Insects as food why the western attitude is important

Food Quality and Preference 44 (2015) 148–156

Contents lists available at ScienceDirect

Food Quality and Preference

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / f o o d q u a l

The psychology of eating insects: A cross-cultural comparison between Germany and China

http://dx.doi.org/10.1016/j.foodqual.2015.04.013 0950-3293/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Consumer Behavior, Department Health Science and Technology, Universitaetstrasse 22, CH-8092 Zurich, Switzerland.

E-mail address: Christina.Hartmann@hest.ethz.ch (C. Hartmann).

Christina Hartmann a,⇑, Jing Shi a, Alice Giusto b, Michael Siegrist a a Department Health Science and Technology, ETH Zurich, Switzerland b School of Journalism and Communication, Peking University, China

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

Article history: Received 9 February 2015 Received in revised form 17 April 2015 Accepted 24 April 2015 Available online 25 April 2015

Keywords: Food neophobia Attitudes China Germany Willingness to eat Insects

Based on their high nutritional value and low production costs, insects are an excellent and sustainable source of animal protein. In contrast to countries such as China, in Western societies, the consumption of insects is not rooted in traditional diet. Data for the present study was collected from adults in Germany (n = 502) and China (n = 443). A cross-cultural comparison was conducted based on consumers’ willing- ness to eat different insect-based, processed (e.g., cookies based on cricket flour) and unprocessed (e.g., crickets) food. The influence of food neophobia on consumers’ willingness to eat insects was examined. The Chinese rated all insect-based food more favourably with regard to taste, nutritional value, familiar- ity and social acceptance compared with the Germans. Also, they indicated greater willingness to eat the tested food products, and no differences were observed between their ratings of processed and unpro- cessed food. The Germans reported higher willingness to eat the processed insect-based foods compared to the unprocessed foods. Further results revealed that low scores for food neophobia, positive taste expectations, high scores for social acceptance and experiences with eating insects in the past were sig- nificant predictors of consumers’ willingness to eat insects in both countries. Consequently, the introduc- tion of insects as a food source in Western societies seems more likely to succeed if insects are incorporated into familiar food items, which will reduce neophobic reactions and negative attitudes towards insect-based foods.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Increases in the global food demand and the accompanying environmental burden necessitate the establishment of food pro- duction systems that are efficient and sustainable. The conven- tional production of animal protein in Western countries consumes significant resources and is ethically questionable. Animal protein is an important dietary component not only for Western societies but also for many demographic groups all over the world. Interest in alternative protein sources of high nutritional value, such as edible insects, has increased remarkably in recent years (Caparros Looy, Dunkel, & Wood, 2014; Megido et al., 2014; Van Huis et al., 2013; Vanhonacker, Van Loo, Gellynck, & Verbeke, 2013). In fact, in varying degrees according to the species, metamorphic stage and feeding, some insects contain high value protein and essential amino acids, have a high vitamin and mineral content, have low cholesterol concentrations compared to some

meat-based animal products and have favourable n�3/n�6 fatty acid and polyunsaturated fatty acid/saturated fatty acid ratios (Belluco et al., 2013; Verkerk, Tramper, van Trijp, & Martens, 2007). At the same time, production requirements for insect farm- ing are low, requiring little water and space, and the biomass con- version rate is better than that of most animals (Van Huis et al., 2013). The combination of high nutritional value, smaller environ- mental footprint and low production costs makes insects particu- larly interesting as ‘mini-livestock’ (De Foliart, 1995) suitable for human and animal nutrition all over the world (Van Huis et al., 2013).

Entomophagy, or the eating of insects, was and still is an impor- tant dietary behaviour in many parts of the world, including Africa, Latin America and Asia (Van Huis et al., 2013). In Western soci- eties’, however, the consumption of insects is not rooted in tradi- tional diet. Thus far, conscious consumption of edible insects is restricted to experimental restaurants as a delicacy (Verkerk et al., 2007); specialised food items based on insect protein (e.g., protein powder for muscle growth); or incorporation into familiar products as a fun ingredient (e.g., insect-embedded lollipops). Previous research on the acceptance of insects as food was

http://crossmark.crossref.org/dialog/?doi=10.1016/j.foodqual.2015.04.013&domain=pdf
http://dx.doi.org/10.1016/j.foodqual.2015.04.013
mailto:Christina.Hartmann@hest.ethz.ch
http://dx.doi.org/10.1016/j.foodqual.2015.04.013
http://www.sciencedirect.com/science/journal/09503293
http://www.elsevier.com/locate/foodqual
C. Hartmann et al. / Food Quality and Preference 44 (2015) 148–156 149

focussed on European consumers (Caparros Megido et al., 2014; Schösler, de Boer, & Boersema, 2012; Vanhonacker et al., 2013; Verbeke, 2015; Yen, 2009), and lacks cross-cultural studies (Shan et al., 2015). Therefore, a central focus in the present study is a comparison of two countries, i.e. Germany and China, with differ- ent cultural exposures to insects as food (Zhang, Tang, & Cheng, 2008), to explore consumers’ perceptions and cognitive associa- tions with insects. By contrasting the two, we can better under- stand the similarities and differences in evolutionary associations, individual experiences and cultural representations (Lockwood, 2013) of insect-based food, which is paramount to establish its widespread consumption.

1.1. Consumption of insects in China and Western countries

The tradition of eating insects in China dates back more than 3000 years and has outlasted various dynasties (Chen, Feng, & Chen, 2009; Zhi-Yi, 1997). Various insect species, based on local preferences, were collected, prepared and used as a nutritive source, medicine and delicacy in mainly rural parts of China (Harris, 1998; Zhi-Yi, 1997). In the present day, the knowledge and tradition of the consumption of insects has been lost; their reg- ular consumption is now restricted to few regions in China such as the minority area of the Yunnan province (Chen et al., 2009). The globalisation of food markets, improvements in food technology and the Western influence on China’s cuisine have resulted in changes in food customs and the traditional food system (Yen, 2009). Following improvements to the economic conditions in China, people can now afford more varieties of different foods, so their nutrition has undergone a transition, especially in cities were people are exposed to a more diversified food culture. Nowadays, the consumption of insects can no longer be considered a common food practice in China (Zhi-Yi, 1997). Nevertheless, the Chinese respondents are likely to be more familiar with the idea of consum- ing insects because it is part of their culinary tradition and thus manifested in their cultural consciousness.

In Western societies’ nutritional evolution, insects were rarely experienced as an edible food source, and they are compulsively rejected as non-food, unclean and a health risk associated with food contamination and filthiness (Kellert, 1993; Looy et al., 2014). These internalised defensive reactions towards insects are reflected in low willingness-to-eat ratings in previous studies. Studies in this area have focused on people’s willingness to substi- tute meat with insects (Schösler et al., 2012; Vanhonacker et al., 2013; Verbeke, 2015). In a Belgian study, more than 65% of meat consumers disagreed with the idea to substitute meat with insects (Verbeke, 2015). In another study about meat substitution options in the Netherlands, attractiveness ratings and the likelihood of actually preparing the meals were the lowest for dishes including locusts or fried mealworms (Schösler et al., 2012). In addition, insect protein received the lowest ratings compared to other sus- tainable food choices by Flemish consumers in terms of what they were willing to buy and pay for (Vanhonacker et al., 2013). The public aversion towards insects in Western societies makes the establishment of insects as a widespread food source, and meat substitute in particular, very challenging.

1.2. Factors influencing willingness to eat insects

A consumer’s acceptance or non-acceptance of a product is determined by various beliefs and perceived attributes of the pro- duct, which are considered in food decision-making (Tuorila, 1997), and thus are involved in the developmental process of pref- erences. Cultural influences on attitudes and beliefs as well as the meaning and social appeal of food also have a significant impact on consumption (Rozin, 1988). Culture, social norms and previous

experiences shape the boundaries of what is considered edible or inedible (Rozin & Fallon, 1980). Items that are considered inedible, such as insects, can elucidate a strongly affect-laden disgust response when people imagine ingesting them. This socio-cultur- ally transmitted defence mechanism (Rozin, 1996) aims to protect the body from contact with potentially noxious substances (Miller, 2004). The visual appearance of a food can also trigger a disgust- based food rejection. Rozin and Fallon (1987) postulated that prominent reminders of a food’s origin as an animal food (its ‘ani- malness’) are core elicitors of a disgust response. The vision of an entire cricket, with its extraordinarily big legs, in someone’s mouth, or a silkworm with its slimy surface, might elicit fear and disgust in consumers. Next to reminders of animalness, perceived mouth feeling and aversive textural properties are the most impor- tant characteristics that lead to food being rejected based on dis- gust (Martins & Pliner, 2006). Therefore, there is reason to believe that people would be more willing to eat products made with processed insects because the origin and disgusting attributes are less prominent.

Numerous other attributes can influence the acceptance of food, such as nutritional value, quality and beneficial health effects as well as expected taste and geographic origin (Barrena & Sánchez, 2013), but are under-researched constructs as antecedents of will- ingness to eat insects. In the present study, we focused on the fol- lowing four attributes of insects as food: nutritional value, taste, familiarity and social acceptance. Attitudes about the health value of substances can be a driver for consumption. However, health motivation and knowledge about the nutritional value of the pro- duct are important prerequisites. Expected or experienced sensory appeal is another dimension of acceptability (Pilgrim, 1957; Raats, Daillant-Spinnler, Deliza, & McFie, 1995). Of course, negative taste expectation is a strong incentive to avoid a substance. Furthermore, the cuisine within a cultural group is the result of adaptations to a particular environment. Again, cuisine determines cultural food rules and thus what can be served and is socially acceptable. Whether or not a food is socially valued and considered as culturally familiar is likely to have an impact on its acceptance as well. By focussing on these four dimensions, further insights can be gained into whether different cultural exposures to insects as food lead to different attitude ratings. Thus, this study will reveal distinguishing factors in the evaluation of insects as food as well as drivers of their appeal as food.

Previous experiences with a product have an impact on the individual’s attitudes towards a product, of course. Also, underly- ing psychological factors connect attitudes towards a product to consumer behaviour. Insects are a novel, unfamiliar food for European consumers, and unfamiliar food that deviates from cul- tural norms can evoke rejection and avoidance. This ‘innate’ response to the unfamiliar is a characteristic feature of eating organisms, and it was once an important survival mechanism to prevent the ingestion of potentially poisonous substances (i.e., the internal gatekeeper) (Pliner & Hobden, 1992; Rozin, 1990). Consumers can vary greatly in their attitudes towards novel foods (Dovey, Staples, Gibson, & Halford, 2008). Food neophobia as an individuals’ behavioural tendency naturally starts in early child- hood. Although it decreases with age through positive food experi- ences, it can be present in adults as well. Food neophobia is associated with decreased levels of willingness to eat a novel food (Tuorila, Lähteenmäki, Pohjalainen, & Lotti, 2001) and plays a cru- cial role in the establishment of insects as foods (Verbeke, 2015). Food rejection can be motivated by negative taste expectations (Fallon & Rozin, 1983), low levels of expected enjoyment (Raudenbush & Frank, 1999) and uncertainty about the origin of the product (Tuorila, Meiselman, Bell, Cardello, & Johnson, 1994). Unfamiliar food, especially of an animal nature, is likely to be rejected due to feelings of disgust, expected distaste and

150 C. Hartmann et al. / Food Quality and Preference 44 (2015) 148–156

dangerousness (Fallon & Rozin, 1983). Although the danger factor is considered to be less important in Western societies because food products on the market are predominantly safe (Pliner, Pelchat, & Grabski, 1993), uncertainty about the product and its attributes is still a driver for rejecting it (Tuorila et al., 1994). Reducing uncertainty about a product by giving information or by evoking associations with familiar products can enhance its acceptance (Tuorila et al., 1994). For example, a study with Belgian visitors of an insectarium showed that insects prepared with known flavours, such as paprika or chocolate, were preferred to baked or boiled mealworms or crickets (Caparros Megido Rudy et al., 2014). Though it should be noted that these results are based on people already interested in insects, adding familiar spices or flavour profiles to a new food seems to help to decrease the neophobic reaction and increase the willingness to taste (Stallberg-White & Pliner, 1999).

1.3. The present study

Given the information above, there is reason to believe that German and Chinese respondents differ in their attitudes towards insects as food. We hypothesised that the Chinese respondents would be more willing than the German respondents to consume insects and would hold more favourable attitudes towards insects as food. Instead of solely asking about peoples’ willingness to eat insects as a meat substitute (Vanhonacker et al., 2013; Verbeke, 2015), the focus was on various processed and unprocessed insect-based food items. We further expected that food neophobia would be an important predictor for willingness to eat insect- based products for the German respondents, but not for the Chinese. Such cross-cultural studies help to understand the con- cept of food neophobia in different countries and cultural regions. Moreover, investigating peoples’ opinions and attitudes is not only assumed to improve the accuracy of predictions regarding con- sumers’ willingness to eat novel foods but also may be the basis for effective marketing strategies in the establishment of food innovations.

1 Except for Qinghai, Hainan. 2 Except for Ningxia, Tibet. 3 In addition, we analysed the data utilising the full food neophobia scale; virtually e same results were observed. The full scale mean value in the German sample was Germany = 3.34 (SD = 0.77) and in the Chinese sample was MChina = 3.62 (SD = 0.71).

2. Methods

2.1. Data collection

Data collection in Germany and China occurred in October 2014. Internet panels from commercial providers of sampling ser- vices were used for recruiting the study participants (Germany: Respondi AG; China: InterfaceAsia Holden). Participants received a small financial reward for filling out the whole questionnaire. Excluded were respondents who did not complete the survey and whose total survey duration was less than half of the median of the total survey duration, which indicates that a respondent did not seriously answer the questions (nChina = 94; nGermany = 27). Quota samples were used with the quota variables gender and age. The two samples did not significantly differ in regard to these two vari- ables after exclusion of the participants as indicated above.

2.2. Participants

The German sample consists of 502 respondents (48% men) ranging in age between 20 and 69 (M = 44.3 years; SD = 14.2). Three categories of education level—low, middle and high—had the following distribution: 7.6% low (primary school, lower sec- ondary school), 60.4% middle (secondary school, vocational school, senior high school) and 31.9% high (vocational academy, college, undergraduate and above). Of the German sample, 67.5% lived in a city, 6.5% in a suburb and 25.9% in a rural area.

The Chinese sample consists of 443 respondents (49% men) ranging in age between 20 and 69 (M = 44.2; SD = 13.1). Education levels were distributed as follows: 2.9% low (elementary school, junior high school), 14.7% middle (secondary vocational school, senior high school) and 82.4% high (college, undergraduate and above). Slightly over half of the sample (53.7%) lived in Shanghai, Beijing (Peking), Guangdong or Shandong. The other half of the sample was distributed among the other provinces,1 auton- omous regions2 and direct-controlled municipalities. Of the Chinese sample, 94.1% lived in a city, 4.7% in a suburb and 1.1% in a rural area (see Table 1).

2.3. Questionnaire

For most of the questions used in the survey, English and German versions were available. The items were not available in Chinese, however. Therefore, the English versions were translated by the third co-author into Chinese. The second author translated the items from Chinese back into English; both are fluent in English and Chinese. The two co-authors worked independently from each other. The few differences between the back-translation and the original wording of the items were discussed by the authors of the paper until there was agreement that the Chinese version had the same meaning as the English version.

2.3.1. Food neophobia All items from the food neophobia scale (Pliner & Hobden,

1992) were translated into Chinese and German. The wording of some items had to be slightly changed in order to make them meaningful for the study participants. The term ‘ethnic food’ is not used in German or Chinese, for example. In the German ver- sion, the term was replaced by ‘food from other cultures’ and in the Chinese version by ‘non-Chinese food (e.g., Italian cuisine, French cuisine)’. The German version has been validated for the German language in a previous study (Siegrist, Hartmann, & Keller, 2013). As in the original, participants answered on a seven-point response scale ranging from �3 (‘do not agree at all’) to +3 (‘totally agree’). The extreme categories were verbally anchored, and the other categories were only numerically anchored. Various authors have found that the food neophobia scale, though it has respectable reliability coefficients, is not unidi- mensional (e.g., Choe & Cho, 2011; Siegrist et al., 2013; Tuorila et al., 2001). Only a subset of the items was found to form a one- dimensional scale in cross-national research (Ritchey, Frank, Hursti, & Tuorila, 2003). The inclusion of invalid items certainly creates the risk of invalid conclusions. The item analysis for the present study suggested that three of the original items (3, 8 and 9) should be excluded in order to have acceptable item-total corre- lations (>.30).3 The final seven items built a unidimensional scale and were averaged into one food neophobia score. Cronbach’s alpha for the German sample was a = .78 and for the Chinese sample was a = .73.

2.3.2. Willingness to eat insect-based products Based on the procedure by Verbeke (2015), participants were

informed prior to the questioning the following ideas about insects: they are a good source for high-value protein; their pro- duction requires little space; their feed conversion is efficient; and the eating of insects provides benefits in terms of sustainabil- ity. Respondents indicated their willingness to eat six different

th M

Table 1 Food neophobia, willingness to eat processed and unprocessed insects (scale) and percentage of respondents who previously consumed insects in the German (N = 502) and Chinese (N = 443) sample.

Range Germany China

M or % SD M or % SD

Age (years) 20–69 44.3 14.2 44.2 13.1 Gender (men) 48% 49% Education

Low 7.6% 2.9% Middle 60.4% 14.7% High 31.9% 82.4%

Living rural 25.9% 1.1% Food neophobia 1–7 3.0 1.0 3.2* 0.9 Willingness to eat

insect-based fooda 1–10 3.3 2.4 5.6** 2.5

Insects consumption previously (yes)

13.3% 67.7%

** p < .001. * p < .01. a Variable is based on averaged items related to willingness to eat insects as meat

substitute, deep-fried silkworm, deep-fried crickets, silkworm drink and cookies made with cricket flour.

C. Hartmann et al. / Food Quality and Preference 44 (2015) 148–156 151

insect-based food items. Respondents indicated (1) their willing- ness to adopt insects as a meat substitute (Verbeke, 2015) as well as their willingness to eat (2) deep-fried silkworms,4 (3) deep-fried crickets, (4) cookies based on cricket flour and (5) chocolate chip cookies based on cricket flour. The questions related to deep-fried silkworms, deep-fried crickets and chocolate chip cookies based on cricket flour were displayed with a picture of the corresponding food item. Then, respondents indicated (6) their readiness to consume a drink containing silkworm protein. Respondents were informed prior to this question that silkworm protein is a by-product of silk production and has various health benefits, such as helping to lower cholesterol and slow the aging of the brain (e.g., Limpeanchob et al., 2010; Sasaki, Yamada, & Kato, 2000). All six items were answered on a ten-point scale ranging from 1 (‘do not agree at all’) to 10 (‘totally agree’). The extreme categories were verbally anchored; the other categories were only numerically anchored. The five items related to insects as a meat substitute, silkworms, crickets, silkworm drinks and cookies made with cricket flour were averaged into the willing- ness-to-eat-insects scale (a = .94).

2.3.3. Previous consumption of insects Participants responded to the following statements: ‘I regularly

eat insects’ (yes/no)5 and ‘I have eaten insects in the past’ (yes/no).6

Participants who answered ‘yes’ to at least one of the two questions were coded as having been exposed to insects in the past regardless of the duration or intensity of the exposure.

2.3.4. Attitudes Respondents’ explicit attitudes towards different insect-based

food items (i.e., deep-fried crickets, deep-fried silkworms) were assessed. The four different attitudes focussed on were nutritional value, taste, social acceptance and familiarity. Ratings were given on a 10-point bipolar adjective scale: low nutritional value – high nutritional value, disgusting – tasty, exotic – familiar and primitive – civilised. The two extreme categories were verbally anchored.

Analyses were performed using the SPSS statistics software package version 22 (SPSS Inc., Chicago, IL). Results with p < .01 were considered statistically significant.

3. Results

3.1. Willingness to eat insects

A two-way mixed ANOVA was conducted with country (Germany, China) as the between-subjects factor and readiness to eat insects in the different food contexts (insects as a meat sub- stitute, deep-fried silkworms, deep-fried crickets, drinks contain- ing silkworm protein, cookies based on cricket flour and chocolate chip cookies based on cricket flour) as the within- subjects factor. Subsequently, simple effects were analysed to examine whether readiness to eat the different foods varied both within and between the countries (see Table 2).

The mixed ANOVA revealed a significant main effect of country (F[1, 943] = 172.67; p < .001), which shows that respondents in China and Germany differ in their willingness to eat the various insect products. There was also a significant main effect of food

4 There are many species of edible insects (e.g., ants, termites and wasps consumed in different developmental stages (e.g., larvae and eggs) (Chen et al. 2009; Zhi-Yi, 1997). In the present study, we focus on two edible insect species crickets and silkworms. The silkworm is one of the most domesticated insects cultivated for thousands of years in China and prepared in many ways (e.g., steamed boiled and deep-fried) (De Foliart, 1995). Crickets are also common edible insects (Chen et al., 2009). The most common preparation methods for insects include deep frying, boiling, stewing, braising and roasting (Chen et al., 2009).

5 Regular insect consumption: China: 12.9%, Germany: 0%. 6 Previous insect consumption: China: 54.8%, Germany: 13.3%.

7 Tested interaction effects of country vs. food neophobia and food neophobia vs previous insect consumption did not explain a significant amount of variance in the willingness to eat.

) , : , ,

product (F[5, 4715] = 158.48; p < .001), which indicates that the willingness to eat differs significantly between the different food products. The interaction of country and food type was also signif- icant (F[5, 4715] = 57.94; p < .001), which indicates that the will- ingness to eat the different foods varied significantly in Germany and China. The most interesting results, however, are based on the simple effects analysis. The simple effects of food type within the level of country were significant (FGermany[5, 4715] = 102.84; p < .001 and FChina[5, 4715] = 112.95; p < .001), showing that par- ticipants’ ratings of their willingness to eat varied by food type within both Germany and China. The mean values and 95% CI for willingness to eat insects in the different food contexts indicate that, within Germany, the willingness to eat was significantly higher for processed food items (i.e., drinks and cookies) than for unprocessed food items (i.e., crickets and silkworms). In contrast, the willingness-to-eat ratings in China did not differ significantly between processed and unprocessed food items, with the excep- tion of silkworm protein, which was rated exceptionally high. The lowest rating occurred for crickets. In addition, the simple effects of country within the level of foods revealed that the will- ingness to eat significantly differed between countries for all foods (Table 2), with the differences being smaller for processed food items than for unprocessed food items.

3.2. The influence of socio-demographic and psychological variables

The impact of gender, age, education, country and food neopho- bia on the willingness to eat insects-based food was examined by multiple linear regression analyses (Table 3). Since a high percent- age of respondents in the Chinese sample indicated that they had already consumed insects in the past, previous insect consumption was included as control variable. Model 1 was significant and explained 33.4% of the variance in the willingness to eat insects- based food (F[6, 944] = 78.23; p < .001). Including food neophobia into the model helped to explain an additional 3.4% of the variance (F[7, 944] = 79.56; p < .001).7 The Chinese respondents indicated a higher willingness to eat insects-based food than the German respondents. Even though insect consumption is much more cultur- ally grounded in China than in Germany, the analysis revealed that

.

Table 2 Means values and 95% confidence intervals for willingness to eat the different insect-based foods. Simple effects indicate significant differences in willingness to eat between countries.

Willingness to eat Germany (N = 502) China (N = 443) Simple effects

M SD 95% CI M SD 95% CI F(df1, df2)

Insects as meat substitute 3.04 2.51 [2.82, 3.26] 5.51 2.77 [5.25, 5.77] F(1, 943) = 206.87*

Silkworm (deep-fried) 2.80 2.49 [2.58, 3.02] 5.69 3.20 [5.39, 5.99] F(1, 943) = 243.27*

Crickets (deep-fried) 2.78 2.53 [2.56, 3.01] 4.72 2.98 [4.45, 5.00] F(1, 943) = 117.19*

Silkworm drink 4.10 2.86 [3.85, 4.35] 6.96 2.45 [6.73, 7.19] F(1, 943) = 269.55*

Cricket cookies 3.93 3.03 [3.66, 4.20] 5.25 2.90 [4.98, 5.52] F(1, 943) = 46.66*

Cricket cookies choco chip 4.20 3.13 [3.92, 4.47] 5.45 2.94 [5.18, 5.73] F(1, 943) = 40.06*

Respondents indicated their willingness to eat on a ten-point scale ranging from 1 (‘do not agree at all’) to 10 (‘totally agree’). * p < .001.

Table 3 Linear hierarchical regression with willingness to eat insect-based food (scale) as dependent variable.

Variables Model 1 Model 2

B SE B b B SE B b

Constant 2.99* 0.42 4.89* 0.48 Gender (1 = men) 0.19 0.14 0.04 0.35 0.14 0.07 Age (years) �0.01 0.01 �0.04 �0.01 0.01 �0.04 Country (1 = China) 0.79* 0.19 0.15 1.16* 0.19 0.21 Education (low vs. high) 0.46 0.34 0.09 0.11 0.33 0.02 Education (low vs. middle) 0.18 0.33 0.03 0.05 0.32 0.01 Insects consumption

previously (1 = yes) 2.49* 0.18 0.45 2.19* 0.18 0.40

Food neophobia �0.59* 0.08 �0.21

R2 = .37 (N = 945, p < .001). * p < .001.

Table 4 Differences in attitudesa towards unprocessed insects in Germany and China.

Germany (N = 502) China (N = 443)

M SD 95% CI M SD 95% CI

Silkworm (deep-fried) Low – high nutritional

value 6.93 2.47 [6.71,

7.15] 8.31* 1.82 [8.14,

8.48] Disgusting – tasty 3.86 2.47 [3.64,

4.07] 6.31* 2.96 [6.04,

6.59] Exotic – familiar 2.66 2.18 [2.66,

2.85] 5.98* 2.96 [5.71,

6.26] Primitive – civilized 4.53 2.16 [4.34,

4.72] 5.70* 2.74 [5.44,

5.95]

Crickets (deep-fried) Low – high nutritional

value 6.54 2.63 [6.31,

6.77] 7.12* 2.24 [6.91,

7.33] Disgusting – tasty 3.70 2.56 [3.48,

3.93] 5.36* 2.90 [5.09,

5.63] Exotic – familiar 2.84 2.28 [2.64,

3.04] 5.09* 2.78 [4.83,

5.35] Primitive – civilized 4.40 2.28 [4.20,

4.60] 5.05* 2.66 [4.80,

5.30]

* p < .001 (independent t-test). a Ratings were given on a 10-point bipolar adjective scale: low nutritional value–

high nutritional value, disgusting–tasty, exotic–familiar, primitive–civilized. The two extreme categories were verbally anchored.

152 C. Hartmann et al. / Food Quality and Preference 44 (2015) 148–156

food neophobia plays a significant role in people’s willingness to eat in both countries. People in both countries who scored higher on the food neophobia scale were less likely to eat insect-based food. No significant effect was observed for gender, age and education.

Attitudes related to the different unprocessed insect-based foods (i.e., deep-fried crickets and deep-fried silkworms) were tested for significant differences between countries (independent t-test, Table 4). Significant differences were observed for nutri- tional value, taste, familiarity and social acceptance for both food items, with more positive attitudes reported by the Chinese. In order to examine whether the four assessed attitudes towards deep-fried crickets and deep-fried silkworms are related to a par- ticipant’s willingness to eat them, a hierarchical linear regression was conducted (Table 5). The dependent variable was the willing- ness to eat deep-fried crickets or deep-fried silkworms. Gender, country, previous insect consumption and food neophobia were entered into the model in the first step and the four attitudes in the second step.