Attitude-behaviour Inconsistency in Environmental Consumerism

The following section provides a short review of the attitude-behaviour inconsistency in environmental consumerism. Next, a brief review of the relevant attitude and social dilemma theory is presented followed by the elements that affect consumers’ cooperation in environmental social dilemma.

Get Help With Your Essay

If you need assistance with writing your essay, our professional essay writing service is here to help!

Find out more

Environmental attitudes and behaviours are definitely an interesting yet challenging area for marketers to explore. According to Yencken, Fien and Sykes (2000, p.41), attitude is defined as “the learned predisposition to respond in a consistently favourable or unfavourable manner with respect to a given object”. The theory of attitude stated that consumer attitude has a direct influence on one’s behaviour and it is the predictor of behaviour (Malhotra, 2005; Kotler and Keller, 2006). However, in the environment consumerism context, mixed result showing positive (Peattie, 1995; Arbuthnot, 1977) and weak (Wicker, 1969; McGuire, 1985) relationship between both attitude and behaviour undermine consumer attitude theory (Martin and Simintiras, 1995; Kraus, 1995). According to a recent research done by Gupta and Ogden (2009) and Mintel (2006), it supports the attitude-behaviour inconsistency in green consumerism and claims that despite concern about the environment (attitude), few consumers engage into green buying (behaviour) regularly. Therefore, resulted into a challenge for all marketers to better position their green products. Hence, it is crucial to look at the literature on consumer attitude theory and the factors that affect consumer attitude in order to obtain better insight on the Malaysian environmental attitude and behaviour. The following section provides a summary of attitude theory for better understanding of attitude-behaviour inconsistency.

Link Between Attitude and Behaviour

Researchers have long emphasised that there is a causal link between attitude and behaviour (Hini, Gendall and Kearns, 1995). First, to enhance the predictability of consumer attitude, some researchers focused on investigating the moderator factors that influence the strength between attitude and behaviour. For instance, it have been shown that stronger attitude-behaviour relationship can be yield when the attitude are formed based on great degree of direct experience (Millar & Millar, 1996; Fazio & Zanna, 1981), high accessibility of attitudes about an issue (Krosnick, 1989), high importance of the attitude on issue (Crano, 1997) and the attitude is compatible with the beliefs (Sanderson, C.A., 2010). In the other stream of research, researchers focused on the methodological explanations for the for the contradictory results. For instance, to predict behaviour, correspondence must be occurred to the extent that both attitudes and behaviours are the same on four elements: action, target, time and context (Ajzen and Fishbein, 1977).

Various theoretical models were formed to explain attitude-behaviour relationship. One of the theories was the Theory of Reasoned Action (Fishbein and Ajzen, 1974). According to the theory, consumer’s intention to perform an act is the direct determinant of his/her actual behaviour (Oskamp and Schultz, 2005), and intention is affected by two components: (1) consumer attitude toward an object, and (2) consumer’s subjective norms (Wang, 2007). It is noted that attitude is an individual’s evaluation (favourable or unfavourable) of a behaviour based on his/her belief about the outcomes of that particular behaviour and subjective norms are individuals’ feelings to perform or not to perform a behaviour resulted from social pressure (Breckler, Olson and Wiggins,2006). In summary, this model proposed that individual involved in a series of cognitive processing that results to the development of attitudes, norms, intentions and finally behaviour. For instance, individual may perform a particular behaviour as a result of high intention to perform the behaviour due to good evaluation and feeling of encouraged by people around him/her (Raoprasert and Islam, 2010). According to a study done by Kim and Hunter (1993, p.355), it proved that behavioural intentions played an intervening role between attitudes and behaviours, therefore showing consistency with Theory of Reasoned Action. However, this theory cannot apply to all instances and therefore limiting the range of application (Ajzen, 1985). As argued by Ajzen (1985), intentions can affect the behaviour in question only if it is under consumer’s perceived behavioural control. Therefore, the Theory of Planned Behaviour with additional factor (e.g. perceived behavioural control) is introduced as an extension of the Theory of Reasoned Action in order to overcome the limitations of the original model (Ajzen, 1985).

Theory of Planned Behaviour is similar to the earlier study done by Bandura (1982) on self-efficacy. Ajzen claimed that consumer’s intentions to carry out a behaviour are normally resisted by a lack of belief in his/her ability to carry out that behaviour (Stiff and Mongeau, 2003). Therefore, this theory states that behavioural intention is predicted from the combination of perceived behavioural control, attitude and subjective norms. As defined by Ajzen and Madden (1986), perceived behavioural control is an individual belief on the difficulty and easiness to perform a particular behaviour. Generally, the likelihood of an individual to perform an perceived easier behaviour over the difficult one will be higher. The perceived behavioural control concept also reflect individual perception of both internal (e.g. skill) and external (e.g. cooperation of others, money) factors (Kraft, Rise, Sutton and Roysamb, 2005). In other words, individual who perceive that they have the resources, skills or opportunities to perform a particular behaviour are likely to build up high intentions and subsequently perform that behaviour. In addition, this model also suggests that there is a direct influence of perceived behavioural control on behaviour (Stiff and Mongeau, 2003). Individual will be expected to perform the intention when he/she has high control on the behaviour (Ajzen, 2002). As noted by Getzner, Spash and Stagl (2005), by adding behavioural control component in the theory it enable marketer to predict complex behaviour like green behaviour which is normally rely on performance of other behaviours. However, the deliberate, intentional and reasoned aspect of individual attitudinal behaviour proposed in the Theory of Planned Behaviour have been questioned by some recent developments (Hermans and Baeyens, 2007). Several researches found that not all individual behaviour is deliberate, intentional and reasoned because humans simply have insufficient information processing capacity to implement it (Hermans and Baeyens, 2007).

The rise of dual-mode processing models such as the MODE model (Fazio, 1990) have suggested that behaviour is spontaneous in nature and even when an individual does not deliberate about the attitude, attitude can lead an individual’s behaviour (Fazio, 1986; Fazio and Roskos-Ewoldsen, 2004; Fazio, Powell and Herr, 1983). As proposed, it is necessary to have some sort of attitude activation before an attitude can lead behaviour in a particular condition (Fazio and Zanna, 1981; Fazio, 1990). Once attitude is activated, it will plays a filtering role through which the attitude object will be perceived (Hermans and Baeyens, 2007). Subsequently, a perception of the object is developed from the selective perception in the immediate situation and ultimately leads the behaviour (Hermans and Baeyens, 2007). According to Fazio and Zanna (1981), the strength of the relationship between both attitude and that particular object will signify the extent to which that attitude is activated and therefore bring influence on behaviour. The strength is dependent on factors like direct experience with object of the attitude and the frequency the attitude has been expressed (Chaiken and Stangor, 1987; Baron and Byrne, 2000). At the utmost, a particular attitude might be unformed (Bell et al., 2001). Inconsistency between attitude and behaviour of a dedicated environmentalist could even because of he/she never thought of the idea that rubbish left after a football game is a form of litter (Bell et al., 2001).

Due to the unresolved and debatable nature of environmental attitude-behaviour consumerism, researchers like Gill, Crosby and Taylor (1986) and Samuelson and Biek (1991) have attempted to suggest that a strong relationship between attitude and behaviour could only be acquired when attitudinal and behavioural measures are correspond in specificity. According to Hines, Hungerford and Tomera’s (1987) analysis on 51 environmental studies, they found that attitude-behaviour correlation decreases when general attitude toward the environmental issues was operationalised toward a particular environmental behaviour. For instance, to better predict individual differences in biodegradable detergent purchase, individual differences in attitude toward the use of biodegradable detergent is measured instead of individual differences in attitude toward environment (Heberlein and Black, 1976). Thus, to forecast specific green behaviour, attitude measures have to be operationalised toward a particular environmental issue (Heberlein and Black, 1976). Furthermore, personal, situational and contextual factors such as knowledge (Engel, Blackwell and Miniard, 1993), norm (Schwartz, 1977), habit (Norman and Smith, 1995) and non-environmental concerns (e.g. desire to save money) (Steg, 2003) also play a significant role in environmental attitude-behaviour relationship.

To conclude, there is mounting evidence from the environmental consumerism literature which question the assumption that general environment attitudes can help to forecast green buying behaviour (Berger and Corbin, 1992; Wicker, 1969) with agreement that attitudes are tend to be unrelated or weakly related to behaviour (Stiff and Mongeau, 2003). Thus, based on the explanations given by the previous researchers, this paper will discuss the attitude-behaviour gap by structuring it as a social dilemma, furthermore, factors was adopted from this literature to differentiate green buyers from non-green buyers. Social Dilemma model is better used to explain the environmental consumerism problem instead of the attitude models because it takes into account the critical attributes of social dilemma situations, the collective outcomes of the behaviour, the conflict between collective and individual interests, and the reality that individual outcomes are rely on others behaviours (Steg, 2003). Next, Social Dilemma model which is pointed specifically on the environment conservation issue and factors that distinguish green buyer from non-green buyer is discussed.

Social Dilemma in Environment Conservation

Social dilemmas is a condition whereby individual interests are at conflicts with collective interests (Dawes, 1980; Stern and Gardner; 1981; Liebrand, Messick and Wilke, 1992; Samuelson, 1990). In situation when a group has access to a limited common resource, the interdependence nature of this group can be seen as a dilemma because they have the choice to act for either their personal benefit (eg. Individual pleasure) or for collective good (eg. Reduce the consumption of scarce resource in order to protect the environment and consequently leads to long term collective good) (Kurz, 2002). This dilemma can be considered as social because the collective decisions of all the group members determine the common resource’s long term condition and these individual decisions are dependent on others’ decisions and expectation of others’ decisions (Foddy et al., 1999). In that sense, social dilemma is a situation where an individual faces two clashing sets of rationality that he/she need consider when making decisions (Kurz, 2002). As suggested, environmental issues such as pollution and resource depletion are caused by the individuals likelihood to make decisions that lead to personal benefit which will harm the collective that depends on the common resource (Kurz, 2002). As suggested by Samuelson (1990), environmental conservation behaviour represent the real world “common dilemmas”. For instance, when citizens are asked to bring their own shopping bag to shopping in order to conserve the environment by reducing the use of plastic, individual who choose to defect in self interest will not cooperate. However, if all of them choose to defect and keep consuming plastic bag, then everyone will be worse off when environmental degrade.

To summarise, social dilemmas are defined by four properties (Steg, 2003; Dawes, 1980; Vlek, 1996; Messick and Brewer, 1983):

a higher payoff is received by each individual for socially defective behaviour (eg. using a lot of plastic bag) as compared to socially cooperative behaviour, regardless of what others do,

outcomes of behaviour decision rely on the decisions that others make,

a defective decision bring harm to others as compared to a cooperative decision,

compared to all defect, all cooperate is better for all individual.

Therefore, social dilemma is a mixed-motive game which involves individuals attempt to maximise personal interests (selfish) by leaving everyone worse off and individuals attempt to cooperate for the sake of everyone in the long run (Foddy et al., 1999).

Factors that affect consumers’ cooperation in environmental social dilemma

This section will explain attitude-behaviour inconsistency in relation to various social dilemma factors. According to the social dilemma framework, green and non green buyers will be distinguished by the social value orientation, trust in others cooperation, perceived efficacy, ingroup identification and individual consequences.

Social Value Orientation

Research has pinpointed that individual social value orientation influenced whether people cooperate or defect in social dilemma (Messick and McClintock, 1968). As defined by Messick and McClintock (1968), social value orientation is the weights people distribute the outcomes to their own and others in situations of interdependence. According to Biel et al. (2008), individuals are different in their disposition to cooperate or defect in social dilemmas. Some individuals are driven by the desire to maximise their relative gains (competitors), while others desire to maximise their absolute gains (individualists) and there is also others who desire to maximise common gains (cooperators) (Biel et al. 2008). Cooperators weight the moral consequences of their choices more and prefer to cooperate in a social dilemma (Van dijk Vormgeving, n.d.). In conditions of environmental degradation, cooperators have the greater tendency to harvest less from a common resource and engage in pro-environmental works (Van dijk Vormgeving, n.d.). Thus, it is proposed that:

H1. There is a significant association between social value orientation and green buyers.

Trust in Others Cooperation

Besides social value orientations, cooperation is more likely when people trust others to cooperate in social dilemma (Mosler, 1993). The cause of social dilemma is that people generally do not trust others will sacrifice their immediate self-interest (selfish) for the good of all in the long-term (Goethals, Sorenson and Burns, 2004). People also have a strong belief in reciprocity and equity (when someone is selfish then he/she will be selfish too), therefore, causes them to reciprocate the others actions (Goethals, Sorenson and Burns, 2004). In other words, high-truster will have confidence in other’s goodwill, therefore engage in reciprocal cooperation, while on the other hand, low-truster will defect in social dilemma for self-interests (Granovetter, 1992; Joireman et al. 1997). Thus, it is proposed that:

H2: There is a significant association between trust in others cooperation and green buyers.

Perceived Efficacy

Research in social dilemma has shown that cooperative behaviour is also determined by individuals’ perceived efficacy, the judgement of one’s capability to achieve collective rational outcomes (Wim, Messick and Wilke, 1992). It is clear that an individual may believe that his/her efficacy level is very limited especially in the large scale social dilemmas. However, social dilemmas which characterised by a clear rule as to whether or not the collectively outcomes is achieved may work as a boomerang effect (Gasparski, Mlicki and Banathy, 1996). As suggested, individuals may sometimes cease to behave cooperatively because they think their own cooperative efforts would not really create any impact and a lot of others are going to cooperate anyway (Klandermans, 1992). Finding across various social dilemma studies suggested that high level of perceived efficacy is associated with greater cooperation (Kerr, 1992). Furthermore, people may have somewhat biased and underestimate their own efficacy level which lead them to think that their cooperative choices do not really matter, and thus cease to cooperate (Gasparski, Mlicki and Banathy, 1996). Therefore, it is proposed that:

H3: There is a significant association between perceived efficacy and green buyers.

Cost of Cooperation: Inconvenience

Social dilemmas can be conceptualised in the social exchange theory (Gupta and Ogden, 2009) which proposed that social interactions depend on individuals judgements concerning the rewards and costs and individuals primarily concern on maximising rewards and minimising costs (Thibaut and Kelley, 1959). Therefore, social dilemma study recognises the important role of rewards and costs in social dilemma cooperation, whereby it is necessary for individuals to make compromise between the rewards for the group and costs derived from their cooperative behaviour (Messick and Brewer, 1983). However, besides the type of costs such as guilty, anxiety and lost of pride, not much research has explicitly investigated the impacts of individual costs on cooperation (Gupta and Ogden, 2009).

Find out how can help you!

Our academic experts are ready and waiting to assist with any writing project you may have. From simple essay plans, through to full dissertations, you can guarantee we have a service perfectly matched to your needs.

View our services

As proposed by Sen, Gurhan-Canli and Morwitz (2001), consumer boycotts can be regarded as a type of social dilemma. Thus, one factor (eg. inconvenience) was borrowed from this literature to assess consumers costs of adopting pro-environmental behaviour. Inconvenience means to how inconvenient individual perceived to cooperate in an ecologically compatible way (Mukherji, 2005). For instance, individual may think that reducing plastic consumption is important, but he/she may not does not reduce their plastic consumption because it takes too much time or require extra effort. This is the case because by performing environmental behaviour, it presents a big cost to the individual, therefore he/she will attempt to minimise this cost by not cooperating.

Expectation of Others to Cooperate

Past analyses of social dilemmas suggest that expectations of others probable behaviours plays an essential role in determining whether individuals will cooperate or defect (Messick and Brewer, 1983). Expectations of others’ cooperative behaviour could be influenced by one’s own choice due to self-justification process or an assumed similarity mechanism (Yamagishi and Sato, 1986). In the other words, a cooperative individual might rationalise his/her choice by assuming that most others would do the same or take their own choice as an anchor for their estimates of others’ choices (Yamagishi and Sato, 1986). Conversely, it is reasonable that a person’s expectations of others’ decisions influence his/her own decisions and perceptions of norms (Gasparski, Mlicki and Banathy, 1996). For instance, to avoid being a “sucker” or the only one who defects, normative conformity pressures to cooperate is introduced when individual learn that most others are going to cooperate and vice versa (Gasparski, Mlicki and Banathy, 1996). Thus, it is proposed that:

H3: There is a significant association between expectation of others to cooperation and green buyers.

Research Limitations

There are several limitations in this research that need to be recognised despite the reported interesting and relevant insight regarding the inconsistency of attitude-behaviour in environmental consumerism. First, the sample size is limited to 300 respondents in Petaling Jaya town area, thus it is perhaps hard to generalise to a larger number of consumer in Malaysia. As noted by Pitta, Fung and Isberg (1999), all the informations about consumers found are arguable unless the emphasis of the research can be reduced into a smaller measurable group. Therefore, future research should increase the number of sample and research area in nation-wide so as to increase the research reliability and validity. Second, the validity of the survey research depends on the ability of the respondents to correctly determine their degree of agreement with the survey questions. Third, non-product (“No Plastic Bag Day” programme) was used to assess green buying. Therefore, future research should examine the consumption of product (green) to increase the validity of the research. Forth, the self-reported survey may affect the accuracy of the results. Respondents who indicate their green behaviour may not behave so in reality (Laroche, Bergeron and Barbaro-Forleo, 2001), thus to strengthen the findings, future research should add behavioural or observational measures as part of the research in assessing the respondents’ green behaviour (Straughan and Roberts, 1999). Fifth, this research was limited to few factors only, therefore additional factors regarding to social dilemma in green buying may be added into future research.

Fieldwork Methodology

Survey Instrument Development

A questionnaire (survey instrument) with scaled measures was developed to examine the hypotheses relationships. Measures used were adapted from the previous studies (e.g. Hair et al. (1998), Scanlon et al. (2004), Van Lange et al. (1997), De Cremer and Stouten (2003), Ellen, Wiener and Cobb-Walgren (1991), McCarty and Shrum (1994)). Besides that, respondents’ demographic data was also gathered in the first part of the questionnaire.

A pilot study was carried out with 20 business undergraduates in SEGi University College. Upon completion of the survey, students’ feedback regarding the layout, instructions, order and questions were collected. Consequently, the survey was refined and question order was re-arranged in order to construct an allow easy to understand survey. Explanations below explain the survey instrument in detail.


Dependent Variable

To categorise green buyers (dependent variable) from non-green buyers, measure from Hair et al. (1998) was adapted for this study. For this measure, the following question were asked: “When you go shopping, how likely do you bring your own shopping bag?” Responses were recorded on a seven point Likert scale ranging from 1 = Always Bring to 7 = Never Bring. Green buyer was measured as a dichotomous dummy variable whereby a variable has only two categories (green or non-green buyer) and takes only a value of 1 or 0 (1 = green buyer; 0 = non-green buyer) (Babbie, 1998). In this study, polar extreme method was used. This method excludes the middle group from further statistical analysis and compares only the two extreme groups (Hair et al., 1998). This method was adopted because it is able to disclose differences that are not obvious in a regression analysis (middle group), thus provides a deeper insight of both green and non-green buyer (extreme groups) (Green, Tull and Albaum, 1994). Respondents who replied 6 and 7 on the seven point Likert scale were categorised as non-green buyer and subsequently coded as “0”, respondents who replied 1 and 2 on the seven point Likert scale were categorised as green buyer and subsequently coded as “0”, while those who replied 3, 4 and 5 on the seven point Likert scale were eliminated from further analysis. 33 respondents from the middle group were eliminated and thus left 240 respondents for the analysis.

Independent Variables

The measures used in the research were adapted from previous studies. Each answer was collected on a Seven-point Likert scale (1 = strongly agree; 7 = strongly disagree) except social value orientation. As noted by Jacoby and Matell (1971), six to seven point Likert scales are proven to be the most optimal.

Social Value Orientation

To assess the differences in respondents’ social value orientation, a cognitive task with nine choice situations that was adopted from Van Lange et al. (1997) was used. Respondents were categorised as cooperator, competitor and individualist if they made at least six constant choices with one of the social value orientations. For this study, competitors, individualists and respondents who were indifferent between the three social value orientations were coded as “0” while cooperators were coded as “1”.


To assess pro-environmental attitude, a semantic differential scales which was adapted from the Scanlon et al. (2004) was used due to its ability to focus on a respondent attitude toward a specific behaviour (Fishbein and Ajzen, 1975). Respondents’ environmental attitude was measured by instructing respondents to evaluate plastic conservation by placing a mark on one of the seven scales between pairs of antonymic adjectives (eg. good – bad, beneficial – harmful, favourable – unfavourable, useful – useless, wise – foolish, of value – worthless and important – unimportant). The seven pairs of adjectives was selected from Marketing Scales Handbook (Bruner and Hensel, 1992) due to its close applicability.


To measure this variable, four scale items that were adapted from De Cremer and Stouten (2003) were used. The statements for the four items were measured using a seven point Likert scale with two items were reverse coded. The statements were: “I trust that others bring their own shopping bag to shopping; I do not trust that others will bring their own shopping bag to shopping; I think that others trust me to bring my own shopping bag to shopping and; I think that others do not trust me to bring my own shopping bag to shopping”.

Expectation of other Consumers to Cooperate

Respondents expect of other consumers to cooperate were measured using two items that were adapted from Gupta and Ogden (2009). The two items were: “Day after day, most of the consumers will take their own shopping bag when they go shopping and; Most consumers are willing to sacrifice for the sake of the environment”. Responses were recorded on a seven point Likert scale.

Perceived Efficacy

Two reverse coded scale items adpated from Ellen, Wiener and Cobb-Walgren (1991) were used to measure consumer perceived efficacy. The two items were: “There is little that any individual consumer can do to conserve the environment and; As long as other consumer refuse to conserve, the environmental conservation efforts of one person are useless”. Using a seven point Likert scale, respondents’ respond were recorded.

Individual Consequences

Three reverse coded scale items were adpated from McCarty and Shrum (1994). Respondents were asked to indicate their (dis)agreement to items: “It is inconvenient to bring bring a shopping bag when I go shopping; There is too much trouble to bring a shopping bag when I go shopping; I hate to bring the shopping bag when I go shopping”. Responses were recorded on a seven point Likert scale.


Hypotheses were examined through self-administrated survey with a convenience sample of 300 consumers at Giant hypermarket where “No Plastic Bag Day” programme was taken place. As noted by Laroche, Bergeron and Barbaro-Forleo (2001), unrealistic answers from the respondents can be avoided by making sure that “No Plastic Bag Day” programme was in place. Besides that, self-administrated survey was conducted due to its speed and reliability of respondents’ answer (Bourque and fielder, 2003). The respondents represented Malaysian population and there were no specific criteria such as ethic group were used in recruiting the sample, thus leading to the higher results validity. 27 respondents were eliminated from the sample due to incomplete of the survey and 33 were eliminated from the sample due to unclear division between green and non-green buyers. This lead to a final sample of 240 where 61 were categorised into green buyer group and 179 into non-green buyer group.



The researcher provided the respondents with a brief oral explanation of the research, length of time, risks and advantages of participation. Respondents were given the chance to indicate their consent to take part. Finally, 300 respondents agreed to take part while 36 respondents disagreed and therefore left the survey. Respondents were given the time to fill in the questionnaire and, translation and further explanations were given to the respondents when they request. Questionnaire was then collected back for a brief screening to ensure that the respondents had answered all the questions. Finally, respondents were thanked for taking part in the survey.


Most Used Categories

EssayHub’s Community of Professional Tutors & Editors
Tutoring Service, EssayHub
Professional Essay Writers for Hire
Essay Writing Service, EssayPro
Professional Custom
Professional Custom Essay Writing Services
In need of qualified essay help online or professional assistance with your research paper?
Browsing the web for a reliable custom writing service to give you a hand with college assignment?
Out of time and require quick and moreover effective support with your term paper or dissertation?