consumers’ concern about environmental issues might not necessarily lead to green purchases, but when they have a strong belief that their behavior is the right one, they are more likely to engage in such behavior. On a similar note, McCarty and Shrum (1994) argued that consumers’ belief about their ability to influence future outcomes influences their pro-environmental behavior. In their meta-analysis of 15 studies, Hines et al. (1987) report that consumers who performed actions that contributed to environmental sustainability were more often those with high perceived effectiveness, whereas consumers who thought that certain changes were due to other factors showed environmentally responsible behavior less often.

Many researchers have explored the unique contribution of perceived environmental effectiveness to the prediction of different pro-environmental behaviors (e.g., Liu et al., 2012; Mostafa, 2007; P. S. Ellen et al., 1991; Vermeir & Verbeke, 2006). However, it seems likely that consumers feel that a single purchase of a green product or the purchases of any one individual do not make a difference to or impact the environment (Gleim et al., 2013). This has been shown in the case of Millennials, where perceived consumer effectiveness, or the belief in solving a specific environmental problem alone did not result in actual purchases (Heo & Muralidharan, 2019). Conversely, perceived consumer effectiveness had a positive impact on green purchase intention in the study of H. Wang et al. (2019), as well as in the case of Kautish and Sharma (2019) and Brochado et al. (2017), who showed that perceived consumer effectiveness has a direct and positive effect on environmentally conscious consumer behavior, while P. S. Ellen et al. (1991) found that perceived consumer effectiveness was a powerful driver of three environmental behaviors: buying, recycling, and contributing to environmental groups. Thus, we hypothesize the following:

H3b: Perceived consumer effectiveness has a positive influence on environmental buying behavior.

Perceived personal relevance

Perceived personal relevance is described as a personal postulate that a particular behavior is related to personal interest and in line with one’s lifestyle habits (Celsi et al., 1992). Related to environmental issues, Kang et al. (2013, p. 445) defined perceived personal relevance “as the extent to which an individual believes that environmentally responsible consumption is consistent with his/her own personal lifestyle, values, social presentations and self-images.” Consumers’ concern for the environment is to a great extent connected with concern for their own health or the future of their children (Howarth & Norgaard, 1995). If consumers are concerned about the environment, then environmental consumption may represent a large part of who they are. In other words, the more concerned they are about the environment, the more they perceive environmental consumption as personally relevant, which leads to the following hypothesis:

H4a: Environmental concern has a positive influence on perceived personal relevance.

In general, consumption patterns are influenced by the way people perceive and evaluate themselves; therefore, people buy products that are compatible with their identity, social status, and values (Belk, 1988). It is important to note that when a certain action is relevant to consumers, they are more motivated to engage in that action (McQuarrie & Munson, 1992). Scholars report that when consumers perceive that buying a green product is relevant and beneficial for themselves or is a health-related concern for them or their family, it becomes a significant factor in choosing that particular product (Prakash & Pathak, 2017). Exploring green consumer behavior with regard to adopting green electricity, Ozaki (2011) demonstrated that consumers concerned about the environment were not willing to adopt this green behavior, with the main reason being the lack of personal relevance to the consumers’ lives. Ozaki (2011) also pointed out that in order for a consumer to adopt some green change, it must accord with their individual identity, self-image, and values. Related conclusions can also be found in other studies by Pickett‐Baker and Ozaki (2008) and Hustvedt and Dickson (2009).

In exploring the influence of perceived consumer effectiveness in their research on green textile and apparel consumption, Kang et al. (2013) concluded that perceived personal relevance influences attitudes, subjective norms, and perceived behavioral control regarding the consumption of green apparel, although they did not specifically test if it influences behavioral intentions or actual buying behavior. We propose that when a consumer sees that a certain behavioral action is personally relevant, he or she is more likely to behave positively toward that action. This provides the foundation for developing the next hypothesis:

H4b: Perceived personal relevance has a positive influence on environmental buying behavior.

Mediating role

In addition to the proposed hypotheses, we aim to address the mediating role of the three main constructs, hence, based on the evidence presented above, we propose the following hypothesis:

H5: The relationship between environmental concern and environmental buying behavior is mediated by (a) environmental knowledge, (b) perceived consumer effectiveness, and (c) perceived personal relevance.

In summary, in order to examine the relationship between environmental concern and buying behavior, we developed the conceptual model presented in Figure 1.

Figure 1. Conceptual model of the relationship between environmental concern and buying behavior.

Research methodology

Data collection

This study was conducted in Slovenia, where the issue of sustainable development is gaining a great deal of attention and people rate themselves as fairly conscious consumers, although an attitude-behavior gap is evident as well (Golob et al., 2017). Data was gathered by distributing a survey online through social media platforms and by e-mail directing respondents to the online questionnaire. Invitations to participate were sent on a voluntary basis first to friends and colleagues, who were asked to share it with other people, taking the form of “snowball sampling” (Saunders et al., 2009). By clicking on the link, respondents were directed to Qualtrics, an online platform for creating and analyzing surveys. The total number of participants was 319, of which 21.9% were male and 78.1% female. The overall mean age was 34.6 years, with the youngest respondent 17 years old, and the oldest 75. Regarding the level of education of all participants, 39.8% had completed at least a bachelor degree. Data about the perceived standard of living show that a majority of the respondents, i.e. 63%, stated that their standard of living was average.

Study instrument

The questionnaire was developed in English using scales from existing literature and then translated into Slovene. During translation, attention was devoted to lexical, idiomatic, and experiential meaning, as well as to grammar and syntax. The level of environmental concern was measured using the scale from Bang et al. (2000) with four items. A sample item is: “How concerned are you about the environment when making purchases?” The scale ranged from 1 (not at all concerned) to 4 (very concerned). While previous studies measured environmental knowledge in two different ways, by using perceived or factual environmental knowledge, the focus of this study was on perceived environmental knowledge, as endorsed by Mostafa (2007). The original scale was developed by P. Ellen et al. (1997). Therefore, environmental knowledge was assessed with five items, such as: “I know how to select products and packages that reduce the amount of waste ending up in landfills.” In order to measure perceived consumer effectiveness, a scale from Kang et al. (2013) was used. The scale consisted of four items, originally developed by Roberts (1996). A sample item is: “When I buy products, I tend to try to consider how my use of them will affect the environment.” Perceived personal relevance was also measured based on Kang et al. (2013), who used five items originally adapted from the self‐relevance scale of Celsi et al. (1992). An example of this scale is “The purchase and/or use of eco-friendly packaged products lets others see me as I ideally would like them to see me.” Consumer buying behavior was explored by adopting a scale from Y. Kim and Choi (2005), as a set of five items, i.e. “When I have a choice between two equal products, I buy the one less harmful to other people and the environment.” The measurement scales used can be found in Table 1.

Table 1. Measurement scales.

Scale/Item	Standardized factor loading	Cronbach’s alpha	Mean value	Standard deviation
Environmental concern (Bang et al., 2000)		.832	M = 4.17*	SD = 0.77
How concerned are you about the environment when making purchases?	.717		4.13	0.94
How concerned are you about air pollution?	.724		4.19	0.89
How concerned are you about water pollution?	.755		4.39	0.87
How concerned are you about land use?	.794		3.98	1.05
Environmental buying behavior (Y. Kim & Choi, 2005)		.839	M = 3.83	SD = 0.75
I make a special effort to buy paper and plastic products that are made from recycled materials.	.647		3.84	0.84
I have switched products for ecological reasons.	.759		3.61	1.03
When I have a choice between two equal products, I buy the one less harmful to other people and the environment.	.734		4.14	0.85
I make a special effort to buy household chemicals such as detergents and cleaning solutions that are environmentally friendly.	.695		3.76	1.08
I have avoided buying a product because it had potentially harmful environmental effects.	.749		3.82	1.02
Perceived consumer effectiveness (Kang et al., 2013)		.752	M = 4.43	SD = 0.57
It is worth it for an individual consumer to make efforts to preserve and improve the environment.	.508		4.62	0.64
Since each individual can have any effect upon environmental problems, what I do can make meaningful difference.	.785		4.26	0.78
By purchasing eco-friendly packaged products, each consumer’s behavior can have a positive effect on the environment and society	.870		4.42	0.68
Perceived personal relevance (Kang et al., 2013)		.845	M = 3.65	SD = 0.78
The purchase and use of eco-friendly packaged products helps me to attain the type of life I strive for.	.790		3.81	0.90
I can make connections or associations between the purchase and use of eco-friendly packaged products and other experiences and/or behaviors in my life.	.738		3.62	0.94
The purchase and use of eco-friendly packaged products is of personal importance to me.	.842		3.94	0.85
The purchase and use of eco-friendly packaged products helps me to express who I am.	.670		3.25	1.08
Environmental knowledge (Mostafa, 2007; P. Ellen et al., 1997)		.830	M = 3.41	SD = 0.70
I know that I buy products with packages that are environmentally safe.	.580		3.13	0.83
I know more about Eco packaging than the average person.	.769		3.25	0.96
I know how to select products and packages that reduce the amount of waste ending up in landfills.	.816		3.64	0.88
I understand the environmental phrases and symbols on product package.	.679		3.52	0.95
I am very knowledgeable about environmental issues.	.668		3.53	0.91

* Environmental concern was measured with a 4-point scale, whereas all other constructs were operationalized with a 5-point scale. For the analysis, the scores obtained for the environmental concern items in the questionnaire were also converted to a 5-point scale.

Common method bias

To account for the potential bias and variance that may result from the instrument used, common method bias was tested. Defined as “variance that is attributable to the measurement method rather than to the construct of interest ‘‘(Bagozzi & Yi, 1991, p. 426), its presence in the data set and consequently the results compromise their validity. Commonly used techniques to assess common method bias are Harman’s single factor test and common latent factor (Eichhorn, 2014). It is assumed that common method bias exists in the data when these techniques show 50% or more of the explained variance. In our analysis, Harman’s single factor test was conducted in SPSS for the items measuring the chosen constructs. The results showed that 38.11% of the variance can be explained by the single factor. Since the single factor does not explain more than 50% of the variance, it can be concluded that common method bias does not create a problem in this analysis. The common latent factor evaluated in AMOS resulted in R² of 13.10%, which confirms that the measurement instruments did not bias the obtained variance.Table 2

Table 2. Hypotheses testing.

Hypothesis	Relationship	Standardized Estimate	Result
H1a	Concern → Behavior	0.144*	Supported
H2a	Concern → Knowledge	0.464***	Supported
H2b	Knowledge → Behavior	0.318***	Supported
H3a	Concern → Effectiveness	0.343***	Supported
H3b	Effectiveness → Behavior	0.103	Not supported
H4a	Concern → Relevance	0.605***	Supported
H4b	Relevance → Behavior	0.509***	Supported

* p < .05, *** p < 0.001

Results

Structural equation modeling was used to test the proposed conceptual model. Our data analysis consisted of acquiring assessments of construct reliability and validity (measurement model analysis) and testing the latent variable relationships proposed in our hypotheses (structural model analysis). This procedure is established in structural equation modeling (Anderson & Gerbing, 1988).

Measurement model analysis

The analysis of the measurement model enabled us to assess the relationship between the indicators and related latent variables through evaluation of the validity and reliability of the proposed construct measures (Diamantopoulos et al., 2000). For this purpose, we calculated the average variance extracted (AVE) and composite reliability (CR) for all constructs in the model. All constructs satisfied the threshold values of AVE (<0.5) and CR (<0.7) (Table 3). Internal consistency and scale reliability were also calculated for each of the constructs. Cronbach’s alpha was used for this purpose; its value for each of the constructs in the model was above 0.7, indicating the adequacy of the scales. The discriminant validity of the involved constructs was assessed using the heterotrait-monotrait ratio (HTMT), as suggested in the existing literature (Henseler et al., 2015; Voorhees et al., 2016). If the HTMT is below 0.85, discriminant validity is established. Performing this analysis on our dataset (see Table 4) gave satisfactory results (Henseler et al., 2015).

Table 3. AVE and CR for the model constructs.

Construct	AVE	CR	Cronbach’s alpha
Environmental concern	0.560	0.836	0.832
Perceived knowledge	0.543	0.773	0.830
Perceived effectiveness	0.500	0.831	0.752
Personal relevance	0.584	0.848	0.845
Buying behavior	0.517	0.842	0.839

Table 4. Discriminant validity overview (HTMT analysis).

	Environmental concern	Perceived effectiveness	Perceived knowledge	Personal relevance	Buying behavior
Environmental concern
Perceived effectiveness	0.551
Perceived knowledge	0.519	0.356
Personal relevance	0.683	0.705	0.667
Buying behavior	0.687	0.612	0.730	0.813

Note: The values reported in this table are the HTMT ratio of correlations between the model constructs. HTMT values lower than 0.85 indicate discriminant validity.

The model analysis conducted in AMOS showed that the measurement model has a good overall fit. The Chi-square equaled 350.064 with 179 degrees of freedom (the Chi-square/d.f. ratio was 1.956). Other fit indices were satisfactory: standardized RMR = 0.0577, CFI = 0.947, IFI = 0.947, NFI = 0.898, and RMSEA = 0.055.

Structural model analysis

The relationships proposed in our hypotheses were tested in the structural model, again by using the AMOS program. While the Chi-square metric was significant (Chi-square = 350.836, df = 180, Chi-square/d.f. = 1.949), the other indicators of overall model fit predominantly show satisfactory values: standardized RMR = 0.0586, CFI = 0.947, IFI = 0.947, NFI = 0.898, and RMSEA = 0.055.

The hypotheses testing (see Table 4) indicated a strong and stable influence of environmental concern on environmental behavior, confirming H1. Furthermore, environmental concern had a positive effect on consumer environmental knowledge (H2a), which further influenced environmental buying behavior (H2b). In addition to this, environmental concern positively influenced perceived consumer effectiveness (H3a); however, the latter did not affect environmental buying behavior. Finally, H4a and H4b were confirmed as higher levels of environmental concern led to higher levels of perceived personal relevance, which resulted in environmental buying behavior.

We conducted additional analysis in order to test H5. The confirmed path and regression weight significance along with the supported direct relationship between environmental concern and behavior imply the existence of partial mediation between environmental concern and pro-environmental behavior through environmental knowledge and perceived personal relevance (Jose, 2013). Model appropriateness and validity were additionally assessed by performing a formal test of mediation. Our proposed model was compared to a model without a direct path from environmental concern to behavior. The determined difference in the chi-square values in the two models (Δχ² = 5.035) with 1 degree of freedom proved to be significant at p < .05, confirming the robustness of our proposed model of partial mediation (Werner & Schermelleh-Engel, 2010). Based on these results, we found partial support for H5, seeing that environmental knowledge and perceived personal relevance were found to mediate the relationship between environmental concern and pro-environmental behavior.

Discussion

The study delved into the links between environmental concern and environmental buying behavior by introducing three potential mediators, i.e. consumer environmental knowledge, perceived personal relevance, and perceived consumer effectiveness. As the results showed, the proposed relationship between environmental concern and buying behavior is partially mediated by consumer environmental knowledge and perceived personal relevance. In addition, we found that environmental concern significantly influences perceived consumer effectiveness, which is not a significant predictor of buying behavior. These results should be viewed through the prism of previous studies.

In examining the results, we confirmed a positive relationship between environmental concern and behavior, which is not surprising. As has been shown in previous studies, higher levels of environmental concern lead to more environmental purchasing among Slovenian consumers, making them comparable to consumers from other countries, such as the USA (Heo & Muralidharan, 2019), Italy (Cerri et al., 2018), India (Kautish & Sharma, 2019; Trivedi et al., 2018), China (H. Wang et al., 2019), Turkey (Aytekin & Büyükahraz, 2013), and Thailand (Maichum et al., 2016), to name a few. The same conclusion has been drawn in other similar contexts, such as in the field of green hotel visits, where environmental concern significantly and positively influenced consumer’s intention to visit green hotels (Verma et al., 2019) or in the context of energy-efficient appliances where higher levels of environmental concern were manifested in willingness to pay a premium price for such products (Li et al., 2019).

Additionally, consumer environmental knowledge has been confirmed as a mediator between environmental concern and buying behavior. While most of the previous studies focused on environmental knowledge as an antecedent to behavior (Joshi & Rahman, 2015), we advance this knowledge and demonstrate its mediating role as well. Liobikienė and Juknys (2016) suggested that a lack of knowledge may be the reason for the gap between environmental concern and behavior. We can confirm this proposition as consumers with a lower level of environmental concern are less knowledgeable about environmental buying options, which is reflected in their avoidance of environmentally friendly buying choices. We echo the findings of Pagiaslis and Krontalis (2014), who also found a similar path in the context of renewable energy and biofuels, while our study focused more on environmentally friendly products.

Perceived personal relevance was also introduced as a mediator between environmental concern and behavior. The literature shows contradictory results for the effect of perceived personal relevance on consumer buying behavior, because, as Follows and Jobber (2000) explain, when consumers are purchasing green products they balance between the consequences for the environment, captured by environmental concern, and the consequences for the individual. The mediating role of perceived personal relevance in our study was shown to be significant and positive, thus linking environmental concern and buying behavior. This conclusion is an extension of previous research carried out by Song and Kim (2019), who showed that motives related to self-centered interests might be better predictors of green buying behavior than altruistic reasons. Similarly, Ozaki (2011) confirmed that the main reason concerned consumers were not willing to adopt green electricity was a lack of personal relevance.

Finally, environmental concern had a positive influence on perceived consumer effectiveness. Surprisingly, the latter did not affect consumer buying behavior, which is comparable to the results of Mishal et al. (2017), who explored the pattern of green purchasing behavior. They confirmed the positive link between environmental concern and perceived consumer effectiveness, but not with buying behavior. Although the positive influence of perceived consumer effectiveness on green purchasing behavior has been confirmed by many authors (e.g., Joshi & Rahman, 2015; Taufique & Vaithianathan, 2018; H. Wang et al., 2019), our study did not support this. One of the reasons may be that the other two mediators, i.e. knowledge and relevance, carry more weight in determining buying behavior. Additionally, Mishal et al. (2017) suggest that the discrepancy with regard to previous studies may be due to cultural differences and the beliefs of consumers. Missing instant gratification might be another reason for the insignificance of this relationship, as it has been expected to demotivate consumers to take environmentally friendly actions (Gaudelli, 2009; Heo & Muralidharan, 2019).

Theoretical contributions

We believe that the findings from our study contribute to the existing pool of literature and research on consumer environmental behavior in several ways. First, the mediating role of two out of the three proposed factors was confirmed, thereby establishing that perceived knowledge and perceived personal relevance are not only antecedents of attitudes, norms and behavioral control, as was the focus of the study by Kang et al. (2013), but they mediate the relationship between environmental concern and behavior as well. Second, adding to previous studies on environmental concern that primarily looked at the antecedents of environmental concern or behavior as its main outcome (see Gifford & Nilsson, 2014), we demonstrated that not only behavior, but also other outcomes of environmental concern, i.e. knowledge, relevance, and effectiveness, are important. Third, we would especially like to point out perceived personal relevance as a novel relevant factor which previously received the least amount of attention in the environmental consumption domain, but was shown to determine behavior and be influenced by concern. Fourth, the testing of the proposed model showed that Slovenian consumers are comparable to their counterparts from other countries, as several established relationships (between three factors and environmental buying behavior) were re-confirmed.

Implications for practice

These research findings have numerous implications for companies, governments, and public policy makers. The results can contribute to the formulation of useful and valid marketing strategies, regulations, and campaigns that can lead to an increase in environmental buying behavior. The main indication stemming from this research is that environmental concern is a fundamental element explaining consumer buying behavior. All involved parties can promote pro-environmental behavior by tapping into consumer environmental concern. Marketers of environmentally friendly or eco packaged products could promote pro-environmental behavior by addressing consumer awareness of an environmental issue and presenting a solution through their product(s). This can be done through mass media campaigns or more personalized approaches by employing opinion leaders who could stress the importance of environmental concern in everyday buying behavior.

As consumers’ environmental knowledge holds implications for marketers in the business and public policy sectors, they need to be certain about the level of consumers’ knowledge of the environment and eco-friendly products in order to build effective communication and raise knowledge. Doing marketing research to capture consumer knowledge needs to be followed by creating and maintaining communication that will be both highly informative and encouraging to help shape consumer decisions and behavior in an adequate, environmentally friendly manner. Providing more information on different ways consumers can engage in environmentally friendly buying behavior and the all-encompassing benefits of environmentally friendly products can subsequently raise the level of such behavior.

Public policy makers and marketers can take advantage of the findings that consumers need to feel empowered in the struggle to achieve sustainable change, i.e. they need to believe that their actions can make a difference. Communication campaigns can focus on pointing out the contribution of each individual’s environmental actions to the greater good with the goal of motivating and inspiring individuals to support sustainable choices. Also, it has been found that when people feel that sustainable consumption is relevant to them personally, they are more likely to buy such products. Directly emphasizing environmentally friendly lifestyles and values – as well as an environmentally friendly self-image – in communication strategies (Kang et al., 2013) can raise the level of personal relevance, which has been shown to have positive effects on behavior as well.

Limitations and directions for future research

While this study contributes to the existing literature on the environmental attitudes of and behavior research on consumers, it is not without limitations. Namely, a social desirability bias which is the “tendency of individuals to present themselves in the most favorable manner relative to prevailing social norms and mores” (King & Bruner, 2000, p. 80) is largely present when discussing environmental issues and reporting behavior. As such was not addressed in this study, it represents a limitation thereof and creates an opportunity for future research to deepen the understanding of the results obtained by including a measure of social desirability bias (see King & Bruner, 2000; Randall & Fernandes, 1991). Since we used a convenience sample, the results cannot be generalized due to the sampling method and sample size used in this study. In this regard, there may be an issue of sample selection bias, as it seems possible that people with strong pro-environmental attitudes are more likely to be overrepresented in the sample, because they were more motivated to participate (Hage et al., 2009). In order to achieve the generalizability of the results obtained, further research using a more random sample, as well as a larger sample size, is necessary. In addition, the study was conducted in a single country. While consumers are becoming more similar in the globalized economy, similar studies carried out in other countries would be required in order to be able to claim the universal validity of the results. Furthermore, if we consider the UN’s Sustainable Development Goals (United Nations, 2021), even Slovenia, which is ranked 12th out of 166 countries, faces major challenges in achieving several of these goals, such as responsible production and consumption, and climate action (Sachs et al., 2020). Therefore, future research opportunities lie in investigating how to combat climate change and ensure sustainable consumption and production patterns in this country. Finally, the focus of the study was on examining the connection between environmental concern and buying behavior and the impact knowledge, relevance, and effectiveness can have on this relationship. While the three mediators were relevant, they cannot be considered to be the only factors in the relationship between environmental concern and buying behavior. Previous literature elaborated on some additional constructs that can be of interest (e.g., Bouscasse et al., 2018; Newton et al., 2015). Further research might include some of these constructs as mediators, and evaluate their influence on the buying behavior of consumers.

Disclosure statement

No potential conflict of interest was reported by the author(s).