https://doi.org/10.1177/2158244020902074
SAGE Open
January-March 2020: 1 –10
© The Author(s) 2020
DOI: 10.1177/2158244020902074
journals.sagepub.com/home/sgo
Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License
(https://creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of
the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages
(https://us.sagepub.com/en-us/nam/open-access-at-sage).
Original Research
Introduction
Climate change, industrial pollution, and excessive personal
consumption have all been increasingly affecting the environ-
ment and bringing progressively harsher effects on human
life (Carfora et al., 2017; Thøgersen, 2009). Meanwhile, peo-
ple’s awareness over the importance of environmental protec-
tion has been growing, so much so that it, alongside the goal
to achieve sustainable development, has been slowly turning
into a consensus among nations worldwide. Governments
and enterprises, as important economic entities, have started
an endeavor toward environmental protection by changing
production methods, developing green products, and adjust-
ing environmental protection policies. However, consumers’
role in environmental protection cannot be ignored; past
research showed that a reduction in environmental hazards
produced by consumers by increasing pro-environmental
consumption behavior was a very significant step toward
environmental protection. Thus, the role of pro-environmen-
tal consumption behavior is very important in the establish-
ment and maintenance of environmental protection.
Mainieri et al. (1997) defined pro-environmental con-
sumption behavior as a type of volunteer behavior that con-
sciously seeks to tackle environmental issues, such as climate
change, global warming, and environmental degradation.
This type of behavior has been shown to minimize the nega-
tive impact of one’s actions in the environment through the
purchase of organic products that are environmentally bene-
ficial (Mainieri et al., 1997). Typical pro-environmental con-
sumption behaviors include the purchase of environmentally
responsible products that minimize environmental impact,
products from firms with good environmental reputations,
and/or products produced using biodegradable, carbon neu-
tral or recycled inputs, and so on (Cleveland et al., 2012).
Many studies have analyzed pro-environmental consump-
tion behaviors from multiple perspectives, and, although
pro-environmental consumer behavior has been more widely
studied in recent years, there are still major research prob-
lems in the subject; we believe that the key to resolve this
problem does not lie within the understanding of what this
type of behavior is or how to activate it—topics often
902074SGOXXX10.1177/2158244020902074SAGE OpenWang et al.
research-article20202020
1Zhejiang University of Finance & Economics, Hangzhou, China
Corresponding Author:
Jian Gao, School of Business Administration, Zhejiang University of
Finance & Economics, Hangzhou 310018, China.
Email: jettgao@zufe.edu.cn
Effect of Green Consumption Value
on Consumption Intention in a Pro-
Environmental Setting: The Mediating
Role of Approach and Avoidance
Motivation
Jianguo Wang1, Jianming Wang1, and Jian Gao1
Abstract
Based on the theory of consumer values, this study aimed to examine the relationship between green consumption values and
pro-environmental consumption intention by establishing a “value-motivation-intention” model and to check the moderation
effect of green involvement. In total, 741 shoppers were recruited. Data analyses showed that (a) green consumption values
positively influenced pro-environmental consumption intention; (b) the behavioral approach system positively influenced
pro-environmental consumption intention, but the behavioral inhibition system did not; (c) the behavioral approach system
positively mediated the relationship between green consumption values and pro-environmental consumption intention; and
(d) green involvement positively moderated the relationship between green consumption values and pro-environmental
consumption intention.
Keywords
green consumption values, pro-environmental consumption intention, behavior activation system, behavior inhibition system
2 SAGE Open
explored in previous literature. We identified the research
gap by this question: How do we make pro-environmental
consumption behavior become not a short-term but a long-
term behavior?
Previous research has found that people’s values are posi-
tively related to pro-environmental consumption behavior
(Stern, 2000). Stern (2000) suggested the values–belief–
norm theory and divided people’s values into three catego-
ries: egoism values, altruistic values, and biological values;
the same author also discussed the relationship between
these values and pro-environmental behavior. Chan (2001)
used evidence from Chinese consumers and discussed the
effect of a man–nature orientation on pro-environmental
behavior. Furthermore, people’s values are a relatively stable
state of mind, as they tend to remain the same for a certain
period of time. Therefore, it is necessary to analyze the rela-
tionship between values and pro-environmental consump-
tion behavior as well as the role of consumer contextual
factors and personal traits.
Thus, this study primarily aimed to explore the mecha-
nisms behind the relationship between green values and pro-
environmental consumption behavior, and to analyze the
mediating effect of approach and avoidance motivation on
consumers’ psychological motivation characteristics when
purchasing pro-environmental products.
This study secondarily aimed to explore the moderating
effect of green involvement (GI) and analyzed the differ-
ences between consumers who have different levels of GI.
Theoretical Background and
Hypotheses
Pro-Environmental Consumption Behavior
Recently, pro-environmental consumption behavior has been
receiving increasing attention in the literature (Lacroix,
2018; Lange & Dewitte, 2019; Mainieri et al., 1997; Maio &
Wei, 2013; Moser, 2015; Steinhorst & Klöckner, 2018;
Urban et al., 2019; Welsch & Kühling, 2009). This focus is
consistent with an increasingly broader interest in under-
standing pro-environmental behavior that has persisted for
several decades (e.g., Hines et al., 1987; Kollmuss &
Agyeman, 2002; Lange et al., 2018). Overall, such studies
mainly focused on achieving harmony between man and
nature through the realignment of consumer behavior, and
they have mostly sought to understand the differences among
individual consumers regarding their pro-environment con-
sumption behavior. Some of these studies have focused on
finding pro-environmental consumers and segmenting them
in the market, something that can be illustrated by some
researches’ description of pro-environmental consumers:
They regard that these consumers are mostly female and
older, and have more income and higher educational levels.
Nonetheless, these same studies were limited in the sense
that they did not analyze why and how pro-environmental
consumption behavior is generated.
Therefore, other studies have tried to solve this problem and
turned their research perspectives into understanding how pro-
environmental consumption behavior can be promoted: One
study identified that both psychological and contextual factors
are important variables affecting pro-environmental consump-
tion behaviors (Ertz et al., 2016). Specifically, the psychologi-
cal factors that predict pro-environmental consumption
behavior include environmental attitude, social norms, motiva-
tion, perceived value (Gifford & Nilsson, 2014; Miao & Wei,
2013), and personal ethics (Bamberg, 2003). Regarding the
contextual factors, they include interpersonal relationships,
laws, and the convenience of recycling facilities; the stimula-
tion of consumers’ perceptions regarding such contextual fac-
tors was shown to be able to change their psychological factors,
thereby producing actual pro-environmental consumption
behavior (Guagnano et al., 1995; Steg & Vlek, 2009; Stern,
2000). In sum, this means that pro-environmental consumption
behavior is a result of the interweaving between psychological
and contextual factors. Notwithstanding, there are few studies
exploring which elements among these factors enable the con-
tinuous promotion of pro-environmental consumption behav-
ior, so it is a topic that requires further examination.
Theory of Consumption Values
Based on the theory of consumption values, green con-
sumption values (GCVs) were defined as one’s tendency to
express its own environmental protection values through
his or her purchases and consumption behavior (Haws
et al., 2014). Many researchers suggested that consumers’
GCVs are an important factor that guides consumer behav-
ior and affects their preference regarding which goods and
services they access in a pro-environmental context
(Candan & Yıldırım, 2013; Gonçalves et al., 2016). Another
study showed that the pro-environmental outcomes of one’s
GCVs are achieved through intrinsic and extrinsic factors
that are associated with the components of a given purchase
(Biswas & Roy, 2015). This finding suggested that GCVs
are part of a more extensive nomological network associ-
ated with conversations beyond environmental resources; it
emphasized three facets of consumer choice behavior: (a)
consumer choice of what to buy and what not to buy, (b)
consumer choice of preferring one type of product to
another, and (c) consumer selection between different
brands. Consumers with stronger GCVs would be more
conscientious in the use of financial and physical resources
(Biswas & Roy, 2015).
Regarding financial resources, past research suggests that
green consumption (or conservation) may be related to con-
cerns regarding spending money (e.g., family size).
Regarding physical resources, consumers that are more
aware of green consumption may try to use products in their
entirety and try not to use more than the necessary amount,
all with the intent for the product to effectively perform its
function (Haws et al., 2014). Furthermore, one’s GCVs have
remained unchanged for a period of time.
Wang et al. 3
Pro-Environmental Consumption Behavior
and GCVs
Many studies suggested that value was a critical influential
factor of pro-environmental consumption behavior. Stern
(2000) subdivided pro-environmental consumption behavior
into two categories according to their environmental impact
during the stages of production and consumption: green con-
sumerism and the purchase of major household goods and
services. These categories denote that pro-environmental
consumption behavior not only relied on the actual behavior
but also guided by one’s inner values.
De Groot and Steg (2015) suggested that individuals with
environmental values, such as apathy for nature, personal
inclination toward preserving the planet, and ecocentric phi-
losophies, were more committed to displaying pro-environ-
mental behaviors. Nguyen et al. (2016) found that biospheric
values can influence pro-environmental consumption behav-
ior; other research supported this same conclusion, for
instance, Qasim et al.’s (2019) research subdivided people’s
values into social value, conditional value, epistemic value,
and emotional value. By doing so, they found that condi-
tional value, emotional value, and epistemic value had a sig-
nificant positive influence on consumers’ behavioral
intention to consume organic food.
Behavioral intention is regarded as an important predictor
of actual behavior (Ajzen, 1991). Pro-environmental con-
sumption behavior is the practical expression of one’s pro-
environmental consumption intention (PCI). Thus, it can be
said that the stronger the PCI, the more likely the actual pro-
environmental consumption behavior will occur. Therefore,
pro-environmental consumption behavior was replaced by
PCI as an outcome variable. In the relationship between
GCV and PCI, GCVs included values of environmental con-
sumption, such as purchasing green or recyclable production
and reducing consumption environmental hazard. Therefore,
we assumed that PCI was influenced not only by biospheric
and environmental values but also by GCVs, mainly because
GCVs are more embedded in the consumption context than
are environmental values. Thus, we hypothesized the
following:
Hypothesis 1 (H1): GCVs will positively influence PCI.
Approach and Avoidance Motivation
When consumers face green products, in terms of decision
making, they need to judge their benefits and compare the
differences between expected and actual results. Consumers
with strong GCVs tend to consider the benefits that purchas-
ing products bring not only to themselves but also to the
environment. In the process of evaluation, such people tend
to make their choices toward a more environmentally friendly
result (owing to its importance for them) and avoid environ-
mentally hazardous results.
Furthermore, approaching happiness and avoiding pain is
the most important nature of human beings. Gray (1987) pro-
posed a behavioral motivation theory (Reinforcement
Sensitivity Theory [RST]), which describes the behavioral
approach system (BAS) and the behavioral inhibition system
(BIS). BAS refers to one’s sensitivity to rewards and positive
stimuli, activating behaviors that produce positive or plea-
surable goals, whereas BIS refers to one’s sensitivity to pun-
ishment cues and negative stimuli, inhibiting behaviors that
produce unpleasant outcomes to avoid them (Carver &
White, 1994; Merchán-Clavellino et al., 2019). Spielberg
et al. (2011) found that the activation approach (BAS, or
approach motivation) was associated with left-lateralized
middle frontal gyrus activation and that the region responds
to stimuli associated with expectations, harvests, and plea-
sure. Conversely, avoidance (BIS, or avoidance motivation)
was associated with right-lateralized middle frontal gyrus
activation and that the region responds to stimuli associated
with disappointment, loss, and pain. Therefore, BAS and BIS
are not related to the same part of the nervous system, mean-
ing that both systems could, potentially, be activated at the
same time.
Recently, these two concepts have been directly studied in
research areas such as emotion, sense of power, risk predic-
tions, sports psychology, and human experiences (Keltner
et al., 2003; Lochbaum & Gottardy, 2015; Updegraff et al.,
2004). These researches usually examine two aspects of
these two systems: (1) How to activate the BAS and BIS,
such as how Keltner et al. (2003), which suggested that high-
power perception activates the BAS; in another example,
Kramer and Yoon (2007) found that both positive and nega-
tive emotional information influenced individuals’ BAS. (2)
The factor which can be affected by the BAS and the BIS,
such as Carver and White (1994), which found that the BAS
can influence and evoke different emotions that makes con-
sumers have different emotional experiences in the con-
sumption process, which are affected by the BAS and the
BIS, findings that are supported by Arnold and Reynolds
(2012).
Moreover, BAS and BIS are closely related to consumer
behavior. Based on the regulatory focus theory (Higgins
et al., 2001), it was found that the BAS and the BIS could
govern how people pursue goals, and could be a chronic pre-
disposition of individuals or could be situational induced
(Aaker & Lee, 2001; Higgins et al., 2001). This finding
denotes that specific situations could activate the BAS and
the BIS. Arnold and Reynolds (2012) discussed the effect
mechanism of approach (BAS) and avoidance motivations
(BIS) on hedonic consumption in a retail setting, and found
that, as fundamental motivational dispositions, BAS and BIS
are cross-situational and could be positively related to
hedonic shopping motivations.
Generally speaking, BAS should be related positively to
PCI for several reasons. (a) Keltner et al. (2003) suggested
that high-power perception would activate the BAS; in the
4 SAGE Open
pro-environmental consumption setting, the consumer could
feel high-power perception by helping in the diminishment
of environmental degradation, thereby activating the BAS,
and the consumers’ environment improvement goal would
subsequently lead to PCI. (b) PCI can bring others’ moral
recognition and provide consumers with positive feedback
for their actions, which helps consumers to perceive the posi-
tive value brought by their behavior; these expectations that
the consumers have toward evoking positive reactions from
the environment by the prospect of a pro-environmental
behavior tend to activate the BAS, ultimately leading to the
PCI.
In the relationship between BIS and PCI, consumers may
come to feel powerlessness about environmental change,
which could evoke low-power perception, thereby activating
the BIS (Keltner et al., 2003); this may help consumers to
avoid behaviors that could produce further negative results
for the environment, leading them to act toward preventing
such occurrences; pro-environmental consumption behavior
is one of the ways to prevent negative results caused by envi-
ronmental changes, subsequently leading consumers to exert
their PCI. Thus, we hypothesized the following:
Hypothesis 2a (H2a): The BAS will positively influence
PCI.
Hypothesis 2b (H2b): The BIS will positively influence
PCI.
BAS and BIS, as fundamental motivation dispositions, are
influenced by cultural norms and values (Carver & White,
1994; Updegraff et al., 2004). Hence, in the pro-environmen-
tal consumption context, GCVs could be one of the important
influence factors that activate the BAS and the BIS. In other
words, consumers with stronger GCVs may more positively
perceive the purchase of green products (BAS activation/
approach motivation) and may help other consumers under-
stand the benefits of such products (BIS activation/avoidance
motivation), which promotes environmental protection and
pollution avoidance. The result of this series of psychological
mechanisms is the generation of pro-environmental con-
sumption behavior. Therefore, the BAS and the BIS can play
a mediating role between GCVs and PCI.
Hypothesis 3a (H3a): BAS will positively mediate the
relationship between GCVs and PCI.
Hypothesis 3b (H3b): BIS will positively mediate the
relationship between GCVs and PCI.
GI
GI refers to the outcome of the interaction between consum-
ers and green products. Anti et al. (1986) defined the concept
of involvement as a state of perceived importance or a state-
ment of one’s interest that was evoked by the stimulus and
the situation. Others have defined it somewhat differently:
Rothschild (1984) defined involvement as a statement of
one’s interest, motivation, or arousal, and subdivided the
concept into two categories, persistent involvement and situ-
ational involvement. Persistent involvement refers to con-
sumers’ persistent attention toward a product, whereas
situational involvement refers to consumers’ short attention
toward a product in a specific consumption situation.
Rothschild (1984) determined that involvement can have a
positive impact on consumers’ brand sensitivity: Consumers
with high involvement levels tend to more thoroughly search
for product information to compare and evaluate the product,
and consumers with low involvement tend to skip the infor-
mation searching process and go straight for the purchase. In
this study, we defined GI as a statement of interest, motiva-
tion, or arousal toward green products.
In the relationship between GCVs and PCI, consumers
with high GI will search for more information about green
products to compare and evaluate them—from functional
and valuable perspectives—to assist them in their purchas-
ing decisions. Consumers who have low GI will ignore
product information and only be affected by the product
advertisements. In a pro-environmental consumption con-
text, consumers with high GI would be able to evaluate the
environmental effect of their pro-environmental behavior,
which demonstrates that GI may be a mechanism influenc-
ing people’s GCVs, thereby making consumers with higher
GCVs more likely to display PCI. Therefore, GI may be a
moderating variable between GCVs and PCI (see Figure 1).
Hypothesis 4 (H4): The positive relationship between
GCVs and PCI will be stronger when consumers’ GI is
high.
Method
Data Collection and Sample Description
Data were collected from actual shoppers in four urban
Chinese shopping streets—including Chongqing Road of
Changchun, Henan Street of Jilin, Taiyuan Street of
Shenyang, and Central Street of Harbin—through face-to-
face intercept surveys. Shoppers were randomly intercepted
and recruited to participate in this study. Data were collected
from 10:00 a.m. to 5:00 p.m. The face-to-face surveys were
administered by research assistants who were well trained
and instructed in intercepts and interviewing techniques.
Participants
Participants were assured that the collected data of this
research would be used exclusively for academic purposes.
Participants were asked to read a paragraph, including the
following content, before proceeding to the surveys:
Wang et al. 5
If you need to buy a refrigerator for your family, there are two
kinds of refrigerators you can choose, which are the energy-
efficient refrigerator and the ordinary refrigerator. Compared
with the conventional refrigerator, the energy-efficient
refrigerator is consistent in freezing and refrigeration, but has
better energy efficiency and a higher price.
Following Bentler and Chou’s (1987) and Hair et al.’s
(2014) suggestions, the sample sizes should be at least 15
times the number of the items in the measure of the observed
variables to ensure good statistical power of the study. In this
study, there were 34 items, so the minimum sample size
required was 510 samples, and 741 out of 1,000 intercepted
shoppers agreed to complete the survey questionnaire, result-
ing in a 74.1% response rate. Therefore, the statistical analy-
sis of this study had strong statistical power.
The demographic profile of these shoppers can be charac-
terized as follows: There were slightly more female partici-
pants (59.2%); approximately 62% of the participants were
aged between 18 and 25 years old; 26% were married and
74% were single; 74.4% had completed a university degree;
and all income ranges were well represented, as 21.9%
reported income between 12,000 and 15,000 Renminbi
(RMB), and 8.8% had an income greater than 80,000 RMB.
Measures
Scale structure and reliability analysis. The measurement scales
employed in this research were developed and validated in a
past study. For GCVs, we used a six-item scale adapted from
Haws et al.’s (2014) study. In this study, the Cronbach’s
alpha was .914. For the BAS and the BIS, we used 13 (BAS)
and seven items (BIS) that were adapted from Carver and
White’s (1994) study. In this study, the Cronbach’s alphas
were .915 and .769, respectively. The PCI scales used three
items adapted from Ajzen’s (1991) study, and GI scales used
five items adapted from Traylor and Joseph’s (1984) study.
In this study, the Cronbach’s alpha for the PCI scale was
.831, and the Cronbach’s alpha for the GI scale was .897. All
items were rated on a 5-point Likert-type scale, and all Cron-
bach’s alpha exceeded .700, exhibiting sufficient reliability.
Validity Analysis
Except for BIS (0.424), other constructs’ average variance
extracted were all above 0.500, and all constructs’ consis-
tency reliability were above 0.700. The chi-square value for
the confirmation factor analysis containing all research con-
struct measures was 2,493.831. Other goodness-of-fit mea-
sures were χ2/df = 6.704, non-normed fit index = 0.814,
comparative fit index = 0.837, and root mean square error of
approximation = 0.088.
We also followed Bollen and Stine’s (1992) model and
analyzed the model fit of this research. The Bollen-stine
bootstrap tests were 2,000 times; the results showed that the
χ2 was 515.49, χ2/df = 1.386, comparative fit index = 0.931,
non-normed fit index = 0.925, and root mean square error of
approximation = 0.023. We also chose five demographic
variables as control variables in this research (gender, age,
income, marital status, and education). Construct analyses
are shown in Table 1.
Intercorrelations among the variables are reported in
Table 2. The results indicated that GCVs were significantly
related to PCI (r = . 697, p < .001), BAS (r = .446, p <
.001) and BIS (r = .299, p < .001) were significantly related
to PCI, GCV was significantly related to BAS (r = .527. p <
.001), GCV was significantly related to BIS (r = .322,
p < .001), and GI was significantly related to PCI (r = . 632,
p < .001).
Results
Total and Indirect Effects
This study used the Partial Least Squares Structural Equation
Model (PLS-SEM) to test the hypotheses. By comparing
with the hierarchical linear regression model (e.g., ordinary
least squares [OLS]) and traditional structural equation
model (e.g., Analysis and Moment of Structure [AMOS]),
PLS-SEM relaxes the restrictions on normal distribution and
Figure 1. Theoretical research model.
Note. GCVs = green consumption values; BAS = behavioral approach system; BIS = behavioral inhibition system; PCI = pro-environmental consumption
intention; GI = green involvement.
