Journal of Environmental Psychology 40 (2014) 218e227

Contents lists available at ScienceDirect

Journal of Environmental Psychology

journal homepage: www.elsevier.com/locate/jep

Who pays more (or less) for pro-environmental consumer goods?

Using the auction method to assess actual willingness-to-pay
Nelson A. Barber*, Melissa Bishop 1, Thomas Gruen 2
University of New Hampshire, Peter T. Paul College of Business and Economics, Peter T. Paul Hall, 10 Garrison Avenue, Durham, NH 03824, USA

a r t i c l e i n f o a b s t r a c t

Article history: The purpose of this study was to measure consumers' actual willingness-to-pay (WTP) for pro-
Available online 8 July 2014 environmental (PE) and non-PE products through a controlled experimental auction. Ninety-eight in-
dividuals from the Northeastern U.S. participated in an auction and were segmented into groups based
Keywords: on whether they would pay a) more, b) about the same, or c) less for a PE product compared to an
Willingness-to-pay equivalent non-PE product. Demographic and psychological group profiles were comprised based on
Experimental auction
perceived product benefits, values, consequences of purchase behavior and demographics. Findings
Market segmentation
showed the majority of consumers would not pay more for PE offerings, suggesting that they may not
Pro-environmental behavior
view PE products as “normal.” Implications for shaping PE behavior of this neutral majority are
addressed. Further, the need for researchers to collect actual behavioral data is emphasized, as this is of
paramount importance in the environmental domain due to the well-known “action gap” between in-
tentions and behavior.
© 2014 Elsevier Ltd. All rights reserved.

1. Introduction 2004; Schultz, Oskamp, & Mainieri, 1995) and household con-
sumption (Gatersleben, Steg, & Vlek, 2002). Across all three oper-
According to the Organization for Economic Co-operation and ationalizations, efforts to motivate people to believe in the
Development (OECD, 2008), consumer choice behavior is crucial to importance of PE behavior have been fairly successful. However,
urging sustainable production and plays an essential role in pro- converting people to incorporate these stated beliefs into their
environmental (PE) development. It explains that most govern- daily behaviors has been much more challenging (Bang, Ellinger,
ment policies focus on curtailing the environmental impact of un- Hadjimarcou, & Trailhal, 2000; Laroche, Bergeron, & Barbaro-
sustainable industrial production, primarily through regulations Forleo, 2001; Ottman, Stafford, & Hartman, 2006).
and taxes. Yet, promoting sustainable consumption is equally Research concerning attitudes in this domain appears to be
important to limit negative environmental and social externalities following a similar path as stated beliefs. Attitudes are heavily
as well as to provide markets for sustainable products (OECD, relied on to predict intentions, which then are used to predict
2008). behavior. These relationships are outlined in the well-established
In this study, we examine PE consumption within the general theory of reasoned action (TRA) (Ajzen & Fishbein, 1980) and the
context of pro-environmental behavior. PE behavior is most more recent and extended theory of planned behavior (TPB) (Ajzen,
commonly defined as intentionally minimizing the negative impact 1991). The first part of the TPB pertaining to the attitudeeintention
that an action can have on the environment (Kollmuss & Agyeman, relationship appears to be strongly supported by many fields of
2002). It has been operationalized in several ways, including daily research (Chan, Wu, & Hung, 2010; Fielding, McDonald, & Louis,
environmental behavior (Tindall, Davies, & Mauboules, 2003), 2008; Greaves, Zibarras, & Stride, 2013; Roseman, Hoon Kim, &
conservation behavior (Monroe, 2003; Poortinga, Steg, & Vlek, Zhang, 2013; Walsh, Shiu, & Hassan, 2012). However, the inten-
tionebehavior relationship has received much less support, espe-
cially in the environmental domain.
This gap between attitudes, intentions, and behavior has been
* Corresponding author. Tel.: þ1 603 862 3571. referred to as the “action gap” for general consumer research
E-mail addresses: nelson.barber@unh.edu (N.A. Barber), melissa.bishop@unh.
(Godin, Conner, & Sheeran, 2005) and for environmental consumer
edu (M. Bishop), tom.gruen@unh.edu (T. Gruen).
1
Tel.: þ1 603 862 3370. research (Blake, 1999; Kennedy, Beckley, McFarlane, & Nadeau,
2
Tel.: þ1 603 862 3360. 2009; Kollmuss & Agyeman, 2002; Niessen & Hamm, 2008;

http://dx.doi.org/10.1016/j.jenvp.2014.06.010
0272-4944/© 2014 Elsevier Ltd. All rights reserved.
 N.A. Barber et al. / Journal of Environmental Psychology 40 (2014) 218e227 219

Vermeir & Verbeke, 2006). The most common explanations for this change, examining critical factors underlying this behavior, and
action gap are the perceived quality of PE products and the higher suggesting corrective action to influence the change in and the
price that these products usually command (Bazoche, Deola, & determinants of the behavior. Based on these suggestions and in
Soler, 2008; Bennett & Williams, 2011; D'Souza, Taghian, & light of the environmental action gap between intentions and
Khosla, 2007; Loureiro, 2003; Ottman et al., 2006). behavior, the two main objectives of this research are as follows:
Because of this action gap, the importance of capturing behav-
ioral data is paramount, especially for products that entail a
normative component, such as PE products. This is because PE of- B The first objective is to use a backward or behavioral segmen-
ferings tend to be viewed as the socially acceptable choice, which tation approach (described in the following section) to cluster
may lead to inflated purchase intentions relative to actual purchase consumers using the price one will actually pay to purchase a PE
behavior (Roozen & De Pelsmacker, 1998). However, behavioral product as opposed to a non-PE product. Data is collected
measures often fall short, as many studies use self-reported through a controlled auction experiment.
behavior or do not distinguish behavior from behavioral in-
tentions. For example, although Thøgersen (2002) measured B The second objective is to profile each of these segments ac-
behavior concerning organic and non-organic wine, essentially cording to a) product benefits sought b) values c) perceived
only behavioral intentions were measured. Harland, Staats, and consequences of purchase behavior, and d) demographics based
Wilke (1999) studied self-reported behavior for five different PE on self-reported survey data.
behaviors (including turning off the faucet while brushing one's
teeth, purchasing energy-saving light bulbs, and using other forms To accomplish these objectives, a survey was administered first,
of transportation than the car). Schultz et al. (2005) measured self- followed shortly after by an auction experiment that assessed
reported behavior (including purchasing products in reusable actual behavior. As a result, the experiment is described first to
containers, composting food scraps, and picking up litter) for a year establish the actual behavior segments (objective #1), followed by
among six different countries. Similarly, segmentation studies by the survey data which was used to profiled within each of the
commercial marketing research firms have also used hypothetical resulting segments (objective #2).
behavior (Grail Research, 2009) and attitudinal and self-reported
behavioral measures (Cotton Inc., 2010) to segment PE con- 2.1. The first objective e backward segmentation
sumers. While studies such as these rely on individuals' self-
reported behaviors, it is important to reiterate that what people Regarding the first objective, Bennett and Williams (2011) argue
“say” may not necessarily represent what they do (Horowitz, that there needs to be a serious shift in theoretical thinking. Rather
McConnell, & Murphy, 2008; Murphy & Stevens, 2004). than spending countless resources (time, money, etc.) changing the
Thus, obtaining actual behavioral data is essential to those intention-behavior connection, there needs to be a focus on shaping
interested in changing behaviors that entail a normative compo- (or changing) behavior, which in turn may shape values and attitudes
nentdso that they may understand certain drivers of behavior (Smith & Mackie, 2007; Van Veen, Krug, Schooler, & Carter, 2009).
rather than stated behavioral intentions or self-reported behavior. Their argument supports a “backward” approach to market segmen-
Therefore, the purpose of this research is to examine e in an actual tation that begins with actual behaviors and is counter to the more
consumption context e what drives people to actually pay more for common “forward” approach that often begins with attitudes and is
PE products compared to non-PE products. We look at these drivers most commonly used by researchers (Andrews & Currim, 2003).
in terms of product benefits, values, perceived consequences of The forward segmentation approach segments consumers by
purchase behavior, and demographics. Specifically, we consider characteristics, such as values and attitudes, and then discriminates
those consumer goods produced in an environmentally friendly by product selection behaviors. However, given the action gap be-
manner as PE products, and those as conventionally produced as tween intentions and behavior in the PE domain, starting with
non-PE products. Using a more specific context such as this allows behavior makes more sense. Unlike attitudes (that are difficult to
us to uniquely focus on this behavior and test willingness-to-pay measure validly and reliably), behaviors can be directly observed,
(WTP) behavior in a controlled context. [It should be noted that thus segmenting by behaviors in a research setting is likely to result
WTP data may be hypothetical in nature (such as with the in segments that mirror the actual population. Hence, the backward
contingent valuation method whereby consumers are simply asked approach involves grouping consumers' responses based on their
to state how much they would pay). It may alternatively be non- similarity in choice of products, services, and other activities (e.g.
hypothetical, requiring some form of economic commitment from recycling), and is followed by discriminating these groups by con-
participants. In this research, “actual” WTP refers to the non- sumer characteristics, such as demographics, values, and attitudes
hypothetical version of WTP.] (Wedel & Kamakura, 2000). In summary, this approach looks at
Further, as one of the biggest hindrances to purchasing PE what people actually do first, groups them accordingly, and then
products is the higher price they usually entail (Bazoche et al., deconstructs these groups to understand what similarities exist
2008; Grail Research, 2009; Loureiro, 2003), it is important to un- within groups, and what differences exist between groups. This
derstand who will actually pay more. By finding out who will pay approach is similar to the one taken by Clark, Kotchen, and Moore
more for PE compared to non-PE consumer goods, we may be able (2003), who compared participants and non-participants in a
to shed light on how other stakeholders may encourage positive premium-priced, green electricity program.
changes in general PE behavior in non-consumption domains (such For a measure of backward segmentation in this research, we
as recycling, using public transportation, etc.). consider the difference between actual purchase behavior for a
conventionally produced product and for a PE produced product. If
2. Study objectives some individuals are willing to pay more (or less) for a PE product,
then they can be profiled based on this behavior to see which
Steg and Vlek (2009) assert that environmental psychologists characteristics may be driving them to pay more (or less) for these
need to participate in the management of environmental problems products. Therefore, using the difference in the price premium (or
by supporting behavioral change. They outlined several issues to discount) one will pay for a PE product compared to a non-PE
consider in this process, including identifying the behavior to product may be a valuable measure for segmenting consumers.
220 N.A. Barber et al. / Journal of Environmental Psychology 40 (2014) 218e227

2.2. The second objectivedpurpose of profiling Dunning (1998) and Schultz and Zelezny (1999) stated, self-
enhancement works to make people feel good about themselves
Once segments have been identified based on WTP behavior, the and becomes particularly noticeable in situations of threat, failure
second objective is to deconstruct the groups in terms of what or blows to an individual's positive self-image.
drives individuals to pay more, pay the same, or pay less for PE Kilbourne, Grünhagen, and Foley (2005) further suggested that
products compared to non-PE products. The profiling measures individuals higher in self-enhancement tend to be more material-
considered here include perceived product benefits, values, con- istic than those higher in self-transcendence, and consequently
sequences of purchase behavior, and demographics. Each are may be less concerned about the impact their consumption has on
considered important consumer influencers when purchase de- the environment. Thus, it is unlikely that an individual high in self-
cisions are made (Barber, Kuo, Bishop, & Goodman, 2012; Laroche enhancement will behave in a manner compatible with univer-
et al. 2001; Stampfli, Siegrist, & Kastenholz, 2010), and under- salism (self-transcendence or altruism) values that exhibit the trait
standing them in light of actual behavior may help marketers to or the practice of concern for the welfare of others (Schwartz, 1992).
develop more sophisticated positioning strategies. However, a self-enhancing individual may still be concerned about
environmental issues because of the negative effects that may
2.2.1. Perceived product benefits result to him/herself. On the other hand, a self-transcending indi-
Benefits that consumers seek, such as price, functionality, social, vidual may also be concerned about environmental issues but for a
emotional, and quality, cannot be overlooked when considering PE different reasondthe potential negative effects to other people
products. Most consumers seek several primary benefits in pur- (which ties into the following section concerning consequences of
chase decisions, and one benefit may need to be sacrificed in favor purchase behavior).
of another in these decisions. For example, many may see the
environmental benefit of choosing organically-grown food; yet 2.2.3. Consequences of purchase behavior
someone who also values thriftiness may feel conflicted when €rling, Fujii, Ga
Ga €rling, and Jakobsson (2003) tested the norm-
presented with organic products if they are significantly more activation theory of altruistic behavior (Schwartz, 1977), which
expensive than the cost of conventionally-produced food. Further, suggests that PE behavior depends on, among other constructs,
D'Souza et al. (2007) found that consumers would be willing to pay awareness of consequences for oneself, for others, and for the
a higher price for “green” goods, but only if the quality was higher environment. They found that social value orientation modifies the
than conventional goods. Another example is the social benefit a relationship between PE behavior and awareness of environmental
product provides, which entails how a product reflects a certain consequences. Social value orientation refers to the distinction
positive social image to others (Orth, Wolf, & Dodd, 2005; Sweeney between pro-self (individual) and pro-social. In other words, con-
& Soutar, 2001). [This ties into self-enhancement (discussed under sumers may incorporate social issues into their purchasing
“values” below), which encompasses social recognition and behavior by evaluating the consequences of their consumption
enjoyment of life (Schwartz, 1992, 1994).] Thus, understanding upon society and themselves (Follows & Jobber, 2000; Fujii, 2006).
consumers' perceived product benefits, why they choose a partic- Thus, the purchase of PE products may be influenced by the attri-
ular product, and the factors that play a role in their selection butes of perceived individual consequences and the environmental
processes is important in understanding purchase behavior consequences of the purchase decision (Barber et al., 2012; Follows
(Kayaman & Arasli, 2007; Kim & Kim, 2005; Orth et al., 2005; & Jobber, 2000; Laroche et al., 2001; Schultz & Zelezny, 1999). In-
Ottman, 2011; Ottman et al., 2006; Sweeney & Soutar, 2001). dividual consequences refer to how the PE behavior is perceived to
affect an individual's personal satisfaction, while environmental
2.2.2. Values consequences consider how consumers incorporate social issues
Values are the most abstract types of social cognitions that can into their purchasing behavior by evaluating the consequences of
be used in determining human behavior by providing a glimpse their consumption behavior upon society.
into an individual's interpersonal world (Follows & Jobber, 2000).
Values guide selection or evaluation of behavior and events (Barber 2.2.4. Demographics
et al., 2012; Follows & Jobber, 2000; Roccas, Sagive, Schwartz, & Laroche et al. (2001) performed a comprehensive literature re-
Knafo, 2002; Schwartz, 1992, 1994). Similar to Follows and Jobber view finding several factors that may influence consumers'
(2000), the values in this research include self-transcendence and willingness-to-pay more for PE products. They classified the factors
self-enhancement. Self-transcendence consists of an active concern into five categories: demographics, knowledge, values, attitudes
for others and the desire to work for the good of society. It is a and behavior. The overall conclusion from this review was that,
personal journey of self-discovery, where one strives for greater although by themselves demographics are not a perfect tool for
perfection, higher perspective, and moves beyond prior concepts of profiling, there is a connection between certain characteristics that
behavioral limitations, which goes beyond ego (Barber et al., 2012; add value to PE marketing strategies. For example, overall, their
Follows & Jobber, 2000; Schwartz, 1992, 1994). Self-enhancement combined results portray a highly socially conscious person as fe-
reflects the extent to which individuals are motivated to enhance male, pre-middle aged, with a high level of education and above
their own personal interests and how they see themselves (Barber ~ as, and
average socioeconomic status. Further, Torgler, García-Valin
et al., 2012; Follows & Jobber, 2000; Schwartz, 1992, 1994). These Macintyre (2008) found that gender and age were important in
self-concepts are believed to develop in response to social experi- determining environmental attitudes and behavior. Thus, this study
ences with the goal to adapt behavior in order to achieve a positive also considers gender, income, education and age in profiling
reaction from their close and significant references (Banister & respondents.
Hogg, 2004; Dunning, 2007; Sedikides, Gregg, Cisek, & Hart, 2007).
Relationships among values can reflect shared motivations or 3. Method
indicate that values are in conflict with each other (Roccas et al.,
2002; Schwartz, 1994). For example, behavior undertaken to 3.1. Study design
satiate values of power and achievement (self-enhancement or
egoism) represent an individual's need for social standing, respect, In order to address the objectives above, the study was run in
or dominance over people and resources. Or as Beauregard and two steps, an online survey followed by an experimental auction.
 N.A. Barber et al. / Journal of Environmental Psychology 40 (2014) 218e227 221

For the first step, participants completed an online survey to assess 3.4. Procedural steps
their perceived product benefits, values, perceived consequences of
purchase behavior, and demographic data. The URL link to the 3.4.1. Procedural sequencing
survey was emailed to participants and also included instructions The auction started four weeks after the survey was closed to
on when and where to meet for the auction. Reminder emails were allow the researchers time to organize with the venue hosting the
sent concerning the survey and the auction. Four weeks after the auction. In terms of the timeframe, Morwitz, Steckel, and Gupta
survey was closed, respondents participated in a controlled (2007) explored the time between intent and behavior measure-
experimental auction. ment, suggesting that shorter intervals are significantly associated
The second step was a controlled auction experiment examining with higher intentebehavior correlations. They also found the time
the price one would actually pay for a PE product compared to a between intent and behavior measurement is correlated with the
non-PE product. Based on the degree to which individuals would type of product under examination because it reflects the typical
pay more or less for a PE product compared to a non-PE product, a inter-purchase time for different types of products. Particularly,
cluster analysis was used to determine the number of segments. they found the strongest correlation for time periods of one month
Each segment was then profiled in terms of the variables outline or less for non-durable goods. Further, an advantage of the survey
above that were collected from the survey, using analysis of vari- being completed before the auction is that responses may not be
ance for pairwise comparisons. influenced by the auction. Alternatively, the survey could prime
behavior at the auction (Chandon, Morwitz, & Reinartz, 2005).
However, priming effects, depending on the product (durable
3.2. Sample selection versus non-durable) tend to be short-term in nature (Fishbein &
Ajzen, 1975; Morwitz et al., 2007). Thus, because wine is a non-
Participants for both the survey and auction were customers of a durable good and for everyday consumption, there should be lit-
local winery, retail store, and a restaurant in western Connecticut tle priming effect (if any) because of the four-week timeframe be-
and were randomly recruited through the intercept method. Po- tween the end of the survey and the start of the auction.
tential participants were asked whether they would be willing to
take part in both steps of the study e the survey and the auction 3.4.2. First objective e assessing actual behavior using the Vickrey
about wine (see “Product stimuli”). Those individuals who agreed auction
to participate had to meet the following criteria: they must (1) be Before the auction process started, the participants were told
21 years of age or older, (2) be a wine drinker (drinking wine at least the $20 (V14.48) participation fee would be paid at the end of the
once a week), and (3) be involved in their household wine pur- auction. Participation fees may be considered endowments that
chases, and (4) not be allergic to sulfites. Each participant was told a may bias actual estimates (Harrison, 1989). Loureiro, Umberger, and
$20 (V14.48) participation fee would be paid at the end of the Hine (2003) found that an endowment similar to the value of the
auction. The researchers approached 225 individuals asking them auctioned item should not significantly impact WTP estimates.
to participate in both parts of this study. Ninety-eight individuals Silva, Nayga, Campbell, and Park (2007) compensated participants
agreed to participate in both steps with this sample size similar to $4 (V2.90) for participating in an auction where the good went for
those used in studies by Barber et al. (2012) and Bazoche et al. about $2 (V1.45). In the present study, the average price consumers
(2008). Reasons for not participating varied, but mainly related to pay in general for a bottle of conventional wine is about $15
not living within the area that would justify making the trip to (V10.86) (confirmed by the local retail store). Therefore, taking into
participate in the auction. There were no withdrawals between the account their travel expenses to the auction and time to complete
survey and the auction. the auction (two hours), the participation fee should not signifi-
cantly bias the results.
To assess actual purchase behavior the second price Vickrey
3.3. Product stimuli auction (Vickrey, 1961) method was used. The Vickery auction is a
commonly used incentive-based elicitation method (Lusk &
The product selected for this study was wine. Wine, as an Shogren, 2007; Soler, Gil, & Sanchez, 2002; Voelckner, 2006)
experiential consumer good, is challenging for a consumer to whereby sealed bids from all bidders are collected and the winner
judge the price/quality value just by looking at the bottle or label. is required to purchase the product at the second highest price. The
Further, very little is known about the actual price premium that auction was conducted using the following three steps (see Fig. 1 for
consumers would be willing to pay for organic wine. For this a summary of the auction process).
study, a PE wine is one that can be designated as following many
different practices either in viticulture (e.g. bio-dynamic, organic, 1. Participants were randomly assigned to five separate sessions
etc.) or enology (e.g. solar power in the winery, gravity feed water, with approximately 20 people per session. As recommended by
etc.). Wine also fits naturally in the bidding process of an auction. Bazoche et al. (2008) and Kagel and Levin (1993), prior to the
Other experimental auctions have used wine (Barber et al., 2012; actual auction (and to ensure the participants understood the
Thøgersen, 2002) or similar consumer goods that people would auction process), a test-run auction was held with an alternative
find normal to bid on, such as fair trade chocolate (Didier & Lucie, product. Participants thereby became familiar with the auction
2008), food products such as potato chips (Bernard, Zhang, & procedure, thus eliminating procedural confusion as an alter-
Gifford, 2006), and meats, fruits, and vegetables (Gil, Gracia, & native explanation of results.
Sanchez, 2000). 2. During the actual auction, participants sat in a room in such a
One PE wine and one conventionally produced wine were way that they could not communicate with each other. The
chosen as the product stimuli for this study. It should be noted that participants evaluated (tasted or visually assessed) and
although two different wines were used in the auction, results from appraised each wine in a pre-established order to control for the
the blind bidding condition (collected first in the auction) revealed impact of the order of presentation of the products on the
no significant differences between the two wines in terms of WTP assessment. The wines selected were a PE merlot wine and one
or likeability. This is further reflected in Table 2 and discussed in the non-PE (conventional) merlot wine, both from California. The
Results section. evaluations were in three different informational situations:
222 N.A. Barber et al. / Journal of Environmental Psychology 40 (2014) 218e227

Fig. 1. Study design.

blind tasting, partial information (the label alone, no brand was controlled for by randomizing each of the wines. Prior to the
name provided) and full information (the label plus tasting, no bidding, participants were informed that only bids from one
brand name provided). All brand names were removed from the situation (partial information or full information), randomly
two informational situations to avoid the influence of brand selected by one participant, would be used. This procedure
recognition during the evaluation process (see Appendix for avoids strategic behavior leading participants to submit a high
label information). To control for any sensory differences and to reservation price only in situations they preferred. The partici-
eliminate any potential impact of taste during the full infor- pant who submitted the highest price became the winner and
mation situation, the blind tasting situation did not provide any had to pay for the product, but at the second highest bid.
information about the two different wines. Rather it was used to
establish the essential equivalence of the wines in terms of Ninety-eight subjects made four bids each in the partial and full
likability based on bid prices and taste without the influence of information conditions, yielding a database of 392 bid prices. The
additional information. If no significant differences existed be- bid prices from the wines for the partial and full information sit-
tween or within groups based upon blind tasting bid prices or uations were averaged and used to represent a composite price
likability scores, then the influence of taste should be minimal score of individuals' evaluations of the products. A “difference” in
(or non-existent) for the full information situation (see Table 2). actual behavior was calculated by subtracting the PE product
3. The bidding followed protocol used by Bazoche et al. (2008) and composite price from the non-PE product composite price.
Lange, Martin, Chabanet, Combris, and Issanchou (2002). Par- As a manipulation check, after all bids were submitted for the
ticipants were provided with a reference price of $15.00 partial and full information situations, participants rated whether
(V10.86). This reference price was developed using the average each of the wines were PE or not (1 ¼ not PE; 7 ¼ very PE) using the
price consumers actually pay for conventional wine as reported following statement “To what extent is this wine pro-environ-
in a study by Barber, Taylor, and Strick (2010), and according to a mental?” Respondents reported the PE wine to be very PE (M ¼ 6.4,
local retail wine store. Reference prices have been used to assess SD ¼ .8). For the non-PE wine, respondents found the wine not to
WTP for fair-trade coffee (De Pelsmacker, Driesen, & Rayp, 2005) be PE (M ¼ 2.1, SD ¼ .5). These differences were significantly
and are common in research involving auctions (Chernev, 2003). different (t(97) ¼ 26.44, p < .05).
After evaluation of each wine, participants wrote down their
maximum bid for the wine evaluated, assuming that the wine 3.5. Survey measures
would be auctioned at the end of the experiment. This way
participants could not revise their valuations in hindsight after The first page of the survey contained an introductory paragraph
experiencing the other wine or situation. The design of the that explained the study and provided a statement of consent.
experiment considered the issue of same subject response. This Following this, each construct used for profiling was measured
 N.A. Barber et al. / Journal of Environmental Psychology 40 (2014) 218e227 223

using a seven-point bi-polar scale to record the responses. CFA and 4. Results and discussion
the Cronbach's alpha were used to estimate the reliability of the
multi-item scales and to test the measurement items. 4.1. First objective e segmentation using cluster analysis

This paper set out to understand PE behavior through the

3.5.1. Perceived product benefits
method of backward segmentation. To do this, a cluster analysis
Following the work of Sweeney and Soutar (2001) and Orth et al.
was run on the composite auction prices (partial and full infor-
(2005), participants were assessed using the statement “My favorite
mation only) for the PE and non-PE wines. A two-step clustering
wine …,” followed by the selection of benefit(s). The 19 items
procedure was adopted: (1) a hierarchical cluster analysis to
assessed the importance of product benefits were divided into five
identify the appropriate number of clusters, and (2) a K-means
separate categories (Orth et al., 2005): 1) Price/value benefits (four
cluster analysis, providing further elaborative information on the
items such as “is a good product for the price”; a ¼ .79; factor
cluster membership. The hierarchical cluster analysis procedure
loading ¼ .79e84), 2) Functional quality benefits (three items such
was applied to the data after computing several solutions for
as “has an acceptable standard of quality”; a ¼ .90; factor
different numbers of clusters (e.g., two, three, four, etc.). A three-
loading ¼ .79e84), 3) Environmental benefits (three items such as “is
group solution was selected based on minimized intra-group er-
a recyclable product”; a ¼ .85; factor loading ¼ .76e80), 4) Social
ror, interpretability and practicality of the group profiles. Using the
benefits (four items such as “makes a good impression on other
K-means, resulting clusters were statistically different from one
people”; a ¼ .86; factor loading ¼ .82e87), and 5) Emotional benefits
another (between clusters for each wine category) for the PE wine
(five items such as “would give me pleasure”; a ¼ .77; factor
price [F(2,95) ¼ 346.97, p < .05] and the non-PE wine price
loading ¼ .76e82).
[F(2,95) ¼ 24.49, p < .05], while within clusters there were signif-
icant differences between the PE and non-PE auction prices
3.5.2. Values (Table 1).
Value measures in this study followed the work of Schwartz Each cluster was subsequently given a descriptive name. The
(1992, 1994), Follows and Jobber (2000), Barber et al. (2012), and clusters were labeled using the difference in PE and non-PE wine
Schultz and Zelezny (1999), using two constructs: self- price, respectively, (1) ‘‘High PE price premium,’’ (2) ‘‘Neutral PE
transcendence (six items) and self-enhancement (six items). Self- price premium,’’ and (3) ‘‘Negative PE price premium.’’ The High PE
transcendence (a ¼ .88; factor loading ¼ .82e.90) included two price premium group (30% of the total) was named as it had the
sub-scales e universalism (equality, unity with nature, and social highest actual price differential between the PE and non-PE wine
justice) and benevolence (helpful, forgiving, and loyalty). Self- products at the auction (M ¼ $3.42; V2.50). The Neutral PE price
enhancement (a ¼ .79; factor loading ¼ .83e89) included three premium group (43% of the total) was named as its price differ-
sub-scales: achievement (ambition, successful), hedonism (plea- ential was just above zero (M ¼ $.39; V.28). Finally, the Negative PE
sure, enjoying life), and power (social power, preserving public price premium group (28% of the total) included individuals who
image). Participants assessed each of the above values by were not at all willing to pay more (and would even pay less) for a
answering, “How important are the following words to you, as PE wine (M ¼ $2.74; V1.98). This group, on average, would pay
guiding principles in your life?” $2.74 less for a PE product than for a conventional product.
As a form of manipulation check, the bid prices for the blind sit-
uation only were compared (Table 2). Statistical testing showed no
3.5.3. Consequences of purchase behavior differences between clusters for the PE wine prices [F(2,95), ¼ 2.73,
Following the work of Follows and Jobber (2000), six items were p > .05] and the non-PE wine prices [F(2,95), ¼ .13, p > .05], while
used to measure environmental consequences (a ¼ .79; factor within clusters there were no significant differences for the PE or the
loading ¼ .79e85). For example, one statement read, “How wine non-PE wine prices [High PE price premium t(28) ¼ .11, p > .05;
production may affect the environment is important to me.” Also Neutral PE price premium t(41) ¼ .87, p > .05, and Negative PE price
from Follows and Jobber (2000), seven items were used to measure premium t(26) ¼ 1.87, p > .05]. When participants were asked
individual consequences of purchase behavior (a ¼ .76; factor whether they liked the wine, statistical testing showed no differ-
loading ¼ .75e84). For example, one statement read, “It is impor- ences between clusters. These results suggest that without the
tant to me whether packaging (e.g., wine bottles) adds to a landfill benefit of information about the wines, participants were willing to
site.” pay nearly the same price for each wine based solely on the products'
taste (likability) (see Lange et al., 2002, for similar findings).
3.5.4. Demographics
The final section collected respondents' demographic informa- 4.2. Second objective e profiling based on actual purchase behavior
tion including age (Kollmuss & Agyeman, 2002), income
(Diamantopoulos, Schlegelmilch, Sinkovics, & Bohlen, 2003), edu- Overall there were a few more males than females in the sample
cation (Guagnano & Markee, 1995), and gender (Hunter, Hatch, & (52 vs. 46). There was a significant difference among the three
Johnson, 2004). groups in terms of the gender mix [c2 ¼ 20.81, p < .05], with the

Table 1
Cluster analysis: Mean composite auction price in U.S. Dollars between and within groups (n ¼ 98).

(Label and Full Information Overall High PE price premium Neutral PE price premium Negative PE price Premium Pairwise Testing
Situations only) (n ¼ 29) (n ¼ 42) (n ¼ 27)

PE wine product $16.28 (V11.78) $19.53 (V14.27) $16.19 (V11.83) $12.89 (V9.42) F(2,95) ¼ 346.97, p < .00
Non-PE wine product $15.84 (V11.47) $16.11 (V11.77) $15.80 (V11.54) $15.63 (V11.42) F(2,95) ¼ 24.49, p < .05
Difference in price $0.44 (V.32) $3.42 (V2.50) $0.39 (V.28) $2.74 (V2.00) F(2,95) ¼ 181.76, p < .05
T-test overall and within t(97) ¼ 1.02, p > .05 (t(28) ¼ 12.94, p < .05) (t(41) ¼ 2.29, p > .05) (t(26) ¼ 13.05, p < .05)
each cluster
224 N.A. Barber et al. / Journal of Environmental Psychology 40 (2014) 218e227

Table 2
Cluster analysis: Mean blind only auction price in U.S. Dollars (Euros) between and within groups (n ¼ 98).

Blind Situation Only Blind Situation High PE price premium Neutral PE price premium Negative PE price premium Pairwise Testing Between
Overall (n ¼ 29) (n ¼ 42) (n ¼ 27) Clusters

PE wine product $15.75 (V11.64) $15.91 (V11.76) $15.77 (V11.64) $15.57 (V11.55) F(2,95) ¼ 2.73, p > .05
Non-PE wine product $15.74 (V11.63) $15.90 (V11.75) $15.71 (V11.75) $15.59 (V11.56) F(2,95) ¼ .13, p > .05
Difference in price $0.01 $0.01 $0.06 ($0.02)
T-test overall and within each cluster t(97) ¼ .23, p > .05 t(28) ¼ .11, p > .05 t(41) ¼ .87, p > .05 t(26) ¼ 1.87, p > .05
Likability (Taste) PE Winea 4.8 4.9 4.7 4.8 F(2,95) ¼ 1.02, p > .05
Likability (Taste) Non-PE Winea 4.7 4.8 4.7 4.7 F(2,95) ¼ .96, p > .05
a
After the blind tasting, participants asked: “How would you rate the taste of this wine?” on a scale from 1 to 5 (1 ¼ Did not like it at all; 5 ¼ Liked it very much).

Neutral and Negative PE price premium groups showing somewhat participants and 25% of the total male participants. 100% of this
more males than the High group. The average age of respondents group had a college degree, and currently spends on average $15.13
was 40 years, and there were no significant differences in age among (V11.15) per bottle on wine.
the groups. Overall the sample was highly educated (77% with a
college degree). However, there was a significant difference among
4.4. Those that will pay less
the three groups in education [F(2,95) ¼ 17.06, p < .05], with the
Negative group reporting a lower overall education level. On
Those that are in the Negative PE price premium group seem
average, the respondents spend $15.09 (V11.15) per bottle of wine
less concerned with overall society (low self-transcendence) rela-
and have been wine consumers for 20 years. The average income of
tive to the High group. This is in line with their high self-
respondents was $66,500 (V48,136), with significant differences in
enhancement values where they are motivated to enhance their
income between the three groups [F(2,95) ¼ 3.82, p < .05], again
own self-interests over the benefits of society. They seem imper-
with the Negative group showing a lower overall income level. There
vious to the influence of those around them who may be exhibiting
were no significant differences regarding years consuming wine or
PE values and behaviors as is evident by the low social benefit or
current spending per bottle of wine between the three groups.
the emotional benefit that a product provides them. They also find
Analysis of variance (ANOVA) with the Scheffe method was used
the individual and environmental consequences of their behavior
to determine which of the variables (consequences, values, and
to be unimportant, as suggested by their low self-transcendence
benefits) differed among the three clusters (Hair, Black, Babin, &
value. Their actual behavior showed they would pay significantly
Anderson, 2007). The results indicated that numerous pairwise
less ($2.74; V1.98) for a PE product than for a conventional (non-
differences were significant (Table 3), and are explained in the
PE) product. The group contains approximately 26% of the total
following section.
sample female participants and 28% of the total sample male par-
ticipants. For level of education, only 32% of this group had a college
4.3. Those that will pay more degree, and they earn significantly less per year ($62,250) than the
High PE group. Since this group currently spends on average $15.08
Per the profile analysis, the High PE price premium group ap- (V11.14) per bottle on wine (about the same as the other groups),
pears very concerned about society in general (self-transcendence), neither education nor income correlates with buying cheaper wine
and less about their own self-enhancement, reflecting the diver- overall. They simply pay less for PE wine.
gence between these values as suggested earlier. They view the
consequences of their behavior, both environmental and individual, 4.5. Those for whom PE makes no difference in WTP
to be very important. Further, they search for environmental and
emotional benefits from a product. This group earns more per year Those in the Neutral PE price premium group are somewhere in
($72,150) than the other two groups (Neutral $66,500 and Negative between the previously described two groups in how they classify
$62,259) and contains approximately 35% of the total female their PE lifestyle and actual behavior. In many ways they appear
open to PE behavior, showing some concern but not leading the
way for others. Despite the importance they place on the individual
Table 3
and environmental consequences of their behavior, they consider
Respondents' values, consequences of behavior, and perceived product benefits.
social product benefits relatively high, suggesting that it is critical
High PE Neutral PE Negative PE Pairwise Testing that a product make a good impression and helps them feel
price price price
acceptable. Interestingly, both self-enhancement and self-
premium premium premium
transcendence scores are nearly the same for this group. That is,
Individual 5.9 4.3 2.9 F(2,95) ¼ 105.82, p < .05
they feel it is important to care about the welfare of society as a
Consequences
Environmental 5.8 4.2 2.8 F(2,95) ¼ 54.39, p < .05 whole, but also view status and self-satisfaction important in their
Consequences lives, which ties into the social importance of a product. Their actual
Self-transcendence 6.1 4.5 2.9 F(2,95) ¼ 49.61, p < .05 behavior, however, suggests thatdat this pointdthey would not
Value pay much more ($.39; V.28) for a PE product than for a conven-
Self-enhancement 3.3 4.4 5.7 F(2,95) ¼ 32.87, p < .05
tional (non-PE) one. Nearly half (46%) of all sample males (n ¼ 24)
Value
Environmental 6.6 4.4 1.9 F(2,95) ¼ 37.66, p < .05 fell into this group, compared to approximately 39% of the females
Benefit (n ¼ 18). 88% of this group had earned a college degree, and the
Emotional Benefit 6.5 4.9 3.1 F(2,95) ¼ 11.86, p < .05 average income was $66,500.
Quality Benefit 6.1 5.6 5.2 NS
Our findings in terms of WTP correspond with the general
Price Benefit 5.2 5.5 5.3 NS
Social Benefit 6.1 5.2 2.2 F(2,95) ¼ 43.11, p < .05
assessment of PE behavior in the U.S. by Bennett and Williams (2011),
who find that most Americans are in this middle ground. They are
Price Difference $3.42 $0.39 ($2.74) F(2,95) ¼ 246.97, p < .05
neither “hard-core green” nor are they unappreciative of issues
 N.A. Barber et al. / Journal of Environmental Psychology 40 (2014) 218e227 225

surrounding sustainability. Therefore, as this neutral group repre- study (which included such countries as Japan, Norway, Great
sents the mainstream market, we offer three key suggestions to in- Britain, the Netherlands, East Germany, West Germany, among
crease product acceptance and shape behavior toward PE products. others), Hunter et al. (2004) found evidence for cross-national
gender differences with females showing greater PE behavior to-
4.5.1. Influencing the neutral majority ward buying organic produce or driving less. Bennett and Williams
To influence the neutral majority, our first suggestion considers (2011) found similar results, claiming that the “feminization” of the
the positioning of PE products or services as trendy or distinctive PE movement may hold men back when it comes to visible PE
that bestows early adopter status on those consumers who join in, behavior. However, companies such as Tesla Motors may become
helping them stand apart. However, as mentioned earlier, most successful at bringing males into the PE fold through creative slo-
who want to become “pro-environmental” and self-identify as gans such as “Burn rubber, not gasoline” used for its new 100%
early adopters have done so already. Yet, many neutral majority electric Roadster.
(mainstream) market consumers are still uncertain, and are not
looking to stand apart from everyone else; they just want to be 5. Limitations and future research
“normal” and blend in Bennett and Williams (2011), Ottman (2011).
Thus, those interested in PE product advocacy should consider A major limitation was that this research was conducted in only
ways to make PE offerings seem “normal.” As Kollmuss and the U.S., as well as only in one area of the U.S., and with a small
Agyeman (2002, p. 256) suggested, “If we want to establish a new sample sizedleading us to strongly caution the generalizability of
behavior, we have to practice it. We might be perfectly willing to the findings. Further, only one product was used to assess WTP
change our behavior but still not do so, because we do not persist behavior (wine), similarly limiting the study's external validity. The
enough in practicing the new behavior until it has become a habit.” context of this research is also limiting in that it addresses con-
An example of this approach has been employed by Glendale Water sumption, rather than political action or regulation, making it
and Power, a large municipal utility company in the U.S. Its goals are difficult to generalize to non-consumption-related PE behaviors.
to use normative comparative feedback by shifting consumers to Another limitation of this study was assessing actual behavior
more energy-efficient PE behavior by communicating (through using an auction method. Voelckner (2006) suggests that researchers
home energy reports) what their neighbors are doing. As part of should not depend on the auction participants' true valuations of the
their energy bill, customers receive a graphic that shows how their good because the participant's goal may be only attaining a winning
energy behavior compares to others in terms of efficiency. This may bid. As a result, valuations obtained from an auction setting may
make PE behavior seem normal; that everyone else may be doing it differ from those in retail settings. Nevertheless, in experimental
(GWP, 2012). research, simulations (such as an auction) are common to assess
A second suggestion considers the important step of pricing and behavior and to test theory because they require the commitment of
how much consumers are willing to pay for a product. Oftentimes real money have been argued to create an incentive to reveal true
the resistance to PE behavior is the cost (Bazoche et al., 2008; reservation prices (Noussair, Stephanie, & Bernard, 2004).
Bennett & Williams, 2011; D'Souza et al., 2007; Grail Research, Although a limitation to an extent, the strength of this research
2009; Ottman et al., 2006). Consumers may not understand why is also represented by the auction process. Going forward, research
a PE product should cost more if it was manufactured with less measuring PE behaviors needs to incorporate creative efforts such
packaging or it was transported less distance. For example, in this as this to capture actual behavior rather than rely on self-reported
study, the Neutral PE price premium group stated price and quality measures. For example, Hanss and Bo €hm (2013) tested actual sus-
(value) are among the most important product benefits they sought tainable grocery purchase behavior by giving individuals a real
when shopping for a particular product e in this case, wine. budget to purchase groceries. Barker, Fong, Grossman, Quin, and
Perhaps the best approach to start this value creation perception is Reid (1994) measured congruence between self-reported recy-
to entice consumers to change their PE behavior. An example of cling behavior and attitudes with actual recycling via a mail room
increasing the value perception could come through loyalty awards experiment on a college campus. Through systematic and unob-
programs, such as the U.K.'s Nectar loyalty card program that trusive observation, they found these to be weakly linked, as most
awards two points for every £1 spent on certain purchases or participants reported positive attitudes and pro-recycling behavior,
through the online store (Nectar, 2012). but few actually engaged in it. Similar methods and findings appear
Our third suggestion considers the importance of “image.” As in Gamba and Oskamp (1994) for a comingled recycling program,
discussed earlier, consumers reporting high self-enhancement who also found an overstatement in actual recycling behavior
values place a larger importance on the social gratification a compared to self-report. Studies such as these that measure actual
product provides and the related image with its ownership. For environmental behavior can overcome response biases such as
those in the Neutral PE price premium group, their self- social desirability (Dono, Webb, & Richardson, 2010), especially for
enhancement was average, yet they placed a moderately high behaviors that entail a normative component.
agreement on the social benefit a product should deliver. Conse- Finally, an interesting finding in this study was that there were a
quently, a key consideration of shaping PE behavior is the product substantial number of people who would actually pay less for a PE
image and how it is related. Image has a commanding impact over consumer good. That is, in the presence of information (vs. the
the way people perceive things, and should be an essential concept blind condition), their actual WTP for a PE product decreased by an
in shaping marketing, advertising, and communications endeavors. average of $2.74. This negative reaction to the PE product may stem
from the overwhelmingly imperative nature of most environ-
4.6. Gender and PE behavior mental messages (Kronrod, Grinstein, & Wathieu, 2012), as well as
their aim to arouse negative emotions such as fear and guilt. For
Finally, our last general suggestion considers gender effects example, some of the most popular environmental slogans from
regarding pro-environmental behavior (Stafford & Hartman, 2012). ThinkSlogans.com are as follows: “Protect our earth today for our
We believe that to encourage PE behavior and reach a wider children's tomorrow” and “Don't let the water run in the sink, our
audience, all consumers need to be engaged. Yet, with few excep- life's on the brink!” Kronrod et al. (2012) found that imperative
tions, males are not targeted as aggressively as females (Zelezny, claims such as these may backfire if the issue is not important to
Chua, & Aldrich, 2000). For example, in a 22 nation cross-national consumers. This is rooted in psychological reactance theory (Brehm
226 N.A. Barber et al. / Journal of Environmental Psychology 40 (2014) 218e227

& Brehm, 1981), which states that people tend to resist being told Bernard, J. C., Zhang, C., & Gifford, K. (2006). An experimental investigation of
consumer willingness to pay for non-GM foods when an organic option is
what to do, and may actually do the opposite. Perhaps the resis-
present. Agricultural and Resource Economics Review, 35, 374e385.
tance to pay more for PE products (and even pay substantially less) Blake, J. (1999). Overcoming the ‘value-action gap’ in environmental policy: Ten-
is based in this psychological reactance theory. However, beyond sions between national policy and local experience. Local Environment, 4(3),
profiling this group, we can only propose the possible underlying 257e278.
Brehm, J. W., & Brehm, S. (1981). Psychological reactance: A theory of freedom and
drivers. Future research should consider this potential issue of control. Mahway, NJ: Erlbaum.
reactance within the PE domain. Chan, D. C. N., Wu, A. M. S., & Hung, E. P. W. (2010). Invulnerability and the intention
In sum, the results of this study are stepping stones toward to drink and drive: An application of the theory of planned behavior. Accident
Analysis & Prevention, 42(6), 1549e1555.
future research in a field that is growing in theoretical and practical Chandon, P., Morwitz, V. G., & Reinartz, W. J. (2005). Do intentions really predict
significance. Future research should continue to investigate how to behavior? Self-generated validity effects in survey research. Journal of Market-
shape PE behavior, both within and beyond a consumption con- ing, 69(2), 1e14.
Chernev, A. (2003). Reverse pricing and online price elicitation strategies in con-
textdand within a framework that is able to capture actual sumer choice. Journal of Consumer Psychology, 13, 51e62.
behavioral data. This is especially important due to the increasing Clark, C., Kotchen, M. J., & Moore, M. R. (2003). Internal and external influences on
global media coverage on environmental issues, the expansion of proenvironmental behavior: Participation in a green electricity program. Jour-
nal of Environmental Psychology, 23, 237e246.
PE products in the market, and the integration of ecological issues Cotton Inc. (2010). Consumer environment survey. Retrieved from http://www.
in educational and political systems. cottoninc.com/corporate/Market-Data/SupplyChainInsights/Shades-of-the-
Green-Consumer/.
De Pelsmacker, P., Driesen, L., & Rayp, G. (2005). Do consumers care about ethics?
Appendix. Information from Wine Labels Used During the Willingness to pay for fair-trade coffee. Journal of Consumer Affairs, 39(2),
Auction Process 363e385.
Diamantopoulos, A., Schlegelmilch, B. B., Sinkovics, R. R., & Bohlen, G. M. (2003).
Can socio-demographics still play a role in profiling green consumers? A review
The following are descriptions of the wines used in the auction, of the evidence and an empirical investigation. Journal of Business Research,
which includes all of the information that appeared on each of the 56(6), 465e480.
Didier, T., & Lucie, S. (2008). Measuring consumer's willingness to pay for organic and
labels except for brand designation.
fair trade products. International Journal of Consumer Studies, 32(5), 479e490.
Dono, J., Webb, J., & Richardson, B. (2010). The relationship between environmental
activism, pro-environmental behaviour and social identity. Journal of Environ-
PE wine label information
mental Psychology, 30, 178e186.
D'Souza, C., Taghian, M., & Khosla, R. (2007). Examination of environmental beliefs
This Merlot wine is produced in Ukiah, California, using organic and its impact on the influence of price, quality and demographic character-
grape growing processes. No pesticides or herbicides are used in istics with respect to green purchase intention. Journal of Targeting, Measure-
ment & Analysis for Marketing, 15(2), 69e78.
the making of these wines. Instead, ladybugs by the millions con- Dunning, D. (2007). Self-image motives and consumer behavior: How sacrosanct
trol pests, naturally. These vineyards have been farmed this way for self-beliefs sway preferences in the marketplace. Journal of Consumer Psychol-
more than forty years. ogy, 17(4), 237e249.
Fielding, K., McDonald, R., & Louis, W. (2008). Theory of planned behaviour, identity
and intentions to engage in environmental activism. Journal of Environmental
Non-PE wine label information Psychology, 28, 318e326.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An intro-
duction to theory and research. Reading, MA: Addison-Wesley.
This wine is a Merlot made from grapes grown in the Livermore Follows, S. B., & Jobber, D. (2000). Environmentally responsible purchase behaviour:
Valley in California. The austere, rocky soil from the Livermore A test of a consumer model. European Journal of Marketing, 34(5e6), 723e746.
Fujii, S. (2006). Environmental concern, attitude toward frugality, and ease of
Valley encourages the vines to concentrate flavors in the ripening behavior as determinants of pro-environmental behavior intentions. Journal of
fruit. Environmental Psychology, 26, 262e268.
Gamba, R., & Oskamp, S. (1994). Factors influencing community residents' partici-
pation in commingled curbside recycling programs. Environment and Behavior,
References 26(5), 587e612.
Ga€rling, T., Fujii, S., Ga
€rling, A., & Jakobsson, C. (2003). Moderating effects of social
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavioral and value orientation on determinants of pro-environmental behavior intention.
Human Decision Processes, 50(2), 179e211. Journal of Environmental Psychology, 23(1), 1e9.
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social Gatersleben, B., Steg, L., & Vlek, C. (2002). The measurement and determinants of
behavior. Englewood Cliffs, NJ: Prentice-Hall. environmentally significant consumer behavior. Environment and Behavior,
Andrews, R., & Currim, I. (2003). Recovering and profiling the true segmentation 34(3), 335e362.
structure in markets: An empirical investigation. International Journal of Gil, J. M., Gracia, A., & Sanchez, M. (2000). Market segmentation and willingness to
Research in Marketing, 20, 177e192. pay for organic products in Spain. The International Food and Agribusiness
Bang, H. K., Ellinger, A., Harjimarcou, J., & Traichal, P. (2000). Consumer concern, Management Review, 3(2), 207e226.
knowledge, belief, and attitude toward renewable energy: An application of the Godin, G., Conner, M., & Sheeran, P. (2005). Bridging the intentionebehaviour gap:
reasoned action theory. Psychology and Marketing, 17(6), 449e468. The role of moral norm. British Journal of Social Psychology, 44(4), 497e512.
Banister, E. N., & Hogg, M. K. (2004). Negative symbolic consumption and con- Grail Research, LLC. (2009). The green revolution. Retrieved from http://
sumers' drive for self-esteem: The case of the fashion industry. European Journal grailresearch.com/pdf/ContenPodsPdf/The_Green_Revolution.pdf.
of Marketing, 38(7), 850e868. Greaves, M., Zibarras, L. D., & Stride, C. (2013). Using the theory of planned behavior
Barber, N., Kuo, P., Bishop, M., & Goodman, R. (2012). Measuring psychographics to to explore environmental behavioral intentions in the workplace. Journal of
assess purchase intention and willingness to pay. Journal of Consumer Market- Environmental Psychology, 34, 109e120.
ing, 29(4), 280e293. Guagnano, G., & Markee, N. (1995). Regional differences in the socio-demographic
Barber, N., Taylor, C., & Strick, S. (2010). Selective marketing to environmentally determinants of environmental concern. Population and Environment, 17(2),
concerned wine consumers: A case for location, gender and age. Journal of 135e149.
Consumer Marketing, 10(1), 64e75. GWP. (2012). Home energy reports. Glendale Water & Power. Retrieved from http://
Barker, K., Fong, L., Grossman, S., Quin, C., & Reid, R. (1994). Comparison of self- www.glendalewaterandpower.com/save_money/residential/home_energy_
reported recycling attitudes and behaviors with actual behavior. Psychological reports.aspx.
Reports, 75(1), 571e577. Hair, J., Black, W., Babin, B., & Anderson, R. (2007). Multivariate data analysis (7th
Bazoche, P., Deola, C., & Soler, L. G. (2008). An experimental study of wine con- ed.). Englewood Cliffs, NJ: Prentice-Hall.
sumers' willingness to pay for environmental characteristics. In 12th Congress of Hanss, D., & Bo € hm, G. (2013). Promoting purchases of sustainable groceries: An
the European Association of Agriculture Economists. Retrieved from http://www. intervention study. Journal of Economic Psychology, 33, 53e3367.
legrenelle-environment.fr/grenelle-environment. Harland, P., Staats, H., & Wilke, H. A. (1999). Explaining proenvironmental intention
Beauregard, K. S., & Dunning, D. (1998). Turning up the contrast: self-enhancement and behavior by personal norms and the theory of planned behavior. Journal of
motives prompt egocentric contrast effects in social judgments. Journal of Applied Social Psychology, 29(12), 2505e2528.
personality and social psychology, 74(3), 606. Harrison, G. W. (1989). Theory and misbehavior of first-price auctions. American
Bennett, G., & Williams, F. (2011). Mainstream green. The Red Papers, 4, 1e131. Economic Review, 79(4), 749e763.
 N.A. Barber et al. / Journal of Environmental Psychology 40 (2014) 218e227 227

Horowitz, J. K., McConnell, K. E., & Murphy, J. J. (2008). Behavioral foundations of Roccas, S., Sagive, L., Schwartz, S., & Knafo, A. (2002). The big five personality factors
environmental economics and valuation (Working Paper). Retrieved from http:// and personal values. Personality and Social Psychology Bulletin, 28(7), 789e801.
home.cerge-ei.cz/richmanova/UPCES%5CHorowitz,%20McConnell,%20Murphy% Roozen, I. T. M., & De Pelsmacker, P. (1998). Attributes of environmentally friendly
20-%20Behavioral%20Foundations%20of%20Environmental%20Economics% consumer behavior. Journal of International Consumer Marketing, 10(3), 21e41.
20and%20Valuation.pdf. Roseman, M. G., Hoon Kim, Y., & Zhang, Y. (2013). A study of consumers' intention to
Hunter, L., Hatch, A., & Johnson, A. (2004). Cross-national gender variation in purchase ethnic food when eating at restaurants. Journal of Foodservice Business
environmental behaviours. Social Science Quarterly, 85, 677e694. Research, 16(3), 298e312.
Kagel, J. H., & Levin, D. (1993). Independent private auctions: Bidder behavior in Schwartz, S. H. (1977). Normative influences on altruism. In L. Berkowitz (Ed.), Ad-
first-, second-, and third-place auctions with varying numbers of bidders. vances in experimental social psychology (Vol. 10). New York: Academic Press.
Economic Journal, Royal Economic Society, 103(419), 868e879. Schwartz, S. H. (1992). Universals in the content and structure of values: Theoretical
Kayaman, R., & Arasli, H. (2007). Customer based brand equity: Evidence from the advances and empirical tests in 20 countries. In M. Zanna (Ed.), Advances in
hotel industry. Journal of Managing Service Quality, 17(1), 92e109. experimental social psychology (Vol. 25). Orlando, FL: Academic Press.
Kennedy, E. H., Beckley, T. M., McFarlane, B. L., & Nadeau, S. (2009). Why we don't Schwartz, S. H. (1994). Are there universal aspects in the structure and contents of
“walk the talk”: Understanding the environmental values/behaviour gap in human values? Journal of Social Issues, 50(4), 19e45.
Canada. Human Ecology Review, 16(2), 151. Schultz, P. W., Gouveia, V. V., Cameron, L. D., Tankha, G., Schmuck, P., & Marek, F.
Kilbourne, W. E., Grünhagen, M., & Foley, J. (2005). A cross-cultural examination of (2005). Values and their relationship to environmental concern and conserva-
the relationship between materialism and individual values. Journal of Economic tion behavior. Journal of Cross-Cultural Psychology, 36, 457.
Psychology, 26, 624e641. Schultz, P. W., Oskamp, S., & Mainieri, T. (1995). Who recycles and when: A review
Kim, H., & Kim, W. (2005). The relationship between brand equity and firms' per- of personal and situational factors. Journal of Environmental Psychology, 15,
formance in luxury hotels and chain restaurants. Tourism Management, 26, 105e121.
549e560. Schultz, P. W., & Zelezny, L. (1999). Values as predictors of environmental attitudes:
Kollmuss, A., & Agyeman, J. (2002). Mind the gap: Why do people act environ- Evidence for consistency across 14 countries. Journal of Environmental Psy-
mentally and what are the barriers to pro-environmental behaviour? Environ- chology, 19, 255e265.
mental Education Research, 8(3), 239e260. Sedikides, C., Gregg, A. P., Cisek, S., & Hart, C. M. (2007). The I that buys: Narcissists
Kronrod, A., Grinstein, A., & Wathieu, L. (2012). Go green! Should environmental as consumers. Journal of Consumer Psychology, 17(4), 254e257.
messages be so assertive? Journal of Marketing, 76(1), 95e102. Silva, A., Nayga, R., Jr., Campbell, B. L., & Park, J. (2007). On the use of valuation
Lange, C., Martin, C., Chabanet, C., Combris, P., & Issanchou, S. (2002). Impact of the mechanisms to measure consumers' willingness to pay for novel products: A
information provided to consumers on their willingness to pay for champagne: comparison of hypothetical and non-hypothetical values. International Food and
Comparison with hedonic scores. Food Quality and Preferences, 13(7/8), Agribusiness Management Review, 10(2), 65e179.
597e608. Smith, E. R., & Mackie, D. M. (2007). Chapter 8 Cognitive Dissonance: changing atti-
Laroche, M., Bergeron, J., & Barbaro-Forleo, G. (2001). Targeting consumers who are tudes to justify behavior. Social Psychology. London: Psychology Press.
willing to pay more for environmentally friendly products. Journal of Consumer Soler, F., Gil, J. M., & Sanchez, M. (2002). Consumers' acceptability of organic food in
Marketing, 18(6), 503e520. Spain: Results from an experimental auction market. British Food Journal, 4(8),
Loureiro, M. L. (2003). Rethinking new wines: Implications of local and environ- 670e687.
mentally friendly labels. Food Policy, 28(5e6), 547e560. Stafford, E. R., & Hartman, C. L. (2012). Making green more macho. The Solutions,
Loureiro, M. L., Umberger, W. J., & Hine, S. (2003). Testing the initial endowment 3(4), 25e29.
effect in experimental auctions. Applied Economic Letters, 10(5), 271e276. Stampfli, N., Siegrist, M., & Kastenholz, H. (2010). Acceptance of nanotechnology in
Lusk, J. L., & Shogren, J. F. (2007). Experimental auctions. Cambridge, MA: Cambridge food and food packaging: A path model analysis. Journal of Risk Research, 13(3),
University Press. 353e365.
Monroe, M. C. (2003). Two avenues for encouraging conservation behavior. Human Steg, L., & Vlek, C. (2009). Encouraging pro-environmental behaviour: An integra-
Ecology Review, 10(2), 113e125. tive review and research agenda. Journal of Environmental Psychology, 29,
Morwitz, V. G., Steckel, J. H., & Gupta, A. (2007). When do purchase intentions 309e317.
predict sales? International Journal of Forecasting, 23(3), 347e364. Sweeney, J., & Soutar, G. (2001). Consumer perceived value: The development of a
Murphy, J. J., & Stevens, T. H. (2004). Contingent valuation, hypothetical bias and multiple item scale. Journal of Retailing, 77(2), 203.
experimental economics. Agricultural and Resource Economics Review, 33(2), Thøgersen, J. (2002). Direct experience and the strength of the personal norm-
82e192. behavior relationship. Psychology and Marketing, 19(10), 881e893.
Nectar. (2012). Rewards program. Available at http://www.nectar.com/about-nectar. Tindall, D. B., Davies, S., & Mauboules, C. L. (2003). Activism and conservation
points. behavior in an environmental movement: The contradictory effects of gender.
Niessen, J., & Hamm, U. (2008). Identifying the gap between stated and actual Society and Natural Resources, 16, 909e932.
buying behaviour on organic products based on consumer panel data. In Torgler, B., García-Valin ~ as, M. A., & Macintyre, A. (2008). Differences in preferences
Cultivating the Future Based on Science: 2nd Conference of the International towards the environment: The impact of a gender, age and parental effect
Society of Organic Agriculture Research ISOFAR, Modena, Italy, June 18e20, 2008. (Working Paper). Retrieved from http://ageconsearch.umn.edu/bitstream/6371/
Available at http://orgprints.org/11998/2/Niessen_11998_ed.pdf Accessed 2/080018.pdf.
22.03.13. Van Veen, V., Krug, M. K., Schooler, J. W., & Carter, C. S. (2009). Neural activity
Noussair, C., Stephane, R., & Bernard, R. (2004). Revealing consumers' willingness- predicts attitude change in cognitive dissonance. Nature Neuroscience, 12(11),
to-pay: A comparison of the BDM mechanism and the Vickrey auction. Jour- 1469e1474.
nal of Economic Psychology, 25(6), 725e741. Vermeir, I., & Verbeke, W. (2006). Sustainable food consumption: Exploring the
OECD. (2008). Promoting sustainable consumption good practices in OECD countries. consumer “attitudeebehavioral intention” gap. Journal of Agricultural and
Retrieved from http://www.oecd.org/greengrowth/40317373.pdf. Environmental Ethics, 19(2), 169e194.
Orth, U. R., Wolf, M. M., & Dodd, T. H. (2005). Dimensions of wine region equity and Vickrey, W. (1961). Counter speculation, auctions and competitive sealed tenders.
their impact on consumer preferences. The Journal of Product and Brand Man- Journal of Finance, 16(1), 8e37.
agement, 14(2e3), 88e97. Voelckner, F. (2006). An empirical comparison of methods for measuring con-
Ottman, J. A. (2011). The new rules of green marketing: Strategies, tools and inspiration sumers' willingness to pay. Marketing Letters, 17(2), 137e149.
for sustainable branding. San Francisco, CA: Berrett-Koehler Publishers, Inc. Walsh, G., Shiu, E., & Hassan, L. M. (2012). Investigating the drivers of consumer
Ottman, J. A., Stafford, E. R., & Hartman, C. L. (2006). Avoiding green marketing intention to buy manufacturer brands. Journal of Product & Brand Management,
myopia: Ways to improve consumer appeal for environmentally preferable 21(5), 328e340.
products. Environment: Science and Policy for Sustainable Development, 48(5), Wedel, M., & Kamakura, W. (2002). Introduction to the special issue on market
22e36. segmentation. International Journal of Research in Marketing, 19(3), 181e183.
Poortinga, W., Steg, L., & Vlek, C. (2004). Values, environmental concern, and Zelezny, L. C., Chua, P. P., & Aldrich, C. (2000). Elaborating on gender differences in
environmental behavior a study into household energy use. Environment and environmentalism. Journal of Social Issues, 56, 443e457.
behavior, 36(1), 70e93.