Acceptance of Functional Foods: A Comparison of French, American, and French Canadian Consumers

Acceptance of Functional Foods: A Comparisonof French, American, and French Canadian

Consumers

JoAnne Labrecque,1 Maurice Doyon,2 Francois Bellavance3

and Jane Kolodinsky4

1Associate Professor, Department of Marketing, HEC Montreal, 3000, chemin de laCote-Sainte-Catherine, Montreal, Quebec, Canada H3T 2A7 (phone: 514-340-6800;

fax: 514-340-5631; e-mail: [email protected]).2Invited Professor, UMR GAEL INRA, Universite Pierre Mendes France, France and

Associate Professor, Department of Agricultural Economics and Consumer Science, LavalUniversity, Canada.

3Associate Professor, Department of Management Sciences, HEC Montreal, Canada.4Professor, Department of Community, Development and Applied Economics,

University of Vermont, Vermont.

Food products have diversified with industry globalization. To market functional foods efficiently, foodmanagers must gauge cross-cultural variance of functional food acceptance. Expanding on previousresearch, we investigate young consumers’ acceptance of functional foods. Data collected in FrenchCanada, United States, and France in 2004 reveal that business students are slightly in favor offunctional foods, and associate health benefits with these foods but very few product-related benefits.Students do not have strong opinions on the trustworthiness of information and expressed a slightinterest in purchasing this type of product. Analyses of cultural differences revealed significant, albeitsmall, divergence in these variables. Statistical analysis performed on the full sample assessed theimpact of food attitudes and other cognitive and attitudinal factors on the general attitude towardfunctional foods. Health and product-related benefits and belief about the credibility of informationare the main positive determinants of the acceptance of functional foods, followed by high knowledge.Apart from the negative impact of Neophobia, none of the other food attitudes influences attitudestoward functional foods. Linear regressions performed on each subgroup indicated similar positivecross-cultural results for health and product-related benefits. However, cross-cultural differences aredetected for knowledge, credibility of information, and food attitudes that influence acceptance offunctional foods.

Les produits alimentaires se sont diversifies avec l’avenement de la mondialisation. Pour commer-cialiser efficacement les aliments fonctionnels, les gestionnaires du secteur alimentaire doivent evaluerl’acceptation des aliments fonctionnels par les diverses cultures. A partir de travaux de rechercheanterieurs, nous avons examine l’acceptation des aliments fonctionnels par les jeunes consommateurs.Des donnees recueillies en 2004 au Canada francais, aux Etats-Unis et en France ont revele que lesetudiants en commerce etaient legerement en faveur des aliments fonctionnels, leur associaient desavantages pour la sante, mais tres peu d’avantages lies aux produits. Les etudiants n’avaient pasune tres bonne opinion concernant la credibilite de l’information et ont manifeste un faible interetpour l’achat de ces produits. Des analyses des differences culturelles ont revele des divergences,petites mais significatives. Les analyses statistiques effectuees sur l’echantillon total ont evaluel’impact des attitudes envers les aliments ainsi que d’autres facteurs cognitifs et attitudinaux sur

Canadian Journal of Agricultural Economics 54 (2006) 647–661

647

648 CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

l’attitude generale envers les aliments fonctionnels. Les avantages pour la sante, les avantages liesaux produits et les croyances concernant la credibilite de l’information sont les principaux facteursfavorables a l’acceptation des aliments fonctionnels, suivis d’un degre de connaissances eleve. Outrel’impact negatif de la neophobie, aucune autre des attitudes envers les aliments n’influe sur les attitudesenvers les aliments fonctionnels. Les regressions lineaires effectuees pour chaque sous-groupe ontindique des resultats positifs similaires. Toutefois, nous avons note des differences interculturellesquant a la connaissance, a la credibilite de l’information et aux attitudes envers les aliments quiinfluencent l’acceptation des aliments fonctionnels.

INTRODUCTION

In most industrialized countries, changes in demographic patterns—especially the in-creasing proportion of women in the labor force over the last 30 years—combined withthe increasing use of more sophisticated technology, have profoundly modified the fooduniverse. These phenomena have prompted researchers to examine how consumers haveadapted to this new environment. Given that food decisions are complicated for con-sumers, research has focused on many factors that influence food choice, ranging fromthe attributes of food itself to attitudes, motives, and intentions, and the influence of theenvironment on decision making. Drawing on this sizable body of literature, the presentstudy adopts a cross-cultural approach to exploring how attitudes influence the accep-tance of functional foods among French, American, and French Canadian young adultconsumers.

Researchers have developed measures of food-related attitudes in order to under-stand better how health-related and non-health-related factors influence dietary choices.Steptoe et al (1995) introduced a multidimensional measure of motives related to foodchoice, including nine factors labeled Health, Mood, Convenience, Sensory appeal, Naturalcontent, Price, Weight control, Familiarity, and Ethical concern. Testing the associationsbetween demographics and their measure of motives, these researchers found differencesin motives for food choice associated with sex, age, and income. Roininen et al (1999)subsequently designed a Health and Tastes Attitudes Questionnaire from which theyidentified three health-related factors—General health interest, Light product interest, andNatural product interest—and three taste-related factors—Craving for sweet foods, Usingfood as a reward, and Pleasure. They also found attitudinal differences based on age andgender, with females being more interested in the health and taste aspects of foods thanmales, and younger respondents being less concerned with health but more interested intaste than older respondents.

Other researchers (Fischler 1990; Rozin et al 1999; Poulain 2002) documented foodfunctions in people’s minds and lives, and how consumers have modified their food habits.In his seminal work L’Homnivore, Fischler (1990) discusses how changes in lifestyles andfood technology have contributed to a de-structuration of French eating habits, inducing anincrease in snacking, among other effects. Poulain (2002) investigated contemporary foodhabits by exploring different aspects of the organization of daily food intake, and describedhow social change and an abundant food supply have impacted French consumers’ foodhabits and generated a shift toward “grazing,” characterized by a combination of solitaryfood intake and structured meals taken in a social context. Rozin et al (1999) studiedhow consumers’ beliefs about different food-related aspects vary between countries. They

ACCEPTANCE OF FUNCTIONAL FOODS 649

found that Americans associate food the most with health and the least with pleasure, andthe French are the most food-pleasure-oriented and the least food-health-related group.

Acceptance and perceived healthiness of functional foods and the determinants ofpurchase intentions of food categories ranging from organic to genetically modified foods,including functional foods, have been studied extensively (Bredahl 2001; Makatouni 2002;Bech-Larsen and Grunert 2003; Cox et al 2004; Urula and Lahteenmaki 2004; Verbeke2005). According to Verbeke (2005), belief in the health benefits of functional foods is themain determinant of acceptance, followed by the presence of an ill family member, butdecreases disproportionately with claimed awareness of the concept. This result refutedthe IFIC (1999) findings identifying knowledge as one of the major motivations to eitherpurchase or consume functional foods (in Verbeke 2005). Moreover, these effects werefound to outweigh socio-demographic variables as potential determinants of functionalfoods acceptance. Cox et al (2004), referring to Protection Motivation Theory, exam-ined middle-aged consumers’ intentions to consume different prototypes of functionalfoods that could improve memory. They demonstrated that perceived “efficacy” and self-efficacy, related to the impact of the consumption of functional foods on preventingmemory loss, are the most important determinants of intentions to consume functionalfoods. Perceptions of functional foods have been compared across cultures. Bech-Larsenand Grunert (2003) studied Danish, Finnish, and American consumers’ attitudes to-ward functional foods and their perception of the healthiness of these foods. Notably,the Finnish had a more positive attitude and acceptance of the healthiness of functionalenrichments than the American consumers. The Danish respondents exhibited the mostnegative attitude.

Nonetheless, few studies have explored cross-cultural variance of acceptance of func-tional foods within different segments of the population and how food attitudes and othercognitive and attitudinal factors condition acceptance of functional foods. Given foodproduct diversification and industry globalization, food managers need more informa-tion on differences in food attitudes in order to target the right type of products tothe right type of consumers in each country. This study expands on previous researchand investigates young consumers’ food attitudes. Two aims motivate this study: (1) toidentify whether acceptance of functional foods varies across cultures; and (2) to verifywhether food attitudes and other cognitive and attitudinal factors influence acceptanceof functional foods.

METHOD

To attain the previously stated objectives, an exploratory study was conducted oncollege/university students from three countries—France, United States, and Canada(French Quebecers).1 The choice of countries is motivated by the fact that North Americaand Europe are the second and third most important markets for the relatively new func-tional foods (Australian Government 2004), along with the accessibility of researchers toconsumers in the selected countries and the expected important contrast between Frenchand American food consumption and food attitudes; French-speaking Quebec consumersare believed to be situated at the halfway point. Although our sample is not representativeof the population of functional food consumers, college/university students represent aninteresting group of consumers because they were identified, together with women and

650 CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

consumers aged 55 years and older, as the segments most likely to adopt functional foodsin their diet (IFIC 2000). The exploratory nature of this study does not require that thesample be representative of the population.

A self-administered questionnaire was distributed to college/university students. Atotal of 611 students enrolled in a business program in Canada (Quebec n = 280), France(n = 170), and the United States (n = 161) completed the questionnaire in 2004. To ensureas homogeneous a sample as possible across countries, questionnaires of students underage 18 years or over age 25 years were discarded. In the final sample, which consists of 545fully completed surveys, French Canadian students predominate, with 227 respondents,while the American and French respondent populations are of similar size, 155 and 163.The average age of the sample is 21.9 years (Canada/Quebec: 22.8; United States: 21.7; andFrance: 20.8) and the proportion of male and female respondents varies somewhat acrosscountries, with the largest proportion of male respondents in the United States (57%) andthe smallest in Canada (42%). The characteristics of the Quebec education system explainwhy French Canadian students were slightly older than the sample average.2 Descriptivestatistics of the sample are presented in Table 1.

In addition to the demographic variables, the questionnaire included a set of gen-eral food-related attitude statements corresponding to scale measures taken from theliterature. Three scales measure eating-related attitudes—Pleasure (Roininen et al 1999)and Cooking enjoyment and Convenience (Kolodinsky and Labrecque 1996). Two scalesmeasure health-related attitudes—Health Consciousness (Kraft and Goodell 1993) andDiet–health link (Rozin et al 1999). Two scales measure novelty-seeking attitudes—FoodNeophobia3 (Pliner and Hobden 1992) and Innovativeness (Goldsmith and Hofacker1991). All items related to these scales were measured on a 7-point or a 10-point Lik-ert scale, with the categories ranging from “strongly disagree” to “strongly agree.” Foreach scale, factor analyses (Oblimin rotation) using the full sample were conducted totest the unidimensionality of the scales. Items with factor loadings below 0.5 were ex-cluded from the analyses. Scale reliability was measured using Cronbach’s Alpha. Highinternal consistency was observed within most food attitude scales, apart from those oneating-related and novelty-seeking-related attitudes, for which Cronbach’s Alphas (α) wereslightly lower than 0.70. Final scores corresponded to the mean of the items defining thescales.

Following this procedure, the original sets of items for the Pleasure, Health Con-sciousness, and Innovativeness scales were reduced from 6 to 3 items. The 10 items fromthe original Food Neophobia scale loaded on two factors, each composed of 3 items and

Table 1. Sample descriptive statistics

Age Sex

Country Number of cases Mean (SD) Male (%) Female (%)

Canada (Quebec) 227 22.9 (1.3) 42 58United States 155 21.7 (1.1) 57 43France 163 20.8 (1.6) 52 48Total 544 21.9 (1.6) 49 51

ACCEPTANCE OF FUNCTIONAL FOODS 651

identified for the purpose of this research as Neophobia and Ethnic Food Enjoyment. Theother items of the original Neophobia scale were deleted because their loading on bothfactors was under 0.5. Table 2 describes these results.

Another section included questions evaluating the respondents’ knowledge of func-tional foods, their overall attitude, their beliefs about the benefits associated with thisproduct category, and their perception of the credibility of the information concerningthese food products. Respondents were first asked to specify their degree of knowledgeof the term functional foods on a 4-point categorical variable with the question: Haveyou heard the term functional foods? (yes, occasionally, very little, never). Responseswere sorted into three categories such that the first corresponded to participants whoreported having a good knowledge of this term, the second corresponded to those havinga partial knowledge of the term, and the third comprised respondents who did not knowthe term functional food at all. This measure of knowledge is somewhat different fromthat used by Verbeke (2005), who evaluated respondents’ knowledge of functional foodsbased on three items measured on a 7-point scale, the sum of which was recoded intothree categories—low, medium, and high knowledge.4

Before answering the other questions related to functional foods, students wereprovided with the following definition of functional food: Functional food products are anew category of product with an added value, created to meet the expectations of consumerswho are more health conscious than ever. Within the food category, functional food productsare designed to offer nutritional elements that promote better health, in addition to thenutritional elements they naturally contain.5

Students’ general attitude and credibility of information were measured by the scalesused by Kozup et al (2003) to evaluate overall attitude toward a food and credibility of nu-tritional information provided. Beliefs concerning health-related benefits were measuredon a 5-point bipolar scale designed by Kolodinsky et al (2003). Product-related benefitsreferred to Bredahl’s (2001) corresponding items, and are associated with improvementof standard of living and quality. Purchase intention was measured with respect to threespecific functional products: eggs with Omega-3, milk with calcium, and orange juicewith calcium. Factor analyses (Oblimin rotation) were also conducted to investigate theunidimensionality of the scales. Table 3 reports the results.

In the last section of the questionnaire, respondents were asked to evaluate theirfrequency of consumption of different products that included milk with Omega-3 andeggs with Omega-3 on a 7-point scale, which ranged from “never” to “many times aweek.” They also indicated the health value that they associated with these products.

Questionnaire TranslationSince most scales were taken from articles written in English, the questionnaire was firstwritten in English, then translated into French and back-translated by a professionaltranslator to ascertain the equivalent meaning in both languages.

RESULTS

Attitudes Toward Functional FoodsBefore discussing how beliefs and attitudes toward functional foods vary across countries,levels of knowledge of functional foods are reported across countries. Results on the

652 CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

Table 2. Description of the food attitude scales

Factor Item Factor loading

Eating-relatedCooking enjoyment(Cronbach’s α = 0.87)

On a scale of 1–10, where 1 means “not close atall” and 10 “very close, ” indicate how closeeach statement matches your point of view

I enjoy cooking 0.897I enjoy cooking when I invite people to my

home0.896

I enjoy trying new recipes 0.872I enjoy looking at cookbooks and food 0.733

Pleasure (Cronbach’s α =0.61)

When I eat, I concentrate on enjoying the tasteof the food

0.661

I finish my meal even when I do not like thetaste of the food

0.784

An essential part of my weekend is eatingdelicious food

0.805

Convenience (Cronbach’sα = 0.64)

On a scale of 1–7, rate the importance of theseaspects when you purchase food in general

Preparation time 0.788Convenience 0.896Usage 0.837

Health-relatedHealth consciousness(Cronbach’s α = 0.79)

Please rate those statements on a 7-point scale

I read more health-related articles than I didthree years ago

0.856

I am interested in information about my health 0.900I am concerned about my health all the time 0.771

Diet-health link(Cronbach’s α = 0.78)

On a scale of 1–7, how much of an effect doyou believe diet has on the following?

Heart disease 0.828Obesity 0.818Good health in general 0.778Cancer 0.736

Novelty-seeking-relatedInnovativeness(Cronbach’s α = 0.60)

Please rate those statements on a 7-point scale

If I heard that a new food product wasavailable through a local store, I would beinterested enough to buy it

0.773

I would consider buying a new food product,even if I hadn’t heard of it

0.759

I know more about new foods than otherpeople do

0.693

(Continued)

ACCEPTANCE OF FUNCTIONAL FOODS 653

Table 2. Continued

Factor Item Factor loading

Neophobia (Cronbach’s α

= 0.67)I don’t trust new food 0.704

Ethnic food looks too weird to eat 0.810I am afraid to eat things I have never tried

before0.804

Ethnic food enjoyment(Cronbach’s α = 0.81)

I like foods from different countries 0.898

At dinner parties, I will try new food 0.790I am afraid to eat things I have never tried

before0.851

degree of knowledge of the term functional foods (Table 4) show that higher proportionsof American students (56.9%) and French Canadian students (45.8%) had heard aboutfunctional foods before the study when compared with the French students (10.6%). Themajority of the French students (63.8%) reported not having heard of the term beforethe study. The proportion of American students who have heard about functional foodsis smaller (57%) than that observed in the American population by the IFIC in its 2000survey, which found that 82% of Americans can identify a functional food but that 72%had no concerns about these products. This variation might be explained by importantdifferences in the samples.

Analyses of variance and Scheffe’s tests were carried out on the attitudinal and beliefscores associated with functional foods and on the credibility of information about thisproduct category. As shown in Table 5, overall, respondents were slightly in favor of func-tional foods (5.05 on a 7-point scale), associated some health benefits with these foods(6.52 on a 10-point scale), but very few product-related benefits (4.21 on a 7-point scale).In addition, students did not have strong opinions on whether they could trust the infor-mation (4.12 on a 7-point scale) and expressed a slightly positive interest in purchasingthis type of product (7.55 on a 10-point scale). In terms of cultural differences, analy-ses revealed significant, albeit small, differences in the scores related to these variables.French Canadian students expressed a more favorable attitude (5.27 on a 7-point scale),associated higher health benefits (6.75 on a 10-point scale), reported having more trust inthe information (4.36 on a 7-point scale), and showed a stronger purchase intention (7.87on a 10-point scale) than French students. Between American students and the otherstudents, no significant differences were detected apart from higher purchase intention(7.65 on a 10-point scale) when compared with the French students (7.01 on a 10-pointscale). Thus, their means are more similar to those of the French Canadian studentsthan those of the French students. Overall, French students represent the subgroup thatassociates the fewest benefits with functional foods and that is least in favor of this typeof products.

To further investigate the acceptance of functional foods among young consumers,students were asked to report their actual frequency of consumption of two functionalfoods (milk with Omega-3 and eggs with Omega-3), together with the health value they

654 CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

Table 3. Description of attitude, belief, and credibility measures

Variable Item Factor loading

General attitude functionalfoods (7-point scale,Cronbach’s α = 0.95)

Based on what you have heard aboutfunctional foods and on the definition given,what is your overall attitude toward thisproduct category?

Favorable-unfavorable 0.960Good-bad 0.964Positive-negative 0.948

Health benefits (10-pointscale, Cronbach’s α =0.87)

We want to know your opinions concerningfunctional foods, compared with traditionalfoods. For each pair of statements, pleaseindicate on a scale of 1 to 10 how close eachstatement matches your point of view. Wouldyou say that functional foods are

Healthier–less healthy 0.851Beneficial for health–harmful to health 0.912Good for the immune system–harmful to the

immune system0.866

More nutritious–less nutritious 0.715Lengthens people’s lifespan–shortens people’s

lifespan0.723

Product-related benefits(7-point scale, Cronbach’sα = 0.86)

Please rate those statements on a 7-point scaleFunctional food products will improve the

standard of living of future generations0.908

Functional food products will increase my ownand my family’s standard of living

0.923

Functional food products are better qualityfoodstuffs than other food products

0.818

Credibility of info (7-pointscale, Cronbach’s α =0.90)

In general, how do you perceive informationabout functional foods?

Undependable–dependable 0.915Untrustworthy–trustworthy 0.951Dishonest–honest 0.874

Purchase intention(10-point scale,Cronbach’s α = 0.82)

Would you like to try such products? 0.862Would you buy such products if you happened

to see them in a store?0.933

Would you actively seek out such products in astore in order to purchase them?

0.781

associated with these two products. This information is then used to validate previousattitudes reported in Table 5. Apart from product-related benefits, the results reportedin parentheses in Table 5 were almost the same for the smaller sample of users and thecomplete sample (users and non-users of functional foods). This fact seems to validatethe quality of the attitudes reported.

ACCEPTANCE OF FUNCTIONAL FOODS 655

Table 4. Knowledge of functional foods

Proportions N Yes Occasionally or very little Never

Total∗ 538 38.5% 30.3% 31.2%Canada 225 45.8% 29.8% 24.4%United States 153 56.9% 35.9% 7.2%France 160 10.6% 25.6% 63.8%

∗Chi-square (d.f. = 4) = 137.4; p < 0.001 (seven observations were missing).

The sample mean on frequency of consumption shows that students who consumemilk with Omega-3 or eggs with Omega-3 do so about once a month. Analyses of varianceon these variables reveal cultural differences. Consistent with the results on purchaseintention, American and French Canadian students reported knowingly eating at leastone of the two functional products more often (about once a month) than the Frenchstudents (less than once a month). These results suggest that functional foods are notpart of young consumers’ regular diet, especially for French students. However, withinthe segment of students who reported eating these two products once every other weekor more, significant proportions of respondents reported higher rates of consumption.In fact, 42% of Canadian students and 51% of American students reported eating thesefunctional food products at least once every two weeks. This proportion was only 7% inthe French group. Moreover, consistent with our results related to beliefs about healthbenefits, French Canadian (5.28) students attributed a higher health value to milk andeggs with Omega-3 than French (4.55) and American (4.73) students did.

Correlations between the food attitude scales and the variables related to functionalfoods were computed to investigate the associations between these variables. Table 6reveals significant, albeit weak, associations between most of the food attitude scales.The strongest statistically significant correlations were positive, observed between thefollowing variables: Attitude, Health benefits, Product benefits, Credibility of informa-tion, and Purchase intention. Accordingly, food attitudes and other cognitive and atti-tudinal factors appear to influence acceptance of functional foods, given that purchaseintention is positively correlated with variables such as attitudes toward functional foods,health benefits, product benefits, and credibility of information.

To further investigate the determinants of the general attitudes toward functionalfoods and to expand on Verbeke’s study of the role of cognitive, attitudinal, and socio-demographic variables in the acceptance of functional foods, multiple linear regressionwas used to assess the impact of beliefs about health and product-related benefits,knowledge, credibility of information, food attitudes, and gender (independent variables)on the general attitude toward functional foods (dependent variable; see definition inTable 3). The results for the full sample, presented in the first column of Table 7, showthat health- and product-related benefits, high knowledge, and the belief about the cred-ibility of information have a positive effect on the general attitude toward functionalfoods, while Neophobia has a negative effect. None of the other food attitudes signifi-cantly influences the attitude toward functional foods when the full sample is examined.The variance explained (R2) by these variables is 40.6%. Similar to Verbeke (2005), we

656 CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

Tab

le5.

Sum

mar

yof

resu

lts

rela

ted

tofu

ncti

onal

food

s

1–7

1–7

1–10

Pro

duct

1–7

1–10

1–7

1–7

Gen

eral

Hea

lth

-rel

ated

Cre

dibi

lity

Pur

chas

eF

requ

ency

ofH

ealt

hat

titu

debe

nefi

tsbe

nefi

tsIn

foin

tent

ion

cons

umpt

ion

valu

e(p

=0.

003)

(p=

0.00

5)(p

=0.

001)

(p=

0.00

0)(p

=0.

000)

N(p

=0.

00)

(p=

0.00

0)

1.C

anad

a5.

273

6.75

34.

413

4.36

37.

873

159

4.35

35.

282,

3

2.U

nite

dSt

ates

5.06

6.55

4.22

4.08

7.65

311

44.

433

4.73

1

3.F

ranc

e4.

741

6.19

13.

941

3.84

17.

011,

215

92.

671,

24.

551

Tota

l5.

05(5

.10)

6.52

(6.5

6)4.

21(3

.77)

4.12

(4.1

6)7.

55(7

.58)

432

3.75

4.87

Supe

rscr

ipts

indi

cate

coun

trie

sw

hose

mea

nsc

ores

diff

erat

p=

0.01

(Can

ada

=1,

Uni

ted

Stat

es=

2,an

dF

ranc

e=

3).

ACCEPTANCE OF FUNCTIONAL FOODS 657

Tab

le6.

Cor

rela

tion

sbe

twee

nfo

odat

titu

des

and

atti

tude

sto

war

dfu

ncti

onal

food

sA

ttit

ude

Die

t–E

thni

cfu

ncti

onal

Hea

lth

Pro

d.C

red.

Pur

ch.

Coo

k.he

alth

Hea

lth

food

food

sbe

n.be

n.in

foin

tent

ion

enj.

Ple

as.

Con

v.lin

kco

nsc

enjo

y.In

nov

Att

itud

efu

ncti

onal

1fo

ods

Hea

lth

ben.

0.51

0∗∗1

Pro

duct

ben.

0.49

8∗∗0.

386∗∗

1C

redi

bilit

yof

info

.0.

420∗∗

0.40

0∗∗0.

339∗∗

1P

urch

ase

inte

ntio

n0.

450∗∗

0.35

9∗∗0.

283∗∗

0.31

3∗∗1

Coo

king

enjo

ymen

t−0

.014

−0.0

180.

085

0.00

4−0

.025

1P

leas

ure

−0.0

03−0

.037

0.09

5∗0.

065

0.00

20.

292∗∗

1C

onve

nien

ce0.

039

0.09

0∗0.

042

0.09

4∗0.

059

−0.0

72−0

.030

1D

iet–

heal

thlin

k0.

077

0.04

20.

117∗∗

0.03

90.

141∗∗

0.15

7∗∗0.

121∗∗

0.01

41

Hea

lth

cons

ciou

s0.

062

0.07

50.

082

0.02

20.

171∗∗

0.26

4∗∗0.

151∗∗

0.05

00.

275∗∗

1E

thni

cfo

od0.

071

0.06

00.

094∗

0.07

50.

166∗∗

0.24

3∗∗0.

156∗∗

0.12

3∗∗0.

084

0.34

5∗∗1

enjo

ymen

tIn

nova

tive

0.05

20.

075

0.08

00.

157∗∗

0.12

8∗∗0.

309∗∗

0.27

8∗∗0.

090∗

0.11

6∗∗0.

310∗∗

0.32

1∗∗1

Neo

phob

ia−0

.129

∗∗−0

.058

0.00

0−0

.071

−0.0

81−0

.125

∗∗0.

028

0.10

1∗−0

.007

0.04

4−0

.165

∗∗−0

.103

∗

∗∗C

orre

lati

onsi

gnif

ican

tat

0.01

.∗ C

orre

lati

onsi

gnif

ican

tat

0.05

.

658 CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

Table 7. Determinants of the general attitude toward functional foods

Full sample French Canadians Americans French(N = 493) (N = 218) (N = 118) (N = 157)

βa p βa p βa p βa p

Health benefits 0.317 <0.001 0.368 <0.001 0.358 <0.001 0.215 0.002Product benefits 0.308 <0.001 0.278 <0.001 0.314 <0.001 0.375 <0.001Knowledge-high 0.111 0.010 0.130 0.077 0.199 0.185 0.068 0.342Knowledge-medium 0.056 0.177 0.047 0.485 0.250 0.089 −0.019 0.782Credibility of info 0.196 <0.001 0.207 0.001 0.147 0.120 0.158 0.027Cooking enjoyment −0.066 0.094 −0.119 0.045 0.153 0.100 −0.138 0.060Pleasure −0.028 0.465 0.005 0.927 0.004 0.964 −0.044 0.546Health consciousness 0.050 0.211 −0.017 0.781 0.135 0.123 0.049 0.485Innovativeness −0.039 0.333 −0.038 0.555 −0.139 0.121 0.022 0.752Neophobia −0.134 <0.001 −0.126 0.065 −0.042 0.615 −0.218 0.011Ethnic food enjoyment −0.017 0.681 0.089 0.222 −0.154 0.094 −0.017 0.826Diet-health link 0.023 0.533 −0.012 0.827 0.061 0.451 0.012 0.859Convenience −0.008 0.829 −0.049 0.347 0.078 0.369 −0.025 0.706Sex −0.009 0.806 −0.059 0.273 0.117 0.172 −0.025 0.709Adjusted R2 0.406 0.445 0.381 0.355Maximum variance 1.52 2.11 4.21 1.71

inflation factorTest of heteroskedasticityb p = 0.6317 p = 0.7363 p = 0.4776 p = 0.8348

aStandardized estimate of β.bWhite’s test (1980).

did not find a significant linear relation between health-related attitudes and acceptanceof functional foods nor between gender and acceptance of functional foods. However,whereas Verbeke (2005) reported a negative association between high knowledge andacceptance of functional foods, our results show a positive association, in accordancewith the IFIC (1999) findings. In addition to knowledge, credibility of information hasa positive effect on acceptance of functional foods, at a magnitude higher than that ofknowledge.

Furthermore, linear regressions were performed separately for each country (lastthree columns of Table 7). The results were similar to those of the model for the full sample,except for high knowledge where the regression coefficients were no longer statisticallysignificant at the 5% level for all three countries and for Neophobia for the models withFrench Canadians and Americans. This is mainly due to lower sample sizes in the separatemodels. However, a more important difference between the model for the full sample andthe three separate models was found for the independent variable “Cooking enjoyment.”While the impact of Cooking enjoyment on the general attitude toward functional foodsis negative and statistically significant for French Canadians at the 5% level and at the10% level for French students, the impact is positive and statistically significant at the10% level for the Americans.

To test whether the determinants of the acceptance of functional foods vary sig-nificantly across cultures, we included in the full model the main effect of country and

ACCEPTANCE OF FUNCTIONAL FOODS 659

the interaction terms of country with all independent variables (the results of this modelare not shown). The only statistically significant interaction was between country andCooking enjoyment (p = 0.0244). The lack of other significant interactions with countryimplies that the relationship between acceptance of functional foods and the other vari-ables is similar across countries. A contrast analysis for the interaction between countriesand Cooking enjoyment gave the following results: French Canadian versus French stu-dents, p = 0.672; French Canadian versus Americans, p = 0.016; French students versusAmericans, p = 0.010.

Multicollinearity and heteroskedasticity tests were also performed. Results of thesetests show that multicollinearity was not a serious problem; all values for variance inflationfactors in the regression models were much lower than 10 (Neter et al 1996, p. 387), whileheteroskedasticity tests (White 1980) were non-significant (Table 7).

SUMMARY AND CONCLUSION

One objective of the present research was to identify whether acceptance of functionalfoods varies across cultures. First, results on the level of knowledge of functional foodsreveal the lower level of knowledge of the term functional foods among French studentscompared with the other students surveyed. Regarding acceptance of functional foods,overall, students slightly favored functional foods, associated some health benefits withthese foods, did not have strong opinions on whether they could trust the information,and expressed only a slight positive interest in purchasing this type of product. Resultson actual behavior supported these findings.

Cultural differences, albeit small, also emerged in the attitudinal and beliefs scoresassociated with functional foods and regarding the credibility of information. The Frenchstudents reported less favorable attitudes toward functional foods and indicated that theytrust less the information on functional foods than the French Canadian students do. Theminor differences in the attitudinal and beliefs scores associated with functional foodsacross cultures suggest that the market development of this product category could beapproached in this perspective as a global market rather than a local market. The adagethink global, act local apparently applies to this market.

To complete this investigation and meet our final objective, general attitude towardfunctional foods was regressed on food attitudes and other variables related to func-tional foods. It is well recognized in the marketing literature that the attitude toward afood product constitutes an antecedent of its purchase intention by consumers, and hasmanagerial implications. Regression results for the full sample showed that health- andproduct-related benefits, credibility of information, and high knowledge have a positiveeffect on the students’ attitude toward functional foods, while Neophobia has a negativeeffect. Accordingly, while food attitudes make a very small contribution to the variabilityof the general attitude toward functional foods, beliefs about benefits, be they health- orproduct-related, together with credibility of information, are the main contributors to thevariability of acceptance of functional foods.

In contrast to Verbeke’s findings, knowledge was found to have a positive impacton acceptance of functional foods. Striking findings from this research are the positiveimpact of product-related benefits and credibility of information on the general attitudetoward functional foods. Other researchers have underscored the positive impact of health

660 CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS

benefits on acceptance of functional foods. Thus, our results suggest that the promotionof other product benefits associated with functional foods, such as high quality, couldalso help optimize market development. Nevertheless, this type of promotion shouldbe undertaken prudently, given the positive impact of credibility of information, which,rather than the quantity of information, seems to have the strongest influence on thegeneral attitude toward functional foods.

In its qualitative study, which explored how to construct effective health messagesthat are understandable, credible, and scientifically valid, the IFIC (2000) concluded thatthere is no magic bullet to develop such messages. However, it provided guidelines thatinclude considering the level of awareness and concern consumers have about specifichealth issues. The type of information, its consistency, and the credibility of the source ofthe information have more influence on consumers’ beliefs about the benefits of functionalfoods than the quantity of the information, and could have a significant impact on themarket development of this product category. Much remains to be understood abouthow consumers react to information on functional foods, and more research shouldbe conducted to understand better how different segments of consumers with differentcharacteristics react to specific information on functional foods, such as food claims.

Combined with the fact that young consumers did not express strong confidence inthe available information on functional foods, these results suggest that to address theyoung consumer market, not only would various stakeholders need to develop productsappealing to this target segment, but they would also benefit from harmonizing theircommunication standards to avoid sending mixed messages.

NOTES1In Quebec, the survey was conducted at Laval University, University of Quebec at Trois-Rivieres,and HEC Montreal. In the United States, it was conducted at the University of Vermont and inFrance at ESSEC Paris. In all cases, subjects were recruited in business-related classes.2In Quebec, as opposed to France and the United States, after high school students have to go toCEGEP for a period of two to three years before being admitted to a university program.3Neophobia, in this context, correspond to the fear of new things or experiences.4The scale was recoded as follows: low (1 and 2), medium (3, 4, and 5), high (6 and 7).5The definition was attributed to a group of University specialists in the questionnaire.

REFERENCES

Australian Government. 2004. Australian Functional Food: A Healthy Choice for Functional FoodInvestment. Invest Australia.Bech-Larsen, T. and K. Grunert. 2003. The perceived healthiness of functional foods: A conjointstudy of Danish, Finnish and American consumers’ perception of functional foods. Appetite 40:9–14.Bredahl, L. 2001. Determinants of consumer attitudes and purchase intentions with regard togenetically modified foods—Results of a Cross-National Survey. Journal of Consumer Policy 24:23–61.Cox, D. N., A. Koster and C. G. Russell. 2004. Predicting intentions to consume functional foods andsupplements to offset memory loss using an adaptation of protection motivation theory. Appetite43: 55–64.Fischler, C. 1990. L’Homnivore. Paris: Editions Odile Jacob.

ACCEPTANCE OF FUNCTIONAL FOODS 661

Goldsmith, R. E. and C. F. Hofacker. 1991. Measuring consumer innovativeness. Journal of theAcademy of Marketing Science 19 (3): 209–22.IFIC. 1999. Functional Foods: Attitudinal Research (1996–1999). IFIC, International Food Infor-mation Council Foundation, Washington.IFIC. 2000. Functional Foods: Attitudinal Research, August: Quantitative and Qualitative Summary.IFIC, International Food Information Council Foundation, Washington.Kolodinsky, J. and J. Labrecque. 1996. The allocation of time to grocery shopping: A compar-ison of Canadian and U.S. households. Journal of Economic and Family Issues 17(3/4): 393–408.Kolodinsky, J., T. DeSisto and J. Labrecque. 2003. Understanding the factors related to concernsover genetically engineered food products: Are national differences real? International Journal ofConsumer Studies 27 (4): 266–76.Kozup, J. C., E. Creyer and S. Burton. 2003. Making healthful choices: The influence of health claimsand nutrition information on consumers’ evaluations of packaged food products and restaurantmenu items. Journal of Marketing 67: 19–34.Kraft, F. B. and P. W. Goodell. 1993. Identifying the health conscious consumer. Journal of HealthCare Marketing 13: 18–25.Makatouni, A. 2002. What motivates consumers to buy organic food in the UK? Results from aqualitative study. British Food Journal 104: 345–52.Neter, J., M. H. Kutner, C. J. Nachtsheim and W. Wasserman. 1996. Applied Linear StatisticalModels. 4th ed. Irwin, California: UCLA Academic Technology Services.Pliner, P. and K. Hobden. 1992. Development of a scale to measure the trait of food neophobia inhumans. Appetite 19: 105–20.Poulain, J. P. 2002. The contemporary diet in France: De-structuration or from commensalisms tovagabond feeding. Appetite 39: 43–55.Roininen, K., L. Lahteenmaki and H. Tuorila. 1999. Quantification of consumer attitudes to healthand hedonic characteristics of foods. Appetite 33: 71–88.Rozin, P., C. Fischler, S. Imada, A. Sarubin and A. Wrzesniewski. 1999. Attitudes to food and therole of food in life in the U.S.A, Japan, Flemish Belgium and France: Possible implications for thediet–health debate. Appetite 33: 163–80.Steptoe, A., T. M. Pollard and J. Wardle. 1995. Development of a measure of the motives underlyingthe selection of food: The food choice questionnaire. Appetite 25: 267–84.Urula, N. and L. Lahteenmaki. 2004. Attitudes behind consumers’ willingness to use functionalfoods. Food Quality and Preference 15: 793–803.Verbeke, W. 2005. Consumer acceptance of functional foods: Socio-demographic. Cognitive andattitudinal determinants. Food Quality and Preference 16: 45–57.White, H. 1980. A heteroskedasticity-consistent covariance matrix estimator and a direct test forheteroskedasticity. Econometrics 48: 817–38.