Don’t come too close: An investigation into the effect
of proximity on social norm effectiveness.
Master Thesis
Name Ayolt Willem de Roos
Student number 10506683
Program MSc Business Studies
Track Marketing
Institution Amsterdam Business School - UvA University First supervisor dhr. drs. J. Labadie
Second supervisor dhr. drs. R.E.W. Pruppers Date of submission February 23rd, 2014
ABSTRACT
Study This study firstly evaluated the relationship between the use of social norms in advertising material and the consumers’ advertisement attitudes, brand attitudes and purchase intentions. Secondly, it was evaluated how proximity towards an introduced reference group influences this relationship.
Method Five samples of 30 respondents were collected. Each sample received a different survey, each showing an advertisement for the chocolate brand Guylian, with a differing social norm. The three variables under investigation (advertisement attitudes, brand attitudes and purchase intentions) were analyzed using statistical analyses (one-way ANOVA), performed on the collected data from the online survey.
Findings A significant main effect of the presence of a social norm in advertising communications was found on advertisement attitudes (H1a) and brand attitudes (H1b), whilst a significant main effect of the inclusion of a reference group in the depicted social norm was only found on purchase intentions (H2c). The effect of proximity was analyzed with three separate sets of hypotheses and firstly showed that only for brand attitudes a significantly higher score was measured between low and medium proximity (H3b). Secondly it was found that a significant drop was measured in all three dependent variables once the reference group moves from medium to high proximity (H4a-c). Lastly, it was found that the drop was only big enough for purchase intentions to make the high proximity condition dip below the value of the low proximity condition (H5c).
Conclusions From the findings of this research, it can be suggested that 1) social norms in advertisements have different effect on intentions and attitudes; 2) intentions and attitudes also react differently to the the inclusion of a reference group in the social norm; 3) not all measures react the same to an increase in proximity, but they do all show a significant drop once proximity gets too close.
TABLE OF CONTENT
1. INTRODUCTION 9
1.1 The influential social environment 9
1.1.1 Power of the masses 9
1.1.2 A crowd you can relate to 10
1.1.3 New insights into social norms 11
1.2. Problem definition 11
1.2.1 Problem statements 11
1.2.2 Subquestions 12
1.2.3 Delimitations of the study 12
1.3 Contributions 13
1.3.1 Theoretical contributions 13
1.3.2 Managerial contribution 14
1.4 Structure/outline 15
2. THE IMPORTANCE OF SOCIAL NORMS IN MARKETING 16
2.1 The theory of conformity 16
2.1.1 Asch’s conformity experiment 16
2.1.2 Conformity through canned laughter 17
2.1.3 Conformity in everyday life 17
2.2 Social norms 18
2.2.1 Types of social norms: injunctive versus descriptive 18
2.2.2 Social norms in intervention marketing 19
2.2.3 Social norms in commercial marketing 20
2.3 The Elaboration Likelihood Model 21
2.3.1 Low involvement products 21
3. THE ROLE OF REFERENCE GROUPS IN SOCIAL NORM MARKETING 23
3.1 Reference groups 23
3.1.1 The use of reference groups in marketing 23
3.1.2 Different reasons to conform to a reference group 23
3.1.3 The use of reference groups in social norms 24
3.2 Different types of reference groups 25
3.2.1 Mental reference groups 25
3.2.2 Physical reference groups 27
3.3 Proximity towards a reference group 29
3.3.1 Mental proximity 29
3.4 Issues regarding proximity 30
3.4.1 The reactance issue 30
3.4.2 Privacy issues 31
4. HYPOTHESES 33
4.1 Hypotheses 33
5. METHOD 42
5.1 Participants and procedure 42
5.1.1 Sample 42 5.1.2 Procedure 42 5.2 Description of measures 46 5.2.1 Dependent variables 46 5.2.2 Control variables 47 5.3 Data analysis 48 6. RESULTS 50 6.1 Preliminary tests 50 6.1.1 Participant characteristics 50
6.1.2 Reliability tests/factor analyses 50
6.1.3 Computing variable means 50
6.1.4 Correlation matrix 51
6.2 Manipulation check 51
6.3 Hypothesis testing 54
6.3.1 The main effect of the inclusion of a social norm 56 6.3.2 The main effect of the inclusion of a reference group in s social norm 58 6.3.3 The effect of increasing proximity towards a reference group 60 6.3.4 The negative return on increasing the proximity towards a reference group 61
6.3.5 The negative result of getting too close 63
7. DISCUSSION 65
7.1 The effect of social norms in advertisements 65
7.2 The effect of reference groups in social norms 70
7.3 The effect of proximity on social norm effectiveness 74
7.3.1 The inverted U-shape 75
7.3.2 Variables falling to an all-time low 79
8. CONCLUSION 84
8.1 Theoretical contributions 84
8.2 Managerial contributions 86
8.3 Limitations and future research suggestions 88
8.4 Final thoughts 91 ACKNOWLEDGEMENTS 92 LITERATURE 93 APPENDICES 99 Appendix 1 99 Appendix 2 100 Appendix 3 101 Appendix 4 104 Appendix 5 109 Appendix 6 109 Appendix 7 110 Appendix 8 110 Appendix 9 111
1. INTRODUCTION
1.1 The influential social environment
Have you ever bought a product or followed a suggestion because you were told that the majority of other people acted in a certain way? If you are like most people, you probably will have. As a child at the age of 8 you may have been convinced to ask Santa Claus for the newest race car because it was said to be the most popular toy of the moment. As an adolescent you may have bought the newest Nike Air Max because all the cool kids were wearing them. And when you bought your first car, the tag that said “best selling car of the year” may have convinced you that this car was just right for you. Companies are fully aware of this urge that consumers have to follow the crowd. They know that we care about the opinions and buying behavior of others in the marketplace, and they employ this knowledge in countless instances.
1.1.1 Power of the masses
Companies use social norms, a descriptive norm that tells an individual about the behavior of others in his/her social environment (Goldstein et al., 2008), to shape the behavior of prospective and existing customers. They show us what everybody else is already doing, and whether we like to admit it or not, we’ve probably all followed this depicted crowd at some point in our life.
In many of our choices as a consumer, this social norm plays a big role. The behavior of others in the marketplace shapes our own decisions. Every day we are confronted with information regarding the marketplace behavior of people in our social environment, in an attempt to get us to jump on the bandwagon with them. The information that is used to convince us comes from all different angles. We get notified that fifteen of our friends like a newly established clothing brand on Facebook, we are told that 82% of the users of a new shampoo are satisfied with the results they got and the book on our nightstand very often has a New York Times Best Seller sticker on the cover, indicating its popularity. We care about these things, since we tend to care about what others think, do and say. Almost everybody has the desire to conform with a group of similar individuals, with whom we can identify ourselves (Martin, 2012).
Organizations recognize the value of social norms, since they can help to increase their credibility. By stating the percentage of consumers that prefer, choose or like the brand, the quality claims made by the producer are backed up by actual consumers. A tea producer can be very satisfied with their new product, but only by claiming that “78% of the people in their test panel graded Pickwick Herbal Goodness between good and very good” (Het Testpanel, 2013a) can they show that their
opinion is shared by the consumer. The same goes for a change in the formula of an established dairy product, which can be justified by stating that “84% of the respondents like the new formula of Yoki-drink more than the old formula” (Het Testpanel, 2013b). The inclusion of such claims in marketing messages can show that it is not just the developers who are satisfied with the new product, but that consumers are just as pleased: a social norm for Pickwick Herbal Goodness tea or Yoki-drink is thereby created.
1.1.2 A crowd you can relate to
A social norm can state what the general population thinks, but it can also speak to a specific group of people, known as a reference group. Such a reference group refers to similar people as the person who is being addressed (Lewis and Neighbors, 2004). A specific age segment can be targeted, a gender split can be created or a shared interest can serve as a reference criteria. Imagine, for example, that you are a 27-year old female, who travels a lot for both private and business related purposes. A shampoo brand can choose to create different slogans, revolving around different reference groups that may appeal to you: they can state that “80% of the consumers between the ages of 25 and 35 prefer the product”, that “78% of the female consumers prefer the product”, or that “81% of frequent travelers prefer the product”. In each of these instances, a different reference group is created and subsequently, a different group of consumers is addressed.
Reference groups are not just limited to age, gender or shared interest. Any kind of population demographic can serve as a criteria (e.g. marital status or nationality), as well as many different psychographic aspects (e.g. personality traits or group membership). Goldstein et al. (2008) introduced yet another interesting reference criteria: physical proximity. The researchers argue that such proximity to the reference group can be of importance and they coined the new term “the provincial norm”. This norm implies that a norm which is set by people in your immediate surroundings can be more effective than a norm demarcated by age or gender. This would imply that a message read by a male Amsterdam citizen would be more effective if it read “the majority of Amsterdam citizens drink Heineken”, than if it read “the majority of men drink Heineken”. Such a technique is already found in direct marketing practices, in which your energy consumption can be compared to that of your neighbors (Burchell et al., 2013) or in which your registration as an organ donor is put next to the rest of your city’s residents.
1.1.3 New insights into social norms
Although the use of social norms is already visible in the commercial marketplace, empirical evidence supporting this marketing technique is very limited. This represents the first gap in current literature that this paper will address. In the vast collection of research that has already been devoted to the phenomenon of social norms, the most often researched context regards intervention techniques. Situations that belong to this category are, amongst others, trying to increase timely tax payments (Martin, 2012), getting people to give up bad habits such as smoking (Dohnke et al., 2011) or drinking (DeJong et al., 2006) and increasing towel re-use in hotel rooms (Goldstein et al., 2008). In such intervention circumstances the general goal is to either minimize socially undesired or maximize socially desirable behavior of consumers. Empirical evidence is needed, that justifies the choices made by brands to use the social norms approach in commercial settings. Settings in which they attempt to influence consumers to purchase their product or service. At this moment, when the success of social norms in such a setting is discussed, experience accounts are presented as the only proof. By addressing this gap in literature, it can be investigated whether the use of social norms really is beneficial in commercial advertising campaigns, and what kind of effect such norms have on consumer behavior.
A second gap that is going to be addressed in this paper concerns proximity of the reference group. When you define a reference group as “a group of likeminded individuals”, the borders are not quite clear. An age segment can be 20 to 40-year olds, but can also exclusively refer to 24-year olds. The former segment then has a lower proximity than the latter, since the demarcations of the reference group are much narrower in the latter instance. Proximity can also be increased by introducing more criteria into the reference group. When a reference group is demarcated by both age and gender, proximity is increased since the group is made smaller. By adding a geographical aspects (e.g. nationality or city of residence) proximity can once again be increased, by moving the chains closer and closer to the target consumer. What this paper will set out to determine is whether proximity should be as close as possible, or whether people might feel that getting too close will become too personal. In other words: can a message get too close?
1.2 Problem definition
1.2.1 Problem statements
With the specified gaps in current knowledge regarding the use of social norms theory in the commercial marketplace and the proximity of such social norms, two problem statements are defined, which will be addressed in this research paper:
1. What is the relationship between the use of social norms in commercial advertisement material and the consumers’ purchase intensions and attitudes?
2. How is the relationship between the use of social norms and the consumers’ purchase intensions and attitudes influenced by the perceived proximity towards the depicted social norm?
By addressing these two problem statements, this paper should contribute empirical evidence that can address the two presented gaps in current literature.
1.2.2 Subquestions
In order to come to a satisfactory conclusion regarding the defined problem statements, a collection of subquestions will need to be addressed:
1. What are social norms?
2. What theories are social norms based on?
3. How are social norms present in (commercial) marketing?
4. What types of reference groups are used in marketing communications? 5. How can we define proximity of a reference group?
These subquestions will help in the evaluation of current literature available on the defined topics. This literature will form the basis of this research, since it will help to create research hypotheses. The relevant literature will be presented in chapter 2 and 3.
1.2.3 Delimitations of the study
This research is clearly set in the commercial marketplace, not in the realm of intervention techniques. This implies that whilst the importance of limiting socially undesirable behavior and maximizing socially desirable behavior is fully noted, other researchers have already shown the value of the social norms approach in that area of expertise. This paper will focus on the commercial marketplace, with the primary focus being on social norms used in printed advertising material. Printed advertising material is chosen since findings in this setting can serve as incremental findings in this specific topic, purely focusing on the relationships under investigation. If the setting would be based around online marketing content, in-store marketing communications or direct marketing practices, the main relationship under investigated would be influenced too
For this research, only social norm based on behavior by fellow consumers is used. A social norm could also read “9 out of 10 dentists advice ABC Toothpaste”, however it is believed that such a norm has more to do with signaling theory. In an instance where the opinion of experts is depicted, the perceived quality of a product is backed up by a professional quality claim, not a social norm. The example given above only functions as a true social norm for other dentists.
In this research, reference groups based on age, gender and geographic location will be used. These types of reference groups are chosen since they are most often found in both research and since results of a performed pretest indicate that these reference groups are also the most often found in practice. Additionally, these reference criteria accommodate the desire to alter proximity of the reference group towards the consumer. Other reference group criteria (i.e. occupation or psychographic values) are omitted, due to the fact that they would complicate the chosen research set-up. Their importance is thus not disregarded, but their potential influence is seen as material to be investigated in future research papers, once the incremental influences of the reference group criteria investigated in this paper are established.
With these specific delimitations regarding this study, it is believed that the most valuable results can be obtained and the clearest discussion can be formed.
1.3 Contribution
1.3.1 Theoretical contributions
The theoretical contribution of this paper is based around three objectives. First, the success of the social norms approach in many intervention circumstances is extended into the field of commercial marketing (specifically printed advertising material). There are several reasons to argue why differences could be found between the intervention setting and the commercial setting. Primarily, an intervention technique revolves around omitting a certain behavior from the individual’s routine (e.g. not lighting up another cigarette of grabbing another beer) or improving the individual’s attitude towards a desired outcome (e.g. getting him/her to file taxes on time). This is different than behavior in the commercial marketplace. In the commercial marketplace the individual is asked to perform other sorts of activities, primarily focused on their private costs and benefits, such as going to a store to purchase a good, substitute a different good in his/her basket for the advertised good or spending more money than before. This implies that there are arising opportunity costs and a difference in activity levels required from the consumer. This may imply a different inclination to follow or reject social norms. Additionally, intervention techniques are put in place to address behaviors which are socially (un)desirable. In the commercial setting, society at large plays no role.
You can, for example, not argue that increased sales of a new lipstick is socially desirable. This may influence the way in which people react to social norms: in instances in which society is of a main concern, it may be perfectly acceptable that the behavior of people is monitored (e.g. measuring electricity usage in your household), whilst this may be frowned upon in commercially driven markets (e.g. measuring how much of a specific toothpaste brand you buy).
A second objective is to determine whether the inclusion of a reference group in the social norm can improve the effectiveness of a social norm in an advertisement. Several authors (Bearden and Etzel, 1982; Englis and Solomon, 1995; Neighbors et al., 2010) claim that marketing messages work more effectively when a market segment or reference group is precisely targeted. If this is the case for social norms as well, it should become clear that a social norm based around a specific reference group is more effective than a general claim about the population.
The third objective of this paper concerns the proximity of the reference group. Whilst some people may argue that the more precise a marketing campaign is, the better it will work, others might argue that the means through which this is achieved may be on the border of privacy invasive (Boyd, 2008; Delo, 2012). By investigating the relationship between proximity of a reference group to the customer and the effectiveness of such a reference group, conclusions can be drawn on which of the two opinions holds most ground: “the closer the better” or “keep appropriate distance”.
1.3.2 Managerial contributions
The managerial contribution of this paper is twofold. First, managers often use social norms in their marketing messages, but current literature suggests that there is little empirical evidence supporting this choice. The findings regarding the success of social norms come primarily from the intervention instances in which they are used, and results from commercial use come primarily from experience accounts. For managers in this area of expertise, findings from this paper might provide validation for the use of social norms in marketing campaigns.
Secondly, the proximity of a reference group used in marketing messages is becoming a more and more important topic for any marketeer. With the development of modern advertising methods (e.g. internet marketing campaigns, localized print media, location tracking applications on mobile devices) location-based and personal information is becoming more easily accessible. By having the means to shape marketing messages according to close proximity, consumer can be more personally approached. Such specific marketing messages could be of great value to a company, since it might give them an edge over their competition in a market defined by immense information clutter. A
generic claim. However, it is also possible that a too personalized message will not work, because people feel as if the information used was gained in a wrongful manner. Finding out whether this diminishing effect is present is one of the main objectives of this paper.
1.4 Structure/outline
In the remainder of this paper, finding answers to the defined problem statements will be of central importance. In the next two chapters current literature on social norms and reference groups will be reviewed. In chapter 4 the proposed hypotheses will be introduced and the reasoning behind them will be discussed. In chapter 5 the methodology of the conducted research is introduced. Chapter 6 is devoted to the collected data, as this will be analyzed using statistical methods. The discussion of the analyzed data, in connection to the stated hypotheses and the proposed problem statement, will be presented in chapter 7. Additionally, limitations posed by the research as well as avenues for future research will be discussed. Finally, chapter 8 will end this paper with a summary of the most important contributions of this paper and a final conclusion tying all the findings together.
2. THE IMPORTANCE OF SOCIAL NORMS IN MARKETING
2.1 The theory of conformity
Before you can start to look at the practice of using social norms in marketing communication, the theory which forms the backbone of this practice needs to be understood. Of prime importance here is the theory of conformity. In its most simplistic form, the theory of conformity suggests that human behavior can be shaped by our understanding of what others do in a similar situation (Burchell et al., 2013). If most people in a group claim that answer A is the correct response to a given question, some others who believe answer C to be correct, will still chose answer A. They
conform to what the majority thinks, while ignoring their own gut feeling. This theory has been
tested in many, often experimental, settings.
2.1.1 Asch’s conformity experiment
One of the most groundbreaking examples of conformity comes from a series of experiments performed by Solomon Asch in the 1950s (Barkan, 2010). In one of these experiments, Asch presented a group of people with a simple problem, which begins by showing them one example line and three option lines (see figure 1). The respondents in this experiment were then asked which of the option lines matched the example line in length. When people were asked to privately write down their answer to this problem, they gave the correct answer 98% of the time, indicating that the task is fairly easy to perform correctly.
The conformity effect becomes visible in the next stage of the experiment. Before the test subject can share his/her answer to the simple question, several confederates give the wrong answer out loud (no longer privately). Once the test subject is
asked to answer the question, he/she has to chose between his/her own intuition and the answer given by the rest of the group (Asch, 1956). In total, in about one-third of the trials the test subject gives the (wrong) answer given by the group, simply because the majority did so (Barkan, 2010). This is a fairly unexpected result, given the fact that the task was initially successfully completed by 98% of the test
The need to conform does however only seem to prevails when opinions are publicly proclaimed (i.e. stating your answer out loud). In an additional test condition, Asch found that if people were asked to write down their answer privately, after confederates gave their answers out loud, that they often gave the correct answer (i.e. a different answer than given by the group). In other words: respondents went against the group’s opinion and followed their own feeling, simply because they did not have to conform out loud (Asch, 1956).
2.1.2 Conformity through canned laughter
As stated, conformity theory revolves around human behavior which is shaped by the behavior of a group. In the case of Asch’s experiment, the group behavior was clearly set in a laboratory settings which caused some critique to arise (e.g. Nosancuk and Lightstone, 1974). Some real life situations can be applied in a monitored setting, to get more generalizable findings. Nosancuk and Lightstone (1974) for example used canned laughter (the prerecorded laugh track heard in many American sitcoms) in an attempt to substitute a very objective measure (i.e. length of lines as in Asch’s experiment), with a much more subjective measure (i.e. the funniness of anecdotes). The main finding in their experiment was that people’s public reaction (i.e. audible laughter) to the anecdote did increase once accompanied by canned laughter, whilst their privately held believe (i.e. written score) to the anecdote was hardly affected by the canned laughter.
These findings are in line with Asch’s results. The increase in public reactions was similar to the trials in Asch’s experiments, when the respondents publicly gave the wrong answer, simply to conform to the crowd. The fact that there was no increase in privately held reactions, was in line with Asch’s last condition, in which respondents were asked to write down their answers privately: conformity dropped, since there was no one to publicly conform to.
The findings by Nosancuk and Lightstone (1974) were one of the first findings which showed that Asch’s experimental study could be applied to something much more subjective and lifelike.
2.1.3 Conformity in everyday life
It is even possible to step further away from the controlled environment of the laboratory, and look into everyday situation which involve conformity theory. Many of the best examples can be found in the works of psychologist Robert Cialdini. His works “Influence: Science and Practice” (2009) and “Influence: The Psychology of Persuasion” (2007) include countless examples of real life situations in which we see conformity at play. One of the best examples he gives concerns cloakroom employees and their tip jar (Cialdini, 2009). In this study it was found that the average
tip of guests increased if a jar was filled with higher value bills. These findings suggest that if we see that the jar is filled with $1 bills, we automatically assume that $1 is the standard tip, whilst we would only leave a quarter if the jar was full of quarters. In such a case we are prone to conform with the norm of tipping, which we believe in such a situation to be a $1 tip. Such a finding is the perfect example that the theory of conformity does not only work in a controlled laboratory environment, but can be very effective in the outside world as well.
Both the findings from the laboratory setting and the everyday examples show that the power of conformity is strongly rooted in the human brain. This makes it a perfect mechanism for marketeers to exploit in their search for customers.
2.2 Social norms
The group opinion which is of importance in the theory of conformity is, in marketing, often referred to as the social norm. In his article regarding the use of social norms and tax payments, Martin (2012) states that the practice of using social norms revolves around the premise that “people’s behavior is largely shaped by the behavior of those around them”. In essence, this is a similar principle to the one found in the theory of conformity. As we will see, social norms are often the tool used in marketing communications, to increase social conformity.
2.2.1 Types of social norms: injunctive versus descriptive
Just as the theory of conformity can be found in many everyday life situations, so can social norms. Burchell et al. (2013) state that norms can be seen as rules that guide and regulate social behavior in certain situations. Something as simple as shaking hands when you meet someone for the first time can be considered a social norm, but also something much deeper such as the high rate of bulimia amongst American cheerleaders can be credited to the power of social norms (Crandall, 1988). According to Cialdini et al. (1990) social norms can be split up into two categories: descriptive norms and injunctive norms. Descriptive norms describe the behavior of the majority and they state which kind of behavior can be considered typical (Cialdini et al., 1990). Such norms often state the behavior of the majority of people in the form of a percentage, as is seen in the article by Martin (2012): “Over 94% of UK citizens pay their taxes on time.” By stating such behavior, it is expected that the people not yet following suit, will be convinced to join the majority in their ways of doing.
The injunctive norm is more aimed at the morality of an action. Such a norm states what is morally approved or disapproved and can come in the form of a plead, such as: “People start pollution, people can stop it” (Cialdini, 2003).
Overall it is believed that the descriptive norm is more successful in changing people’s behavior than the injunctive norm (Berkowitz, 2004). It is said that this is the case because people get an example to follow when confronted with descriptive norms, whilst they are asked to put in a lot more effort to set the leading example when confronted with an injunctive norm. It is therefore that we see more descriptive norms put into place when social norms are used as a marketing tool: they ask the audience to perform a much simpler and clearer task than when they are confronted with an injunctive norm. People will only have to follow an example, they will not have to set it.
2.2.2 Social norms in intervention marketing
The main weapon of choice in marketing is thus the descriptive norm: the behavior of the majority is shown as the most sensible thing to do (Cialdini et al., 1990). Such a majority is often stated as a percentage of customers who act in the desired way, such as we can see in a campaign by Evanston Township High School, in their efforts to limit smoking behavior of their students: “I DON’T SMOKE. Just like 88% of ETHS students” (Burchell et al., 2013). By stating a (high) percentage it can be shown that it is not just a “majority” that acts in a certain way, since a majority could imply just 51% of the population. The practice of using such percentages is widespread through both commercial and intervention marketing, but it is the latter situation in which the vast majority of the empirical evidence regarding the power of social norms marketing can be found.
In intervention marketing the aim is to minimize socially undesirable behavior and/or increase socially desirable behavior (Lewis and Neighbors, 2004; Wechsler et al., 2003). Such behaviors include decreasing smoking habits (Dohnke et al., 2011) or alcohol intake (DeJong et al., 2006), increasing timely tax payments (Martin, 2012), increasing towel re-use in hotel rooms (Goldstein et al., 2008) and decreasing littering in public places (Cialdini et al., 1990; Cialdini, 2003).
According to Berkowitz (2013) the use of social norms can be most successful in situations in which individuals have an incorrect perception of the behavior of the community. As an example he states that the average college student believes that his/her peers drink more than they actually do. The individual will thus drink more than he/she would normally do, to keep up with the perceived level of alcohol intake. By stating what the actual norm of drinking is (i.e. less alcoholic beverages than the student anticipates), the alcohol consumption of the student can be limited. This tactic can be found in countless marketing campaigns, used by many different educational facilities, including
Hobart and William Smith Colleges. They’ve used the following slogan in their Alcohol Education Project (Perkins and Craig, 2002): “In reality, most HWS students drink one to four drinks or do not drink at all when partying.” With this slogan they want to show their students that their peers drink less than what they most likely perceive themselves to be the case.
Similar slogans can be found in other instances. Martin (2012) for example shows that timely tax payments increase dramatically once the slogan “Over 94% of UK citizens pay their taxes on time” was introduced on the tax forms. Apparently UK citizens thought it was perfectly acceptable to postpone tax payments (the perceived norm), until they realized that most citizens actually do pay on time (the actual norm).
So far, the use of social norms in intervention marketing campaigns has been documented in countless papers, with most often positive results concluded. The use of the specified technique in commercial marketing is however not discussed very often. Far less empirical evidence has been presented regarding the use of social norms in the latter setting (Burchell, 2013). The next section will discuss the use of social norms in this commercial setting, and the sparse collection of results regarding this setting will be presented.
2.2.3 Social norms in commercial marketing
It seems to be that the success of social norms used in commercial marketing is build purely on a foundation of experience accounts: whenever an article discusses the success of social norms used in marketing campaigns, successful examples from the field of marketing are given instead of empirical findings from colleague academia.
Many different articles concerning the use of social norms, refer to different campaigns in which the practice was used. Burchell et al. (2013) for example refer to a campaign for Whiskers cat food, in which the norm read: “Eight out of 10 owners who expressed a preference said their cats preferred it”. This shows that the majority of cat owners prefer Whiskers, creating a norm to which non-users might want to conform. Martin (2012) states that his favorite use of a social norm was the simple rewriting of the infomercial tagline “operators are standing by” to “if operators are busy, please call again”. Higher popularity for the product was sketched, once again creating a norm for people who had not yet purchased the NordicTrack exercise machine (Martin, 2012).
Flip through any magazine, and you’re bound to come across an advertisement that uses a social norm to their advantage. Especially magazines aimed at female readers are full of social norms, with cosmetic brands leading the pack. One such example can be found in an advertisement for
Another example can be seen in an advertisement for Estée Lauder’s Revitalizing Supreme, which reads: “After just 4 weeks the skin of 92% of the women was visibly enhanced” (see appendix 2). It is interesting to see that in both academia and in everyday life, the examples of social norms being used in commercial advertising are countless, whilst empirical proof for their success is missing. The continuing use of social norms in commercial marketing is most likely due to the undoubted success of the theory of conformity and past experiences with the practice. Brands know that people will follow the majority and since other brands have successfully shown this to be true, new advertisements are crafted in similar fashion. There is however another theory which may be able to help explain the success of social norms in commercial advertising, and this is the Elaboration Likelihood Model (ELM), as discussed by Petty and Cacioppo (1983).
2.3 The Elaboration Likelihood Model
What the ELM states is that there are two routes to persuading a consumer to buy your product: a central route and a peripheral route (Petty and Cacioppo, 1983). Which of the two routes is followed is determined by answering two questions: 1) is the consumer motivated to process the marketing message?, and 2) is the consumer able to process the marketing message? Motivation is determined by both internal (i.e. need for cognition) and external (i.e. personal relevance) factors. This is also the case for ability: internal factors encompass aspects such as familiarity and external factors encompass aspects such as distractions, complexity and repetition (Petty and Cacioppo, 1983).
If a consumer is both motivated and able to process the message, the central route is followed. When this is the case, aspects such as argument quality are of prime importance. When the consumer is either not motivated or not able to process the message (or both), the peripheral route is followed. When this is the case, peripheral cues are able to persuade the consumer. These cues include, amongst other things, source expertise (i.e. a man in a white coat is perceived as an expert and thus more trustworthy), endorser fame (i.e. a celebrity’s opinions can weigh heavily on the consumer) and quantity of argument (i.e. nine arguments are better than three, regardless of their content) (Petty and Cacioppo, 1983).
2.3.1 Low involvement products
One instance in which the peripheral route is (almost) always followed is when the advertised product does not grab the consumer’s attention (Petty and Cacioppo, 1983) and thus involvement is low. This could for example be an advertisement for female hygiene products, being observed by a
man. Or in a more general sense: a product which is assessed to have low risks associated with the purchase of it (Rossiter et al., 1991). A product which has low involvement scores is one of which consumers often have a “try the brand and see” attitude, meaning that there is no harm in trying an unknown brand for once (Rossiter et al., 1991). This attitude is often driven by low financial costs, low potential harm and ease of replacement. A selection of product categories which belong to the low involvement level are salty snacks, barbecue sauce, beer, paper towels, donuts and shampoo (Ratchford, 1987).
In advertisements for such products as the ones listed above, peripheral cues can be very effective. A famous baseball player can be shown putting the advertised brand of barbecue sauce on his hamburger, a long list of arguments stating why the advertised brand of paper towels is most effective can be given or a man in a white lab coat (representing an expert) can be shown explaining why the advertised shampoo brand gets your hair the cleanest. Another way in which the audience may be persuaded is by showing a norm to which they can conform. By stating the percentage of customers who enjoy Hunt’s Original Barbecue Sauce or the percentage of customers who only drink Heineken, the customer with the “try the brand and see” attitude may be persuaded into buying the advertised product. In such an instance, a social norm can be said to act as a peripheral cue, trying to persuade consumers who are not motivated and/or not able to process the advertisement properly. When a consumer is however highly motivated and able to process the advertisement, the social norm may be ignored or even seen as a weak argument, since it does not come from a valued source or a highly credible panel.
The success of social norms in commercial advertisements is generally accepted. However, a reasoning behind this assumption is missing. The theory of conformity may explain why people are interested in the opinions of others. The Elaboration Likelihood Model may show in which instances a social norm may be successful. And the success of social norms in intervention techniques clearly shows that it is a powerful advertising tool.
It would however be very valuable to have some proof of the power of social norms in commercial settings, in a similar fashion to research done on social norms in intervention techniques. Some proof which shows that the use of a social norm in an advertisements is truly effective, and drives the attitudes of consumers towards the advertisement and the brand, as well as the consumers’ purchase intentions.
3. THE ROLE OF REFERENCE GROUPS IN SOCIAL NORM MARKETING
3.1 Reference groups
The term reference groups was first introduced in 1942 by Herbert Hyman (Bearden and Etzel, 1982), when he investigated to which subsets of the population an individual monitors him/herself. Many studies have followed involving this notion of reference groups, and across them the definition has not differed a lot. In most cases a general definition of the term comes down to the attitudes, behaviors and preferences of a group of likeminded individuals, to which others can monitor their own attitudes, behaviors and preferences. In simple terms: a group of likeminded individuals, to which individuals compare and contrast themselves.
3.1.1 The use of reference groups in marketing
Reference groups are very often used in marketing communications. They are a product of market segmentation, as a specific segment can serve as a reference group in the creation of marketing material (Englis and Solomon, 1995). The most straightforward way in which marketeers use reference groups in an advertisement is by showing a group member as a spokesperson, using the product and endorsing its results (Bearden and Etzel, 1982). This could for example be as simple as showing a housewife using and praising a certain brand of laundry detergent, hoping that other housewives will follow along. By showing a member of the desired segment to the audience, a demarcation for a specific reference group is created by the company, through which they try to show for who the product is meant and who should use it.
The success of using a reference group in a marketing message is almost taken as a given in marketing. This attitude comes from the fact that market segmentation is believed to be of great value, with a subsequent reference group following the same line of thought. By creating a specific segment (i.e. a reference group) the target audience fitting this description is more precisely targeted, making compliance to the stated group norm a greater possibility (Neighbors et al., 2010).
3.1.2 Different reasons to conform to a reference group
Why people tend to conform more easily to reference groups than to the general population can have three reasons, according to Bearden and Etzel (1982). Firstly, it can be due to informational
influences. In the above given example of the laundry detergent it would imply that a housewife is a
good source of information for other housewives. The fact that the housewife in the commercial is satisfied with the product might take away some risks for the housewives at home, helping them to
make an informed purchase. A second reason for compliance may be due to utilitarian influences. This is a case of gaining rewards, or avoiding punishment by complying to the behavior depicted by the reference group. Imagine that you find yourself in a foreign city, during an important football match: by joining in with the local population you might gain a reward (i.e. acceptance by the local supporters) and avoid punishment (i.e. physical harm inflicted to you by the locals). A third type of influence is value-expressive influence. This can be seen as a classic case of peer pressure: you act like them, because you want to be like them and you want to be accepted by them. Your first ever cigarette could come from this: you liked the people that smoke and you wanted them to like you too, so you lit up a cigarette with them.
In advertising, the first and third kind of influence are most often seen (Bearden and Etzel, 1982). An advertisement can show a certain spokesperson or demarcate a certain reference group, in order to make the information provided more precise and thus more convincing, giving way to informational influences. The example of the housewife endorsing the laundry detergent covers this. When a company wants to use value-expressive influences in their advertisements, they can show an endorser or reference group as being trendsetting, hoping that others will want to join this trend and purchase the corresponding item. An example of this could be showing a group of street basketball players wearing the newest Reebok shoes, hoping that other (aspiring) players will follow their lead.
3.1.3 The use of reference groups in social norms
As discussed above, marketeers can use a certain image (i.e. a spokesperson being shown using the product) to link their product to a reference group. However, it is also possible to use a reference group in text: either in the slogan or the promotional text in the advertisement. One way to do this, is by implementing a reference group into a stated social norm.
A social norm can state what the entire population thinks, does or says, but it can also show how a specific group acts. A certain percentage of students buys product X for example, or a certain amount of women between the ages of 35 and 45 uses product Y. Reference groups are used in such instances, for the same reason reference groups might be used in other contexts (e.g. imagery): it is believed that the likelihood of norm conformity can be increased when the similarities between the group and the individual are enhanced (Goldstein et al., 2008). Just like showing a picture of a reference group or a person belonging to this group, mentioning the group in the text can increase the perceived similarities between the group and the consumer.
Whether a reference group used in a social norm is successful or not in part depends on whether the reference group is appropriate. This point is made in a paper by Burchell et al. (2013), in which the authors state “appropriateness of the reference group depicted” as one of the four characteristics needed, to make a campaign based around a social norm successful.
The authors state that the most successful reference group is one which the target group most closely identifies itself with. Or in simpler terms: people most like themselves (Berkowitz, 2004). Some of the most well known examples concern social norms created around a demographic reference group such as gender (Lewis and Neighbors, 2004) and race (Neighbors et al., 2010), or around a psychographic reference group such as group membership (Perkins and Craig, 2002). Such aspects can act as the demarcation criteria around which the reference group is based. These different types of reference group criteria will be discussed next.
3.2 Different types of reference groups
The border on which a social norm is demarcated can have dramatic effects on its effectiveness, just like Burchell et al. (2013) state in their analysis. It is therefore of prime importance for a marketeer to realize which reference group can best be used in the marketing campaign they are designing. There are many different criteria around which a reference group can be formed. Within these different types of criteria, a distinction can be made between mental reference groups and physical reference groups.
3.2.1 Mental reference groups
Reference groups which are formed around a mental basis include reference groups based on demographic, lifestyle or psychographic factors. Such a reference group is based on demarcations which are only present in the minds of the consumer. An example of this could be a demographic segment, which is based on important population statistics such as age, gender or ethnicity of the consumer (Nolan et al., 2008). The demarcation of such a segment exists only in the mind of the consumer, since he/she groups together all people of a certain age, gender or ethnicity, into one single segment.
Countless studies have been performed using demographic reference groups or segmentation. Lewis and Neighbors (2004) for example found that college students often have misperceptions regarding gender-specific drinking norms. The research shows that students, both male and female, overestimate the drinking patterns of their same-sex peers. Both sexes estimate that their peers drink more in volume and in frequency than is actually the case. The boys and girls of the college used
were not grouped together by the researcher, but the respondents formed this group in their minds. The reference group was not actually separated from the rest, but mentally it was seen as a separate entity.
A similar result was found by Neighbors et al. (2010). Next to the above described gender results, they found similar results regarding race. They found that perceived drinking norms were overestimated by about 200%, and that a stronger perceived relationship with same-race peers resulted in more compliance with this perceived drinking norm. Again, the same-race reference group was a group formed in the minds of the respondents, not in a physical research setting.
Both these findings could give way to an intervention campaign, with gender- and/or race-specific norms as intervention messages. Such intervention messages are based on mental reference groups, since the consumer targeted with a message such as “85% of all male UvA-students drink between zero and two beers when at a party” will be asked to mentally form a reference group of all male UvA-students.
That the inclusion of a mental reference group can increase the success of a social norm in marketing communications was shown by Goldstein et al. (2008). In this study, the researchers wanted to see which type of social norm was most effective in inducing towel reuse in hotel rooms. The results showed that a social norm based on gender identity was more effective than the standard environmental message as seen in most hotel rooms. This implies that the inclusion of a gender reference group could be more effective than a generic norm or statement.
Another way to create a mental reference group is by working with psychographic segments or segments based around certain lifestyles and lifestyle choices (Martin, 2011). Such segmentation efforts imply segmentation along the lines of different lifestyles, personality traits or social classes of the consumer (Lesser and Hughes, 1986). Such a split in the consumer base is still unexplored territory when it comes to social norms, but it is used in other market segmentation efforts.
The use of psychographic segmentation has one main benefit: you can create a customer segment which has diverse demographics, but still has some common ground (Lin, 2002). This is specifically useful for companies that want to produce a broad range of products, but still have a segment to focus on. A company can, for example, produce goods for both young and old customers, both men and women, but as a psychographic segment they might focus only on people in a higher social class or with a wealthier lifestyle.
beyond the scope of the experimental set-up. Further research, building upon the incremental findings this paper will put forth, may want to include psychographic elements in their experimental set-up, since reference groups shaped around certain lifestyle traits may be very successful.
3.2.2 Physical reference groups
A reference group can be said to be formed around a physical basis if the boundaries around which it is demarcated are actual physical boundaries. Such boundaries are of prime importance in the practice of geographic segmentation (Lin, 2002). In such segmentation, a market is demarcated by geographical criteria, most often including national, regional or city borders. With such segmentation, demographic and psychographic factors are ignored, and only geography plays a part. In recent literature, reference groups based around certain geographic factors have gotten increased attention. Especially location-based and proximity-based reference groups have been topics which are now starting to get more and more attention, quite possibly due to the rise of interactive media (i.e. social media and internet marketing), which gives more possibilities to employ such reference group criteria. One of the most interesting topics regarding these types of reference groups is discussed by McAleney et al. (2010). These authors discuss the topic of a sense of locale. What they mean by this, is that a social norm can be created which portrays the norm of a certain neighborhood. Such a sense begins with a norm based on local figures, such as “Your energy consumption was above average in your neighbourhood” (Burchell et al., 2013). However, to really give it a sense of locale McAlaney et al. (2010) suggest that the message has to be accompanied by more local information. They suggest using real members of the given target group as models, including logos of local institutions and using photographs of local landmarks in the messages. When a sense of locale is given to the message, the reference group people will relate to is their neighborhood and the people comprising it. This reference group is demarcated with a physical boundary, since the attention of consumers is pointed towards their specific neighborhood and the people comprising it.
Another geographically related type of reference group regards physical proximity. In their 2008 paper, Goldstein et al. discuss the difference between the provincial norm and the global norm, and their different attitudes towards towel reuse in hotel rooms. The global norm implies the towel reuse of all the guests in a specified hotel, whilst the provincial norm they applied regards the towel reuse of the guests who have stayed in a particular room. Their findings suggest that in this particular setting, the provincial norm outperformed the global norm (as well as the gender-specific norm and the standard environmental message). The researchers state that this is not as surprising as it may
seem, since people comply to the provincial norm in many occasions. People for example act very differently at a fraternity party than at a country club dinner, as well as differently in a public library than in a commuter train (Aarts and Dijksterhuis, 2003). These examples are in essence examples of provincial norms being obeyed.
Although the sense of locale approach and the provincial norm are both examples of physical and geographic reference criteria, their workings differ a bit. The provincial norm is flexible and moves along with the consumer. The group to whom the consumer relates changes as their whereabouts change. The location the group is based on can be in close proximity at a particular instance, but foreign at other times (e.g. a hotel room or a public library). Take for example a businessman in a hotel room in New York City. In this instance, the physical reference group can be the other guests who have stayed in that room or the people currently staying in the hotel. The consumer may only be in New York for one night, but according to the provincial norm the reference group he/she will relate to still revolves around that one hotel room and its direct surroundings.
The sense of locale approach is much more static. This approach implies creating a marketing message based around a certain community, or a given location which the consumer inhibits on a regular basis, such as their neighborhood or possibly their workplace. This may mean that the businessman staying in the New York hotel room, but who is actually from the Dutch town of Leiden, will see an advertisement in one of his magazines which states “90% of the inhabitants of Leiden subscribe to this magazine”. This slogan may be accompanied with the official city logo, a photo of the city’s mayor with in the background the landmark Hooglandse church. By creating this whole message around his hometown, the consumer may be intrigued to purchase a subscription to the magazine he is reading, even though he is currently not in the location around which the norm is based.
The distinction between these two different types of physical reference groups is very important. The provincial norm is purely physical, and the demarcations of the reference group move along as the consumer moves between places. The sense of locale approach on the other hand is more of a mental reference group: you monitor yourself to a certain set of people who may be demarcated with physical borders (e.g. the city limits of Leiden), but who you group together in your head (i.e. even when sitting in a New York hotel room you may see the city of Leiden as one big reference group). Whilst both type of reference group rely on geographic aspects, they differ in their reliance on physical presence and proximity.
3.3 Proximity towards a reference group
When creating a reference group to base a marketing message around, you are faced with a decision regarding proximity (Goldstein et al., 2008). Such proximity refers to how focused the demarcations of the reference group are. You can have a very broad scope, focusing on a large portion of customers. Or you can have an extremely narrow scope, making the message as personal as you possibly can. When looking at proximity there are two types of proximity to consider: mental and physical proximity.
3.3.1 Mental proximity
Mental proximity can be investigated in relationship to mental reference groups. This type of proximity regards the scope of the characteristics which are portrayed. A low proximity (i.e not close to the consumer we’re focusing on) regarding age, can for example be “people between the age of 20 and 40-years old”. A high proximity (i.e. very close to the consumer we’re focusing on) is much more detailed: “people of the age of 25-years old” for example. A reference group with high proximity is one which speaks to a much smaller audience, making it more specific but limiting the people that can be reached with it.
Another way in which mental proximity can be narrow down is by introducing more than one criteria along which the reference group is created. This implies that apart from stating “people between the age of 20 and 30-years old”, you add additional criteria: “who are of Dutch decent, work in finances and are divorced”. By increasing the number of criteria from one (i.e. age) to four (i.e. additionally nationality, occupation and marital status) the demarcation of the reference group becomes much more specific.
One article in which mental proximity is discussed is that by Borsari and Carey (2003). In their article, the authors state that in literature concerning college drinking habits the reference groups used have differed severely in proximity. Whilst one discusses drinking habits of “your best friend” (i.e. high proximity), others simply state “most students” (i.e. low proximity). These two examples serve as an additional depiction of how a reference group can be based around either very narrow descriptions or very broad ones.
3.3.2 Physical proximity
Physical proximity has more to do with location-based and thus physical reference groups. With physical proximity, the area around which the reference group is formed is made smaller. Low proximity could entail “consumers in The Netherlands”, whilst an example of high proximity could
be “consumers in De Jordaan (a district in Amsterdam)” or even “consumers on the Nieuwe Achtergracht (a street in Amsterdam)”.
Next to geographical areas, other locations such as rooms or buildings can be used as well. Goldstein et al. (2008) for example find that a message that reads “75% of the guests who stayed in
this room participated in our new resource savings program” led to a higher percentage of towel
reuse than the message “75% of the guests participated in our new resource savings program”. This study implies that a closer geographical reference group can have an increased effect on conformity. Another way in which this could be applied is in research concerning college drinking habits (e.g. Borsari and Carey, 2003; Perkins and Craig, 2002). A marketing slogans could read “students in this dormitory drink a maximum of four beers while at a party”, instead of claiming “student on this campus drink a maximum of four beers while at a party”. In this way, the physical proximity to the reference group is narrowed down, possibly making it more efficient.
3.4 Issues regarding proximity
When trying to establish a high proximity for the target group, it could be that privacy becomes an issue or that a reference group becomes unrealistic or not properly demarcated (Burchell et al., 2013). Such issues can decrease the effectiveness of the social norm on the chosen audience. In this section two of such issues are discussed: the reactance issue and privacy issues.
3.4.1 The reactance issue
In a paper from 1966, Venkatesan looked at the relationship between conformity and independence. His first finding supported the notion of conformity: when we cannot see a difference between several options presented to us (i.e. in this experiment it were three identical suits) we have a tendency to go along with the group norm. However, his second finding introduces something known as the reactance issue. This notion implies that once we get a feeling that we are being pressured into picking a certain option, people will go against group norm. This second finding is supported by showing that once multiple confederates stated that their choice was purely based on the choice of the person before them (i.e. “Well, you guys chose B. Although I am not sure, I am just going along to be a good guy. I choose B too.” (Venkatesan, 1966)) the test person chose any of the three options presented to him/her a statistically equal amount of times. This finding implies that when a person feels pressured towards a certain option, he/she will refuse to take this as a point of advice or possibly even deliberately go against the group’s choice. The target audience will thus
go against the group norm, since it is being forced onto them and they don’t want to conform to such pressures.
The topic of reactance is untouched territory with regard to social norms and reference groups. However, when we translate the findings into a reference group scenario, it could mean that if a reference group gets too specific and if the norm implies no other choice available than the one provided, that the customer might turn its back on the message. It might imply that a 32-year old male lawyer, of Dutch decent, living in Amsterdam will feel that the claim “the majority of 32-year old Dutch male lawyers in Amsterdam prefer brand X” is too specific. He might feel as if he is being forced into a certain norm, instead of leaving him with a choice. If this is the case, he might decide to go against the norm, and settle for another brand instead.
3.4.2 Privacy issues
With the rise of the (mobile) internet, two interesting topics have arisen for marketeers. First of all, the internet has opened the door to personalized marketing (Chang and Vowles, 2013). With such efforts, marketeers can craft marketing communications which are specifically made for a tiny niche market or even a single individual. All the publicly available data on users (e.g. age, gender, interests) can be used to target them with specific offers and tailor-made communications. The second topic which has arisen for marketeers is more of a burden than a blessing, since it concerns the never ending discussion on online privacy. This is the second issue regarding proximity.
For years now, critics claim that the internet can be an unsafe place. In a recent online article for the Harvard Business Review, Wang (2013) discusses the trend of trading privacy for convenience. People leave large amounts of private data on their phones, tablets and PCs (i.e. passwords, credit card numbers, contact information), to make their transactions easier than before. This convenience comes with the price of sacrificing privacy, since all off this data can be accessible, either because it is stored on public domains or because technology savvy individuals can easily crack the storage system. It is the data in the public domain which is of most importance to marketeers. This is the data which they can use to create customized marketing communications, to target ever smaller niches. Marketeers can increase proximity by for example analyzing the age, gender, occupation or interest of a person, whilst they can come closer geographically by looking at their address or their current location (i.e. gained through mobile activity). This data, used to narrow down proximity, is also the data about which a lot of people complain, since they argue that it is not there to be used for business, but only for pleasure.
The information gained by marketing organizations is often collected through a method of improper
collection (Wang et al., 1998). This implies that information from users is being stored, without
given them notice of this action. The collected data is than used to pinpoint certain niche segments, with specific marketing communications. Facebook, the social media website most often in the spotlights when privacy issues are discussed (Boyd, 2008), even goes as far as to monitor users’ status updates to see whether they are pregnant (Delo, 2012). If one of the two partners in a relationship posts multiple messages concerning pregnancy, Facebook will start to show advertisements for brands such a Huggies diapers and other baby-related items.
Although such efforts as the one described above can be seen as perfect market research and very effective market segmentation, it is often not greeted as such. Most internet users see it as an invasion of their privacy, once they start to receive such personalized marketing efforts, although Facebook’s privacy terms (as well as those of other social network websites) clearly state that your information is available for such usage. Once pinpoint precise marketing communications are perceived as an invasion of privacy, it could lead to a reactance reaction (see section 3.4.1): customers might opt to buy another brand (e.g. Pampers diapers), since they feel that the advertised brand (e.g. Huggies diapers) is reaching out to them with wrongly acquired information and is thus forcing themselves onto the customer. This could make the efforts of the most effective way of marketing segmentation worthless, since people will be more often scared away from the product than attracted to it.
In the past two chapters, all relevant literature on the topics of conformity, social norms, reference groups and proximity of these reference groups has been addressed. With this literature in mind, the gaps in the scientific literature concerning these topics can be addressed. These gaps function as the main inspiration for the hypotheses which will be tested in this paper. These hypotheses will be introduced in the next chapter.
