- Email: [email protected]
Consumer liking for sausages affected by sensory quality and information on fat content
Appetite,
1992, 19, 285-292
Consumer Liking for Sausages Affected and Information on Fat Content RAGNHILD
by Sensory Quality
SOLHEIM
MATFORSK-Norwegian
Food Research Institute
The sensory quality of sausages containing 12 and 20% fat (the usual level) was assessed by descriptive profiling in a trained sensory panel. The effects of information on fat level and sensory differences on liking or dislike by consumers were assessed in 347 eaters of sausages recruited in food stores. The knowledge of a 40% reduction in fat content did not affect the hedonic responses to sausages with different sensory quality, whereas the sensory differences had major effects on liking scores. Fat-reduced sausages were liked as much as standard sausages when their texture and saltiness were similar to the more familiar quality. When sensory quality was similar, false information that the fat content of the 20%-fat sausages was 12% increased the liking for them although correct information on fat content of the sensorily similar 12%-fat sausages decreased liking. Thus, to avoid reducing the acceptability of reduced-fat sausages, it seems necessary to have a profile similar to the sausages containing usual levels of fat.
INTRODUCTION Both marketers of food products and health information ogranizations are interested in how consumer response is affected by the sensory quality of a product and information on its nutritional quality. To contribute to a further reduction of the fat content in the Norwegian diet, 1l-12% fat sausages have been introduced as a supplement to the regular l&20% fat products. For comparison, sausages on the U.S. market have a fat content in the range 20-30%. Sausages containing 12% and 20% fat were chosen for a study of the effect on consumer response of information on the fat content of sausages. When there are changes in the nutritional status of a product, it is essential to find whether consumers’ knowledge of this is enough to change acceptability, and if changes in sensory attributes are important. In past studies, consumer response has been measured separately to the sensory quality of products or to given information. In this study, consumer response was measured to both these factors together. Consumers hedonic responses to the two types of sausages were tested both with and without provision of correct and incorrect information on the fat content. To aid interpretation of differences in hedonic response to the sausages, sensory quality was analysed descriptively.
The author is grateful to E. Risvik and T. Naes for methodological discussions, and to the staff at the sensory laboratory at MATFORSK for performing the sensory profilings and the consumer tests. Address correspondence to: R. Solheim, MATFORSK, Osloveien 1, N-1430 As, Norway. 0195-6663/92/060285+08
%08.00/O
0
1992 Academic
Press Limited
286
R. SOLHEIM METHODS
Materials
Two variants of a commercially made type of sausage (12 and 20% fat) were each produced by two different companies. This type of sausage is mostly used for dinner. The two companies used the same recipe for the low-fat sausages (weight % of minced meat batter): 45% beef, 10% pork fat, 5% potato flour, 3% milk powder, 35.6% water/ice and 1.4% sodium chloride; 0.2 g of spices and 0*04g of ascorbic acid were added per kg of minced meat batter. The spices were a mixture of ginger, pepper and nutmeg; each Company used their own mixture of spices. The production procedures varied between the companies. The 20%-fat sausages were produced from the Company’s standard recipe, using the same raw materials as in the low-fat sausages. To obtain the same intensity of saltiness in the low-fat and high-fat product, the amount of sodium chloride was 0.2% less in the minced meat batter for the low-fat sausages. The chemical composition of the sausages is given in Table 1. The sausages were stuffed in natural casings, 30mm in diameter, then cooked, vacuum packed, and kept at 4°C until the sensory profiling 24 h later. Before being tested, the sausages were heated in a water bath to core temperature of 72°C. Samples were served in 2-cm thick slices. Descriptive Projiling by Trained Panel
The four sausage products were analysed by conventional profiling using 12 trained assessors. The assessors developed a vocabulary by describing differences between extreme samples and agreed on a consensus list of attributes for profiling. Before profiling, the assessors were trained in the use of the rating anchors and the sensory attributes (Table 2) with extreme samples. This was done to ensure panel calibration in the use of the scale. To avoid possible confusion in the definition of attributes, pre-testing and discussion after the assessments were performed during the training period. The colour attributes were evaluated on fresh cross-sections of the sausages and defined following the NCS-system (SSI, 1979). The evaluation took place in a laboratory designed for sensory analysis. Each assessor evaluated the sample at individual speed on a computerized system for direct recording of data (SENSTEC, Tecator AB, Sweden). Responses were made by touching a point on an
TABLE
Chemical
composition
(%
1
by weight) of two sausages produced commercial companies
by different
Contents
Sausage nominal % fat
Water
Fat
Protein
Carbohydrate
Sodium chloride
A
12 20
67.1 60.3
12.4 18.9
11.9 12.3
5.7 5.0
1.7 1.9
B
12 20
67.6 61.7
12.5 20.6
11.7 10.1
5.4 5.6
1.4 1.8
Company
CONSUMER LIKING FOR FAT IN SAUSAGES
287
TABLET Dejinition of sensory attributes
used in profiling of sausages
Attribute
Low intensity
High intensity
Whiteness Colour Colour strength Smell of smoke Taste of smoke Saltiness Spiciness Meat flavour Hardness
black sample yellow/red sample little colour no smell of smoke no taste of smoke no salty taste no taste of any spice no taste of beef/pork no force necessary to attain deformation dry feeling when chewing no covering of oil or fat in mouth when chewing not granular testure
white sample red/blue sample much colour strong smell of smoke strong taste of smoke strong salty taste strong taste of any spice strong taste of beef and/or pork much force necessary to attain deformation juicy when chewing mouth-feel of oil or fat when chewing very granular texture
Juiciness Greasiness Graininess
unstructured line, anchored on the left end with “low intensity” and on the right end with “high intensity”. The descriptor was displayed above the response line. The computer transformed the responses into numbers from 1.0 (low intensity) to 9.0 (high intensity).
Consumer Hedonic
Tests
The acceptance tests were performed in four centrally located food stores, believed to cater to similar socio-economic strata. The assessors were customers aged 18 years and older who were regular eaters of sausages. Consumers in two of the stores tasted sausages from Company A and consumers in the two other stores tasted sausages from Company B. The companies were the main suppliers of sausages to the two stores where their sausages were tested. The sausage samples were coded with 3-digit random numbers when served. The first set of samples was served as a blind test. The second set was served with the fat content of the sausages written on the attached forms. Half of the consumers in each store were given correct information on the fat content of the samples and the other half the false information that the low-fat sample was the high-fat sample and vice versa. Each consumer tasted the samples twice and marked their acceptance on a 15cm continuous line anchored at the left end with “dislike very much” and at the right end with “like very much”. Afterwards, their marks were linearly transformed into numbers keyed into the SENSTEC system, with “dislike very much” scored as 1.O and “like very much” scores at 9.0. Each consumer took 5- 10 min to perform the evaluations. Data Analysis The significance levels of differences in sensory analysis of variance (ANOVA). A cross-classification and the first order interactions included was applied.
attributes were calculated by model with the main effects The assessors were treated as
R. SOLHEIM
random effect. The hedonic data were compared by t-tests and ANOVA, analysing separately the responses to sausages produced by the two companies (Table 3).
RESULTS
Sensory Profiles
For all sensory attributes except saltiness and juiciness and graininess for Company A’s sausages, differences between the sausages were functions of fat content (Table 4). The low-fat sausages had significantly higher intensities of colour, colour strength, smell of smoke, taste of smoke, meat flavour and hardness. The sausages containing the standard level of fat had significantly higher intensities of whiteness, spiciness and greasiness. The sausages produced by Company B differed also in juiciness and graininess between fat content levels. The explanation of the differences between the companies may be variety of quality of raw material and processing procedures. Sausages produced by Company B had significantly higher colour, colour strength, spiciness and hardness scores than Company A’s sausages, while the sausages produced by Company A had significantly higher whiteness, juiciness and graininess. Consumer
Likings
Provision of false or correct fat level affected the hedonic responses to the sausages produced by Company A, but did not affect the hedonic responses to the sausages produced by Company B (Figure 1; Table 5). The two types of sausages produced by Company A were quite similar in sensory profile. These sausages were not significantly different in hedonic response in the blind test. Correct information about the fat level of the 20% fat sausages did not affect the hedonic response (Table 6: comparison D). However, the false information that fat level of the 20%-fat sausages from this Company was 12% raised the liking ratings (Table 6: comparison E). In contrast,
TABLE 3 Experimental design of consumer liking for sausages affected by sensory quality and nutritional information. The design was usedfor testing 12%-fat and 20%-fat sausages produced by different companies. Similar letters indicate paired comparisons Consumer number
Blind test 20% fat sausages
12% fat sausages
B
1 2
12% fat sausages Correct False D
given
20% fat sausages Correct False D
B
A
A n
Test with information
C
E
C
E
CONSUMER LIKING FOR FAT IN SAUSAGES
Sensory projles
289
TABLE 4 of the four sausages, showing main effects of fat level and producer Company in ANOVA Mean attribute score Company A
Attributes
Whiteness Colour Colour strength Smell of smoke Taste of smoke Saltiness Spiciness Meat flavour Hardness Juiciness Greasiness Graininess
Company B
12% fat
20% fat
12% fat
20% fat
Fat content difference
Company difference
5.5
6.1 3.3 4.1 4.8 5.1 5.9 6.0 4.4 2.4 6.6 6.5 5.5
4.5 6.7 6.1 6.5 5.9 5.2 6.0 5.3 6.8 4.7 4.9 3.6
5.5 4.3 5.1 5.4 5.7 6.0 7.0 4.2 3.8 6.2 6.6 4.1
o&o7 < 0~00005 < 0~00005 0.002 0.01 0.1 0.05 0.016 < 0~00005 0.16 0.001 0.1
0/003 0.0003 0.0006 0.1 1.0 0.06 0.025 0.69 < 0~00005 0.000 1 0.77 0.004
6.2 5.6 6.5 6.6 6.1 5.5 5.3 4.9 6.8 5.1 5.1
The means are calculated for 12judges in 3 replicates. The standard error of means were in the range 0.13-0.35. correct information about the fat level of this Company’s 12%-fat sausages decreased the mean hedonic response, while the false information that their fat level was 20% had no effect on the hedonic responses (Table 6; Figure 1). From the sausages produced by Company B, the samples containing 20% fat got the higher mean hedonic score in the blind test (Table 6: comparison A). This difference did not change when the consumers were informed about the fat content. It should be remembered that the low-fat sausages and the standard sausages produced by Company B were not similar in their sensory profiles. A few consumers reacted negatively to tasting four samples in rapid succession. Others commented on the information, that they felt it must be false. This occurred when we gave them false information on the fat contents of the sausages from Company B. For the samples from Company A, no one questioned the information given.
DISCUSSION The experimental design was chosen to emphasize effects of sensory quality of a product and/or nutritional information on consumer responses. The higher hedonic score for 20% than 12% fat sausages from Company B may be explained by the difference in texture that arose in that Company’s product from the different fat levels. For meat and meat products, as for other foods, texture attributes such as tenderness and juiciness contribute to evaluations of freshness and nutritional quality, according to Szczesniak (1990) and Drewnowski (1987). In the present tests, the juicier a sausage was, the greater was the liking score. This was
290
R. SOLHEIM
Company
A
I
2Or
40
15
s 30 PC ; c 20 z 2 IO
c OD $8 IO c ‘; 2 5 ”
0
25 20 3 I& E a -
Correct
IO 5
information
Company
False
information
B
2571
Blind test
I5
0
0 Blind test
20r-
15,
Correct
informatlon
False
Information
FIGURE 1. Distribution of frequencies of hedonic responses to sausages, separately for blind tests and for tests when information was given. The frequency counts are plotted for equal ranges of 0.9 of a point with 1.0 as lowest and 9.0 as highest score.
291
CONSUMER LIKING FOR FAT IN SAUSAGES
TABLE 5 Effects of provision of information (correct or false) on hedonic response to sausages produced by different companies with 12% and 20% fat content Mean liking rating
ANOVA
Information given (correct and false)
Mean square
f
P
Company A (n = 174) 20% fat 5.80 12% fat 5.84
6.21 6.09
32.67 19.94
7.31 4.50
0.007 0.035
Company B (n = 169) 20% fat 6-14 12% fat 5.60
5.81 5.52
13.33 5.43
2.94 1.20
0.088 0.275
Sausage assessed
Blind test
independent of information given on fat content. Consumers might want to be consistent in their assessments. This could have affected the scores given to the second pair of samples, for which correct or false information given, when the samples were not similar in their sensory profiles, and accounts for some consumers questioning the false information.
TABLE 6 Hedonic response to sausages with different level offat content, measured in blind tests and with either correct or false information on the fat content given. Results from t-tests on paired comparisons. (letters in brackets corresponding to the letters in Table 3).
Information given
Comparison
n
Mean difference in liking rating
t
P
Company A’s sausages Blind 20% minus 12% Blind 20% minus informed 20%
None correct False
176 87 89
- 0.045 -0.138 -0641
- 0.22 -0.55 -2.41
0.83 0.58 0.02
Blind 12% minus informed 12%
Correct False
87 89
0649 0.185
-2.61 0.68
0.01 0.50
None Correct False
171 98 73
0.532 0.312 0.135
2.31 1.46 064
0.02 0.15 0.53
correct False
98 73
- 0.072 0.133
-0.27 0.59
0.79 0.56
Comany B’s sausages Blind 20% minus 12% Blind 20% minus informed 20%
g:
Blind 12% minus informed 12%
0
(4
@)
292
R. SOLHEIM
When, however, a texture attribute such as juiciness is similar in sausages of different composition, as for those from Company A, factors other than sensory quality might be able to affect hedonic responses. The liking scores for the 20%-fat sausages from Company A increased when the false information was given that they were reduced in fat content. These results indicate that at least some consumers react positively to sausages with reduced fat content. Lundgren (1981) found that identical sausages served with different labels (low-fat and high-fat) were perceived as nondifferent for one-third and as different for two-thirds of a group of 72 consumers. Nutritional information has been found to increase consumers’ hedonic response to some products (Byrd-Bredbenner et al., 1988; Cheng, 1990), but not all (Eiser et al., 1984), and has not been shown generally to affect consumers’ buying habits. Russo et al. (1986) found that lists of vitamins and minerals increased nutritional knowledge, but it had no influence on actual purchase. A list of products with added sugar (focused on negative nutrients) increased the market share of low-sugar breakfast cereals at the expense of high-sugar brands. They concluded the display of “negative” nutrients such as calories, cholesterol, sodium, sugar, and various chemical additives to be of importance. Thus, in the cases of a negatively evaluated nutrient such as fat, one may hypothesize that information on fat reduction should increase a product’s buying potential. However, it is not known whether information on reduced fat in sausages will affect consumers’ choice between fat-reduced and today’s standard sausages. From the results of this study, the sensory quality of the sausages seems to be more important to the hedonic response than does nutritional information. In conclusion, it seems to be important that sensory quality of fat-reduced sausages is quite similar to today’s standard sausages in order to be accepted by the consumers. If this is achieved, it may be possible to motivate some consumers to buy such a product by giving information about the fat reduction. REFERENCES
Byrd-Bredbenner, C., Martin, R., Lewis, C. J. & Shannon, B. (1988) The effectiveness of two methods of delivering nutrition information to the general public. Journal of Nutrition Education, 20, 63-69. Cheng, H. W., Clark, A. D. & Heymann, H. (1990) Influence of selected marketing factors on consumer response to restructured beef steaks: a conjoint analysis. Journal of Sensory Studies, 4, 165-178. Drewnowksi, A. (1987) Fats and food acceptance: sensory, hedonic and attitudinal aspects. In J. Solms, D. A. Booth, R. M. Pangborn & 0. Raunhardt (Eds.): Food Acceptance and Nutrition, New York: Academic Press. Eiser, J. R., Eiser, C., Patterson, D. J. & Harding, C. M. (1984) Effects of information about specific nutrient content on ratings of “goodness” and “pleasantness” of common foods. Appetite, 5, 349-359. Lundgren, B. (1981) Effect of nutritional information on consumer responses. In J. Solms and R. L. Hall (Eds.), Criteria of Food Acceptance, Zurich: Forster Verlag AG. Russo, J. E., Staelin, R., Nolan, C. A., Russell, G. J. & Metcalf, B. L. (1986) Nutrition Information in the supermarket. Journal of Consumer Research, 13, 48-70. SSI “Color Atlas” (1979) Swedish Standard SS 01 91 02. Stockholm: Swedish Standards Institution. Szczesniak, A. S. (1990) Texture: is it still an overlooked food attribute? Food Technology, September, 86-95. Received
1 October
1991, revision 20 February 1992
1992, 19, 285-292
Consumer Liking for Sausages Affected and Information on Fat Content RAGNHILD
by Sensory Quality
SOLHEIM
MATFORSK-Norwegian
Food Research Institute
The sensory quality of sausages containing 12 and 20% fat (the usual level) was assessed by descriptive profiling in a trained sensory panel. The effects of information on fat level and sensory differences on liking or dislike by consumers were assessed in 347 eaters of sausages recruited in food stores. The knowledge of a 40% reduction in fat content did not affect the hedonic responses to sausages with different sensory quality, whereas the sensory differences had major effects on liking scores. Fat-reduced sausages were liked as much as standard sausages when their texture and saltiness were similar to the more familiar quality. When sensory quality was similar, false information that the fat content of the 20%-fat sausages was 12% increased the liking for them although correct information on fat content of the sensorily similar 12%-fat sausages decreased liking. Thus, to avoid reducing the acceptability of reduced-fat sausages, it seems necessary to have a profile similar to the sausages containing usual levels of fat.
INTRODUCTION Both marketers of food products and health information ogranizations are interested in how consumer response is affected by the sensory quality of a product and information on its nutritional quality. To contribute to a further reduction of the fat content in the Norwegian diet, 1l-12% fat sausages have been introduced as a supplement to the regular l&20% fat products. For comparison, sausages on the U.S. market have a fat content in the range 20-30%. Sausages containing 12% and 20% fat were chosen for a study of the effect on consumer response of information on the fat content of sausages. When there are changes in the nutritional status of a product, it is essential to find whether consumers’ knowledge of this is enough to change acceptability, and if changes in sensory attributes are important. In past studies, consumer response has been measured separately to the sensory quality of products or to given information. In this study, consumer response was measured to both these factors together. Consumers hedonic responses to the two types of sausages were tested both with and without provision of correct and incorrect information on the fat content. To aid interpretation of differences in hedonic response to the sausages, sensory quality was analysed descriptively.
The author is grateful to E. Risvik and T. Naes for methodological discussions, and to the staff at the sensory laboratory at MATFORSK for performing the sensory profilings and the consumer tests. Address correspondence to: R. Solheim, MATFORSK, Osloveien 1, N-1430 As, Norway. 0195-6663/92/060285+08
%08.00/O
0
1992 Academic
Press Limited
286
R. SOLHEIM METHODS
Materials
Two variants of a commercially made type of sausage (12 and 20% fat) were each produced by two different companies. This type of sausage is mostly used for dinner. The two companies used the same recipe for the low-fat sausages (weight % of minced meat batter): 45% beef, 10% pork fat, 5% potato flour, 3% milk powder, 35.6% water/ice and 1.4% sodium chloride; 0.2 g of spices and 0*04g of ascorbic acid were added per kg of minced meat batter. The spices were a mixture of ginger, pepper and nutmeg; each Company used their own mixture of spices. The production procedures varied between the companies. The 20%-fat sausages were produced from the Company’s standard recipe, using the same raw materials as in the low-fat sausages. To obtain the same intensity of saltiness in the low-fat and high-fat product, the amount of sodium chloride was 0.2% less in the minced meat batter for the low-fat sausages. The chemical composition of the sausages is given in Table 1. The sausages were stuffed in natural casings, 30mm in diameter, then cooked, vacuum packed, and kept at 4°C until the sensory profiling 24 h later. Before being tested, the sausages were heated in a water bath to core temperature of 72°C. Samples were served in 2-cm thick slices. Descriptive Projiling by Trained Panel
The four sausage products were analysed by conventional profiling using 12 trained assessors. The assessors developed a vocabulary by describing differences between extreme samples and agreed on a consensus list of attributes for profiling. Before profiling, the assessors were trained in the use of the rating anchors and the sensory attributes (Table 2) with extreme samples. This was done to ensure panel calibration in the use of the scale. To avoid possible confusion in the definition of attributes, pre-testing and discussion after the assessments were performed during the training period. The colour attributes were evaluated on fresh cross-sections of the sausages and defined following the NCS-system (SSI, 1979). The evaluation took place in a laboratory designed for sensory analysis. Each assessor evaluated the sample at individual speed on a computerized system for direct recording of data (SENSTEC, Tecator AB, Sweden). Responses were made by touching a point on an
TABLE
Chemical
composition
(%
1
by weight) of two sausages produced commercial companies
by different
Contents
Sausage nominal % fat
Water
Fat
Protein
Carbohydrate
Sodium chloride
A
12 20
67.1 60.3
12.4 18.9
11.9 12.3
5.7 5.0
1.7 1.9
B
12 20
67.6 61.7
12.5 20.6
11.7 10.1
5.4 5.6
1.4 1.8
Company
CONSUMER LIKING FOR FAT IN SAUSAGES
287
TABLET Dejinition of sensory attributes
used in profiling of sausages
Attribute
Low intensity
High intensity
Whiteness Colour Colour strength Smell of smoke Taste of smoke Saltiness Spiciness Meat flavour Hardness
black sample yellow/red sample little colour no smell of smoke no taste of smoke no salty taste no taste of any spice no taste of beef/pork no force necessary to attain deformation dry feeling when chewing no covering of oil or fat in mouth when chewing not granular testure
white sample red/blue sample much colour strong smell of smoke strong taste of smoke strong salty taste strong taste of any spice strong taste of beef and/or pork much force necessary to attain deformation juicy when chewing mouth-feel of oil or fat when chewing very granular texture
Juiciness Greasiness Graininess
unstructured line, anchored on the left end with “low intensity” and on the right end with “high intensity”. The descriptor was displayed above the response line. The computer transformed the responses into numbers from 1.0 (low intensity) to 9.0 (high intensity).
Consumer Hedonic
Tests
The acceptance tests were performed in four centrally located food stores, believed to cater to similar socio-economic strata. The assessors were customers aged 18 years and older who were regular eaters of sausages. Consumers in two of the stores tasted sausages from Company A and consumers in the two other stores tasted sausages from Company B. The companies were the main suppliers of sausages to the two stores where their sausages were tested. The sausage samples were coded with 3-digit random numbers when served. The first set of samples was served as a blind test. The second set was served with the fat content of the sausages written on the attached forms. Half of the consumers in each store were given correct information on the fat content of the samples and the other half the false information that the low-fat sample was the high-fat sample and vice versa. Each consumer tasted the samples twice and marked their acceptance on a 15cm continuous line anchored at the left end with “dislike very much” and at the right end with “like very much”. Afterwards, their marks were linearly transformed into numbers keyed into the SENSTEC system, with “dislike very much” scored as 1.O and “like very much” scores at 9.0. Each consumer took 5- 10 min to perform the evaluations. Data Analysis The significance levels of differences in sensory analysis of variance (ANOVA). A cross-classification and the first order interactions included was applied.
attributes were calculated by model with the main effects The assessors were treated as
R. SOLHEIM
random effect. The hedonic data were compared by t-tests and ANOVA, analysing separately the responses to sausages produced by the two companies (Table 3).
RESULTS
Sensory Profiles
For all sensory attributes except saltiness and juiciness and graininess for Company A’s sausages, differences between the sausages were functions of fat content (Table 4). The low-fat sausages had significantly higher intensities of colour, colour strength, smell of smoke, taste of smoke, meat flavour and hardness. The sausages containing the standard level of fat had significantly higher intensities of whiteness, spiciness and greasiness. The sausages produced by Company B differed also in juiciness and graininess between fat content levels. The explanation of the differences between the companies may be variety of quality of raw material and processing procedures. Sausages produced by Company B had significantly higher colour, colour strength, spiciness and hardness scores than Company A’s sausages, while the sausages produced by Company A had significantly higher whiteness, juiciness and graininess. Consumer
Likings
Provision of false or correct fat level affected the hedonic responses to the sausages produced by Company A, but did not affect the hedonic responses to the sausages produced by Company B (Figure 1; Table 5). The two types of sausages produced by Company A were quite similar in sensory profile. These sausages were not significantly different in hedonic response in the blind test. Correct information about the fat level of the 20% fat sausages did not affect the hedonic response (Table 6: comparison D). However, the false information that fat level of the 20%-fat sausages from this Company was 12% raised the liking ratings (Table 6: comparison E). In contrast,
TABLE 3 Experimental design of consumer liking for sausages affected by sensory quality and nutritional information. The design was usedfor testing 12%-fat and 20%-fat sausages produced by different companies. Similar letters indicate paired comparisons Consumer number
Blind test 20% fat sausages
12% fat sausages
B
1 2
12% fat sausages Correct False D
given
20% fat sausages Correct False D
B
A
A n
Test with information
C
E
C
E
CONSUMER LIKING FOR FAT IN SAUSAGES
Sensory projles
289
TABLE 4 of the four sausages, showing main effects of fat level and producer Company in ANOVA Mean attribute score Company A
Attributes
Whiteness Colour Colour strength Smell of smoke Taste of smoke Saltiness Spiciness Meat flavour Hardness Juiciness Greasiness Graininess
Company B
12% fat
20% fat
12% fat
20% fat
Fat content difference
Company difference
5.5
6.1 3.3 4.1 4.8 5.1 5.9 6.0 4.4 2.4 6.6 6.5 5.5
4.5 6.7 6.1 6.5 5.9 5.2 6.0 5.3 6.8 4.7 4.9 3.6
5.5 4.3 5.1 5.4 5.7 6.0 7.0 4.2 3.8 6.2 6.6 4.1
o&o7 < 0~00005 < 0~00005 0.002 0.01 0.1 0.05 0.016 < 0~00005 0.16 0.001 0.1
0/003 0.0003 0.0006 0.1 1.0 0.06 0.025 0.69 < 0~00005 0.000 1 0.77 0.004
6.2 5.6 6.5 6.6 6.1 5.5 5.3 4.9 6.8 5.1 5.1
The means are calculated for 12judges in 3 replicates. The standard error of means were in the range 0.13-0.35. correct information about the fat level of this Company’s 12%-fat sausages decreased the mean hedonic response, while the false information that their fat level was 20% had no effect on the hedonic responses (Table 6; Figure 1). From the sausages produced by Company B, the samples containing 20% fat got the higher mean hedonic score in the blind test (Table 6: comparison A). This difference did not change when the consumers were informed about the fat content. It should be remembered that the low-fat sausages and the standard sausages produced by Company B were not similar in their sensory profiles. A few consumers reacted negatively to tasting four samples in rapid succession. Others commented on the information, that they felt it must be false. This occurred when we gave them false information on the fat contents of the sausages from Company B. For the samples from Company A, no one questioned the information given.
DISCUSSION The experimental design was chosen to emphasize effects of sensory quality of a product and/or nutritional information on consumer responses. The higher hedonic score for 20% than 12% fat sausages from Company B may be explained by the difference in texture that arose in that Company’s product from the different fat levels. For meat and meat products, as for other foods, texture attributes such as tenderness and juiciness contribute to evaluations of freshness and nutritional quality, according to Szczesniak (1990) and Drewnowski (1987). In the present tests, the juicier a sausage was, the greater was the liking score. This was
290
R. SOLHEIM
Company
A
I
2Or
40
15
s 30 PC ; c 20 z 2 IO
c OD $8 IO c ‘; 2 5 ”
0
25 20 3 I& E a -
Correct
IO 5
information
Company
False
information
B
2571
Blind test
I5
0
0 Blind test
20r-
15,
Correct
informatlon
False
Information
FIGURE 1. Distribution of frequencies of hedonic responses to sausages, separately for blind tests and for tests when information was given. The frequency counts are plotted for equal ranges of 0.9 of a point with 1.0 as lowest and 9.0 as highest score.
291
CONSUMER LIKING FOR FAT IN SAUSAGES
TABLE 5 Effects of provision of information (correct or false) on hedonic response to sausages produced by different companies with 12% and 20% fat content Mean liking rating
ANOVA
Information given (correct and false)
Mean square
f
P
Company A (n = 174) 20% fat 5.80 12% fat 5.84
6.21 6.09
32.67 19.94
7.31 4.50
0.007 0.035
Company B (n = 169) 20% fat 6-14 12% fat 5.60
5.81 5.52
13.33 5.43
2.94 1.20
0.088 0.275
Sausage assessed
Blind test
independent of information given on fat content. Consumers might want to be consistent in their assessments. This could have affected the scores given to the second pair of samples, for which correct or false information given, when the samples were not similar in their sensory profiles, and accounts for some consumers questioning the false information.
TABLE 6 Hedonic response to sausages with different level offat content, measured in blind tests and with either correct or false information on the fat content given. Results from t-tests on paired comparisons. (letters in brackets corresponding to the letters in Table 3).
Information given
Comparison
n
Mean difference in liking rating
t
P
Company A’s sausages Blind 20% minus 12% Blind 20% minus informed 20%
None correct False
176 87 89
- 0.045 -0.138 -0641
- 0.22 -0.55 -2.41
0.83 0.58 0.02
Blind 12% minus informed 12%
Correct False
87 89
0649 0.185
-2.61 0.68
0.01 0.50
None Correct False
171 98 73
0.532 0.312 0.135
2.31 1.46 064
0.02 0.15 0.53
correct False
98 73
- 0.072 0.133
-0.27 0.59
0.79 0.56
Comany B’s sausages Blind 20% minus 12% Blind 20% minus informed 20%
g:
Blind 12% minus informed 12%
0
(4
@)
292
R. SOLHEIM
When, however, a texture attribute such as juiciness is similar in sausages of different composition, as for those from Company A, factors other than sensory quality might be able to affect hedonic responses. The liking scores for the 20%-fat sausages from Company A increased when the false information was given that they were reduced in fat content. These results indicate that at least some consumers react positively to sausages with reduced fat content. Lundgren (1981) found that identical sausages served with different labels (low-fat and high-fat) were perceived as nondifferent for one-third and as different for two-thirds of a group of 72 consumers. Nutritional information has been found to increase consumers’ hedonic response to some products (Byrd-Bredbenner et al., 1988; Cheng, 1990), but not all (Eiser et al., 1984), and has not been shown generally to affect consumers’ buying habits. Russo et al. (1986) found that lists of vitamins and minerals increased nutritional knowledge, but it had no influence on actual purchase. A list of products with added sugar (focused on negative nutrients) increased the market share of low-sugar breakfast cereals at the expense of high-sugar brands. They concluded the display of “negative” nutrients such as calories, cholesterol, sodium, sugar, and various chemical additives to be of importance. Thus, in the cases of a negatively evaluated nutrient such as fat, one may hypothesize that information on fat reduction should increase a product’s buying potential. However, it is not known whether information on reduced fat in sausages will affect consumers’ choice between fat-reduced and today’s standard sausages. From the results of this study, the sensory quality of the sausages seems to be more important to the hedonic response than does nutritional information. In conclusion, it seems to be important that sensory quality of fat-reduced sausages is quite similar to today’s standard sausages in order to be accepted by the consumers. If this is achieved, it may be possible to motivate some consumers to buy such a product by giving information about the fat reduction. REFERENCES
Byrd-Bredbenner, C., Martin, R., Lewis, C. J. & Shannon, B. (1988) The effectiveness of two methods of delivering nutrition information to the general public. Journal of Nutrition Education, 20, 63-69. Cheng, H. W., Clark, A. D. & Heymann, H. (1990) Influence of selected marketing factors on consumer response to restructured beef steaks: a conjoint analysis. Journal of Sensory Studies, 4, 165-178. Drewnowksi, A. (1987) Fats and food acceptance: sensory, hedonic and attitudinal aspects. In J. Solms, D. A. Booth, R. M. Pangborn & 0. Raunhardt (Eds.): Food Acceptance and Nutrition, New York: Academic Press. Eiser, J. R., Eiser, C., Patterson, D. J. & Harding, C. M. (1984) Effects of information about specific nutrient content on ratings of “goodness” and “pleasantness” of common foods. Appetite, 5, 349-359. Lundgren, B. (1981) Effect of nutritional information on consumer responses. In J. Solms and R. L. Hall (Eds.), Criteria of Food Acceptance, Zurich: Forster Verlag AG. Russo, J. E., Staelin, R., Nolan, C. A., Russell, G. J. & Metcalf, B. L. (1986) Nutrition Information in the supermarket. Journal of Consumer Research, 13, 48-70. SSI “Color Atlas” (1979) Swedish Standard SS 01 91 02. Stockholm: Swedish Standards Institution. Szczesniak, A. S. (1990) Texture: is it still an overlooked food attribute? Food Technology, September, 86-95. Received
1 October
1991, revision 20 February 1992