Determinants of Sustainable Food Consumption

Supervisor: Tommy D. Andersson Master Degree Project No. 2013:69 Graduate School

Master Degree Project in Marketing and Consumption

Determinants of Sustainable Food Consumption

Moving consumers down the path of sustainability by understanding their behavior

Andreas Persson

Determinants of sustainable food consumption

Moving consumers down the path of sustainability by understanding their behavior

Andreas Persson

Increasing consumption of environmentally friendly goods is important for dealing with the climate challenge. The main purpose of the present study is to investigate which factors determine consumption of this type of goods. Primary data is collected in Gothenburg through a questionnaire and analyzed using a SEM model based on the Theory of Planned Behavior. The empirical findings confirm several relationships suggested by the theoretical framework. Attitude and perceived consumer effectiveness are found to be the most important predictors of environmentally friendly food consumption while availability is found statistically insignificant. It is recommended that marketing efforts should focus on these areas, attempting to increase consumer trust in brands and labels. An addition of new variables should increase predictive power of the model and further investigation of different attitudes and behavioral control factors is recommended for future research.

Introduction

According to population forecasts, food production needs to double to meet expected demand, while the harmful environmental effects of food production need to be drastically reduced (Foley et al., 2011) (Carlsson-Kanyama, 1998; Tilman, Balzer, Hill, & Befort, 2011; Tubiello &

Fischer, 2007). To meet the climate challenge, consumption patterns need to change. Market forces will then force production to change when demand for sustainable food products increase.

The total sales of food and drinks in Sweden during 2011 were worth 228 billion SEK. Out of these, roughly 4 %, or about 9 billion SEK, were ecological, a number that has been relatively constant since 2010 (SCB, 2012). In the same period, 1,2 billion SEK worth of Fair Trade labeled products were sold, about 0,5% of the total market sales (Fairtrade-Sverige, 2013). Conventional products have a market share of about 95,5%, and still leave a substantial carbon footprint, while organic product market share growth is stagnating (Carlsson-Kanyama, 1998;

Hillier et al., 2009).

Environmentally friendly products need to have a larger market share to have any practical significance. Practical meaning that the CO

emission from food production would be lowered significantly, which is why understanding consumer behavior regarding sustainable food consumption is important.

The above discussion can be summed up as the following research question:

What factors determine consumption of environmentally friendly food?

This question has been researched before (Verain et al., 2012), but further research will help expand the knowledge in the field, needed to explain what causes this behavior. Given that there might be many local differences between consumers, it is important to investigate the matter in a local context to be able to provide suggestions for more accurate changes (cf.Verain et al., 2012).

One main challenge is that in general, even if positive environmental attitudes are present, they do not translate well into pro- environmental behavior. Typically a combination of different attitudes can explain between 20-30% of the variation in behavior (Armitage & Conner, 2001). In consumer behavior research, this phenomenon is called the gap between attitude and behavior (de Barcellos,

Krystallis, de Melo Saab, Kügler, &

Grunert, 2011; Vermeir & Verbeke, 2006).

Investigating which barriers that exist between attitude and behavior and comprehending the way consumers decide to consume sustainable food products is a part the of process of reducing carbon footprints and in the long run a better theoretical understanding might lead to new information for policy makers. At this point, it is not clear what exact changes that need to be done, but only that drastic change is required. (P. Smith & Gregory, 2013).

Theoretical Framework

The Theory of Planned Behavior (TPB) has been applied in areas as diverse as technological acceptance and religious implications in human behavior (Hobbs, Dixon, Johnston, & Howie, 2013;

Muhamad & Mizerski, 2013; Turhan, 2013).

Furthermore, TPB has been used successfully in numerous studies dealing with sustainable food consumption in both different countries and with different types of products (de Barcellos et al., 2011;

Dean, Raats, & Shepherd, 2008; J. R.

Smith et al., 2008; Urban, Zverinova, &

Scasny, 2012; Zhu, Li, Geng, & Qi, 2013;

Zia, Akram, & Ali, 2010). The

combination of theoretical and practical

stability as well as a proven track record makes this a preferred theoretical framework.

The Theory of Planned Behavior

Theory of planned behavior is developed by Icek Ajzen (1991) using the Theory of reasoned action as a base (Fishbein &

Ajzen, 1975). TPB aims not to just predict, but also explain, human behavior. This is achieved by studying factors that affect behavioral intention which in turn explains behavior. The following section will briefly explain each construct in the framework as depicted in figure 1.

A fundamental assumption of the framework is that intention combines with perceived behavioral control (PBC) to

explain behavior. The purpose of these relationships is to better explain a person’s behavior, in an attempt to capture a broader spectrum of reasons (Ajzen, 1991).

Perceived behavioral control, along with attitude and social norms explain intention.

Attitude toward the behavior is whether the person has a positive or negative attitude to the behavior, for example if a person deems that purchasing sustainable products is a good thing to do, the attitude is positive towards the behavior. The concept Subjective norm is essentially what society thinks about the behavior, and whether or not it is socially acceptable. Finally Perceived behavioral control measures the degree to which an individual believes to be in control of whether or not they can FIGURE 1

Theory of planned behavior as developed by Ajzen (1991)

perform the behavior. The three concepts together play a varying role in explaining the behavioral intention, and some might not be relevant depending on the type of behavior (Ajzen, 1991).

The following sections will elaborate further on the theoretical constructs included in the model.

Attitudes

The word attitude - in the framework of TPB - was originally defined as: “attitude toward the behavior and refers to the degree to which a person has a favorable or unfavorable evaluation or appraisal of the behavior in question.” (Ajzen, 1991, p.

188) In other words, attitudes measure how people feel about the behavior in question and according to Ajzen (1991), attitudes are formed based on the expected outcome of behavior. Because the outcome is something we think is good or bad, we also acquire an attitude towards the behavior, forming a simplified base of the attitude – behavior relationship.

But the above conclusion by Ajzen is very general. Stern, Kalof, Dietz & Guagnano (1995) argue that traditional attitude theory is not well equipped to describe how attitudes are formed for behaviors that are not clearly defined, or still in the process of being defined and changing. Such as sustainable food consumption behavior is,

given that there is no clear dominant label for every type of product to assure consumers that the product is 100%

sustainable. The reason is because traditional attitude- behavior theory uses assumptions about how behaviors or objects exist that are very simplified. I.e.

that behavior exists without any interaction with social processes and do not change in the period of measurement. While in reality it is difficult to precisely define the behavior that the attitude is formed for.

Four general lines of research on attitudes are explored. The first one is linking demographic and personal factors with attitude to environmental action. The second line is based on the perceived risk that is associated with the behavior or object, claiming that attitudes are formed based on how risky the behavior is seen to be, but it fails to account for individual differences, and is more general in nature (Stern, Kalof, Dietz, & Guagnano, 1995).

The third line of research leans on need

theories such as Maslow’s, to claim that

people with their basic needs fulfilled, such

as accommodation and food, are more

likely to care about green behavior and

have a positive attitude towards it. The

forth link is based on moral norms and the

will to help others and care for the greater

good of society; essentially, values are the

reason for attitudes, e.g. when an

individual has a moral value that drives him or her to perform actions that will lead to a greater good, although they might be less convenient for the individual (Stern et al., 1995).

TPB has also been subject to critique. One point, related to attitudes, is that TPB assumes that attitudes are formed from cognitive beliefs. The critique to this is that the affective aspects are ignored, leading to a simplification of attitudes that is not optimal. Researchers such as Arvola et al (2008) argue that both types must be measured in order to achieve the best results. Affective is defined as feelings and emotions that individuals hold about a behavior or an object, cognitive is defined as rational thoughts held about an object (Arvola et al., 2008) (Trafimow &

Sheeran, 1998).

Despite this critique, several studies follow the guidelines of the original TPB, and measure attitudes based on cognitive beliefs. E.g. (Vermeir & Verbeke, 2008) (Vermeir & Verbeke, 2006) (Sparks &

Shepherd, 1992) (J. R. Smith et al., 2008).

Perceived Behavioral Control

Perceived behavioral control can be divided into several variables to cover a greater range of control factors. One alteration is to divide it into two variables:

perceived consumer effectiveness (PCE) and perceived availability. PCE is a measurement for how effective consumers perceive their actions to be. Perceived availability measures how easily accessible consumers believe the goods to be.

Theoretical support for this alteration is provided by studies on the area (Sparks &

Shepherd, 1992; Vermeir & Verbeke, 2008).

Roberts (1996) argue that a high PCE is needed for consumers to act on their intentions. If a consumer does not believe that his or her actions have an effect on the outcome then the consumer will not perform the behavior despite having a positive attitude towards it (Roberts, 1996).

Furthermore, it is argued by Gilg, Barr &

Ford (2005) that PCE in general leads to a higher level of green consumption behavior (Kinnear, Taylor, & Ahmed, 1974). With this in mind, PCE could be a contributing factor to green consumption behavior, and thus should be controlled for in the model (Gilg, Barr, & Ford, 2005).

Similarly, the addition of Perceived

availability represents the Vermeir and

Verbeke (2008) argument that low availability is a contributing reason for why people do not consume sustainable products (Vermeir & Verbeke, 2006, 2008).

By definition in TPB, perceived behavioral control factors predict behavior directly, as well as predicting intentions. Ajzen further mentions that there are often perceived barriers, such as availability, to hinder the performance of the behavior. In this sense, by also including perceived availability in a model, it is possible to attempt a division of the effects from PCE and perceived availability on behavior (Ajzen, 1991;

Vermeir & Verbeke, 2008).

Social norms

Social norms, or subjective norms as they are originally called in TPB, are described mathematically as the sum of each normative belief multiplied with the subject’s motivation. A normative belief in this context reflects the probability that important people or reference groups agree or not with the behavior in question. The motivation is the motivation of the subject to actually adhere to the perceived opinion of said reference group (Ajzen, 1991).

In other terms, social norms are included based on the commonly accepted conclusion that people influence each other’s behavior. Social norms can be

defined as being knowledge about what type of behavior is allowed, frowned upon or judged as forbidden – from a society’s standpoint. Furthermore theory suggests that social norms are learnt naturally, providing evidence of how important they are in acting as predictors for behavior (Ostrom, 2000).

It has been claimed that attitude and social norms are mutually correlated (Terry &

Hogg, 1996; Terry, Hogg, & White, 1999).

Attitudes are formed based on the results of the behavior, which are in turn influenced by what significant others think about the subject performing the behavior.

This correlation is not reflected in TPB, which assumes the individual effects are independent of each other. In an attempt to justify the assumption, Terry et al provided evidence that subjective norms are not a significant predictor of behavior intention.

However the framework used featured several additional variables not found in TPB (Terry & Hogg, 1996; Terry et al., 1999)

Despite the findings of Terry et al, several studies have shown significant results for subjective norms having a positive predictive effect on intention. (see e.g.

(Vermeir & Verbeke, 2008) (J. R. Smith et

al., 2008) (Dean et al., 2008)

There is an inherent problem with the norms measured in a study such as this one, as they are in fact not the social norms but rather the perceived social norms, as seen by members of the society. Social norms then exist only on a societal plane, and are very hard to measure, since if asked, a person can only provide their perception of the norm. The implications is that the findings related to social norms in the current study should not be used as a basis for defining the actual social norms, as they are likely to be wrongly represented (Lapinski & Rimal, 2005).

Intention and Behavior

Ajzen defines intention as how willing people are to perform the behavior in question. Intention, combined with behavioral control, is assumed to be a good indication or prediction of behavior.

(Ajzen, 1991) However, Ajzen has also acknowledged that there is a large gap between intention and behavior. One explanation could be that there is a discrepancy between a hypothetical situation, such as one imposed by a questionnaire, and a real life situation. It is hence expected that only people with either positive or negative extremes respond similarly in both hypothetical and real situations. People with neutral opinions will perhaps be more positive in a hypothetical situation, but less so in a real

situation (Ajzen, Brown, & Carvajal, 2004) (Sutton, 1998). This is a generally sound argument, as there are certainly barriers between intention and behavior, and actions speak louder than words.

One important point to consider when assessing the importance of intention for predicting behavior is the general difficulty to explain human behavior using theoretical models. Using Meta analysis, several studies have found that TPB models explain roughly 20-30% of the variance in behavior, which is limited.

Nevertheless, within the field of understanding behavior, it is a reasonable number. The prediction for intention was between 40-50%, again emphasizing the difference between what a person intends to do, and what they actually do (Sutton, 1998) (Abraham & Sheeran, 2003) (Cooke

& Sheeran, 2004) (Armitage & Conner, 2001).

Intention should be seen as having a

mediating effect on the attitude and

behavior relationship. It is most often the

case that the relationship between intention

and behavior is greater than the direct

relationship between attitude and behavior,

and also that the attitude to behavior

relationship is weaker than the attitude to

intention relationship. Theoretical

implications of this finding include, most

importantly, that there are other highly

significant factors that affect the intention to behavior relationship (Kim & Hunter, 1993).

Model and Hypothesis

The research question “What factors determine consumption of environmentally friendly food?” can be answered through testing of several hypotheses. TPB has been proven to explain a reasonable amount of the variance in behavior in other studies on environmentally friendly food consumption. Perceived behavioral control will be represented by the two measures

perceived consumer effectiveness and perceived availability in the current study.

The division is based on findings from previous research that high PCE will increase green consumer behavior and that lack perceived availability is a barrier preventing consumption (Roberts, 1996;

Vermeir & Verbeke, 2008). All the variables will be measuring opinions about environmentally friendly food and to examine each relationship several hypotheses will be used. Together the answers to these can provide an understanding of possible determinants of FIGURE 2

The proposed model used for measuring environmentally friendly food consumption with the hypotheses listed.

the behavior. The individual hypotheses are listed below along with an illustration of the model in figure 2.

H

: There is a causal relationship between attitude towards environmentally friendly food and intention to consume such food.

H

: There is a causal relationship between social norms regarding environmentally friendly food consumption and intention to consume such food.

H

: There is a causal relationship between perceived consumer effectiveness and intention to consume environmentally friendly food.

H

: There is a causal relationship between perceived consumer effectiveness and consumption behavior of environmentally friendly food.

H

: There is a causal relationship between perceived availability of environmentally friendly foods and intention to consume such food.

H

: There is a causal relationship between perceived availability of environmentally friendly foods and consumption behavior of such food.

H

: There is a causal relationship between intention to consume environmentally friendly foods and consumption behavior of such food.

Methodology

It is typical for studies using TPB as a theoretical framework, to use a survey methodology (Vermeir & Verbeke, 2008) (Ajzen, 1991). The same is done in the current study. The aim of the survey is to collect data that can be used to represent the population, in this case defined as adults of age 18 or higher, living in the greater Gothenburg region. The survey was distributed electronically and respondents received an e-mail with a link to a webpage where they filled in the survey. A structural equation model method is used to analyze the findings and the interrelations of the variables.

Sampling method

To find E-mail addresses for respondents, a web-based telephone directory service, Eniro, was used. The Eniro webpage provides an online open access telephone directory, where you can search for people, based on different variables much like in a traditional telephone directory. To find respondents, repeated searches were made based on surname and postal address. The surnames were taken from a list of the 100 most popular surnames in Sweden, starting with the most popular name. In these searches, any given number of people would also have an e-mail address listed.

When this was the case, the e-mail

addresses were collected by the researcher.

Each search listed a maximum of 1000 people, based on what appears to be a random internal technical process.

By using the described technique, it was possible to ensure that the entire sample resides in the correct geographical area.

The collected e-mail addresses (in total 2330) then received an invitation to participate in the survey along with a short description of the topic and why it was relevant for them to participate. A maximum of two reminders were sent out to non-responders. The main argument for their participation was to aid the research in the area, with the issue of sustainable food consumption mentioned only briefly in an attempt to avoid bias. A response rate of 17,9% was achieved, with a total sample of 417 respondents (n=417).

The sample

A problem with the applied data collection method is that it is possible that several of the collected e-mail addresses were in fact automatically generated e-mails supplied by internet suppliers and service providers.

It might also be e-mails that are seldom checked or not used for other reason. Table 1 provides an overview of how the sample demographic data compares to the population. Based on this comparison, it can be argued that the sample is reasonably

representative of the population, albeit not a true random sample.

Questionnaire design & scales

The questionnaire was divided into a group of questions each measuring a certain variable, followed by some demographic questions. For all seven point Likert scale questions, the numbers 1, 4 and 7 were given the meaning “disagree completely”,

“neutral” and “agree completely”.

Furthermore the term environmentally friendly was used as a common term for all products labeled with Fair Trade, KRAV, EU Eco. Cert. and similar certifications and labeling standards, which was explained to respondents.

Attitude was measured using a seven point semantic scale. Respondents were asked to state to which degree they agree to the

Sample Population

Gender distribution

Male 40,4% 49,5%

Female 57,6% 50,5%

Age distribution

18-25 7% 16%

26-35 21% 18%

36-45 12% 18%

46-55 18% 16%

56-65 19% 14%

66-75 19% 11%

75+ 3% 9%

Average income (age 20-64) 200 000 - 300 000 263 200

*percentages that do not add up are explained by non-respondents or the alternative "I do not want to say"

TABLE 1

Comparison of the sample and the population for key demographics

statements listed. First they were asked to rate the statement: To eat environmentally friendly food is … and then foolish / wise, good / bad, harmful / beneficial, unenjoyable / enjoyable, unpleasant / pleasant. It was followed by a general statement In general, my attitude towards eating environmentally friendly food is:

unfavorable/favorable. Each endpoint was marked by adding the word ‘extremely’.

Cronbach’s alpha was 0,895 for this scale.

The scales used for attitude are based on Sparks & Shephard (1992).

Perceived consumer effectiveness was measured using a seven point Likert scale.

To reiterate, PCE is a measure for how efficient people believe their actions to be in terms of affecting the world’s environmental issues. Cronbach’s alpha was measured to 0,57 in this scale, which is a bit on the low side, but the scale has been used in numerous studies and is considered to be well proven. It includes four questions and questions 1 and 3 were reversed in the analysis due to the phrasing of the questions (Roberts, 1996).

Perceived availability was measured using a seven point Likert scale, asking the respondents three questions about how easy it would be for them to buy environmentally friendly food products in general as well as in their neighborhood and if they think environmentally friendly

food alternatives are easily accessible.

Cronbach’s alpha was 0,905 for this scale.

The scale is constructed by Vermeir &

Verbeke (2008).

Social norms were also measured using a seven point Likert scale, asking for level of agreement to five statements about the effect on your behavior from how society/

friends/ family/ significant others/ people that affect your buying behavior, think you should behave. Cronbach’s alpha was 0,854The particular scale was based on Vermeir & Verbeke (2008), but similar scales are normally used to measure this variable (Dean et al., 2008) .

Behavior intention was measured using a seven point Likert scale, asking respondents to state their agreement to four statements about how likely it is that they would purchase the environmentally friendly alternative of four types of common food products the next time they went shopping instead of conventional products, with the condition that it would be available locally. The four products used were bananas, milk, coffee and chocolate. The products were chosen in an attempt to use common products that most people would at some point purchase.

Dean argues that adding the condition, if

available locally, will remove bias from

lack of availability (Dean et al., 2008).

Cronbach’s alpha for behavior intention was 0,915.

Behavior or self reported past behavior, as it is, was measured by asking the respondents how many times a month; in general, they purchase an environmentally friendly version of each of the above mentioned four food products. Self- reported behavior can provide TPB with more accurate results. (Armitage &

Conner, 2001) A five point scale was used with the end points marked by “Never” and

“Always”. The construct is based on Smith (2008) (J. R. Smith et al., 2008) Cronbach’s alpha was 0,827.

Regarding the Cronbach’s alpha values, only the value from PCE was remarkably low, and should be considered in the further analysis. The remaining values were in a good range.

The complete list of measures along with mean values and standard deviations are listed in Appendix A.

Data cleaning

To avoid problems with missing data, the variables attitude measurement 4 and 5 were removed (with 10,1% and 8,9%

missing data, respectively). The amount of cases remaining with more than 10%

missing data was 9. The 9 still above 10 % were subsequently removed. The reasoning

is that 4 measures for attitude still remain, and should be sufficient and accurate.

In general missing data was only a potential issue on the attitude measures.

The remaining variables had 0-0,7%

missing data. After these operations, all variables and cases have less than 10%

missing data, making it possible to choose freely which imputation method to use - while maintaining a sufficient sample size of 408.

The mean substitution method for imputation of missing data was used to deal with the remaining missing data issues. It is recommended for use when there are relatively low levels of missing data and when there are relatively strong relationships between the variables. It is reasonable to believe that these two conditions are met, with the discussion above in mind.

Results

Measurement model

The measurement model features

correlation relationships between error

terms for the different constructs. The

inclusion of such relationships is under

debate, and it is questioned by some

(Anderson & Gerbing, 1984, 1988). Other

research argues that residual correlations

such as the above are in some cases

necessary and should be encouraged (Cole, Ciesla, & Steiger, 2007). The measurement model is displayed in figure 3

The overall fit of the model is indicated by the chi-square value and the associated p- value. In this case the chi-square was 806,1 with 231 degrees of freedom, giving a Cmin/df equal to 3,490. A significant p- value is expected in this case, due to the large sample size and the large number of measured variables (Hair Jr., Black, Babin,

& Anderson, 2010) .

The CFI value is 0,907 for this model, which is above the recommended value of 0,9, and therefore indicates good fit. The

next type of index is an absolute fit index, RMSEA, which is 0,078 in this case. There is no clear cutoff point for RMSEA values, but a number between 0,05- 0,08 is generally accepted as good fit.

Construct validity

Out of 24 factor loadings, three were slightly under the recommended 0,5 value, while the majority was over 0,7. Out of the three, two were for PCE and one for SN.

Average variance extracted (AVE) was FIGURE 3

The measurement model

higher than 0,5 for all constructs except PCE (AVE = 0,24) Construct reliability was between 0,81-0,91 for all constructs except PCE (0,52). Convergent validity is achieved overall, with the PCE construct perhaps being an issue.

Discriminant validity is tested through comparing AVE with the squared interconstruct correlations (SIC). For attitude, social norms and perceived availability, discriminant validity is proven. For PCE, four SIC values are higher than AVE, and for behavior and behavior intention, one SIC value

respectively is higher than the corresponding AVE value. It is by now evident that there are validity issues with the PCE construct. But overall discriminant validity can be argued for.

Nomological validity is also confirmed as all interconstruct correlations are positive and statistically significant. The relationships are also supported by the vast theoretical evidence supporting the TPB model, providing face validity.

Additionally the use of validated scales in the current study increases the nomological validity.

FIGURE 4

The final structural model

Overall, construct validity is argued to be achieved, with some weaknesses, especially regarding the PCE construct.

Structural model

The structural model was then developed, and checked for fit. The chi-square value is 815,5 with 233 degrees of freedom, resulting in a cmin/df equal to 3,5. The CFI value is 0,906, again reasonably close to the recommended 0,9. The absolute fit index, RMSEA, is 0,078. It is expected that Chi-square is larger in the structural model, than in the measurement model, which is also the case here.

The structural model is displayed in figure 4.

The results from the structural model are summarized in table 2.

H

, H

and H

are significant at a level of 0,01. (p=, 0,006; p=0,003; p<0.001) H

is significant at a level of 0,05. (p=0,034) H

is significant at a 0,1 level (p=0,065), making it the weakest out of the significant hypotheses. H

and H

are both insignificant and the null hypothesis for these cannot be rejected.

DISCUSSION

The empirical results suggest that attitude, social norms and perceived consumer effectiveness are the main determinants for consumption of environmentally friendly food alternatives in a Gothenburg context.

Out of the three, attitude and PCE had the strongest positive effect on the purchase behavior, while social norms had a lower but still significant effect. Mean values for

Hypothesis

Unstandardized

coefficient p-value

H

: Attitude influences behavior intention 0,485 0,006

H

: Social norms influences behavior intention 0,128 0,034

H

: Perceived consumer effectiveness influences behavior intention 1,57 0,003 H

: Perceived consumer effectiveness influences behavior 0,087 0,67 H

: Perceived availability influences behavior intention -0,018 0,822

H

: Perceived availability influences behavior 0,077 0,065

H

: Behavior intention influences behavior 0,508 <0,001

TABLE 2

List of results from hypothesis testing

attitude and PCE are 5,66 and 5,44 respectively, and 4,19 for social norms. A strong causal relationship was confirmed between intention and behavior.

Furthermore perceived availability was found to be insignificant, and no relationship was established to behavior intention, although a weak link between perceived availability and behavior was indicated. No direct link between PCE and behavior could be confirmed.

To put the current findings in perspective, the regression weight of 0,485 for attitude is similar to what can be expected based on other studies. An international study of organic apples and organic pizza purchase behavior in some European countries found results ranging from 0,29 – 0,51 (Arvola et al., 2008).

A study in Sweden from 2001 found that 76% of the respondents bought organic milk less often than ‘sometimes’, which was the middle point. In the current study the same number is 40%, signaling a rather large increase in organic milk sales. Only 10 % of respondents intended to purchase ecological milk in 2001, compared to 30 % in the current study. The attitude measures in the same study had means that were slightly lower than in the current study (Magnusson, 2001). The low increase in the attitude mean value and the high

increase in behavior and intention imply that other significant factors have changed in the twelve year time period that has passed since the referenced study.

The direct link between attitude and behavior has also been explored further in some studies. Theory suggests that the attitude measure had a lower effect on behavior in those studies than the present study where intention is used as a mediator (Aertsens, Mondelaers, Verbeke, Buysse,

& Van Huylenbroeck, 2011) (Pieniak, Aertsens, & Verbeke, 2010). Despite this, other studies have found attitude effects higher than in the current study, in this case on organic yoghurt (B=0,56) (Van Loo, Diem, Pieniak, & Verbeke, 2013).

The above findings further verify that attitudes play an important role in determining environmentally friendly behavior, as is also seen from the results of the current study.

The current study found parts of the perceived behavioral control factors to be insignificant (only PCE was significant for intention and perceived availability was only significant for behavior at a 0,1 level).

Other studies have produced similar

results, albeit on other variations of food

products (Arvola et al., 2008; Dean et al.,

2008; J. R. Smith et al., 2008).

The reason for perceived behavioral control variables often being found to be insignificant could be because in order to maximize the accuracy of the measurement, PBC measures should be phrased and framed closely to the behavior in question. (Sheeran, Trafimow, &

Armitage, 2003). This might explain the insignificant relationships for PBC found, as especially the PCE scale used in the current study, is very general. Evidence exists of PBC being weak or strong depending on the behavior, it follows that environmentally friendly food consumption might be a behavior where it falls short as a predictor (Kraft, Rise, Sutton, & Roysamb, 2005). The current results however contradict this slightly, as PCE was found to have a large impact (regression weight 1,57) on intention.

The above finding is hampered by the fact that the PCE measure in the current study had issues with construct reliability, convergent validity and discriminant validity. However nomological validity is high as it is a scale that has been used successfully several times in the past. It might be that the scale does not perform well in the current study due to geographical or cultural reasons – or it is simply outdated.

In the current study, 89% of respondents chose 4 or higher on the perceived

availability measure, indicating that the distribution is heavily skewed to the right.

This result could be contrary to the popular conclusion that availability is a barrier to increased organic consumption, however it might just mean that availability is generally high in Gothenburg and therefore no longer seen as a barrier (Pearson, Henryks, & Jones, 2011). The finding is further supported by other empirical studies (Tarkiainen & Sundqvist, 2005).

On the other hand perceived availability is significant at a 0,1 level in the relationship with behavior. Maybe this simply shows the logical conclusion that you will not buy a product if it is not available.

Social norms regression weights ranged from 0,12 – 0,46 in the same studies, compared to 0,128 in the current study.

This implies that social norms are less important in Gothenburg than in other places, while attitudes remain important regardless of national context. The findings for social norms are in line with that is expected from cultural theory, with Sweden being an individualistic country, which implies that the findings are consistent with the values of the general population (Hofstede, 1997).

The means for buying intention were for

bananas 5,07, milk 4,97, coffee 4,82 and

chocolate 4,52. While the mean values for

behavior were 3,01, 3,01, 2,74 and 2,35

respectively. Behavior was measured on a five point scale, and intention on a seven point scale, but by comparing the middle points it is still possible to make conclusions based on the means.

Additionally the means for attitude and PCE were also relatively large. These numbers combined suggest that consumers are willing to purchase environmentally friendly products. They believe that their actions will have the intended consequences, and they intend to purchase the goods. But still they do not.

Studies in the UK and Italy alike have found that many consumers associate organic products mostly with greens and fruits, and that they are a gateway product, meaning that it is a product that if purchased, will lead to additional purchases of other sustainable food products. The role of organic vegetables and fruits as a gateway product into organic consumption is verified by a study confirming the relationship between first purchase and additional purchases (Gottschalk & Leistner, 2013). Up to 50%

of all organic products have been found to be fruits and vegetables in some studies (Chinnici, D’Amico, & Pecorino, 2002).

Furthermore, it raises the idea that increases in awareness of the fact that organic/eco-friendly products really are not just vegetables, might increase overall

consumption of all organic food goods (de Magistris & Gracia, 2008; Padel & Foster, 2005; Pearson et al., 2011). This might explain why a positive attitude does not lead to greater consumption – because the attitude is mostly considering vegetables and fruits. If this is the case for the current study is mostly speculation at this point, however, the mean for bananas is the highest, both for intention and self- reported behavior (m=5,07 and 3,01 respectively). The higher mean might also be because bananas, out of the four products, are perceived as the most natural, and therefore have the highest credibility as environmentally friendly. The opposite is true for chocolate as it has a higher carbon foot print in terms of production.

The mean for chocolate is also the lowest (m=2,35 and 4,52) , adding further strength to the prospect.

One possible reason is the lack of trust in

labels (Cliath, 2007). For a high value in

PCE to have an impact on behavior, it must

be possible to purchase products that the

consumer believes are actually

environmentally friendly. To elaborate,

consumers can safely and easily compare

organic bananas to conventional bananas

(Tobler, Visschers, & Siegrist, 2011). Milk

is mostly produced locally, and therefore is

inherently less harmful for the

environment. Additionally the labels used

on milk are using Swedish standards which mean consumers might be more inclined to trust them. For coffee and chocolate, the main ingredients are always imported, usually from far away. With a multitude of environmental and sustainable claims shown by several different labels, comparability is arguably low for these products, and confusing and distrust is high (Engels, Hansmann, & Scholz, 2010;

Gadema & Oglethorpe, 2011; Sirieix, Delanchy, Remaud, Zepeda, & Gurviez, 2013). Simply put, the current findings suggest that the consumers are not sure if they can trust that the products will perform as promised, making them little more than more expensive versions of conventional products.

Theory suggests that a common, perhaps state-supported, label that would apply for all types of products in combination with intense marketing efforts to really communicate a guarantee to the consumers, would help increase the trust of labels (Engels et al., 2010). However, although a national label can be beneficial it is no guarantee of success (Thogersen, 2010). Since trust in labels depend on personal values as well as trust in the organization behind the label, perhaps a lack of trust in the government will work against a national label (Grankvist,

Lekedal, & Marmendal, 2007; Thogersen, Haugaard, & Olesen, 2010).

In addition to the potential distrust in labels, consumers should be educated in what constitutes environmentally friendly food products, to enable them to act on their intentions.

Theoretical implications

Based on the current findings, perceived consumer effectiveness is only predicting intention, and perceived availability is only predicting behavior at a 0,1 level.

Furthermore the perceived consumer effectiveness measure has issues with construct validity and reliability.

An attempt to combine the perceived availability measure with the perceived consumer effectiveness measure into one variable was tested through the use of a chi-square significance test. Out of the two, the current model proved to be the best representation of the data.

The relationship between intention and

behavior was strong. However, it needs to

be pointed out that the behavior measure

was actually self-reported past behavior,

and as such it is not possible to guarantee

the reliability of it. The value might be

inflated compared to if actual behavior had

been measured.

Managerial implications

Although social norms are found to be significant, the actual increase in behavior from change in social norms is quite low. It follows that marketing efforts should not put much effort on this particular subject if the aim is to maximize results.

For producers of environmentally friendly food products looking to increase their market share, the current findings supported by previous findings, suggest that the focus should be on associating the product with a prominent and established certification label. The challenge is choosing the right one. Until a product wide labeling standard is established and has gained a good reputation with consumers, a selection of current labels will have to suffice. On the other hand, using too many labels could have the opposite effect. This also implies that companies should stay away from designing their own sustainability labels, as it might fall in the category of suspicious relatively unknown labels (Sirieix et al., 2013).

Companies dealing in food products that are not vegetables or fruits should also emphasize that organic products can be much more than those. While attitudes are significant in increasing the behavior, attitudes are already generally high, so the

main emphasis in this regard should be to focus marketing on transferring the positive attitudes to all types of food products. Companies should then compete on factors such as quality and taste, but also along with other important factors.

Suggestions for further research

The higher mean for the fruit product of the current study compared to the others, indicates that, a theoretical challenge is to further determine how attitudes differ between groups of environmentally friendly products. It further implies that additional attitude measures might be added with positive results.

Furthermore a redevelopment of TPB models aimed at studying consumption of environmentally friendly food products is suggested. The addition of new variables that measure aspects like price and perceived quality could explain further what determines environmentally friendly food purchases.

A further area of interest follows from the

lack of significance of the perceived

availability measure. It would be

interesting to study whether there are

differences between different retailers in

terms of how reliable they are perceived to

be in their role as providers of sustainable

food products. It might be the case that

availability is perceived high in some retail chains while low in others.

Conclusion

The aim of this study is to increase knowledge about the determinants of environmentally friendly food consumption. To this end, TPB is used and the findings include that perceived consumer effectiveness and attitudes are the most important factors for determining environmentally friendly food consumption behavior. I argue that

perceived consumer effectiveness is related to labels in the sense that through labels, consumers can decide how effective the purchase will be. This finding leads to implications about a need for increasing trust in labeling and changing the image of organic products. Furthermore it provides ideas for future research regarding the use of TPB in this specific research area regarding the adding of additional attitude measures variables and adapting measures to the specific products, as well as adding altogether new variables.

References

Abraham, C., & Sheeran, P. (2003). Acting on intentions: The role of anticipated regret. British Journal of Social Psychology, 42, 495-511. doi: 10.1348/014466603322595248

Aertsens, J., Mondelaers, K., Verbeke, W., Buysse, J., & Van Huylenbroeck, G. (2011). The influence of subjective and objective knowledge on attitude, motivations and consumption of organic food. British Food Journal, 113(10-11), 1353-1378. doi: 10.1108/00070701111179988 Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision

Processes, 50(2), 179-211. doi: http://dx.doi.org/10.1016/0749-5978(91)90020-T Ajzen, I., Brown, T. C., & Carvajal, F. (2004). Explaining the discrepancy between intentions and

actions: The case of hypothetical bias in contingent valuation. Personality and Social Psychology Bulletin, 30(9), 1108-1121. doi: 10.1177/0146167204264079

Anderson, J. C., & Gerbing, D. W. (1984). The effect of sampling error on convergence, improper solutions, and goodness-of-fit indexes for maximum-likelihood confirmatory factor-analysis.

Psychometrika, 49(2), 155-173. doi: 10.1007/bf02294170

Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice - a review and recommended 2-step approach. Psychological Bulletin, 103(3), 411-423. doi: 10.1037/0033- 2909.103.3.411

Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40, 471-499. doi: 10.1348/014466601164939 Arvola, A., Vassallo, M., Dean, M., Lampila, P., Saba, A., Lähteenmäki, L., & Shepherd, R. (2008).

Predicting intentions to purchase organic food: The role of affective and moral attitudes in the Theory of Planned Behaviour. Appetite, 50(2–3), 443-454. doi:

http://dx.doi.org/10.1016/j.appet.2007.09.010

Carlsson-Kanyama, A. (1998). Climate change and dietary choices - how can emissions of greenhouse gases from food consumption be reduced? Food Policy, 23(3-4), 277-293. doi:

10.1016/s0306-9192(98)00037-2

Chinnici, Gaetano, D’Amico, Mario, & Pecorino, Biagio. (2002). A multivariate statistical analysis on the consumers of organic products. British Food Journal, 104(3), 187-199. doi:

10.1108/00070700210425651

Cliath, A. G. (2007). Seeing shades - Ecological and socially just labeling. Organization & Environment, 20(4), 413-439. doi: 10.1177/1086026607309406

Cole, D. A., Ciesla, J. A., & Steiger, J. H. (2007). The insidious effects of failing to include design-driven correlated residuals in latent-variable covariance structure analysis. Psychological Methods, 12(4), 381-398. doi: 10.1037/1082-989x.12.4.381

Cooke, R., & Sheeran, P. (2004). Moderation of cognition-intention and cognition-behaviour relations: A meta-analysis of properties of variables from the theory of planned behaviour.

British Journal of Social Psychology, 43, 159-186. doi: 10.1348/0144666041501688

de Barcellos, Marcia Dutra, Krystallis, Athanasios, de Melo Saab, Maria Stela, Kügler, Jens Oliver, &

Grunert, Klaus G. (2011). Investigating the gap between citizens' sustainability attitudes and food purchasing behaviour: empirical evidence from Brazilian pork consumers. International Journal of Consumer Studies, 35(4), 391-402. doi: 10.1111/j.1470-6431.2010.00978.x de Magistris, T., & Gracia, A. (2008). The decision to buy organic food products in Southern Italy.

British Food Journal, 110(8-9), 929-947. doi: 10.1108/00070700810900620

Dean, Moira, Raats, Monique M., & Shepherd, Richard. (2008). Moral Concerns and Consumer Choice of Fresh and Processed Organic Foods1. Journal of Applied Social Psychology, 38(8), 2088- 2107. doi: 10.1111/j.1559-1816.2008.00382.x

Engels, S. V., Hansmann, R., & Scholz, R. W. (2010). Toward a Sustainability Label for Food Products:

An Analysis of Experts' and Consumers' Acceptance. Ecology of Food and Nutrition, 49(1), 30- 60. doi: 10.1080/03670240903433154

Fairtrade-Sverige. (2013, 2013-03-21). Försäljningsstatistik 2012. Retrieved 19/5, 2013

Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intentions and behaviour: An introduction to theory and research. Reading, MA: Addison-Westley.

Foley, J. A., Ramankutty, N., Brauman, K. A., Cassidy, E. S., Gerber, J. S., Johnston, M., . . . Zaks, D. P.

M. (2011). Solutions for a cultivated planet. Nature, 478(7369), 337-342. doi:

10.1038/nature10452

Gadema, Z., & Oglethorpe, D. (2011). The use and usefulness of carbon labelling food: A policy perspective from a survey of UK supermarket shoppers. Food Policy, 36(6), 815-822. doi:

10.1016/j.foodpol.2011.08.001

Gilg, Andrew, Barr, Stewart, & Ford, Nicholas. (2005). Green consumption or sustainable lifestyles?

Identifying the sustainable consumer. Futures, 37(6), 481-504. doi:

http://dx.doi.org/10.1016/j.futures.2004.10.016

Gottschalk, I., & Leistner, T. (2013). Consumer reactions to the availability of organic food in discount supermarkets. International Journal of Consumer Studies, 37(2), 136-142. doi:

10.1111/j.1470-6431.2012.01101.x

Grankvist, G., Lekedal, H., & Marmendal, M. (2007). Values and eco- and fair-trade labelled products.

British Food Journal, 109(2-3), 169-181. doi: 10.1108/00070700710725527

Hair Jr., Joseph F., Black, William C., Babin, Barry J., & Anderson, Rolph E. (2010). Multivariate data analysis : a global perspective (7th ed.). Upper Saddle River, N.J.: Pearson Education.

Hillier, J., Hawes, C., Squire, G., Hilton, A., Wale, S., & Smith, P. (2009). The carbon footprints of food crop production. International Journal of Agricultural Sustainability, 7(2), 107-118. doi:

10.3763/ijas.2009.0419

Hobbs, N., Dixon, D., Johnston, M., & Howie, K. (2013). Can the theory of planned behaviour predict the physical activity behaviour of individuals? Psychology & Health, 28(3), 234-249. doi:

10.1080/08870446.2012.716838

Hofstede, Geert. (1997). Cultures and organizations : software of the mind. New York: McGraw-Hill.

Kim, M. S., & Hunter, J. E. (1993). Relationships among attitudes, behavioral intentions, and behavior - a metaanalysis of past research .2. Communication Research, 20(3), 331-364. doi:

10.1177/009365093020003001

Kinnear, T. C., Taylor, J. R., & Ahmed, S. A. (1974). Ecologically concerned consumers - who are they.

Journal of Marketing, 38(2), 20-24. doi: 10.2307/1250192

Kraft, P., Rise, J., Sutton, S., & Roysamb, E. (2005). Perceived difficulty in the theory of planned behaviour: Perceived behavioural control or affective attitude? British Journal of Social Psychology, 44, 479-496. doi: 10.1348/014466604x17533

Lapinski, M. K., & Rimal, R. N. (2005). An explication of social norms. Communication Theory, 15(2), 127-147. doi: 10.1111/j.1468-2885.2005.tb00329.x

Magnusson, Maria K.; Arvola, Anne; Koivisto Hursti, Ulla-Kaisa;Åberg, Lars; Sjödén, Per-Olow. (2001).

Attitudes towards organic foods among Swedish consumers. British Food Journal, 103(3).

Muhamad, N., & Mizerski, D. (2013). The Effects of Following Islam in Decisions about Taboo Products. Psychology & Marketing, 30(4), 357-371. doi: 10.1002/mar.20611

Ostrom, E. (2000). Collective action and the evolution of social norms. Journal of Economic Perspectives, 14(3), 137-158. doi: 10.1257/jep.14.3.137

Padel, S., & Foster, C. (2005). Exploring the gap between attitudes and behaviour - Understanding why consumers buy or do not buy organic food. British Food Journal, 107(8), 606-625. doi:

10.1108/00070700510611002

Pearson, D., Henryks, J., & Jones, H. (2011). Organic food: What we know (and do not know) about consumers. Renewable Agriculture and Food Systems, 26(2), 171-177. doi:

10.1017/s1742170510000499

Pieniak, Z., Aertsens, J., & Verbeke, W. (2010). Subjective and objective knowledge as determinants of organic vegetables consumption. Food Quality and Preference, 21(6), 581-588. doi:

10.1016/j.foodqual.2010.03.004

Roberts, James A. (1996). Green consumers in the 1990s: Profile and implications for advertising.

Journal of Business Research, 36(3), 217-231. doi: http://dx.doi.org/10.1016/0148-

SCB. (2012, 2012-10-01). Försäljningen av kött fortsätter att stiga. Nr 2012:865. Retrieved 19/5, 2012 Sheeran, P., Trafimow, D., & Armitage, C. J. (2003). Predicting behaviour from perceived behavioural

control: Tests of the accuracy assumption of the theory of planned behaviour. British Journal of Social Psychology, 42, 393-410. doi: 10.1348/014466603322438224

Sirieix, L., Delanchy, M., Remaud, H., Zepeda, L., & Gurviez, P. (2013). Consumers' perceptions of individual and combined sustainable food labels: a UK pilot investigation. International Journal of Consumer Studies, 37(2), 143-151. doi: 10.1111/j.1470-6431.2012.01109.x

Smith, Joanne R., Terry, Deborah J., Manstead, Antony S. R., Louis, Winnifred R., Kotterman, Diana, &

Wolfs, Jacqueline. (2008). The Attitude–Behavior Relationship in Consumer Conduct: The Role of Norms, Past Behavior, and Self-Identity. Journal of Social Psychology, 148(3), 311-334.

Smith, P., & Gregory, P. J. (2013). Climate change and sustainable food production. Proceedings of the Nutrition Society, 72(1), 21-28. doi: 10.1017/s0029665112002832

Sparks, Paul, & Shepherd, Richard. (1992). Self-Identity and the Theory of Planned Behavior: Assesing the Role of Identification with "Green Consumerism". Social Psychology Quarterly, 55(4), 388-399. doi: 10.2307/2786955

Stern, Paul C., Kalof, Linda, Dietz, Thomas, & Guagnano, Gregory A. (1995). Values, Beliefs, and Proenvironmental Action: Attitude Formation Toward Emergent Attitude Objects1. Journal of Applied Social Psychology, 25(18), 1611-1636. doi: 10.1111/j.1559-1816.1995.tb02636.x Sutton, S. (1998). Predicting and explaining intentions and behavior: How well are we doing? Journal

of Applied Social Psychology, 28(15), 1317-1338. doi: 10.1111/j.1559-1816.1998.tb01679.x Tarkiainen, A., & Sundqvist, S. (2005). Subjective norms, attitudes and intentions of Finnish

consumers in buying organic food. British Food Journal, 107(10-11), 808-822. doi:

10.1108/00070700510629760

Terry, D. J., & Hogg, M. A. (1996). Group norms and the attitude-behavior relationship: A role for group identification. Personality and Social Psychology Bulletin, 22(8), 776-793. doi:

10.1177/0146167296228002

Terry, D. J., Hogg, M. A., & White, K. M. (1999). The theory of planned behaviour: Self-identity, social identity and group norms. British Journal of Social Psychology, 38, 225-244. doi:

10.1348/014466699164149

Thogersen, J. (2010). Country Differences in Sustainable Consumption: The Case of Organic Food.

Journal of Macromarketing, 30(2), 171-185. doi: 10.1177/0276146710361926

Thogersen, J., Haugaard, P., & Olesen, A. (2010). Consumer responses to ecolabels. European Journal of Marketing, 44(11-12), 1787-1810. doi: 10.1108/03090561011079882

Tilman, D., Balzer, C., Hill, J., & Befort, B. L. (2011). Global food demand and the sustainable

intensification of agriculture. Proceedings of the National Academy of Sciences of the United States of America, 108(50), 20260-20264. doi: 10.1073/pnas.1116437108

Tobler, C., Visschers, V. H. M., & Siegrist, M. (2011). Organic Tomatoes Versus Canned Beans: How Do Consumers Assess the Environmental Friendliness of Vegetables? Environment and Behavior, 43(5), 591-611. doi: 10.1177/0013916510372865

Trafimow, D., & Sheeran, P. (1998). Some tests of the distinction between cognitive and affective beliefs. Journal of Experimental Social Psychology, 34(4), 378-397. doi:

10.1006/jesp.1998.1356

Tubiello, F. N., & Fischer, G. I. (2007). Reducing climate change impacts on agriculture: Global and regional effects of mitigation, 2000-2080. Technological Forecasting and Social Change, 74(7), 1030-1056. doi: 10.1016/j.techfore.2006.05.027

Turhan, G. (2013). An assessment towards the acceptance of wearable technology to consumers in Turkey: the application to smart bra and t-shirt products. Journal of the Textile Institute, 104(4), 375-395. doi: 10.1080/00405000.2012.736191

Urban, J., Zverinova, I., & Scasny, M. (2012). What Motivates Czech Consumers to Buy Organic Food?

Sociologicky Casopis-Czech Sociological Review, 48(3), 509-536.

Van Loo, E. J., Diem, M. N. H., Pieniak, Z., & Verbeke, W. (2013). Consumer attitudes, knowledge, and consumption of organic yogurt. Journal of Dairy Science, 96(4), 2118-2129. doi:

10.3168/jds.2012-6262

Verain, Muriel C. D., Bartels, Jos, Dagevos, Hans, Sijtsema, Siet J., Onwezen, Marleen C., & Antonides, Gerrit. (2012). Segments of sustainable food consumers: a literature review. International Journal of Consumer Studies, 36(2), 123-132. doi: 10.1111/j.1470-6431.2011.01082.x Vermeir, Iris, & Verbeke, Wim. (2006). Sustainable Food Consumption: Exploring the Consumer

“Attitude – Behavioral Intention” Gap. Journal of Agricultural and Environmental Ethics, 19(2), 169-194. doi: 10.1007/s10806-005-5485-3

Vermeir, Iris, & Verbeke, Wim. (2008). Sustainable food consumption among young adults in Belgium: Theory of planned behaviour and the role of confidence and values. Ecological Economics, 64(3), 542-553. doi: http://dx.doi.org/10.1016/j.ecolecon.2007.03.007

Zhu, Q., Li, N., Geng, Y., & Qi, Y. (2013). Green food consumption intention, behaviors and influencing factors among Chinese consumers. Food Quality and Preference, 28(1), 279-286. doi:

10.1016/j.foodqual.2012.10.005

Zia, M., Akram, M., & Ali, I. (2010). Ethical consumption and consumer purchase intentions: a study

of uk organic food buyers. Actual Problems of Economics, 144-151.

APPENDIX A

Attitude measure 1-3 and 6 Mean

Std.

Deviation Eating environmentally friendly food is: extremely foolish - extremely wise 5,90 1,210 Eating environmentally friendly food is: extremely bad - extremely good 5,86 1,194 Eating environmentally friendly food is: extremely harmful - extremely beneficial 5,39 1,227 In general, my attitude towards eating environmentaly friendly food is: extremely

unfavorable - extremely favorable

5,50 1,314

Perceived availability measure 1-3

It is easy for me to acquire environementally friendly food 5,09 1,368

It is easy for me to acquire environementally friendly food in my neighborhood 4,86 1,482

I think environmentally friendly food is easily available 4,80 1,414

Social norms measure 1-5

People who are important to me think I should buy environmentally friendly food products 4,34 1,727 My family thinks I should buy environmentally friendly food products 4,19 1,731 Society thinks I should buy environmentally friendly food products 4,38 1,550 My friends think I should buy environmentally friendly food products 4,04 1,566 People who influence my buying behavior think I should buy environmentally friendly food

products

4,00 1,666

Behavioral intention measure 1-4

If it is available locally, I intend to buy environmentally friendly bananas instead of conventional bananas the next time I go shopping

5,07 1,917

If it is available locally, I intend to buy environmentally friendly milk instead of conventional milk the next time I go shopping

4,97 1,921

If it is available locally, I intend to buy environmentally friendly coffee instead of conventional coffee the next time I go shopping

4,82 1,871

If it is available locally, I intend to buy environmentally friendly chocolate instead of conventional chocolate the next time I go shopping

4,52 1,889

Behavior measure 1-4

How often per month do you buy environmentally friendly bananas? 3,01 1,444

How often per month do you buy environmentally friendly milk? 3,01 1,397

How often per month do you buy environmentally friendly coffee? 2,74 1,408 How often per month do you buy environmentally friendly chocolate? 2,35 1,173 Perceived consumer effectiveness measure 1-4 (1 and 3 are reversed)

It is worthless for the individual consumer to do anything about pollution (reversed ) 5,49 1,941 When I buy products, I try to consider how my use of them will affect the environment and

other consumers

4,56 1,547

Since one person cannot have any effect upon pollution and natural resource problems, it doesn't make any difference what I do (reversed)

5,87 1,567

Each consumer's behavior can have a positive effect on society by purchasing products sold by socially responsible companies

5,84 1,535

Each individual measure with mean value and standard deviation