Debate Paper

Research Report

Choice of organic foods is related to perceived consequences for human

health and to environmentally friendly behaviour

Maria K. Magnusson a,*, Anne Arvola

a , Ulla-Kaisa Koivisto Hursti

a ,

Lars Åberg b , Per-Olow Sjödén

a

a Department of Public Health and Caring Sciences, Section of Caring Sciences, Uppsala University, Uppsala Science Park, Uppsala SE-751 83, Sweden

b Department of Psychology, Uppsala University, Box 1225, Uppsala SE-751 42, Sweden

Received 28 March 2001; revised 13 December 2002; accepted 19 December 2002

Abstract

We designed a questionnaire concerned with attitudes and behaviour towards organic foods, environmentally friendly behaviour (EFB),

and perceived consequences of organic food choice in terms of human health, the environment and animal welfare. It was mailed in 1998 to a

random nation-wide sample of 2000 Swedish citizens, ages 18 – 65 years, and 1154 (58%) responded. Self-reported purchase of organic foods

was most strongly related to perceived benefit for human health. Performance of EFBs such as refraining from car driving was also a good

predictor of purchase frequency. The results indicate that egoistic motives are better predictors of the purchase of organic foods than are

altruistic motives.

q 2003 Elsevier Science Ltd. All rights reserved.

Keywords:

Introduction

Several studies have provided evidence that consumers

consider the sensed characteristics of food to be the most

important factors in their choice of food (e.g. Magnusson,

Arvola, Koivisto Hursti, Åberg, & Sjödén, 2001; Torjusen,

Lieblein, Wandel, & Francis, 2001; Wandel & Bugge,

1997). However, it appears that non-sensory attributes of

foods are becoming increasingly important (Torjusen et al.,

2001; Wandel, 1994). Among the most notable are the

absence of food additives, preservatives and residues

(Wandel, 1994; Wilkins & Hillers, 1994), nutritional

value (Jolly, 1991; Torjusen et al., 2001; Wandel, 1994;

Wandel & Bugge, 1997), and how the food was produced

(Land, 1998; Torjusen et al., 2001). Examples of concerns

about food production are animal welfare and the specific

production system used (e.g. conventional vs organic

production). A majority (61%) of Norwegians consider

animal welfare in food production to be an important aspect

of the quality of foods (Torjusen et al., 2001). More than

30% state that they are willing to pay 10% more for meat

produced in accordance with ethical animal care principles

(Wandel & Bugge, 1997).

Consumers also express interest in issues relating food to

health (Fagerli & Wandel, 1999; Rozin, Fischler, Imada,

Sarubin, & Wrzesniewski, 1999; Wandel, 1994). Healthi-

ness is an important criterion for purchase and a parameter

of quality for many consumers (Land, 1998; Magnusson

et al., 2001; Wandel & Bugge, 1997). Although there is no

unambiguous evidence that organic foods are healthier than

conventional foods (Torjusen, Nyberg, & Wandel, 1999),

consumers perceive foods labelled as organic to be healthier

than conventional foods (Grankvist & Biel, 2001; Magnus-

son et al., 2001; Torjusen et al., 1999). Concern for the

environment is another reason for purchasing organic foods

(e.g. Schifferstein & Oude Ophuis, 1998). Thus, concerns

for one’s health and for the environment are the two most

commonly stated motives for purchasing organic foods

(Wandel & Bugge, 1997) with personal health being more

important than concerns for the environment (Tregear,

Dent, & McGregor, 1994; Wandel & Bugge, 1997).

Health and environmental motives differ in the sense that

concern about health can be regarded as egoistic (benefits

the individual or his/her family) while consideration for the

environment and animal welfare are more altruistic (benefits

0195-6663/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.

doi:10.1016/S0195-6663(03)00002-3

Appetite 40 (2003) 109–117

www.elsevier.com/locate/appet

* Corresponding author. Tel.: þ46-18-471-63-13; fax: þ46-18-471-34-

90.

E-mail address: [email protected] (M.K. Magnusson).

society rather than the individual). Altruistic considerations

often carry personal behavioural or even economic costs

(Antonides & van Raaij, 1998) and the majority of

consumers are unwilling to forego much personal benefit

in order to contribute to the benefit of the community (The

Hartman Group, 1997; Wandel & Bugge, 1997).

Several studies of environmental concerns and activities

have found rather weak correlations between attitudes and

behaviour (Ebreo, Hershey, & Vining, 1999; Grankvist &

Biel, 2001; Tracy & Oskamp, 1983). For example, Ebreo

and co-workers (1999) found weak correlations (0.22 – 0.32)

between on the one hand the frequency of recycling of

several materials (e.g. cardboard and plastic containers), and

on the other concerns about product attributes related to the

conservation of nature (e.g. ‘made from recyclable

materials’ and ‘energy conserving’). However, there is

evidence that relationships between items of behaviour

within the same domain are stronger than between

behaviours across domains. Grankvist (2001) found stron-

ger correlations between the self-reported frequency of

purchase of different eco-labelled foods (r ¼ 0:30 – 0:56)

than between the purchase of eco-labelled foods and

recycling of several fractions (e.g. bottles, batteries,

packages etc.) (r ¼ 0:14 – 0:28). Few studies have investi-

gated to what extent environmentally friendly behaviours

(EFBs) relate to attitudes to and the purchase of organic

foods. Since organic foods are marketed as being envir-

onmentally friendly, it is of interest to investigate the

relationships between EFB and attitudes/behaviour towards

organic foods. The present study includes an attempt to

investigate these relationships.

The overall objective of the present study is to investigate

the importance of perceived environmental, animal welfare

and human health consequences of organic food purchase

for consumer attitudes and self-reported purchase of organic

foods. More specifically, the aims are to investigate: (1)

consumer perceptions of the likelihood of occurrence and

importance of three types of consequences of organic food

choice: environmental, human health consequences, and the

well-being of farm animals; (2) the possibility to predict

consumer attitudes to and purchase of organic foods on the

basis of such perceptions; and (3) on the basis of self-

reported recycling and other EFBs. In addition, the role of

demographic variables will be explored. The third aim

concerns the extent to which recycling and other EFBs

covary with attitudes to and the purchase of organic food.

Methods

Respondents

Two thousand addresses ages 18 – 65 years (60% of the

Swedish population is in this age-range) were selected

randomly from the national population register (Sema

InfoData, 1998). Questionnaires were mailed to them

during the spring of 1998, and those not responding were

sent two reminders. Below, the most recent data available

for the entire Swedish population (Statistics Sweden,

2001a,b) are given after the abbreviation Pop. The total

number of respondents was 1154 (58%). Forty-six percent

of those were men (Pop 49%), and 54% women (Pop 51%).

Mean age (S.D.) was 40.6 (13.2) years (Pop 40.2 years).

There were three educational-level response options:

elementary school or nine-year compulsory education

(21%, Pop 27%); a degree from folk high-school (post

compulsory education for adults), or two or three years of

upper secondary school (52%, Pop 45%); and university or

university college education (27%, Pop 26%). Respondents

were offered a lottery ticket (value approximately US $2.5)

if they completed the questionnaire or could donate the

same amount to charities (Swedish Save the Children, or the

Swedish Children’s Cancer Foundation).

Questionnaire

Since the organic food group contains a large variety of

products, we decided to focus most questions on a few types

instead of organic foods in general. Four foods were

selected as targets of questions; milk, meat, potatoes and

bread, to provide examples of organic foods of differing

types, prices and availability. They are all staple foods in the

Swedish diet. At the time of the study, milk was the most

widely available organic food, and potatoes were almost

equally available. Organic bread and meat had just begun to

become available in supermarkets. However, their prices

were much higher than those of conventionally produced

counterparts. A majority of questions concerned the organic

varieties of the four target foods and were focused on buying

rather than on consumption.

The questionnaire consisted of two sections. The first

section (see Magnusson et al., 2001 for details) comprised

questions about the frequency of purchase of organic milk,

meat, potatoes, and bread, general attitudes towards and

perceived availability of these products, and beliefs about

their characteristics relative to their conventionally pro-

duced alternatives. There were two versions of this first

part of the questionnaire. In one, the questions concerned

organic milk and meat and, in the other, organic potatoes

and bread. Half of the sample (n ¼ 1000) received the milk

and meat version (52% response rate), and the other half

(n ¼ 1000) the potatoes and bread version (48%). Descrip-

tive data and demographic comparisons for the questions of

the first section have been reported earlier (Magnusson

et al., 2001).

The second section asked about the likelihood and

importance of 17 possible consequences of buying organic

foods in general. These questions did not deal with specific

foods since they were intended to assess beliefs about the

possible consequences of the use of organic methods of

food production. Respondents were also asked about the

frequency of their performance of several EFBs.

M.K. Magnusson et al. / Appetite 40 (2003) 109–117110

Questions

Attitudes towards buying organic foods. Respondents

were asked to rate how good, important, and wise they think

it is to buy organic milk, meat, potatoes and bread. Five-

point bipolar scales were used, ranging from ‘very bad’ (1)

to ‘very good’ (5), ‘very unimportant’ to ‘very important’,

and from ‘very foolish’ to ‘very wise’. Attitude ratings were

computed as means of these three items.

Perceived importance of ‘organic’ as a purchase

criterion. Respondents were asked to estimate the import-

ance of the purchase criterion ‘organically produced’ for

milk, meat, potatoes and bread. Response formats were: ‘not

at all important’ (1), ‘somewhat important’, ‘rather

important’, ‘important’, and ‘very important’ (5). This

variable will be referred to as ‘organically produced’.

Intention to buy organic. The intention to buy organic

milk, meat, potatoes and bread was assessed on five-point

unipolar scales ranging from ‘not at all likely’ (1) to ‘very

likely’ (5), and the question asked was ‘The next time you

buy…, how likely is it that you will choose organic…?’ This

variable is termed ‘purchase intention’.

Purchase frequency. Purchase frequency of the organic

foods was assessed by the question ‘When you buy

milk/meat/potatoes/bread, how often do you buy organic

milk/meat/potatoes/bread?’ Seven-point unipolar scales

were used, ranging from ‘never’ (1) to ‘always’ (7). This

variable is referred to as ‘purchase frequency’.

Perceived consequences of buying organic foods in

general. Respondents were asked to estimate how likely,

and how important they think it is that specific conse-

quences result/would result from their purchase of organic

foods. The questions asked were ‘By purchasing organic

foods, I help/would help to…’, and ‘When/if you purcha-

se/would purchase organic foods, how important is it to you

that this helps to…’. The same 17 types of consequences

were listed for likelihood and importance judgements.

Responses were given on five-point unipolar scales ranging

from ‘not at all likely/important’ ( ¼ 1) to ‘very like-

ly/important’ ( ¼ 5). A sixth response option was ‘unde-

cided’. In the analyses, this option has been regarded as

neither positive nor negative and has, therefore, been

assigned a value corresponding to the central value of the

scale ( ¼ 3). The consequences described were ‘improve the

general state of the environment’, ‘improve circumstances

and health of the farm animals’, ‘improve my own or my

family’s health’, ‘give myself a good conscience’, ‘avoid

risks that may be associated with eating non-organic foods’,

‘reduce the use of artificial fertilisers in agriculture’, ‘reduce

the eutrophication of lakes and watercourses’, ‘reduce the

pollution of the soil’, ‘reduce the transportation of foods’,

‘reduce the use of petrol and other non-renewable

sources of energy’, ‘reduce the amount of waste’, ‘reduce

the ozone-hole in the atmosphere’, ‘preserve biodiversity in

nature’, ‘reduce the use of herbicides and pesticides in

agriculture’, ‘reduce the medication of farm animals’, ‘give

my children better food’, and ‘reduce the risk for illness in

my family’.

Self-reported recycling and other EFBs. Respondents

were asked how frequently they recycled glass, paper/plas-

tic/metal packages, newspapers/paper, and batteries, and

how often they performed eight other EFBs: ‘avoid

purchasing products in environmentally non-friendly

packages’, ‘purchase environment-friendly labelled pro-

ducts’, ‘save electricity’, ‘donate money to environmental

organisations’, ‘discuss environmental problems with my

friends or family’, ‘refrain from car driving to spare the

environment’, ‘avoid purchasing new products to spare the

environment’, and ‘compost or leave domestic refuse for

composting’. The questions asked were ‘When you are about

to dispose of the following items, how often do you leave them

to be recycled?’ and ‘How often do you perform the following

actions?’ Responses were given on five-point unipolar scales

ranging from ‘never’ (1) to ‘always/very often’ (5).

Statistical methods

Descriptive statistics, factor analysis, bivariate corre-

lations, multiple regression analysis and independent t-tests

were used. For all analyses, the level of significance was set

to ,0.01. Separate exploratory factor analyses (principal

component, varimax rotation, oblique solution, factor scores

$0.50) were performed on the items concerned with

possible consequences of buying organic foods, the

recycling frequency of glass/packages/batteries, and the

performance of other EFBs.

Factor analyses

Perceived consequences of buying organic foods. The

ratings of likelihood and importance of consequences of

buying organic foods in general were subjected to separate

principal component analyses (PCA). The solution for the

importance ratings did not conform to a logical pattern. The

solution for the likelihood ratings had five interpretable

factors. The three likelihood factors with Eigenvalues .1

gave the best interpretation using an oblique rotation. The

first factor (environment) reflected environmental pollution,

the second (Transportation/Waste) transportation and waste

handling, and the third (Health) perceived health aspects

(Table 1). Each factor showed good homogeneity (Cronba-

ch’s a-coefficients; Table 1). The intercorrelations between

the factors were 0.44 (Environment and Transportation/

Waste), 0.56 (Environment and Health) and 0.43 (Trans-

portation/Waste and Health), respectively.

The two statements regarding animal welfare did not

load highly on any of the mentioned factors nor did they

form a separate factor. Thus, further study of the

significance of animal welfare concerns was made

impossible.

Self-reported EFBs and recycling. PCA yielded three

factors (Table 2), and an oblique rotation resulted in the best

M.K. Magnusson et al. / Appetite 40 (2003) 109–117 111

solution. The first factor, EFB was based on performance of

EFBs. The Easy Recycling factor included the recycling of

items disposable at easily available collection points, and

Advanced Recycling involved the recycling of items that

require more effort on the part of the consumer (e.g. fewer

collection points). The intercorrelations between the factors

were 0.32 (EFB and Easy Recycling), 0.39 (EFB and

Advanced Recycling) and 0.43 (Easy and Advanced

Recycling), respectively. The reliability of the factors was

examined in terms of homogeneity (Cronbach’s a-coeffi-

cients; Table 2). All coefficients were between 0.69 and

0.75, which indicates good homogeneity.

Bivariate correlations

Correlations were computed between, on the one hand,

each of the three factors concerned with the consequences of

organic food purchase, and on the other, the variables

attitude, the importance of the purchase criterion ‘organi-

cally produced’, purchase intention, and reported purchase

frequency of organic foods. All correlations were significant

at p , 0:0001 (r ¼ 0:17 – 0:57) except those between

Transportation/Waste, and purchase intention/purchase

frequency of organic milk (detailed data not shown).

Correlations were also computed between each of the

three factors concerned with EFB, and the four variables

regarding attitude/behaviour for each food (Table 3). All

correlations involving EFB, and most of those concerned

with Advanced Recycling were significant at p , 0:0001.

However, only a minority of the correlations with Easy

Recycling were significant at p , 0:0001 most likely

because these behaviours were reported to be performed

very often by most respondents resulting in a reduction of

variance. Since the majority of factors were significantly

related to the attitude/behaviour variables, all were entered

in the multiple regression analyses.

Table 1

Principal component analysis of the rated likelihood of perceived consequences of organic food purchase

Factor name % a

a Items Loading

Environment 51 0.90 Improve the state of the environment 0.63

Reduce the use of artificial fertilisers in agriculture 0.97

Reduce the eutrophication of lakes and watercourses 0.93

Reduce the pollution of the soil 0.75

Reduce the use of herbicides and pesticides in agriculture 0.65

Transportation/Waste 10 0.88 Reduce the transportation of foods 0.92

Reduce the use of petrol and other non-renewable sources of energy 0.90

Reduce the amount of waste 0.54

Health 7 0.89 Improve my own or my family’s health 0.78

Give myself a good conscience 0.74

Avoid risks that may be associated with eating non-organic foods 0.80

Give my children better food 0.68

Reduce the risk for illness in my family 0.63

a % of explained variance.

Table 2

Principal component analysis of self-reported EFBs and recycling

Factor name % a

a Items Loading

EFB behaviour 31 0.75 Avoid purchasing products in non-environmentally friendly packages 0.68

Purchase environmentally friendly labelled products 0.70

Save electricity 0.49

Donate money to environment organisations 0.58

Discuss environmental problems with my friends or family 0.67

Refrain from car driving to spare the environment 0.68

Avoid purchasing new products to spare the environment 0.69

Easy recycling 12 0.69 Recycle glass 0.76

Recycle paper-packages 0.65

Recycle newspapers and paper 0.87

Recycle batteries 0.53

Advanced recycling 8 0.72 Recycle plastic packages 0.87

Recycle metal packages 0.89

Compost or leave domestic refuse for composting 0.57

a % of explained variance.

M.K. Magnusson et al. / Appetite 40 (2003) 109–117112

Results

Likelihood and importance of consequences

For most items, a majority of respondents perceived it to

be quite or very likely and important that the stated

environmental, health and animal well being consequences

will be influenced by their choice of organic foods (Table 4).

Only a minority (1 – 11%) stated that it is not at all likely or

important that the given consequences will result from their

purchase of organic foods.

Frequency of recycling and other

environmental behaviours

The majority of the consumers reported recycling all the

given items regularly. Almost all stated that they often or

always recycle glass (95%), newspapers/paper (90%), and

batteries (89%). Corresponding figures for paper/plastic/

metal packages were 74, 47 and 52%. The EFBs that most

respondents performed often or very often were save

electricity (54%), purchase environmentally friendly pro-

ducts (46%) and compost or leave domestic refuse for

composting (42%). Only a minority declared that they often

or very often donate money to environmental organisations

(6%) and avoid purchasing new products to spare the

environment (10%).

Perceived consequences and EFBs vs attitudes

and behaviour towards organic foods

Multiple regression analyses were performed for each of

the four organic foods separately to investigate the

predictability of the variables attitude, ‘organically pro-

duced’, purchase intention and purchase frequency, on the

basis of the demographic variables (age, gender (man ¼ 0,

women ¼ 1) and education) and the factor-analytically

derived factors. The three education-level responses were

converted into a dichotomous variable (without university

education ¼ 0, with university education ¼ 1). The six

factors and the demographic variables explained between 21

Table 3

Bivariate correlations between, on the one hand factors EFB behaviour,

Easy Recycling and Advanced Recycling and, on the other attitude,

perceived importance of the purchase criterion organically produced

(organic), purchase intention (intention), and purchase frequency (pur-

chase) of organic milk, meat, potatoes, and bread

EFB behaviour Easy recycling Advanced recycling

Attitude—milk 0.39*** 0.17*** 0.18***

Attitude—meat 0.35*** 0.13* 0.14**

Attitude—potatoes 0.40*** 0.18*** 0.22***

Attitude—bread 0.36*** 0.15** 0.18***

Organic—milk 0.36*** 0.13* 0.17***

Organic—meat 0.44*** 0.18*** 0.20***

Organic—potatoes 0.44*** 0.12 0.23***

Organic—bread 0.45*** 0.16** 0.27***

Intention—milk 0.40*** 0.10 0.19***

Intention—meat 0.40*** 0.16** 0.19***

Intention—potatoes 0.41*** 0.11 0.20***

Intention—bread 0.37*** 0.15** 0.22***

Purchase—milk 0.37*** 0.12* 0.17***

Purchase—meat 0.39*** 0.15** 0.16**

Purchase—potatoes 0.44*** 0.16** 0.26***

Purchase—bread 0.40*** 0.17** 0.26***

*p , 0.01, **p , 0.001, ***p , 0.0001.

Table 4

Perceived consequences: percent of respondents who stated that it is rather or very likely/important and not at all likely/important that their choice of organic

foods will result in the listed consequences, mean scores and S.D.

Consequence Very/rather

likely

Not at all

likely

Mean

score

S.D. Very/rather

important

Not at

all important

Mean score S.D.

Reduce the use of herbicides and pesticides in agriculture 76 2 4.1 0.99 85 1 4.4 0.82

Reduce the use of artificial fertilisers in agriculture 72 2 4.1 1.03 79 1 4.2 0.90

Improve the general state of the environment 72 2 4.0 0.97 82 1 4.3 0.86

Improve circumstances and health of the domestic animals 72 2 4.0 1.03 83 1 4.3 0.88

Reduce the eutrophication of lakes and watercourses 71 2 4.1 1.04 83 1 4.3 0.86

Reduce the medication of domestic animals 66 2 3.9 1.05 83 1 4.3 0.87

Give my children better food 66 3 3.9 1.08 80 1 4.3 0.92

Reduce the pollution of the soil 65 2 3.9 1.01 82 1 4.3 0.88

Avoid risks that may be associated with eating

non-organic foods

56 4 3.7 1.13 62 3 3.8 1.04

Give myself a good conscience 56 7 3.6 1.27 53 7 3.6 1.23

Improve my own or my family’s health 52 3 3.6 1.10 79 1 4.2 0.94

Preserve biodiversity in nature 51 4 3.6 1.11 75 1 4.1 0.89

Reduce the risk for illness in my family 46 4 3.4 1.15 79 2 4.3 0.96

Reduce the transportation of foods 36 10 3.1 1.23 67 3 3.9 1.07

Reduce the amount of waste 35 8 3.1 1.17 70 2 4.0 1.00

Reduce the ozone-hole in the atmosphere 33 11 3.0 1.19 70 3 4.1 1.06

Reduce the use of petrol and other non-renewable

sources of energy

32 10 3.0 1.21 67 3 3.9 1.06

M.K. Magnusson et al. / Appetite 40 (2003) 109–117 113

Table 5

Results of multiple regression analysis (standardised beta coefficients, unless otherwise indicated)

Milk Meat

Attitude Organic Intention Purchase Attitude Organic Intention Purchase

n ¼ 528 n ¼ 481 n ¼ 534 n ¼ 509 n ¼ 533 n ¼ 483 n ¼ 528 n ¼ 489

R 2

0.32 0.22 0.26 0.21 0.32 0.38 0.28 0.23

F (df) 26.9 (9, 518) 15.0 (9, 471) 20.6 (9, 524) 14.4 (9, 499) 27.5 (9, 523) 32.5 (9, 473) 22.4 (9, 518) 16.1 (9, 479)

Age 20.21*** (0.04) – – – 20.12* (0.01) – – –

Gender – – – – – – – –

Education – – – 0.14** (0.02) – – – –

EFB 0.21*** (0.03) 0.23*** (0.03) 0.28*** (0.05) 0.28*** (0.05) 0.15* (0.01) 0.23*** (0.03) 0.20*** (0.02) 0.23*** (0.03)

Easy recycling – – – – – – – –

Advanced recycling – – – – – – – –

Environment 0.20*** (0.02) – – – 0.14* (0.01) – – –

Transportation/Waste 20.16* (0.02) – 20.16** (0.02) – 20.13* (0.01) – – –

Health 0.33*** (0.05) 0.37*** (0.06) 0.35*** (0.06) 0.23*** (0.03) 0.45*** (0.09) 0.52*** (0.12) 0.40*** (0.07) 0.26*** (0.03)

Potatoes Bread

Attitude Organic Intention Purchase Attitude Organic Intention Purchase

n ¼ 486 n ¼ 411 n ¼ 485 n ¼ 432 n ¼ 480 n ¼ 463 n ¼ 483 n ¼ 464

R 2

0.41 0.33 0.31 0.28 0.39 0.38 0.31 0.24

F (df) 37.3 (9, 476) 21.4 (9, 401) 23.2 (9, 475) 18.2 (9, 422) 34.0 (9, 470) 30.7 (9, 453) 23.9 (9, 473) 16.3 (9, 454)

Age 20.23*** (0.05) 20.11* (0.01) 20.18*** (0.03) 20.11* (0.01) 20.20*** (0.04) – 20.16*** (0.03) –

Gender 0.10* (0.01) – – – – – – –

Education – – – – – – – –

EFB 0.15** (0.01) 0.28*** (0.05) 0.23*** (0.04) 0.27*** (0.05) 0.12* (0.01) 0.21*** (0.03) 0.19*** (0.02) 0.22*** (0.03)

Easy recycling – – – – – – – –

Advanced recycling – – – 0.14* (0.01) – – – –

Environment 0.18** (0.02) – – – 0.21*** (0.02) – – –

Transportation/Waste – – – – – – – –

Health 0.36*** (0.05) 0.37*** (0.06) 0.34*** (0.05) 0.30*** (0.04) 0.35*** (0.05) 0.40*** (0.07) 0.37*** (0.06) 0.19* (0.02)

*p , 0.01, **p , 0.001, ***p , 0.0001. Significant contributions by demographic variables, self-reported behaviour and perceived consequences to the prediction of attitudes, perceived importance of the

purchase criterion organically produced (organic), purchase intention (intention) and purchase frequency (purchase). (Squared semipartial correlations).

M .K

. M

a g

n u

sso n

e t

a l.

/ A

p p

e tite

4 0

(2 0

0 3

) 1

0 9

– 1

1 7

1 1

4

and 41% of the variance. Health was the most important

predictor of attitudes, ‘organically produced’, and purchase

intention for the four foods (Table 5). The health factor was

also an important predictor of the purchase frequency of the

four target foods (Table 5). However, Health and EFB

appeared to be equally important predictors of purchase

frequency (Table 5). EFB was the second most important

predictor of attitudes (only meat), ‘organically produced’

(all four foods), and purchase intention (all four foods). Age

was a significant predictor of attitudes (all four foods),

‘organically produced’ (potatoes), purchase intention (pota-

toes, bread), and purchase frequency (potatoes). The

younger the respondents the more likely they were to be

positive towards organic foods. The factor Environment

contributed to the prediction of attitude for all four foods. In

a few cases, gender (attitude, potatoes), education (pur-

chase, milk), Transportation/Waste (intention, milk; atti-

tude, milk and meat), and Advanced Recycling (purchase,

potatoes) made significant contributions (Table 5).

Comparisons between groups with strong vs weak intentions

to buy organic foods

Since the factor analysis of importance ratings regarding

the perceived consequences of buying organic foods did not

conform to a logical pattern, no consistent factors could be

found. Instead, respondents with strong vs weak intentions

to purchase each of the four organic foods were compared

with regard to their likelihood and importance ratings of the

environmental, human health and animal welfare conse-

quences. This was done to make possible comparisons

between the two types of ratings. Those who had intention

ratings above the median value (three for milk; four for

meat, potatoes, bread) were classified as having a ‘strong’

intention, and those with ratings below as having a ‘weak’

intention. Overall, those having a strong intention perceived

it more likely and stated that it was more important that the

environmental, human health and animal welfare conse-

quences would result from their choice of organic foods

compared to those with a weak intention. Nearly all of the

likelihood ratings for the four target foods (62 of 68,

tð541 – 590Þ ¼ 3:6 – 9:6, p , 0:0001) and most of the

importance ratings for organic milk, meat and potatoes

differed significantly between the two groups (42 of 51,

tð541 – 590Þ ¼ 4:0 – 7:2, p , 0:0001). However, only five

differences of seventeen possible for importance ratings

were statistically significant between the groups with strong

and weak intentions to purchase organic bread. These

findings indicate that the likelihood and importance ratings

yielded a similar pattern of results.

Discussion

Concern for one’s own or family health was the strongest

predictor of attitudes, the importance of the criterion

‘organically produced’, and purchase intention for the

investigated organic foods. Further, health was an important

predictor of purchase frequency. In regression analyses

including only the three factors based on the likelihood

ratings of perceived consequences of buying organic foods

(Health, Environment, Transportation/Waste), Health was

the strongest predictor of attitudes, purchase intention and

purchase frequency (data not shown). This finding corre-

sponds with results from previous research. Grankvist and

Biel (2001) demonstrated that the belief that eco-labelled

foods are better than conventional foods for one’s own

health was positively correlated with purchase frequency for

eco-labelled milk, meat, potatoes and bread.

Environmental concern is another often stated motive for

purchasing organic foods. In the present study, EFB was a

predictor of attitudes, the importance of the criterion

‘organically produced’, purchase intention, and appeared

to be equally important as health for purchase frequency.

This latter finding may at least partly be a reflection of the

fact that both variables concern self-reported behaviour. Our

results are in accordance with findings from another recent

Swedish study (Grankvist & Biel, 2001) that found a

positive and moderately strong correlation between the

purchase frequency of eco-labelled foods and the perceived

importance of environmental consequences as a purchase

criterion.

In the present study, age was also a significant

contributor primarily to the prediction of attitudes, but

also to the prediction of the importance of the purchase

criterion ‘organically produced’, purchase intention, and

purchase frequency. These results suggest that younger

respondents are more likely to be positive towards organic

foods.

Respondents with a strong intention to purchase organic

foods gave higher likelihood and importance ratings for

perceived positive environmental, human health and animal

welfare consequences than did those with weaker intentions.

This indicates that those with strong intentions perceive

more benefits from purchasing organic foods than do those

with weak intentions.

Our finding that health concern is a better predictor

than concern for the environment indicate that egoistic

motives seem to be stronger than altruistic motives.

Uusitalo (1990) argues that activities in favour of

environmental protection generally increase if a person

can achieve some private side-benefits from the activity at

the same time that (s)he is contributing to the collective

goal. The importance of the egoistic (health) motives may

be due to their perceived individual benefit. Consumers

believing that organic foods are healthier than conven-

tional foods may perceive short-term health consequences

when consuming organic foods. On the contrary, the

effects of performing more EFBs are mostly of a long-

term character and the individual may not even benefit

from them during his/her lifetime. Further, egoistic

(health) benefits can be achieved by the individual him/

M.K. Magnusson et al. / Appetite 40 (2003) 109–117 115

herself, whereas the goal of improving the environment

presupposes a collective effort.

A majority of the respondents reported frequent

recycling of packages, newspapers/papers and batteries.

However, we have previously reported that only a

relatively small proportion (8 – 13%) reported purchasing

organic foods frequently (Magnusson et al., 2001). Thus,

consumers appear to be more willing to perform altruistic

behaviours in the area of recycling than in the area of food

choice. This finding is in accordance with results from

another Swedish survey (Grankvist, 2001). There are some

obvious differences between these categories of behaviour

that may explain the demonstrated discrepancy. Recycling

implies negligible economic and only small behavioural

costs and the alternative to recycling (leaving everything in

the garbage) may appear morally unacceptable. Also, in

Sweden, money is refunded at the return of glass/PET

bottles and aluminium cans. Organic foods may have

premium prices and may not be readily available, and

alternative behaviours represent established purchase and

consumption habits. We have previously demonstrated that

‘organically produced’ was stated to be the least important

criterion for the purchase of milk, meat, potatoes and

bread, and that the most important purchase criteria (‘good

taste’, ‘long shelf-life’, and ‘healthy’) do not match the

most common beliefs about organic foods (‘more expens-

ive’ and ‘healthy’) very well (Magnusson et al., 2001).

Thus, it is not likely that a majority of consumers are

willing to pay more for foods they do not think taste better.

Other factors that may help explain the discrepancy

between recycling and food choice are that eating implies

incorporation of material into our bodies and that eating is

strongly influenced by affective factors. Rozin (1990) has

pointed out that eating is a highly personal, potentially

threatening activity since it involves incorporation of

material from the outside world into our bodies. Birch

(1981) argues that eating behaviour is very resistant to

change because it is characterised by powerful affective,

apparently non-cognitive components that are not easily

influenced by information, not even information about

contingencies or consequences. Thus, establishment of

‘environmental concerns’ as a purchase motive in the food

sector is made difficult by the fact that food-related

behaviours differ in fundamental respects from e.g.

recycling behaviour.

The same methodology (length of questionnaire, remin-

ders, incentive) in other studies of random samples of the

Swedish population has resulted in a 12 – 15% higher

response rate (unpublished data). One reason for the lower

response rate in the present study may be that a large

proportion of people are not particularly interested in or

have limited knowledge of organic foods. In previous

studies, the themes have dealt with more familiar topics

(e.g. attitudes to milk).

In conclusion, perceived health benefits were demon-

strated to be more strongly related to attitudes and behaviour

towards organic foods than were perceived environmental

benefits. This supports the notion that egoistic motives are

better predictors of the purchase of foods than are altruistic

motives. However, the frequency of performance of EFBs

also contributed to the prediction of purchase of organic

foods. Thus, behaviour – behaviour correlations seem to be

stronger than ‘belief’-behaviour correlations in the context

of environmental concerns.

Acknowledgements

The authors are indebted to Dr Birgitta Johansson for her

very valuable statistical advice. This research was supported

by grants from the Swedish Foundation for Strategic

Environmental Research to the programme FOOD-21:

Sustainable Agriculture.

References

Antonides, G., & van Raaij, W. F. (1998). Consumer behaviour—a

European perspective. Chichester: Wiley.

Birch, L. L. (1981). A call for the explicit recognition of affect in models of

human eating behaviour. Journal of Nutrition Education, 13(Suppl.),

49 – 53.

Ebreo, A., Hershey, J., & Vining, J. (1999). Reducing solid waste. Linking

recycling to environmentally responsible consumerism. Environment

and Behavior, 31, 107 – 135.

Fagerli, R. A., & Wandel, M. (1999). Gender differences in opinions and

practises with regard to a ‘healthy diet’. Appetite, 32, 171 – 190.

Grankvist, G. (2001). Purchase criteria, beliefs and habits as

determinants of choice of eco-labeled products. Licentiate disser-

tation, Sweden: Department of Psychology, Göteborg University,

ISSN 1101-718X.

Grankvist, G., & Biel, A. (2001). The importance of beliefs and purchase

criteria in the choice of eco-labeled food products. Journal of

Environmental Psychology, 21, 405 – 410.

Jolly, D. A. (1991). Differences between buyers and nonbuyers of organic

produce and willingness to pay organic price premiums. Journal of

Agribusiness, 9, 97 – 111.

Land, B., 1998. Consumers’ dietary patterns and desires for change.

Working paper no 31/March 1998. Århus, Centre for Market

Surveillance, Research and Strategy for the Food Sector.

Magnusson, M. K., Arvola, A., Koivisto Hursti, U.-K., Åberg, L., & Sjödén,

P.-O. (2001). Attitudes towards organic foods among Swedish

consumers. British Food Journal, 103, 209 – 226.

Rozin, P. (1990). Development in the food domain. Developmental

Psychology, 26, 555 – 562.

Rozin, P., Fischler, C., Imada, S., Sarubin, A., & Wrzesniewski, A. (1999).

Attitudes to food and the role of food in life in the USA, Japan, Flemish

Belgium and France: possible implications for the diet-health debate.

Appetite, 33, 163 – 180.

Schifferstein, H. N. J., & Oude Ophuis, P. A. M. (1998). Health-related

determinants of organic food consumption in the Netherlands. Food

Quality and Preference, 9, 119 – 133.

Sema InfoData, 1998. Sema InfoData AB, Box 34101, SE-100 26

Stockholm, Sweden.

Statistics Sweden, 2001a. Available on http://www.scb.se/statistik/be0101/

be0101tab2helar01.asp.

Statistics Sweden, 2001b. Available on http://www.scb.se/statistik/uf0506/

uf0506tab1.xls.

M.K. Magnusson et al. / Appetite 40 (2003) 109–117116

The Hartman group (1997). Food and the environment: A consumer’s

perspective. Phase II (The Hartman Report), Bellevue. Torjusen, H., Lieblein, G., Wandel, M., & Francis, C. A. (2001). Food

system orientation and quality perception among consumers and

producers of organic food in Hedmark County, Norway. Food Quality

and Preference, 12, 207 – 216. Torjusen, H., Nyberg, A., & Wandel, M (1999). Økologisk produsert mat:

Forbrukernes vurderinger og bruksmønster (Organic food: Consumers’

perceptions and dietary choices). SIFO Report 5:1999. Lysaker,

Norway, Statens institutt for forbruksforskning. Tracy, A. P., & Oskamp, S. (1983). Relationships among ecologically

responsible behaviours. Journal of Environmental Systems, 13, 115 – 126.

Tregear, A., Dent, J. B., & McGregor, M. J. (1994). The demand for

organically grown produce. British Food Journal, 96, 21 – 25.

Uusitalo, L. (1990). Are environmental attitudes and behaviour inconsist-

ent? Findings from a Finnish study. Scandinavian Political Studies, 13,

211 – 226.

Wandel, M. (1994). Consumer concern and behaviour regarding food and

health in Norway. Journal of Consumer Studies and Home Economics,

203 – 215.

Wandel, M., & Bugge, A. (1997). Environmental concern in

consumer evalutation of food quality. Food Quality and Preference,

8, 19 – 26.

Wilkins, J. L., & Hillers, V. N. (1994). Influences of pesticide residue and

environmental concerns on organic food preference among food

cooperative members and non-members in Washington State. Journal

of Nutrition Education, 26, 26 – 33.

M.K. Magnusson et al. / Appetite 40 (2003) 109–117 117