The effect of scarcity on the valuation and buying behavior of embarrassing products.

(1)

The effect of scarcity on the valuation and buying behavior of embarrassing

products.

University of Groningen Faculty of Economics and Business

(2)

Abstract

The present study is built upon previous research in the domain of persuasion techniques. With the focus on making a product more valuable by either making it less available due to supply or due to demand. The study explains the effect of scarcity on the so called non-embarrassing and embarrassing products. An experimental 2x2 partly between (level of embarrassment) and partly within (scarcity) subject-design was used to test the conceptual model. The results show that scarcity has an effect on both non- and embarrassing products, and also increases the valuation of embarrassing products. Contrary to expectations, the analysis shows that scarcity works better for the age group older than 24, compared to the group below 24. This might be because of the chosen products in this study. In the end, the author discusses the implications for enterprises and academia, to conclude with future research directions.

Keywords: Scarcity, Abundance, Scarcity due to Demand, Scarcity due to Supply, Persuasion

(3)

1. Introduction

Multiple times a day people have to choose between different options like what they want to wear that day, if they really need to buy a new product, what they are going to eat and so forth. Past research shows that because of all these choices the human brain gets depleted which leads to impulsive buying behavior and other impulsive behavior (Vohs & Faber, 2007). Marketing departments of companies use the depletion of the human brain to steer consumers to the direction the company wants them to go. An example of depletion is the use of heuristics also known as lay theories. For instance: when something has a high price consumers’ expectation and behavioral performance increases. Wright, Hernandez, Sunder, Dinsmore and Kardes (2013) found that heuristics also works for non-price variables like set size, availability of the product, packaging and taste. Research has shown that scarcity works for products, opportunities and information (Cialdini, 2001; Cialdini & Goldstein 2004).

The central interest in this paper is the scarcity principle from Cialdini (2001) in combination with embarrassing products. Scarcity says something about the availability of a product and can have two different forms: scarcity due to supply: limited edition or scarcity due to demand: nearly sold out (Cialdini, 2001; Cialdini & Goldstein, 2004; Gierl & Guettl, 2010). Scarcity is known for making a non-embarrassing products more valuable, which will lead to more sales (Cialdini, 2001; Cialdini & Goldstein 2004). What is yet still unknown is the applicability to embarrassing products. Brehm (1966) mentioned that when people can have less from something, they feel like their hands are being tied and they have the urge to do it instead. More or the less the same as we all did in our childhood, nothing was so much fun as the things forbidden by our parents. The psychological reactance theory is in line with the scarcity principle of Cialdini (2001), people will react strongly when they think or get information about their options being lessened in the near future.

(5)

Embarrassment is known in the literature as a reaction to, or the anticipation of, a negative evaluation of oneself by a real or imagined audience. People will feel embarrassed when they have to buy or use, even online, services, products which are not standard in society or at least kept secret from each other because they want people to think good about them (Dahl, Manchanda & Argo, 2001; Folkman & Lazarus, 1988; Miller, 1995; Miller & Leary, 1992; Sabini et al., 2001). When people know they have to buy an embarrassing product they will try to minimize the embarrassment by going to stores with less customers, waiting for other customers to leave the aisle, certainly not asking for help from employees and adding more items to their basket to distract the attention from the embarrassing item (Blair and Roese, 2013). As mentioned before scarcity can have two forms: due to supply or due to demand, where supply stand for limited editions and demand for other customers bought it multiple times. Scarcity due to demand is interesting in this case because the researcher wants to find out if an embarrassing product can be made less embarrassing by scarcity. In the way that people might think: “others have bought it as well, why feel embarrassed?’. The researcher wants to find out if scarcity due to demand can decrease the feeling of embarrassment and lead to a higher buying intention for those products.

Interestingly, there is also a difference of compliance between age groups, this is due to the fact that the coping mechanism of younger people are still at the developmental stage (Frydenberg, 1997; Grant and Walsch, 2009). When making a decision at a younger age the decision might differ from adults because the difference of relevant information and the amount of information they actually use when forming a decision. This might lead to less thorough thinking about marketing tactics and more compliance with lay theories (Lewis, 1981; Tester et al., 1987). Furthermore, adolescents are believed to be more receptive to peer influence (Berndt, 1979; Steinberg & Silverberg, 1986). This could be interesting whether these effects still occur in the present research, as this is not tested with embarrassing products yet. One could argue that if adolescents are more receptive to peer influence, it will lead to a stronger effect of scarcity, which could lead to a higher purchasing behavior. In this situation they might feel that when lots of people bought the product it might not be as embarrassing as they thought initially.

(6)

find an effect that makes embarrassing products less embarrassing for consumers. Specifically, the objective here is to find when and if scarcity works with embarrassing products and when, if at all, it ceases to work.

The study focused on online and offline ordered products in a store, the products will be shampoo as non-embarrassing product versus lice shampoo as an embarrassing product. Hence, I hypothesize that embarrassing products and age of the consumer may affect the buying behavior of the consumer while the product is available in scarcity. To summarize, this study will provide insights on how to act as marketeer, to increase the buying behavior of consumers with embarrassing products. To investigate this particular topic, the research question is formulated as follows:

How can marketeers utilize the psychological persuasion technique scarcity to influence consumers’ buying behavior for embarrassing products?

Following sub-research questions are derived from the main research question:

1) Does the persuasion technique scarcity work better for younger age groups?

(7)

2. Literature Review

In the next chapter, the author briefly reviews existing research on the topics of embarrassment, persuasion technique scarcity and age of the consumer. This study focuses on the effects of scarcity on the buying behavior of the consumers with embarrassing products. Furthermore, it will be studied if at all and how this effect might be moderated by embarrassing versus non-embarrassing products and different age groups.

2.1 Persuasion technique scarcity

The scarcity principle of persuasion by Cialdini (2001) shows that people want more of what they can have less of. Scarcity gives the consumers information about the product. A study of Gierl and Huettl (2010) considers two types of scarcity: due to supply (e.g. limited edition) or due to demand (e.g. only a few units remain). They found that scarcity also works with conspicuous consumption, because it signals to limited supply and will make the products look even more valuable. As mentioned before, scarcity says something about the availability of the product and can infer quality because of the short supply or high demand. When scarcity is caused due to high demand, consumers think that the product is popular, of high quality and want to be part of the many others who already own the product as part of social validation (Bearden & Rose, 1990; van Herpen et al., 2009; van Herpen et al., 2014). When scarcity is caused due to limited supply it infers to consumers that the product is exclusive and of high quality, which can stimulate a desire to be quick and to be one of the happy few to own it (Fromkin, 1970; van Herpen et al, 2014).

To quote another paper by Aggarwal, Youl Jun and Ho Huh (2011): ‘Constraining the opportunity to own or experience an object signifies a loss of freedom. In an attempt to negate this loss, people have a tendency to desire objects on which such limitations are placed’. A result of their research is that scarcity messages due to demand raise a product’s perceived value and influences the consumer’s intention to purchase the product (Aggarwal et al., 2011). This is due to the heuristic cue that enables them to infer that if there is less of something, it is more valuable (Cialdini, 2001; Cialdini and Goldstein, 2004; Brehm, 1966).

(8)

cookbooks (Verhallen, 1982). In both studies the scarce product was rated with a higher perceived value and willingness to buy.

Cialdini showed that products, opportunities and even information can be seen as more valuable when they become less available. First, scarcity can be explained as a mental shortcut or lay theory, it provides between something’s availability and its quality (Cialdini, 2001; Cialdini & Goldstein, 2004). The mental shortcut is the lay theory people have when something is limited available (e.g. due to supply, like a limited edition) they think it must be of higher quality (Parker & Lehmann, 2011). Second, in terms of the idea when opportunities become less available we lose freedom. According to Brehm (1966) and his theory of psychological reactance, when our freedoms are restricted or threatened at some point we will do a lot to get this freedom back, which is in line with in the scarcity principle. With a specific focus on that what was being limited for us in the first place.

Interesting and relevant to this paper is the research of Wright, Hernandez, Sundar, Dismore and Kardes (2013) which used the study of Shiv, Carmon & Ariely (2005) as a basis. The Shiv et al. (2005) study showed that high price increases consumers’ expectation and enhances behavioral performance. The study of Wright et al. (2013) showed that they found the same effects as Shiv et al. (2005) but with non-price variables like set size, scarcity (i.e. availability), packaging and taste. With the scarcity study they manipulated the shelf with energy drinks equal to that of competitors versus half of it with a fictitious American energy drink. Then they asked 109 Americans to consume the beverage before completing a word puzzle. They found that participants randomly assigned to the scarce condition outperformed (i.e. made more and longer words in the puzzle) participants assigned to the abundant condition (Wright et al., 2013). This is relevant for the present paper in front of you, because they found that people have a lay theory about scarcity as well. Because of the scarcity heuristic participants thought the scarce energy drink was better than the condition with abundance and as result they performed better in the puzzle (i.e. more and longer words) than the condition with the energy drink in abundance.

When consumers need to buy a product they have no preference for, they are likely to use heuristics. This is because they have no information about the product being good or bad. They will look at things like shown in the Wright et al. (2013) and Shiv et al. (2005) studies: the cheapest one out there, most stylish package, biggest set size, availability compared to other products and so on.

(9)

with multiple wines and one of the wines, a Bordeaux with a classic aroma and pleasant tannins, is almost but not completely sold out. It stands on a nearly empty shelf, because others apparently bought this particular wine today (van Herpen et al., 2009). Consumers will probably have a preference for this scarce wine from Bordeaux because of the bandwagon effect/herd behavior, this effect occurs when consumers follow the behavior of others (Banerjee, 1992; Bikhchandani, Hirschleifer, & Welch, 1998; Corneo & Jeanne, 1997). Such behavior may even occur when consumers do not directly observe behavior of others, a nearly sold out shelf is also an indicator for consumers to behave in the same way as others did before them (van Herpen et al., 2005; van Herpen et al., 2009).

Former research has proven that scarcity can influence behavior of consumers for products which are non-embarrassing. Still, no answer on the question whether this also works for embarrassing products. Though, the researcher believes that for embarrassing products scarcity will work as good or even better than for non-embarrassing products. Mainly because scarcity uses the bandwagon effect / herd behavior of consumers and this could lead to less embarrassment for the customers. The feeling of embarrassment might be lower when choosing a product which is almost sold out: ‘others have bought it, why feel embarrassed?’. Consequently the purchase might be easier, less embarrassing and less stressful for consumers. Therefore, this paper focuses on finding an effect of scarcity with embarrassing products.

To conclude, with both scarcity due to supply and scarcity due to demand people feel that they are missing out if they do not purchase it, which might be a good thing if we speak about embarrassing products. In this paper the researcher chose to focus on the scarcity due to demand. So, when the availability is scarce due to demand people might have a lay theory that it will be of higher value, more popular and people feel less embarrassed to purchase the product because many people bought it before and therefore sales will increase of embarrassing products. Yet, too little research is done about the consumer buying behavior comparing the purchases of embarrassing product and non-embarrassing products while available in scarcity (e.g. due to demand) to prove this effect. In this research we expect to find that scarcity due to demand increases buying behavior for both the embarrassing and the non-embarrassing product and is the same for both conditions in abundance. Furthermore, it is expected that scarcity will lower the embarrassment for both products in scarcity compared to the products in abundance.

Hence, it is assumed that:

(10)

2.2 Embarrassment

During the day we sometimes feel awkward because of what we did, suddenly we start to blush or smile to cover up the feeling that we have made a fool of ourselves. In this exact moment we are experiencing embarrassment. Embarrassment is known in the literature as a reaction to -or the anticipation of- a negative evaluation of oneself by a real or imagined audience, leading to an ‘aversive and awkward emotional state’ and is associated with situations that are perceived to be threatening and often involves fearful or anxious feelings (Dahl, Manchanda & Argo, 2001; Folkman & Lazarus, 1988; Miller, 1995; Miller & Leary, 1992). Embarrassment is an emotion that people feel at some stage, given the fact that people are influenced by what others think of them about what they do (Dahl et al., 2001; Folkman & Lazarus, 1988; Miller, 1995).

Dahl, Manchanda and Argo (2001) identified this emotion of embarrassment in three different situations, during purchase, usage situation and during disposition. For all three situations the feeling of embarrassment is driven by a concern for what others think about the person who feels embarrassed (Miller, 1996; Modigliani, 1971). The feeling involves a threat to the individual’s self-image resulting from unwanted evaluations from real or imagined audiences (Edelmann, 1987; Miller and Leary, 1992). First, one can feel embarrassed during the purchase of embarrassing products like condoms. This is because people are thinking about what others might think of them in a negative way and people always want others to think positive about them. They do not see others buy condoms regularly, so they will think it is weird to buy condoms. Second, one might feel embarrassed in a situation when the credit card ceases to work while paying the bill in a fancy restaurant. People do not want to be in a situation that people think they do not have -enough- money to pay the bill. People would have the feeling about their self that they are not good enough. Or lastly, during disposition of something embarrassing like returning an adult video. People feel embarrassed that they for instance did hire an adult movie and will think that other people might think that it is not normal to do something like that. In this study we are focused on the emotion during purchase of an embarrassing product.

(11)

Verhallen (1982) stated that the scarcity heuristic led to a higher perceived value and likelihood to purchase. People think when a product is scarce it is bought often, then they will think it is of high quality and in the end are more likely to buy it themselves. Furthermore, it shows the social validation of the product: others have bought it so the product is safe to buy, the bandwagon / herd behavior (Banerjee, 1992; Bikhchandani, Hirschleifer, & Welch, 1998; Corneo & Jeanne, 1997). For embarrassing products, the researcher expect that this could be positive. When the availability is scarce due to the high demand I suspect to find a positive effect in the buying behavior of consumers.

In this paper the goal is to find a solution to make people feel less embarrassed when buying embarrassing products and simultaneously to increase the sales of embarrassing products. This is important because the scarcity heuristic is widely adopted by marketeers for non-embarrassing products and multiple research has shown that this leads to more sales. Yet, such tactics are, as far as the researcher is concerned, not used for embarrassing products. This is interesting, because it is expected that scarcity can decrease the perceived embarrassment and increase the buying behavior.

The researcher suspects that scarcity due to demand will lead to a more favorable position to the embarrassing product because people will think: ‘I am not the only one, so why feel embarrassed’. If people see that the embarrassing product is bought multiple times compared to other products on the shelf there is reason to believe that the embarrassing product is not embarrassing anymore. Why feel embarrassed if multiple other people bought it as well? Therefore, I expect that the emotion of embarrassment people have with embarrassing products versus non-embarrassing products will decrease.

Hence, the following hypothesis:

H2a: The scarcity manipulation weakens the negative effect of embarrassment on the willingness to buy

H2b: The scarcity manipulation leads to an increase in willingness to buy

2.3 Age group

(12)

make more risky choices compared to adults (Gardner & Herman, 1991). This implies again that the process of decision making is different for younger people. Furthermore, research of Frydenberg (1997) has shown that customers at a younger age are more likely to be affected by embarrassment than their older counterparts. Frydenberg (1997) states that this is because of the coping mechanisms of younger people are at the developmental stage. Furthermore, stress related to embarrassment is felt more intensive by younger people then adults (Grant and Walsh, 2009).

In the book ‘Development of Life’ written by Newman & Newman (1975) they have made different widely adopted group names for all ages. The three most important groups for this research are: later adolescence (18-24), early adulthood (25-34) and middle adulthood (35-60). Because of the developmental stage of the coping mechanisms of younger people and the susceptibility for peer influence I suspect that scarcity will have a greater effect on later adolescence consumers. Thereupon, the researcher believes that because of the peer influence the perceived embarrassment will be lower for the age group 24 and below. They will see that the product is bought earlier by others and will jump on the proverbial ‘bandwagon’ (Corneo & Jeanne, 1997; van Herpen et al., 2005; van Herpen et al., 2009).

As mentioned above, past research states that younger people behave different than adults in their decision making. They are more prone to peer influence and their coping mechanisms are in the developmental stage, which could be positive for marketeers when using heuristics. However, as Grant and Walsch (2009) mention: ‘stress related to embarrassment is felt more by younger people compared to adults’. Furthermore, the researcher thinks the reason why it is more likely later adolescence compared to adults will fall for marketing tricks easier is because of susceptibility for peer influence and later adolescents possess and use less information in their decisions making. So, later adolescents might be more sensitive to heuristics like scarcity, but on the other hand more receptive to embarrassment. Peer influence have an effect on both: scarcity with peer influence means that later adolescents might think that it is a good product and value it higher because it is bought more often. For embarrassment and peer influence it could mean that later adolescents feel embarrassed in a potential embarrassing situation, because they think they have to feel embarrassed because others do so. When combined, the scarcity might lead to a lower embarrassment because the younger people see that the product is bought often, so it might be not embarrassing after all.

(13)

and they will feel that others have approved the product.

Hence, the following hypothesis:

H3: The effect of scarcity on the willingness to buy embarrassing products is higher for young adolescents than adults.

2.4 Consumer buying behavior

To be able to compare the outcome of different hypotheses, consumer buying behavior is used. This is an measure to compare different options of products and will say something about the product offered. If a product is not good enough it will not be bought and vice versa. The dependent variable consumer buying behavior refers, according to Solomon (1996), to “the processes involved when individuals or groups choose, buy, utilize or dispose products, services, concepts or experiences to satisfy needs and desires”. Consumers will behave in a way of searching for, paying for, using, assessing and disposing of products and services that they anticipate will satisfy their needs (Schiffman et al., 2012). Furthermore, consumers buying behavior is believed to be a concept that answers questions like what, why, how, when and where an individual’s makes a purchase (Green, 1992). For firms and marketeers is it extremely important to understand the buying behavior pattern of consumers, as it helps to plan and implement superior business strategies (Khaniwale, 2015). In this research the definition of consumer buying behavior contributed by Solomon (1996) will be used.

To conclude a conceptual model is proposed (figure 1) that integrates the concepts of scarcity, embarrassing versus non-embarrassing products, age of consumers with the consumers buying behavior. The aim is to explain the impact of the moderators on the effect of embarrassment on consumer buying behavior.

(14)

3. Methodology

In the current chapter, the researcher will elaborate on the research plan, with the experimental setup used to test all the hypotheses. At first, the sample will be discussed. Second, the procedure used to do the manipulation of the independent variable will be explained. Third, the measurement of the dependent variable consumer buying behavior will be presented. At last, the scales of the moderators will be presented.

3.1 Research Plan

In this study we use the research of Wright et al. (2013) as example. They did the research with an made up brand for an energy drink with two famous brands (Monster & Red Bull) next to it in the refrigerators. In this research the researcher wants to do the same, but then with shelves like one would see in his/her favorite drug store. Same as in the research of Wright et al. (2013) the other brands are always in the same quantity available (in abundance) to create the same environment for every respondent. In this research, the decision was made to make brand A always an lice shampoo brand and brand C always a regular shampoo brand. Brand B is changing between respondent between scarcity vs. abundance and regular vs. lice shampoo. This research used a 2x2 partly between (level of embarrassment) and partly within (scarcity) subject design. Hence, the participants see two scenario’s (first one in block two and the second in block four) in total of which the first one is always in scarcity and the second always in abundance. Due to the randomization in those two scenario’s the product can be embarrassing and afterwards non-embarrassing, first non-embarrassing and in the second scenario embarrassing, two times embarrassing or two times non-embarrassing. See table 1 for all possible scenarios. As already mentioned before, in those four conditions the shelf next to the products will always contain the same products in the same amount of abundance, as seen in the study of Wright et al. (2013).

Level of scarcity Level of embarrassment

Scarcity Abundance

Embarrassing Scenario 1 Scenario 2

Non-embarrassing Scenario 3 Scenario 4

Table 1 all possible scenarios in study

(15)

participant to each manipulation leads the possibility of causal relationships to be determined with a greater certainty and it gives more assurance that personality and/or moods do not have an effect on the results (Aronson et al, 1998).

With a self-made digital survey, the respondents are asked to imagine that they are in a store to buy either an embarrassing product for daily use or to imagine that they have to buy a non-embarrassing product. When seeing the shelf of the product, in the second block, the shelf is either near empty due to scarcity (due to demand), or for the other scenario, in the fourth block, the product is in abundance on the shelf. When the availability of the product is scarce due to demand this will be made clear to the respondent via phrases like: ‘nearly sold out’ or ‘almost sold out’ (Gierl & Huetll, 2010). The respondent is than asked to answer several questions about the buying intention of the product.

3.2 Product choice and category choice

This research uses two products that are known by every participant, so that in any case the respondents will know what kind of product they are hypothetically are going to buy. For the non-embarrassing products, this study used a regular daily used shampoo and for the non-embarrassing counterpart a lice shampoo was used. Prior to the research a pre-test was conducted among 10 people with two statements, each on a 5-point Likert scale from strongly disagree to strongly agree.

The results of the pre-test (figure 2 and 3) show that from the 10 respondents nobody thinks regular shampoo is embarrassing. From those respondents eight think lice shampoo is embarrassing, besides two that are undecided. When asked why, the respondents said that: ‘regular shampoo is something everybody uses’ and ‘Lice shampoo only when you have louse and then you feel embarrassed’. In the daily lives shampoo is a necessity and it is expected that nearly every respondent has bought and/or used regular shampoo. Furthermore, they will know the effect of lice shampoo and when to buy the product. 0 2 4 6 8 Strongly disagree

Disagree Undecided Agee Strongly agree

Regular shampoo is

embarrassing

0 2 4 6 8 Strongly disagree

Disagree Undecided Agee Strongly agree

Liceshampoo is embarrassing

(16)

3.3 Survey procedure

The study is focused on people who buy their own groceries from the shops on a regular basis instead of buying online and getting it delivered at home. The study is in English, as the survey will be spread on English websites and Dutch websites. First, in the first block respondents were asked to complete the demographic questions. Second, in the second block they have to read the introduction of the survey and their shopping scenario. In this introduction the scenario is explained that they are buying shampoo or lice shampoo in scarcity in a shop and on the shelf, they see other regular shampoo or lice shampoo in abundance (on the top and bottom shelve) but the product they are buying is manipulated in scarcity as in the study with energy drinks of Wright et al. (2013). See figure 4 underneath.

Figure 4 example of second block in scarcity

(17)

Figure 5 example of second block in abundance

They were again asked in this block to fill in questions about the buying intention, valuation and the feeling towards the product they are buying. After this block another manipulation check, this one asked which product was central in the survey.

3.4 Materials

The survey was distributed via Qualtrics and includes 15 questions divided in different blocks. Those blocks were: first block contains intro, demographic & generic questions. In the second block two scenarios were possible: either a lice shampoo in scarcity or a regular shampoo in scarcity. The respondents will see only one out of two, due to random assignment. Third block contained the first attention check. Fourth block was about the abundance scenario, two scenarios were possible: either a lice shampoo in abundance or a regular shampoo in abundance. The respondents will see only one out of two, due to random assignment. Lastly the fifth block was again an attention check. The survey completes with a customized end message which explains what was being surveyed.

(18)

(1= extremely unlikely, 7 extremely likely) they would feel during purchasing process (M = 2.97, SD = 1.28, a = .72).

A higher buying behavior in different scenarios could also be because of a higher perceived quality when available in scarcity, according to the scarcity heuristic (Cialdini, 2001; Cialdini & Goldstein, 2004). To check this claim, a scale from previous research was used (Dodds, Monroe & Grewal, 1991; Grewal, Monroe & Krishnan, 1998). Participants were asked if product B appears to be of good quality, reliable and dependable (1= strongly disagree, 7 strongly agree) for the buying situation they were in (M = 4.23, SD = 0.90, a = .93).

The third scale asked the participants about the buying intention they had when buying product B in the type of situation the were in. This scale was chosen because it appeared to be of good quality in past research (Dodds, Monroe & Grewal, 1991; Grewal, Krishnan, Baker & Borin, 1998, Grewal, Monroe & Krishnan, 1998). Participants were asked if product B appears to be of good quality, reliable and dependable (1= extremely unlikely, 7 extremely likely) for the buying situation they were in (M = 4.23, SD = 1.10, a = .95).

After block two and four the respondents got attention check questions. In block two this was about whether the previous question was about scarcity or abundance. In the fourth block the respondents got the second attention check. This questions asked for the product category of the products asked in the survey.

(19)

next to make the questions more clear, to set a maximum of possible answers to 1, as I did to all the other questions. Furthermore Qualtrics had automatically deleted 278 partial responses from the database as they did not finish the survey, because participants did not like to fill in the exact same questions twice. See the limitation section for more information about these mistakes.

After adjustment the sample contains of 29 females and 234 males. The sample shows that the majority of the respondents are between 25 and 34 years old with 35 to 60 years old as the second biggest group. The third group is 18-24 years old and contains 67 respondents. 57 percent of the respondents are employed full time, a total op 150 people, and the second biggest group is student with 68 people or 25.9 percent.

The question how often do you buy shampoo is quite divided, the biggest group is 3-4 bottles a year with 82 respondents / 31.2 percent, second 1-2 bottles a year with 53 / 20.2 percent, third 4-6 bottles a year with 49 / 18.6 percent, fourth >8 bottles a year with 47 / 17.9 percent and lastly 6-8 bottles a year with 32 respondents or 12.2 percent.

The biggest group of the respondents have never used lice shampoo in their life: 64.3 percent or 169 people. 80 participants almost never use lice shampoo with an percentage of 30.4 percent and only 11 people say they have to use it occasionally, around 4.2 percent. To have a clear overview of how the descriptive statistics of the sample are combined in one summarized table, see table 2. See appendix 3 for all the outcomes including frequencies and percentages, presented apart from each other.

Gender Age group Employment

status

How often do you buy shampoo?

Ever used lice shampoo in your life?

Male = 234 (89%)

Under 13 = 1 (0.4%)

Employed full time = 150 (57%)

1-2 bottles per year = 53 (20.2%) Never use = 169 (64.3%) Female = 29 (11%) 13-17 = 4 (1.5%) Employed part time = 18 (6.8%)

3-4 bottles per year = 82 (31.2%) Almost never = 80 (30.4%) N = 263 18-24 = 67 (25.5%) Unemployed looking for work = 15 (5.7%)

4-6 bottles per year = 49 (18.6%) Occasionally/Sometimes = 11 (4.2%) 25-34 = 116(44.1%) Unemployed not looking for work = 7 (2.7%)

6-8 bottles per year = 32 (12.2%)

(20)

35-60 = 70 (26.6%)

Retired = 1 (.4%)

More than 8 bottles per year = 47 (17.9%) Frequently use = 2 (.8%) Above 60 = 5 (1.9%) Student = 68 (25.9%) Disabled = 4 (1.5%)

Table 2 descriptive statistics of the sample

3.5 Analysis plan

The present study aimed to check whether scarcity can be used to influence the buying behavior of consumers for embarrassing products. Together with the two sub questions about if it works more or less for different age groups and if scarcity can make a difference for the valuation of the embarrassing product. To prepare the dataset for analysis the collected data was gathered, screened and after that cleaned. Hereafter, the data was checked for abnormalities and being normal distributed. Next, the dataset was prepared for the analytical analysis.

First, demographic questions will be measured with descriptive statics like mean, mode and median to determine the central tendency. Questions about the scenarios are interval, independent and therefore they will be compared with an ANOVA. All respondents see one questions about each scenario which they are randomly assigned to.

Second, to test the first hypothesis an ANOVA was conducted. Afterwards to be able to make it more visible and consistent with the other hypotheses an mediation model 1 was conducted via the Process Macro method of Hayes (2013). The researcher made two dummy variables with all four scenarios outcomes to check whether scarcity has influence on the buying behavior compared to the scenarios in abundance. The buying behavior as the dependent variable and the new dummy variable of manipulated embarrassment as the independent variable in the model. Scarcity was the second dummy variable as a moderator. The dummy variable of manipulated embarrassment was a 0/1 variable: 0 corresponds with regular shampoo and 1 with lice shampoo. The other dummy variable for scarcity was again a 0/1 variable where 0 stands for abundance and 1 for scarcity. This will imply that scarcity lowers the perceived embarrassment of the so called ‘embarrassing’ products.

(21)

for a moderated mediation analysis, because model 8 checks not only for an effect of the moderator on the mediation but also if the moderator has an effect on the relation between the independent variable and the dependent variable.

Fourth, for the third hypothesis the Process Macro of Hayes (2013) was used with an extra variable to check whether there was an difference between age groups. Manipulated embarrassment as independent variable, buying behavior as dependent variable, scarcity as a moderator and age used for the moderation effect. To derive answers for the last hypothesis and sub questions the researcher used model 3. This is a moderated moderation, because the researcher assumed that the age group below 24 would be more sensitive for persuasion techniques like scarcity. To test this mediation properly, the same dummy variables were used as in the first hypothesis.

(22)

4. Results

This chapter describes the results of the possible relation of scarcity (scarce vs. abundance) and manipulated embarrassment (embarrassing vs. non-embarrassing) on the buying behavior for the products regular shampoo and lice shampoo. After this, the results of the two other models whether scarcity leads to a higher valuation and if age matters when a product is available in scarcity will be presented.

4.1 Missing data and Outliers

To make sure that no data was lost in the gathering and preparation process the data-set has been checked for odd and missing values. No missing data was found, this is due to the validation in the survey. Every questions had to be answered to be able to continue. Furthermore, no continuous scales were used so odd answers of free estimates were not found.

4.2 Hypothesis 1; scarcity as predictor of buying behavior

For the first hypothesis the type of embarrassment in product (regular shampoo vs. lice shampoo) was used to predict an overall rating of buying behavior and scarcity as persuasion technique (scarcity vs. abundance) was expected to moderate this relationship in a sense that it would make a embarrassing product less embarrassing when it was available in scarcity. To got a clear overview of the descriptive statistics see table 3 underneath.

In order to analyze whether the relationship between manipulated embarrassment and buying behavior is moderated by scarcity, a 2x2 ANOVA test was performed on buying behavior. This yielded a significant main effect for scarcity F(1,522) = 12.556; p= .000; Mscarcityregularshampoo = 4.27, SD= 1.25; Mscarcityliceshampoo = 4.54, SD = 1.20. The other main effect manipulated embarrassment was insignificant F(1,522) = ,014; p = .907; Mabundanceregularshampoo = 4.19, SD = ,86; Mabundanceliceshampoo = 3.95, SD = .96. The interaction between manipulated embarrassment and scarcity appeared to be significant F(1,522) = 7,118; p = .008. See figure 6 and 7 for a clear overview of the estimated marginal means of buy.

Level of scarcity Level of embarrassment Scarcity Abundance Embarrassing N= 123, M= 4.54, SD= 1.20 N= 134, M= 3.95, SD= .96 Non-embarrassing N= 140, M= 4.27, SD= 1.25 N= 129, M= 4.19, SD= .86 Total N= 263, M= 4.40, SD= 1.23 N= 263, M= 4.07, SD= .92

(23)

Afterwards to be able to make it more visible and consistent with the other hypotheses an mediation model 1 was conducted via the Process Macro method of Hayes (2013). See figure 8 for the conceptual diagram of the moderated relationship for hypothesis 1 and figure 9 for the statistical diagram of hypothesis 1. Overall model summary for the a paths: F(3,522)= 6.41, p = .0003, R2_{= ,036. Which means}

that the model is significant and the adjusted R2 _{had a value of .036, which is low. Hence, 3.6 percent}

of the variance of buying behavior can be explained by the proposed variables.

First, scarcity was an insignificant predictor of buying behavior (the b1 pathway), b= .0826, t(522)= .63, p= .5314. Which means that it is the opposite of what we found in the AVOVA, this could be due to the interaction effect is stronger and leads to more variance. Second, manipulated embarrassment was almost significant and negative, what was expected (the b2 pathway), b= -,2408, t(522)= -1.81, p= .0715. Which means that the more embarrassing the product the less likely one would buy. However, officially it is not significant (significant when it is <.05), so the researcher cannot say this holds for sure. Most interesting to see is that the interaction effect of scarcity and manipulated embarrassment is significant (the b3 pathway) b= .5036, t(522)= 2.66, p= .0079. Which means that when the product is embarrassing and at the same time available in scarcity it would lead to a higher buying behavior.

3,9 4,1 4,3 4,5 4,7 Abundance Scarcity

Estimated Marginal Means of BUY

Regular Lice 3,9 4,1 4,3 4,5 4,7 Regular Lice

Estimated Marginal Means of BUY

Abundance Scarcity

(24)

Figure 8 conceptual diagram hypothesis 1

Figure 9 statistical diagram hypothesis 1

Hypothesis 1 is about: ‘Scarcity has positive effect on buying behavior and the feeling of embarrassment.’ In the moderated analysis by Hayes (2013) scarcity is not significant, this could be due to the interaction effect with manipulated embarrassment is stronger and leads to more variance. However, with ANOVA, the analyses confirms that scarcity has an influence on the buying behavior which is as well in line with former research of Worchel et al. (1975), Cialdini (2001) and Verhallen (1982). Next, we see in both ANOVA and in Process Macro method of Hayes (2013) that scarcity moderates the feeling of embarrassment (lowers the embarrassment feeling) and afterwards it increases the buying behavior of the respondents. Manipulated embarrassment on its own does not affect the buying behavior at all, there is a small effect which is insignificant.

4.3 Hypothesis 2; scarcity as predictor of valuation

(25)

divided in two questions and tries to find an answer on the following: the scarcity manipulation weakens the negative effect of embarrassment on the willingness to buy (H2a) and the scarcity manipulation leads to an increase in willingness to buy (H2b). The type of embarrassment in product (regular vs lice) was used to predict the overall rating of buying behavior, the valuation of the product as mediator and scarcity as persuasion technique (scarcity vs. abundance) was expected to moderate this relationship in a sense that one would value a product higher and therefore felt that an embarrassing product was less embarrassing, when it was available in scarcity. See figure 10 for the conceptual diagram of the moderated relationship for hypothesis 1 and figure 11 for the statistical diagram of hypothesis 1.

Figure 10 conceptual diagram hypothesis 2

(26)

In order to analyze whether the relationship between manipulated embarrassment and buying behavior is a moderated mediation by scarcity and valuation, the macro process of Hayes (2013) on buying behavior was conducted via the steps of Baron and Kenny (1986). Model 8 was used for this type of moderated mediation analysis instead of model 7, this is because model 8 checks for an conditional effect of scarcity on the relation between the IV and DV.

Overall model summary for the a paths: F(3,522)= 5,73, p = .0007, R2_{= ,032. Overall model summary}

for the b and c’ paths: F(4,521)= 28,27, p = .0000, R2_{= ,18. Which means that the overall model is}

significant and the adjusted R2 had a value of .032, which is low. Hence, 3.2 percent of the variance in value can be explained by the proposed variables. Furthermore, the overall model for the b and c’ paths is significant and the adjusted R2 had a value of .18, which is relatively low. Hence, 18 percent of the variance in buying behavior can be explained by the proposed variables. First, manipulated embarrassment was an insignificant predictor of valuation (the a1 pathway), b= -,1171, t(522)= -1.07, p= .2874. Which means that the more embarrassing the products is, the lower the valuation. However this is not significant, so this does not hold. The analysis yielded an significant result for the effect of scarcity on valuation (a2 pathway) b= ,2370, t(522) = 2,18 p= .0298. However, the interaction effect of scarcity and manipulated embarrassment is not significant (pathway a3) b= ,1470, t(522)= ,3451 p= .3451.

Next, valuation has a positive and significant relation with the buying behavior (b pathway), b= ,4891, t(521)= 7.51, p= .0000. So when the valuation was higher, people would buy it more often. Then, the relationship between manipulated embarrassment and buying behavior was negative and not significant (the c1’ pathway), b= -,1862 ,t(521)= -1,51, p= .1318. The relationship between scarcity and buying behavior is also negative and not significant (the c2’ pathway), b= -,0280 ,t(521)= -,2284, p= .8194. Last, the relationship between the interaction of scarcity and manipulated embarrassment to buying behavior is significant and positive (the c3’ pathway), b= ,4350 ,t(521)= 2,49, p = .0130. Which means that when the product is scarce and embarrassment it leads to a higher buying behavior.

As seen in figure 11, scarcity is a significant predictor of a higher valuation of the embarrassing product. Furthermore, one can see an positive significant interaction effect between scarcity and manipulated embarrassment leading to a higher buying behavior. Yet, the interaction effect of scarcity and manipulated embarrassment does lead to a higher valuation but this effect is not significant.

(27)

of embarrassment and scarcity is positive and will lead to a higher willingness to buy. However, I cannot accept the H2b hypothesis: ‘the scarcity manipulation leads to an increase in willingness to buy’. Same as in the moderated analysis by Hayes (2013) of hypothesis 1 and contrary to the ANOVA, scarcity on its own does not lead to more willingness to buy.

4.4 Hypothesis 3; age as predictor of buying behavior

After finding out that scarcity weakens the negative effect of embarrassment on the willingness to buy and strengthens the valuation of embarrassing products it is interesting to know whether this differs for different age groups. For the third hypothesis, “The effect of scarcity on the willingness to buy embarrassing products is higher for young adolescents than adults”, three dummy variables are used. The type of embarrassment in product (regular vs lice) was used to predict the overall rating of buying behavior, the availability of the product as moderator (abundance vs scarce) and age (<24 vs 24 and above) as moderation of the moderator was expected to moderate this relationship in a sense that one would have a higher buying intention when they were in the group of <24.

The macro process of Hayes (2013) was used with model 3 for this type of moderated moderation. First, see table 4 for the descriptive statistics table extended with the statistics of age. Second, see figure 12 for visual diagram of the conceptual diagram of model 3 for hypothesis 3 and figure 13 for the statistical diagram of model 3 for hypothesis 3.

Level of scarcity Level of embarrassment Scarcity Abundance Embarrassing N= 123, M= 4.54, SD= 1.20 N= 134, M= 3.95, SD= .96 Non-embarrassing N= 140, M= 4.27, SD= 1.25 N= 129, M= 4.19, SD= .86 Total N= 263, M= 4.40, SD= 1.23 N= 263, M= 4.07, SD= .92 Scarcity Abundance Embarrassing <24 N=38, M= 4.21, SD= 1.36 N= 39, M= 4.10, SD=1.01 Embarrassing >24 N= 85, M= 4.68, SD= 1.10 N=95, M=3.90, SD= .94 Non-Embarrassing <24 N= 34, M= 4.70, SD= 1.04 N= 33, M= 4.28, SD= .94 Non-Embarrassing >24 N= 106, M= 4.14, SD= 1.29 N= 96, M= 4.16, SD= .83 <24 >24 Total N= 144, M= 4.31, SD= 1.11 N= 382, M= 4.20, SD= 1.09

(28)

Overall model summary: F(1,518)= 4,71, p = .0000, R2_{= ,06. Which means that the overall model is}

significant and the adjusted R2 had a value of .06, which is relatively low. Hence, 6 percent of the variance in buying behavior can be explained by the proposed variables. First, manipulated embarrassment was an insignificant predictor of buying behavior (the b1 pathway), b= -,2685, t(518)= -1,73 p = .0839. As the P value is only ,0839, we can say it is only by a small margin not significant. If argued in that way, one can see that the more embarrassing the product is, the lower the buying behavior. The second path between scarcity and buying behavior (the b2 pathway) is insignificant, b= -,0213, t(518)= -,1410, p= .8879. The third path between age and buying behavior (the b3 pathway) is again insignificant, b= ,1230, t(518)= ,5691, p= .5696.

After that I checked the different interaction effects starting with the fourth path. The path about the interaction effect between manipulated embarrassment and scarcity on buying behavior (the b4

(29)

pathway) is significant, b= ,8124, t(518)= 3.69, p=.0002. This means that when products are available in scarcity and are embarrassing the buying behavior is higher. The fifth path between the interaction effect of manipulated embarrassment and age on the buying behavior (the b5 pathway) is insignificant,

b= ,0804, t(518)= 2707, p= .7867. The sixth path between the interaction effect of scarcity and age on

the buying behavior (the b6 pathway) is not significant, b= ,4344, t(518)= 1,44, p= .1512. The last path between the interaction effect of manipulated embarrassment, scarcity and age on buying behavior (the b7 pathway) is significant, b= -,1,11, t(518)= -2,64, p= .0085. This means that when the product is in scarcity, embarrassing and the age is below 24, the buying behavior is negative with a value of -1.11.

This is not what I had expected. Literature says that younger people are more prone to peer influence, therefore I expected to see a greater effect for people below 24. However, I checked with an Independent-samples T-test ANOVA to compare the means of the buying intention and the groups below 24 (N=144) had a higher M= 4.31 and SD = 1.11 than the group of 24 and above (N=382), M= 4.20 and SD= 1.09. So this opposite effect could be due to the extra interaction with scarcity and manipulated embarrassment. To see this effect in more detail I have made a table for the different age groups and whether it is in scarcity or not.

Age group Scarcity b t(518) p

24 and higher Abundance -,2685 -1,73 ,0839

24 and higher Scarcity ,5438 3,49 ,0005

Below 24 Abundance -,1881 -,74 ,4583

Below 24 Scarcity -,4851 -,192 ,0557

Table 5 the effect of age and scarcity individually on buying behavior

As can be seen above in table 5, both age groups are insignificant when available in abundance. When the availability is scarce, the group of 24 and higher is significant and has a positive effect. When below 24 and available in scarcity it is almost significant (,0557) and has a negative effect on the buying behavior, which was not expected.

(30)

4.5 Extra analysis; perceived embarrassment as the predictor of buying behavior

For this extra analysis, the macro process of Hayes (2013) was conducted on buying behavior mediated by feeling of embarrassment via the steps of Baron and Kenny (1986) to see if embarrassment has an effect on buying behavior. Model 4 was used for this type of mediation analysis. Overall model summary for the a paths: F(1,524)= 40.83, p = .0000, R2_{= ,072. Overall model summary for the b and c’}

paths: F(2,523)= 6.53, p= .0016, R2_{= ,02. Which means that the overall model for the a path is significant}

and the adjusted R2 had a value of .072, which is relatively low. Hence, 7.2 percent of the variance in perceived embarrassment can be explained by the proposed variables. For the overall model for b and c’ path the model is significant and the adjusted R2 had a value of .02, which is low. Hence, 2 percent of the variance in buying behavior can be explained by the proposed variables. See figure 14 for the conceptual diagram and figure 15 for the statistical diagram of this extra analysis.

First, manipulated embarrassment was an insignificant predictor of buying behavior (the c pathway),

b= -,0032, t(524)= ,9387 p= .9731. The analysis yielded an significant result for the effect of

manipulated embarrassment on the perceived embarrassment (a pathway) b = ,6858, t(254)= 6,39, p= .0000. Which means that when the product was embarrassing, the respondent felt more embarrassing as well. With perceived embarrassment high, the buying behavior was negative and significant (b pathway) b= -,1394, t(523)= -3,61, p= .0000. This means that when respondents felt more embarrassing they had a lower buying intention. Next, the relationship between manipulated embarrassment and buying behavior was increased by perceived embarrassment as an adjustor but insignificant (the c’ pathway) b= ,0923, t(523)= ,9387, p= .3483. A partial mediation effect of perceived embarrassment is found, without any direct effect. Officially, the researcher should have had stopped after seeing that X and Y (c path) is not significantly associated according to the steps of Baron & Kenny (1986). However, it is legitimate to conclude that the mediator mediates the association between X and Y even is the total effect is not significant, according to Hayes (2013).

(31)

The central question in the thesis is: ‘How can marketeers utilize the psychological persuasion

technique scarcity to influence consumers’ buying behavior for embarrassing products?’. During this

thesis we have found that scarcity has its influence on buying behavior, an effect already found in other studies (Cialdini,2001; Worchel et al., 1975; Verhallen, 1982). However, this effect disappears when adding age to the model. According to hypothesis 2 we have found an partial mediation of the valuation of the embarrassing product which is higher in scarcity and this was followed by a lower embarrassment feeling and a higher buying behavior. Yet, the direct effect of scarcity on manipulated embarrassment was not significant. For the first sub question if scarcity works better for younger age groups the answer can be short as well. The researcher does see a minor difference but again this effect is not significant so we cannot say this effect holds. The second sub-research questions can be accepted. This question is about if scarcity has its influence on the consumers’ perception of an embarrassing product. Scarcity leads to a higher valuation of the product: b = ,2370, t(522) = 2,18 p= .0298. With every unit increase the overall rating of lice shampoo in scarcity was ,2370 higher compared to lice shampoo in abundance on a scale of 1-7.

(32)

5. Discussion

Companies nowadays use the depletion of the brain to steer people in the beloved direction of the company to sell more products. They use remarkable proportions of their marketing budget to research the best way to sell product X. Literature already said that scarcity has a advantage for non-embarrassing products, I wanted to find a working environment for this effect with non-embarrassing products as well. In this chapter, the results of the study are presented side by side for every hypothesis, for practical and theoretical meaning. Together with the limitations, future research and in the end there will be a summary.

5.1 Conclusion and recommendation hypothesis 1

For the first hypothesis the main effect is the type of embarrassment used to predict an overall rating of the buying behavior and scarcity was used as a moderation to this relationship. Scarcity on its own had a positive effect and showed that the more scarce the product is the more it will be sold. This relation confirms what is already known from former research of Worchel et al. (1975), Cialdini (2001) and Verhallen (1982). Manipulated embarrassment on its own was not a predictor of buying behavior, when combined with scarcity as an interaction it became a predictor. The interaction effect was positive which means that when the product is embarrassing and in scarcity it will lead to more sales.

5.2 Conclusion and recommendation hypothesis 2

After the researcher confirmed that scarcity has its influence on the buying behavior, he continued to check whether this also accounts for valuation of the product. The second hypothesis is divided in two questions and tries to find an answer on the following: the scarcity manipulation weakens the negative effect of manipulated embarrassment on the willingness to buy (H2a) and the scarcity manipulation leads to an increase in willingness to buy (H2b).

(33)

Only scarcity leads to a higher valuation on itself and together with an embarrassing product this effect is gone. Companies might not want to use scarcity to make a product look more valuable. However, the researcher suggest to use this effect for an increase in sales. The interaction effect of scarcity and manipulated embarrassment might not lead to a higher valuation, but is does lead to a higher buying behavior.

5.3 Conclusion and recommendation hypothesis 3

For the third hypothesis the main effect is the type of embarrassment used to predict an overall rating of the buying behavior and scarcity was used as a moderator and age as an moderation of the moderator to this relationship. Again the model showed that manipulated embarrassment is not a positive predictor and not proven as well in this model for buying behavior, albeit that it was almost significant. Then the model would have found the negative effect which would be what one would expect to find: embarrassing product leads to less buying behavior compared to a non-embarrassing counterpart. The interaction effect of manipulated embarrassment and scarcity in this model is positive and leads to a higher buying behavior. The effect of age cannot be found directly nor indirectly via manipulated embarrassment.

The interaction effect of manipulated embarrassment, age and scarcity on the buying behavior is found but negative. Research suggested that younger people are more prone to peer influence, therefore I expected to see a greater effect for people below 24. Instead in this model it leads to a lower buying behavior. When checking for the averages via an ANOVA the mean of the group below 24 is higher, contrary to the outcome of the model you would still expect that this effect remains in the interaction effect as well. As this is not the case it could be that scarcity leads to the negative effect, as the interaction between manipulated embarrassment and age was positive but not proven. For companies this hypothesis does not really tell much. Yes, age has a higher mean when checking for compliance with buying behavior. However, this effect is gone in the moderated moderation with age included.

5.4 Extra analysis

Furthermore, the researcher checked whether the perceived embarrassment for embarrassing products were different from non-embarrassing products. The results were clear: manipulated embarrassment leads to a higher feeling of perceived embarrassment and this leads to a lower buying behavior. Which is an effect one would expect. Together with the results of scarcity it could be beneficial for companies to make a scarcity marketing strategy around those embarrassing products. This as an indication next to other heuristics for consumers.

(34)

and when it is available in scarcity the buying behavior increases. This could be because people think that when others bought the product as well, they might feel that the product is less embarrassing. The researcher therefore proposes that offering an embarrassing product in scarcity will be beneficial for lowering the embarrassment feeling and together with a higher buying intention it will lead to more sales.

5.5 Limitations and implications for future research

Although, the present research provides interesting insights for companies the study does has several limitations. Those limitations offer opportunities for future research about this topic. In the following, the limitations of the study are discussed in brief and presented along with some implications for future research.

First, due to deadlines and limited time for data collection the study needed a quick and effective method for data collection. The researcher made an self-administrated digital survey used on forums with lots of males (technical and game forums). From past research the researcher knew that they would be willing to take part of the survey. Before the deadline the researcher did meet the minimum criteria of respondents. However, the data is not diverse. More than 85% was male, so the results should be treated with necessary care. For future research it might be advisable to use a more diverse (e.g. age and gender) sample as a starting point. This to avoid sampling bias, when participants do not reflect the targeted population. An unrepresentative sample can have an impact on the quality of the results/findings.

Second, after feedback from multiple respondents it became clear that some survey questions were not clear enough. For instance: the second manipulation check question was not right. The question was: “In this survey, you are asked about two products. Please choose the product category of those products”. Most of the people knew they had to check the option shampoo, but because of the ‘two products’ part they thought they had to choose a second one. So in the end a lot of people chose shampoo and a second answer. This could have been prevented: first, make the questions more clear and second, to set a maximum of possible answers to 1. The manipulation check was used to check whether people are paying attention when filling in the survey. This is important because, otherwise one would get outcomes which might not be true. In this case I cannot be 100 percent sure the participants were serious in the second part, contrary to the first manipulation check. Therefore, the results should be treated with necessary care.

(35)

valuation, perceived embarrassment and buying behavior. In the fourth block they saw the product (regular or lice shampoo) in abundance with the questions about valuation, perceived embarrassment and buying behavior. This means that people saw two times identical questions and sometimes the same product, the only difference was the availability. As it was randomized it could be the case that people saw two times regular shampoo once in scarcity and abundance. This led to a lot of feedback in the topic on the website tweakers.com from people who stopped the survey halfway. In the end it led to a enormous amount of people who stopped: 278 in total. With a total view counter of 3908 views on only the topic on the forum of tweakers.com, this is a big letdown. And the total participants could be a lot higher.

Fourth, the researcher got also a lot of feedback about other heuristics missing in the pictures of the survey. This was chosen to prevent the study from having a lower external validity. The researcher had to make the decision on purpose to steer participants in the direction that they would focus on the scarcity heuristic. This research is about scarcity, if I would have had other heuristics like price, brand names, logos and so forth it would not be clear what exactly caused the effect. If I for instance saw a clear effect from the scarcity group compared to control group, but the group was affected by the price heuristic as well, the data is not usable. However, to make it more applicable to a real life situation the same prices could be added for all three products and make the packages more appealing with colors or logo’s. With this approach the respondent will see the products more like they would see them in the local drugstore, with the exception that they still cannot have a specific product in favor due to price, colors and logos, other than scarcity.

Fifth, more than 64% of the participants have never used lice shampoo in their life. Which is in general nice for them, but this could be bad for the study. It might be the case that they have never thought about headlice/lice shampoo until I asked them to. However, probably everyone knew headlice/lice shampoo and have an opinion whether it is embarrassing to use/buy or not. With participants in the sample who never have used lice shampoo in their life they might not have a high buying behavior now. This might lead to participants having more or the less the same opinion and that will lead to a lower explained variance. Looking at the adjusted R2 _{of the models this could be the case in the study.}

This could be the reason that relation between the manipulated embarrassment and buying behavior was insignificant in every model. Though it is something to take in to account, the results should be treated with necessary care.

(36)

of the times the parents would buy these kind of products. For further research it is best to do another pre-test with embarrassing products. It might be an idea for a next study to compare hemorrhoid cream with sun cream. Or maybe even better to make scenario’s were one would have to buy a cream for certain sexual transmitted diseases (STD) and in the control group they have to buy sun cream.

5.6 Summary

(37)

6. References

Aggarwal, Praveen, Sung Youl Jun, and Jong Ho Huh (2011), "Scarcity Messages: A Consumer Competition Perspective," Journal of Advertising, 40 (3), 19–30.

Argo, J., Dahl, D., & Manchanda, R. (2005). The influence of a mere social presence in a retail context.

Journal of Consumer Research, 32(2), 207-212.

Aronson, E., Wilson, T., & Brewer, M. (1998). Experimentation in social psychology. In D. Gilbert, S. Fiske, & G. Lindzey (Eds.), The Handbook of Social Psychology (99-142). New York, New York: McGraw-Hill.

Banerjee, A. V. (1992). A simple model of herd behavior. The Quarterly Journal of Economics, 107, 797−817.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. Journal of Personality

and Social Psychology, 51, 1173-1182.

Bearden, W., & Rose, R. (1990). Attention to social comparison information: An individual difference factor affecting consumer conformity. Journal of Consumer Research, 16(4), 461-471.

Berndt, T. (1979). Developmental changes in conformity to peers and parents. Developmental

Psychology, 15(6), 608-616.

Bikhchandani, S., Hirshleifer, D., & Welch, I. (1998). Learning from the behavior of others: Conformity, fads, and informational cascades. Journal of Economic Perspectives, 12(3), 151−170.

Blair & Roese (2013). Balancing the basket: The role of shopping basket composition in embarrassment. Journal of Consumer Research, 40(4), 676-691.

Brehm, J. (1966). A theory of psychological reactance (Social psychology). New York: Academic Press.

Cialdini, R. (2001). Influence : Science and practice, 4th ed. Boston, Mass: Allyn and Bacon.

Cialdini, R., & Goldstein, N. (2004). Social influence: Compliance and conformity. Annual Review of

Psychology, Vol. 55 (2004), P. 591.

Corneo, G., & Jeanne, O. (1997). Snobs, bandwagons, and the origin of social customs in consumer behavior. Journal of Economic Behavior and Organization,32 ,333−347.

The effect of scarcity on the valuation and buying behavior of embarrassing products.