Advances in Consumer Research Volume 29, 2002 Pages 168-172
HOW DESCRIPTIVE MENU LABELS INFLUENCE ATTITUDES AND REPATRONAGE
Brian Wansink, University of Illinois at Urbana-Champaign
James Painter, University of Illinois at Urbana-Champaign
Koert van Ittersum, Wageningen University, University of Illinois at Urbana-Champaign
How do descriptive menu labels influence customers? In a six-week field experiment involving 140 customers, descriptive menu labels (such as "Grandmas zucchini cookies" or "succulent Italian seafood filet") increased sales by 27% and improved attitudes towards the food, attitudes towards the restaurant, and intentions towards repatronage. Such labels did not, however, directly increase the amount a person is willing to pay for the labeled item. If descriptive labels are used sparingly and appropriately, they can improve sales and post-consumption attitudes of the food and the restaurant.
$ A recent uproar in Australia over a potentially misleading menu description of a "grilled" chicken sandwich . . . serves as a reminder . . . that accurate product labeling remains critical to maintaining quality assurance and consumer confidence (Zuber 1999, p. 45)
$ The names and descriptions on the menu sounded promising, but three of the four dishes our group ordered were mediocre to unsatisfactory (Craddock 2000, p. 60)
Descriptive labeling is growing in use among restaurateurs, but it is not clear how these labels influence consumers. Do they add a positive power-of-suggestion or halo to a menu item, or do they unmeritiously raise expectations and result in disappointment? Although the effect of labeling on consumer evaluation is examined extensively, most research focuses on nutritional labels, health labels or warning labels (Bucklin and Asam 1973, deTurck and Goldhaber 1989, Hackleman 1981, Lenahen et al. 1973, Miller et al. 1998). Although the use of descriptive labels may increase the attractiveness of menu items, insights into the mechanisms underlying this effect are limited (Cradock 2000).
Because the use of descriptive labels is an increasingly common phenomenon (See, Restaurant Business 1997a, 1997b, 1998), the general assumption might be that they positively influence consumer evaluation. However, considering the quotations at the beginning of this article, gaining an understanding of the actual effect of descriptive labels on consumer evaluation is important. If descriptive labels provide an unmerited halo or power-of-suggestion, industry efforts might better monitor their use. Conversely, if descriptive labels unfairly raise consumer expectations about a menu item (cf., Craddock 2000), using these labels might backfire and negatively influence a consumers attitudes towards a menu item and a restaurant. The central questions examined in this research are the following: First, do descriptive menu labels influence consumer attitudes towards the labeled menu item, towards the restaurant, and intention to repurchase the labeled menu item? Second, do descriptive labels influence consumers intention to repurchase the labeled item indirectly, based on consumer cognitions or also directly based on affective beliefs triggered by the label-cue?
The paper is structured as follows. First, a theoretical overview of the mechanism underlying the effect of descriptive labels is given and hypotheses are formulated. The hypotheses, subsequently, are tested by means of a six-week field study of six menu items involving 140 adults. The results indicate that descriptive menu labels (such as Grandmas zucchini cookies or succulent Italian seafood filet), improved attitudes towards the food, attitudes towards the restaurant, and intentions towards repatronage. Such labels do not, however, significantly increase the amount a person is willing to pay for the item. Implications for increasing consumer satisfaction are then discussed.
THE EFFECT OF DESCRIPTIVE LABELS ON MENU ITEM EVALUATION
To understand the effect of descriptive labels on evaluation and intention to repurchase, insights into the mechanisms underlying this process must be obtained. To decide which specific menu item to buy, consumers use evaluative criteria that are determined by consumers desires (e.g., high quality, authenticity). The match between consumers product attribute beliefs and desires form the substrate for consumers evaluative judgements (Huffman and Houston 1993).
To establish whether the menu items identified, provide the benefits needed to satisfy consumers desires, consumers use cues. When in a restaurant, the most important cue a person can use (other than perhaps price) is the name of the menu item. Next, consumers interpret and provide meaning to the cue (Schellinck 1983, Steenkamp 1990). This (coded) meaning largely is a function of 'previously acquired meanings contained in the activated knowledge structures (Anderson and Bower 1973). By encoding and elaborating on the cue, using pre-existing knowledge, which is stored in memory in schemas (Fiske and Taylor 1984), consumers develop an evaluative judgement about the menu items identified (Olson 1981). In Figure 1, a simplified fictitious representation of this process is given for Cajun beans with rice and chocolate pudding. The name of both menu items thus triggers consumers (attribute) beliefs related to them (e.g., tasty, many calories) which are evaluated and combined to one overall evaluation of the menu item (e.g., [+] related to chocolate pudding).
THE IMPACT OF DESCRIPTIVE MENU LABELS ON ATTITUDES AND INTENTION TO REPURCHASE
When consumers are provided with a second cue, a descriptive label, a similar process occurs. Consumers associations with the descriptive label are triggered, and the related beliefs are retrieved from memory. However, next consumers will try to integrate the mental schema related to the menu item under consideration and that related to the descriptive label. Depending on how well both schemas match, consumers start adjusting their menu item attribute beliefs (Schmitt and DubT 1992). They will try to "bring the beliefs in both schemas in line with each other".
FICTITIOUS MENTAL NETWORKS FOR CAJUN BEANS WITH RICE AND CHOCOLATE PUDDING
For instance, if consumers associate "Grandmas" with "an awful lot of sugar", the product attribute beliefs "a lot of sugar", "sweet" and "tasty", related to chocolate pudding, will be adjusted for "Grandmas Chocolate Pudding". Instead of "tasty", consumers may believe that the taste of this 'version is (close to) perfect. Consumers beliefs related to the labeled menu items determine their attitude towards the menu item.
H1: Descriptive menu labels influence the consumers attitude towards the labeled menu item.
Like consumers attitude towards the labeled menu item is determined by their related beliefs, consumers attitude towards the restaurant, which serves the labeled item, is determined by consumers beliefs related to the restaurant. Next to speed of service, prices and for instance the atmosphere, the attitude towards the menu item served determines consumers attitude towards the restaurant. As it is expected that descriptive labels influence consumers attitude towards the labeled menu item, we expect that descriptive labels also affect consumers attitude towards the restaurant.
H2: Descriptive menu labels influence the consumers attitude towards the restaurant.
The match between consumers pre-consumption expectations and their actual consumption experience with the labeled menu item largely determines consumers post-purchase attitude towards the menu item, and consequently their post-purchase attitude towards the retaurant (Oliver 1980). Consumers post-purchase attitude towards the labeled menu item and the restaurant determine consumers intention to repurchase the labeled item and hence to return to the restaurant (Fishbein and Ajzen 1975). With descriptive labels influencing consumer attitudes (Hypothesis H1 and H2), and consumers attitudes influencing consumers intention to repurchase (Fishbein and Ajzen 1975), we expect that descriptive labels influence consumers intention to repurchase a labeled menu item through consumers attitude towards the labeled menu item. Next to their intention to repurchase, we examine if and how descriptive labels influence the amount of money consumers are willing to pay for the labeled menu item.
H3: Descriptive labels influence [a] consumers intention to repurchase and [b] the amount of money consumers are willing to pay for the labeled menu item, indirectly through consumers attitude towards the labeled menu item.
If and how descriptive labels directly influence consumers intention to repurchase is difficult to predict. It largely depends on whether consumers have affective associations related to descriptive labels. If consumers have affective beliefs related to the descriptive labels, a direct effect will be found.
To determine how consumers respond to descriptive labels, we conducted a six-week field experiment in a faculty cafeteria at a major Midwestern university. After reviewing the past sales of products in the cafeteria, we selected six products that were popular enough to offer twice a week and which represented a wide variety of foods (see Table 1). Descriptive labels included a mix of geographic labels, nostalgia labels, or sensory labels, which were presented next to the items in the cafeteria line.
During the Tuesday and Friday lunch of each of the six test weeks, two of the items were presented with a basic label (e.g., grilled chicken); two items were presented with a descriptive label; and two items were not offered. For the next two weeks, the items and the conditions were systematically rotated until all menu items were present in all conditions. In the fourth week, the rotation was repeated. The rotation was planned in order to minimize any unexpected variations that might affect either preferences or participation (such as blizzards, religious holidays, or Illini game days). During a six-week period, each item was available six times. [Reasonable efforts were made to control the production of the items, and there was no reason to believe there were production variations across days or across conditions. In addition, since every product was offered at least six times, an unnoticed problem with quality control would have been moderated by the five serving occations.]
Everyone who selected one of the six target menu items was asked to complete a questionnaire. Approximately 56% of the customers in the sample had selected menu items with descriptive labels and 44% had selected their regular-labeled counterparts. Ninety-eight percent of our customers (140) completed and returned their questionnaires upon finishing their meal and leaving the cafeteria. [While it is not clear as to why three individuals did not hand in their questionnaires, past experience has indicated that this can largely be attributed to them forgetting to do so, or because they misplace either the survey or the pencil.] In each questionnaire, customers were asked single item questions about their attitudes towards the target menu item and the restaurant (see Table 2).
We measured their attitudes about the target food by asking about their perception of the quality and their perception of the value of the menu item. The questionnaire also asked them to rate how well this restaurant followed trends and maintained quality of food compared to similar dining establishments. Most of the questions were asked in the form of Likert-scaled "strongly disagree" or "strongly agree" questions (strongly disagree = 1 and strongly agree = 9). The willingness to pay was measured based on the amount of money customers said would be the most they would be willing to pay for their menu item. Of those participating, 87% were faculty or staff, 9% were graduate students, and 5% were visitors from off-campus. The average age of the respondents was 43 years old. There were no differences between those who bought the descriptive-labeled menu items and those who bought the regular-labeled menu items. Both groups were analyzed on demographic characteristics (including age, gender, education), on the basis of the menu items they selected, on the basis of how healthy they perceived themselves to be, and on the extent to which they were watching their weight. Nothing was significantly different between the two groups (p<.05), and the only variable that remotely neared significance was that of dieting (p=.21). Consumers who purchased the descriptively labeled desserts (Grandmas Zucchini Cookies and Satin Chocolate Pudding) were directionally more likely to be on a diet than those who bought them under the regularly labeled versions of these products.
Analyses of variance were conducted to examine whether consumers attitudes towards the (labeled) menu items and the restaurant, consumers intention to repurchase and willingness to pay for the menu items differed in the two conditions (control group versus descriptive label). Table 3 shows the differences in attitudes, intention repurchase the menu items in the different conditions and the amount of money consumers were willing to pay for the different menu items.
Recall it was expected that descriptive labels affect consumers attitude towards the labeled menu item (H1). As predicted, descriptive labels were found to influence consumers evaluation of the overall quality of the menu item (6.19 versus 6.91; F1, 138 = 5.92). Next, consumers evaluation of the value of the menu item is positively affected by descriptive labels (6.33 versus 7.14; F1, 138 = 6.17). In line with the second hypothesis, it was further found that descriptive labels influence consumers attitude towards the restaurant. It was found that consumer thought that the restaurant was more up to speed with the latest food trends (5.71 versus 6.50; F1, 138 = 6.47), and the overall quality perception of the restaurant was higher (4.84 versus 5.86; F1, 138 = 8.08). Finally, consistent with our expectations, descriptive labels were found to affect consumers intention to repurchase the labeled item (5.96 versus 7.19; F1, 138 = 11.80). No effect regarding the price premium consumers are willing to pay was found ($3.08 versus $3.03; F1, 138 = .05). While this was unexpected, one possible explanation of this can be attributed to the anchoring effect that the purchase price had on their estimate of how much they would be willing to pay. Typically, when the purchase price is still salient, this ends up being the biggest predictor of how much people believe they should be willing to pay (Wansink, Kent, and Hoch 1998).
As shown at the bottom of Table 3, descriptive labels seem to influence consumers intention to repurchase the meal as well. No effect appears to be present regarding the amount of money consumers ere willing to pay. To examine, if and how descriptive labels intention to repurchase and willingness to pay, next, regression analyses were conducted. First, the intention to repurchase was regressed on the average of the evaluative beliefs and a dummy variable for the effect of a descriptive label (0 = no label, 1 = descriptive label). Next, following the same procedure, consumers willingness to pay was used as the dependent variable. Multicollinearity among the evaluative beliefs was not found to be a problem (VIF < 10). Further, dummy variables, representing the different types of meal examined, did not influence the outcomes of the regression results. Consult Table 4 for the regression results regarding hypothesis H3.
Recall that it was expected that descriptive labels only influenced consumers intention to repurchase and willing to pay indirectly, through consumers overall evaluation of the labeled menu item. Contrary to what was predicted, however, it was found that descriptive labels also directly influence consumers intention to repurchase the labeled menu item (bdescriptive label = .124, p < .01). Thus, next to the indirect effect of descriptive labels, through the perceived quality and value of the labeled menu item (bquality = .359, p < .001, bvalue = .401, p < .01), a direct effect is present. This direct effect may be due to affective, hedonic associations with descriptive labels. To examine this notion in more detail, next, we analyzed the open-ended questions. As described, consumers were asked to give their opinion about the menu item through an open-ended question. The responses were classified by each of the three authors as being either utilitarian or hedonic. Table 5 shows the results.
The consumers in the descriptive label condition were more likely to evaluate the menu item in hedonic terms than in utilitarian terms (p < .01).
Consumers affective beliefs related to descriptive labels did not affect consumers willingness to pay a premium price. In line with our expectation, no direct effect of descriptive labels on consumers willing to pay for the labeled menu item was found (bdescriptive label = -.076, p > .10). Descriptive labels thus indirectly influence the amount of money consumers are willing to pay for the labeled menu item, through the perceived value of the menu item (bvalue = .187, p < .05).
RESULTS OF ANOVA (ACROSS PRODUCTS)
REGRESSION RESULTS (STANDARDIZED PATH ESTIMATES)
NUMBER OF UTILITARIAN VERSUS HEDONIC MENU-ITEM RESPONSES
CONCLUSIONS AND IMPLICATIONS
Several conclusions can be drawn from our findings. First of all, descriptive labels positively influence consumers attitude towards the labeled meal. Second, descriptive labels influence consumers intention to repurchase the labeled meal directly and indirectly, through consumers overall evaluation of the labeled meal. Consumers affections, triggered by descriptive menu labels, probably cause this direct effect on intention to purchase.
The name of a menu item is an important criterion for decision making. By providing additional cues, such as descriptive labels, restaurateurs thus not only enhance the perceived attractiveness of menu items, but also they may favorably influence the actual consumption experience. Several conclusions from our findings can help managers improve the reputations of their restaurants. Descriptive labels, such as "Grandmas zucchini cookies" . . .
+ Increased sales by 27%
+ Increased post-trial evaluations of quality and value
+ Increased restaurant-related attitudes
+ Increased restaurant repatronage intentions
There are a number of different ways to generate descriptive labels, including the use of geographic labels (Cajun and Italian), nostalgia labels (homestyle or Grandmas), or sensory labels (tender or satin), or a mix. The type of label that would be most effective depends on the product it describes. Certain types of labels work better with certain foods than others.
In this study, the descriptive labels that were used had all been pretested to evoke favorable associations with the food. Clearly, using labels that have unfavorable associations with them would not be wise. It is less clear what impact labels would have if they had only neutral associations or if they had no associations (such as might be the case with foreign words).
All of the food used in this study was of reasonably high quality. If food of only average or below average quality was used, descriptive labels may have had less of an impact. By using descriptive labels, restaurateurs may raise customer expectations about the quality of the menu item (Schmitt and DubT 1992). These expectations need to be met. If a restaurateur is unable to do so, using an unmerited descriptive label might backfire and negatively influence customers attitudes about the item and, ultimately, the restaurant (Oliver 1980). Restaurant managers should monitor their use of descriptive labels in order to avoid unjustifiably inflating their customers expectations. But beware of the temptation to label yesterdays goulash as "Royal Hungarian Top Sirloin Blend." It will generate first time sales, but they may be the last.
DISCUSSION AND FUTURE RESEARCH
This research was conducted in a life setting for a limited number of meals. More research needs to be devoted on other types of meals and products to be able to generalize our findings. Further, a broader range of descriptive labels needs to be examined.
Additionally, specific attention needs to be given to whether product- and/or label specific effects are present. For instance, some descriptive labels may match some food products better than others (e.g., Grandmas may better match Zucchini cookies than for instance chocolate tofu cheese cake). Since the number of observations for each product separately were too small to test this in our research, future research may address this aspect into more in-depth. The examination of potential product- and label-specific effects increases our understanding on the effect of descriptive labels.
Anderson, John R., Bower, G. H. (1973). Human Associative Memory, Washington D. C.: V. H. Winston & Sons
Bucklin, L., Asam, E. (1973), Nutritional Labeling for Canned Foods: A Study of Consumer Response, Journal of Marketing, Vol. 37, 32-37
Craddock, Catherine (2000) Italian cliche, Business Mexico 10, no. 5 (May 2000): 60
deTurck, M.A., Goldhaber, G.M. (1989), Effectiveness of Product Warning Labels: Effects of Consumers' Information Processing Objectives, Journal of Consumer Affairs, Vol. 23, 111-126
deTurck, M.A., Rachlin, R.A., Young, M.J. (1994), Effects of a Role Model and Fear in Warning Label on Perceptions of Safety and Safety Behavior, Advances in Consumer Research, Vol. 21, 208-212
Fishbein, Martin, and Ajzen, Icek. (1975), Beliefs, Attitude, Intention, and Behavior: An Introduction to Theory and Research. Addison-Wesley Publishing, Reading, MA
Fiske, Donald W., Pavelchak, Mark A, (1986), Category-Based versus Piece- Meal- Based Affective Responses: Development in Schema-Triggered Affect, in: Sorrentino, Richard M., Higgins, Tory E. (eds.), The Handbook of Motivation and Cognition, New-York: Guilford Press
Fiske, Donald W., Taylor, Shelley E. (1984), Social Cognition, Reading, MA: Addison-Wesley, 139-181
Hackleman, E.C. (1981), Food Label Information: What Consumers Say They Want and What They Need, Advances in Consumer Research, Vol. 8, 477-483
Huffman, Cynthia, Houston, Michael J. (1993), Goal-Oriented Experiences and the Development of Knowledge, Journal of Consumer Research, Vol. 20 (2), September, 190-207
Lenahen, R.J., et al., (1973), Consumer Reaction to Nutritional Labels on Food Products, Journal of Consumer Affairs, 7, 1-14
Miller, Debra L., Castellanos, Victoria H., Shide, David J., Peters, John C, Rolls, Barbara, (1998), Effect of Fat-Free Potato Chips With and Without Nutritional Labels on Fat and Energy Intakes, The American Journal of Clinical Nutrition, Vol. 68, 282-290.
Oliver, Richard L. (1980), A Cognitive model of the Antecedents and Consequences of Satisfaction Decisions, Journal of Marketing Research, 17 (November), 460-469.
Olson, J.C. (1981). The Importance of Cognitive Processes and Existing Knowledge Structure for Understanding Food Acceptance. in: J. Solms & R. L. Hal (eds.), Criteria of Food Acceptance: How Man Chooses What He Eats. Forster, Zurich, 69-80.
Restaurant Business (1997a), Boutique Brands, Restaurant Business; Brand power Supplement, p. 20-21.
Restaurant Business (1997a), Signature Dishes, Restaurant Business; Brand power Supplement, p. 36-37
Restaurant Business (1998), Pork: Cut to the Profits: Fruit of the Loin, Restaurant Business; Brand power Supplement, p. 118-119.
Schellinck, Douglas A. (1983), Cue choice as a function of time pressure and perceived risk, Advances in Consumer Research, Vol. 10, 470-475.
Schmitt, Bernd H., DubT, Laurette, (1992), Contextualized Representations of Brand Extensions: Are Feature Lists or Frames the Basic Components of Consumer Cognition, Marketing Letters, 3:2, 115-126
Steenkamp, Jan-Benedict E.M. (1990): Conceptual Model of the Quality Perception Process, Journal of Business Research 21 (August): 309-333.
Wansink, Brian, Robert J. Kent, and Stephen J. Hoch (1998), An Anchoring and Adjustment Model of Purchase Quantity Deciions, Journal of Marketing Research, 35:1 (February), 71-81.
Zuber, Amy (1999), Dispute over 'Grilled Label Raises Regard for Truth-in-Menu Protocol, Nations Restaurant News 33, no. 31 (August 2): 45-50.