“DOES NOVELTY SEEKING BEHAVIOUR INCREASE WITH BEER CONSUMPTION?”

“DOES NOVELTY SEEKING BEHAVIOUR INCREASE

WITH BEER CONSUMPTION?”

Master Thesis, Msc Marketing Management

“DOES NOVELTY SEEKING BEHAVIOUR INCREASE

WITH BEER CONSUMPTION?”

Abstract

Beer consumption is hypothesized to have a positive influence on consumers’ novelty seeking

Table of contents

1 Introduction ... 4 2 Theoretical framework ... 5

Alcohol consumption and novelty seeking behaviour Defining beer consumption

Priming people with beer consumption The moderating role of the prime

3 Research design ... 8 Participants

Independent variables – level of beer consumption Independent variables – primed with beer consumption The Novelty Seeking Behaviour task

Control variables]

4 Results ... 12 Characteristics of variables per condition

Results of the statistical tests

5 Conclusions and recommendations ... 14 Limitations and suggestions for future research

1

Introduction

Although the average beer consumption in the Netherlands declined with 3.6% in 2013 compared to 2012, the Dutch still consume an average of seven glasses or bottles of beer per week. Of these consumers, 46% drank one to five beers per week in 2013, 26% six to ten, and 17% consumed more than ten glasses or bottles of beer per week (Biernet 2014). These facts imply that small to moderate doses of beer play a central role in every Dutch citizen’s life. Nevertheless, not much knowledge exists about how alcohol affects consumer decision making. Zuckerman (2012) suggests that sensation or novelty seeking as personality traits are predictors of individuals’ drug and alcohol consumption levels. However, the aim of this experiment is to test the effect of beer dosage on novelty seeking behaviour, that is, to test the relationship between the two constructs the other way around.

Finding a significant effect of beer intake on novelty seeking behaviour will have important implications for marketers; since the majority of citizens of the Netherlands drink beer on a daily basis, commercials and other marketing outings should be adjusted to this newfound knowledge. On top of that, not only the Dutch drink a lot of beer. As a matter of fact, the Netherlands takes place 15 in the top 20 beer drinking countries. Amongst the countries that know a higher beer intake per capita are the US, Australia, and the UK (CNBC 2015). Consequently, the outcome of this experiment has the potential to affect marketing practices worldwide.

2

Theoretical framework

Novelty seeking behaviour (also known as variety seeking behaviour – e.g. Ratner & Kahn 2002) is described by Chuang et al. (2013, p. 917) as “a mental mechanism that is used to maintain an optimal level of stimulation, which is used to measure the stimulus around an individual”. Legoherel, Dauce and Hsu (2012, p.213) define variety seeking as “a tendency for the consumer to alternate between the different makes of the same product or the quest for diversity in the choice of goods and services”. In essence, consumers are believed to desire variety to become satiated and to prevent boredom (Hoyer, MacInnis and Pieters 2012). To ensure consistency and clarity in this research, ‘novelty seeking behaviour’ is defined as the level of variety that consumers prefer in their choices in order to be optimally stimulated. Novelty seeking behaviour has been identified by several studies as being of major influence on a person’s alcohol- and drug seeking (and/or abusing) behaviour (e.g. Benjamin et al. 1996; Nowak et al. 2000). However, the purpose of this experiment is to identify the relationship between alcohol, or more specific beer, usage and one’s novelty seeking behaviour the other way around. Not much research has been conducted in this field yet, which is surprising since alcohol plays a significant role in many consumers’ lives and might therefore be of great influence on the decisions consumers make on a daily basis.

Alcohol consumption and novelty seeking behaviour

According to Euser et al. (2011) and McMillen, Smith and Wells-Parker (1989) a higher level of alcohol intake results in more risk seeking behaviour compared to no alcohol intake. In other words, individuals take greater risk when they have consumed alcohol (Abrams et al. 2006). Additionally, variety seeking behaviour, novelty seeking behaviour, and risk taking behaviour are all positively associated with each other (e.g. Sharma, Sivakumaran and Marshall 2010; Toledo and Sandi 2011). Thus, if a person’s risk seeking behaviour increases, that person’s level of novelty seeking behaviour will also rise. Since as a result of alcohol consumption consumers are proven to become more risky in their behaviour (Abrams et al. 2006), their novelty seeking behaviour is expected to be affected in a similar way.

These assumptions are translated into to the following first hypothesis of this experiment: H1: A high level of beer consumption leads to more novelty seeking behaviour, compared to a low level of beer consumption.

Defining beer consumption

In order to analyse whether this hypothesis holds or not, the definitions of ‘high’ level of beer consumption and ‘low’ level of beer consumption need to be established. According to the website of Jellinek, a leading organization in addiction treatment in the Netherlands, persons generally feel tipsy after drinking 3-7 glasses of alcohol. They become more careless and less introspective, and start doing things they would not have done if they were sober (Jellinek 2011). Trimbos, another prominent Dutch addiction treatment organization, states that a standard glass of alcohol leads to an alcohol level of 0.2 promille for men (Trimbos Instituut 2013). Moreover, Dutch law dictates that the maximum level of alcohol for car drivers is 0.5 promille, which implies that drinking three or more glasses of alcohol has a noticeable effect on men and women (Rijksoverheid 2015). Therefore, three or more beers are considered a ‘high beer dosage’. In addition, participants that have consumed zero to two glasses of beer fall into the low beer dosage category in this experiment. This line of thinking follows the research by Abrams et al. (2006), in which the safe limit of alcohol intake for driving is used as an estimate of low and high alcohol dosages as well.

Moreover, the focus of this experiment lies on beer intake, and not alcohol intake in general, for several reasons. First of all, every individual’s reaction to alcohol intake is influenced by the

concentration of alcohol in the beverage; different types of alcoholic consumptions contain different percentages of alcohol and may therefore cause individuals to respond to the consumption in different ways (University of Rochester Medical Center 2015; Health Promotion Agency 2014). More specifically, not every type of alcoholic drink will have the same effect on novelty seeking behaviour. Secondly, different types of alcoholic consumptions are expected by individuals to affect them in different ways (Pedersen, Neighbors and Larimer 2010). This will especially affect the second and third hypotheses, which focus on the effect of priming people with consumption. Since beer is the most consumed alcoholic beverage in the Netherlands (Jellinek 2014) and the second most

consumed alcoholic beverage worldwide (World Health Organization 2014), this research explicitly focuses on beer consumption.

Priming people with beer consumption

Van Koningsbruggen and Stroebe (2011, p.807) argue that extensive studies have shown that

have consumed alcohol, regardless of whether they actually consumed an alcoholic drink”. Furthermore, when participants are primed with certain behaviour or stereotypes, they are more inclined to behave consistent with the characteristics of that behaviour or stereotype (Bargh, Chen and Burrows 1996). Furthermore, Oyserman and Lee (2008, p. 313) state that “social cognition as a field has highlighted that human judgment and behaviour are influenced by what comes to mind at the moment of judgment and how this content is interpreted. This salient and accessible information can be content, process, or goal oriented. (…) Content, procedures, and goals can be cued via priming techniques.”

Combining these definitions and insights leads to the assumption that participants who are primed with beer consumption will actually behave in such a way as if they had actually consumed beer even if that is not the case. In addition, McMillen, Smith and Wells-Parker (1989) state that people who are led to believe that they consumed alcohol demonstrate higher levels of risk-taking behaviour – regardless of actual consumption – than do people who are not primed with consumption. As stated before, the concepts of risk taking behaviour and novelty seeking behaviour are positively related to each other (Sharma, Sivakumaran and Marshall 2010; Toledo and Sandi 2011); this implies that people who are led to believe that they consumed alcohol will not only demonstrate higher levels of risk-taking behaviour, but will also engage in more extreme novelty seeking behaviour.

However, Sayette et al. (2012) examined the effects of alcohol consumption and priming people with alcohol consumption on risk seeking behaviour, and found that not the alcohol dosage or the prime affected risk-taking, but the context in which the risk seeking task occurred – in a group context, groups primed with alcohol consumption turned out the be significantly more risk seeking. When participating in isolation, on the other hand, no results were found. Nevertheless, since the major part of the existing literature argues that priming does have a positive effect on risk seeking, and novelty seeking behaviour, the second hypothesis is formulated as follows:

H2: Priming people with beer consumption leads to more novelty seeking behaviour, compared to not priming people with beer consumption.

The moderating role of the prime

Furthermore, ‘primed with beer consumption’ is hypothesized to have an interaction effect on the effect of beer dosage on novelty seeking behaviour. Oyserman and Lee (2008) argue that priming participants with individualism or collectivism moderated the effect of difference in individualism and collectivism on values, self-concept content, relational assumptions, and cognitive style. Evidently, the constructs that were examined in their research are different from novelty seeking behaviour. In addition, Bargh and Chartrand (2000) state that the effects of priming are the same in all situations: priming people with a certain emotion, goal or motivation will inevitably result in people behaving in a way similar to that emotion, goal or motivation. They also state that priming people with mental representations and processes will mediate and moderate their behaviour. Projecting the findings of Bargh and Chartrand (2000) on the outcomes of the Oyserman and Lee (2008) research, leads to the assumption that although the constructs used in the priming experiment of Oyserman and Lee (2008) are different from the constructs used in the present experiment, the effect of the prime will be the same. According to Bargh and Chartrand (2000), it does not matter whether the responses to the prime are impressions, evaluations, goals, or

behaviour. In addition, Harris, Bargh, and Brownell (2009) also found a positive moderating effect of their priming variable on the relationship between their independent and dependent variable. These three studies are amongst many others that hypothesize (and find evidence that supports) such a positive moderating effect of priming participants. Therefore,since the effect of beer dosage on novelty seeking behaviour was established to be positive in the prior section, priming the participants with beer consumption is expected to increase this positive effect even more. As a result, the third hypothesis of this experiment is formulated as follows:

H3: Priming people with beer consumption increases the positive effect of a high level of beer dosage on novelty seeking behaviour.

3

Research design

(Oyserman and Lee 2008; van Koningsbruggen and Stroebe 2011; Bargh, Chen and Burrows 1996). Moreover, it is hypothesized that priming people with beer consumption increases the positive effect of a high level of beer dosage on novelty seeking behaviour (Harris, Bargh and Brownell 2009;

Oyserman and Lee 2008; Bargh and Chartrand 2000). These expected results are summarized in the conceptual model in Appendix 1.

Participants

96 Participants (54 males and 42 females) were randomly assigned to one of the four conditions to control for possible biased results caused by differences between and within groups. To be exact, 22 participants (11 males, 11 females, average age 25.05) were assigned to the non-primed/low beer dosage condition, 23 participants (16 males, 7 females, average age 29.78) fell into the primed/low beer dosage condition, 25 participants (11 males, 14 females, average age 25.28) were assigned to the non-primed/high beer dosage condition, and 26 participants (16 males, 10 females, average age 27.15) were allocated to the primed/high beer dosage condition. All participants were people enjoying the sunny weather on one of the city of Groningen’s (the Netherlands) many terraces. Furthermore, everybody participated on a voluntary basis and people were not notified of the purpose of the research until after all tasks had been completed and all questions had been answered.

Participants were approached by the researchers (Weijers, L – writer of this thesis, and De Boer, D – a colleague examining the effect of beer dosage on consumers’ risk seeking behaviour) while sitting in the city centre of Groningen, and asked to take part in the experiment. The researchers would first introduce themselves and explain that the experiment was part of their final thesis of the Marketing Management master at the University of Groningen. They also gave an indication of the time it would take for the participants to finish the experiment (i.e. about five minutes) and participants were asked to complete all tasks individually. The exact script that was used for this introduction is presented in Appendix 2. In addition, after completing the scrambled sentence test and the Novelty Seeking Behaviour tasks, participants were asked to answer several questions, to control for several variables as well as to measure the participants’ common level of novelty seeking behaviour (i.e. novelty seeking as a personality trait). Subsequently, after answering all questions, participants were fully debriefed by means of a statement written on the last page of the

Independent variables – level of beer consumption

Participants in the low beer dosage condition (both primed and non-primed) were approached while enjoying their time on a terrace in the afternoon, whereas participants in the high beer dosage condition (likewise both primed and non-primed) were approached late in the evening. As a

consequence, all participants that were approached in the early afternoon inevitably fell into the low beer consumption condition, regardless of actual beer consumption. More specifically, participants that were approached in the afternoon were approached between 14:00 and 18:00 hours.

Accordingly, people that participated in the evening were by definition assigned to the high beer consumption condition – between 20:00 and 00:00 hours is meant by evening. The Pearson correlation between dosage (as defined by the time of the day) and actual consumption was 0.362 (significant at a .01 level, 2-tailed).

Independent variables – primed with beer consumption

The second independent variable – primed with beer consumption – comprises two different conditions; participants were either primed with beer consumption, or they were not. In this

research, the priming method that was used is the so called scrambled sentence test, which signifies that all participants were requested to create twelve sentences out of twelve different combinations of words presented in a scrambled order (Bargh 2000). In more detail, participants were presented with twelve different combinations of five words written in a scrambled order, whereupon they were asked to form complete sentences by using four of the five words. Participants that were assigned to the ‘primed with beer consumption’ condition were presented with five sentences including beer related words and seven neutral sentences (see Appendix 3.1); participants in the ‘not primed with beer consumption’ condition were presented with twelve neutral sentences (Appendix 3.2).

The Novelty Seeking Behaviour task

In order measure the effects of priming and beer consumption on novelty seeking behaviour, participants were asked to fulfil two tasks as described by Hong and Lee in their 2009 article. That is, first of all participants were shown a picture on which nine differently coloured M&M’s were

Subsequently, participants were shown pictures of nine different types of candy bars and asked to choose any five candy bars of any combination by writing down the numbers corresponding to the candy bars of their preferences. The scorecard used to administer the participants’ choices can be seen in Appendix 5. Following this task, the level of novelty seeking behaviour was identified as the number of different candy bars participants chose.

To summarize, two different tasks were used to determine the level of novelty seeking behaviour; in other words, novelty seeking behaviour was split into two different dependent variables – novelty seeking 1 was measured as the number of different colours of M&M’s chosen by participants and novelty seeking 2 was measured as the number of types of candy bars chosen by participants.

Control variables

In addition, participants were asked to provide the following information: their age and gender, how many beers they had actually consumed at the time of completing the M&M task, how much they were craving a beer at the time of participation, how many glasses of alcohol they consume on average per day of the week, how they expect the consumption of two beers to impact their novelty seeking behaviour, and their common level of novelty seeking behaviour as a personality trait (the procedure used to measure this construct will be explained in more detail in the next paragraph). This information was translated into control variables. Age, gender, to what extent the participants were craving a beer, and the actual amount of beer participants had consumed at the time of the research are of importance because these variables validate the effect of random assigning

participants to the four different conditions. The control variable ‘to what extent participants crave a beer’ was measured on a 7-point Likert scale ranging from ‘not at all’ to ‘a lot’. Similarly, the impact expectancy (which was defined by assessing how participants expected the consumption of two beers in one hour to impact their ability to solve choice- and agility puzzles) was measured on a 9-point Likert scale, ranging from ‘much worse’ to ‘much better’. The scorecard that was used to record these values can be found in Appendix 7.

seeking as a personality trait’-variable by adding up the scores on all the different eight items and dividing this total score by eight (the Cronbach’s Alpha of the CNS scale was 0.85). Since the CNS scale is an English scale and the experiment took place in the Netherlands, the items in Appendix 6 have been translated into Dutch; Appendix 8 shows the original English items as designed by Manning, Bearden and Madden (1995).

The actual sequence in which participants were required to complete all tasks is the following: first of all, participants were asked to complete the scrambled sentence test; Appendix 3.1 shows the form that was presented to participants in the ‘not primed with consumption’ condition and in Appendix 3.2 the form for participants that were in the ‘primed with consumption’ condition can be found. After this, participants had to accomplish the novelty seeking behaviour tasks; first of all the M&M’s test was conducted (as shown in Appendix 4), and secondly the candy bar test was presented to participants (Appendix 5). Subsequently, novelty seeking as a personality trait was measured by means of the form in Appendix 6, followed by the other control variables (Appendix 7). To conclude, participants were debriefed and provided with the option to note their e-mail addresses in order to receive an overview of results after completion of the research (Appendix 8 shows the corresponding form).

4

Results

First of all, the general characteristics of the dependent, independent, and control variables were analysed. Of the total of 96 participants, 54 were male and 42 were female. Their average age was 26.81 (SD = 8.08).

Characteristics of variables per condition

Furthermore, in condition 2 of the experiment (low beer dosage; primed with consumption), 23 respondents filled out the questionnaire (16 male, 7 female) with an average age of 29.78 (SD = 12.09). On average, these participants’ actual beer consumption at the time of the experiment was 0.57 (SD = 1.08), their usual daily consumption of alcohol was 2.43 (SD = 1.35). Moreover, their average level of craving a glass or bottle of beer at the time of participation was 4.70 (SD = 2.16). And finally, their average impact expectancy level was 4.13 (SD = 1.79) and the average level of novelty seeking behaviour as a personality trait was 3.95 (SD = 0.99).

In addition, 25 participants (11 male, 14 female) with an average age of 25.28 (SD = 4.65) were recorded in condition 3 (high beer dosage; not primed with consumption). Their average actual beer consumption was 2.76 (SD = 2.71), On top of that, participants in condition 3 usually consumed an average of 3.21 (SD = 2.16) glasses of alcohol per day and they were actually craving a glass of beer quite a lot at the time of the experiment (M = 5.44, SD = 1.71). Their average impact expectancy level was 4.56 (SD = 1.92) and the average level of novelty seeking as a personality trait of this group was 4.10 (SD = 1.27).

Finally, condition 4 of the experiment (high beer dosage; primed with consumption) contained 26 participants (16 male, 10 female, M age = 27.15, SD age = 6.75) who had consumed an average of 3.35 (SD = 5.07) beers at the time of observation. Additionally, on average they usually consumed 2.84 (SD = 1.85) alcoholic beverages per day. Their average level of craving a glass of beer at the time of the experiment was 4.77 (SD = 2.34), their average impact expectancy level was 4.31 (SD = 2.00), and their average level of novelty seeking as a personality trait was 3.91 (SD = 1.26).

Results of the statistical tests

A 2 (condition: low beer dosage vs. high beer dosage) x 2 (primed with beer consumption: yes vs. no) ANOVA on consumers’ novelty seeking behaviour (as measured by means of the candy bar task) demonstrated no significant results. Implementing the M&M task scores as the definition of novelty seeking behaviour did not change this outcome, neither did a combination of the two measures of novelty seeking behaviour.

Adding the control variables as covariates (gender, age, usual daily alcohol consumption, level of craving a beer, impact expectancy, novelty seeking as a personality trait) converted the analysis into a 2 x 2 ANCOVA. First of all, the analysis was carried out on consumers’ novelty seeking behaviour as measured by the candy bar task; this did not produce any significant results (F < 1). Subsequently, the M&M task test scores were implemented as being the dependent variable – this did generate

second independent variable – primed with consumption – generated significant results. However, the control variables ‘gender’ (F(1.85) = 5.57, p = 0.02) and ‘level of craving a glass or bottle of beer at the time of the experiment’ (F(1.85) = 4.83, p = 0.03) did produce significant outcomes. The effect of gender on novelty seeking behaviour indicated that males are more included to engage in novelty seeking behaviour (M = 4.41, SD = 1.17) than females (M = 3.93, SD = 1.35). Furthermore, the effect of the level of craving a glass or bottle of beer at the time of the experiment (which will be referred to as needstate in the remaining part of this research) implied that participants with a needstate of 2 (on a 7-point Likert scale) were extreme novelty seekers (M = 5.00, SD = 0.00); participants with a needstate of 3 were least displaying novelty seeking behaviour (M = 3.25, SD = 1.71). The exact order in which needstate affected novelty seeking behaviour (from most influential to least influential) is as follows: Needstate 2 (M = 5.00, SD = 0.00) – Needstate 4 (M = 4.67, SD = 0.58) – Needstate 1 (M = 4.63, SD = 0.83) – Needstate 7 (M = 4.23, SD = 1.38) – Needstate 6 (M = 4.04, SD = 1.28) – Needstate 5 (M = 3.95, SD = 1.47) – Needstate 3 (M = 3.25, SD = 1.71). The other control variables were not proven to have significant effects (F < 1). Likewise, the moderating effect of priming people with consumption on the (positive) effect of beer dosage on novelty seeking behaviour failed to reach significance as well (F < 1).

5

Conclusions and recommendations

The main objective of this study was to explore how beer consumption is related to consumers’ novelty seeking behaviour. The focus was on analysing test results of consumers in a low beer dosage category (i.e. people that were enjoying the sun on a terrace in the afternoon) and consumers in a high beer dosage category (i.e. people relaxing on a terrace in the evening), in combination with the effect of priming people with beer consumption. Accordingly, the corresponding hypotheses were formulated as follows:

H1: A high level of beer consumption leads to more novelty seeking behaviour, compared to a low level of beer consumption,

H2: Priming people with beer consumption leads to more novelty seeking behaviour, compared to not priming people with beer consumption, and

These hypotheses were tested using a 2 x 2 between subjects factorial design, first by means of an ANOVA analysis and subsequently by means of an ANCOVA analysis (using the following control variables as covariates: gender, age, usual daily alcohol consumption, level of craving a beer, impact expectancy, and novelty seeking as a personality trait). Furthermore, the dependent variable Novelty Seeking Behaviour was tested using two separate tasks; the candy bar task and the M&M task (Hong and Lee 2009).

Regrettably, the ANOVA analyses did not produce any significant results; both the candy bar task and the M&M task failed to reach significance. This implies that, without adding the control variables to the analysis, both ‘beer dosage’ and ‘priming participants with beer consumption’ do not influence consumers’ novelty seeking behaviour. Put differently, without controlling for other potential influences, i.e. without controlling for covariates that might explain significant portions of variance in the dependent variable, novelty seeking behaviour does not seem to be affected by whether people have consumed beer or not , nor by whether people think they have consumed beer or not.

The ANCOVA analysis, however, did demonstrate some significant results (when novelty seeking behaviour was defined by the M&M task). More specifically, the results indicated that although both independent variables did not generate significant outcomes, the control variables gender and needstate did in fact significantly influence consumers’ novelty seeking behaviour.

In terms of the theoretical application of these outcomes, several comments can be made. First of all, beer dosage is not confirmed to have a positive influence on novelty seeking behaviour (H1). This is not in line with the work of Euser et al. (2011), McMillen, Smith and Wells-Parker (1989) and Abrams et al. (2006), who state that alcohol intake results in more risk taking behaviour. Risk taking

seeking tasks need to be completed individually in order for the results to be reliable. Even though participants were approached while sitting on the terrace (which is usually a social event), they were still asked to complete the experiment individually. As stated in chapter 2, Sayette et al. (2012) argue that priming people with consumption is only effective when tasks are completed in a group context, which was not the case in the current study. Thus, this could be an important aspect of why no significant effects of priming on novelty seeking behaviour have been found. In addition, Bargh and Chartrand (2000) propose that the more total primes and the greater the concentration of relevant primes, the stronger the effects of the prime. In the present research, participants in the primed conditions were presented with twelve scrambled sentences, of which five sentences contained cues towards beer consumption. Perhaps those five sentences were not enough to truly prime

participants with beer consumption. Moreover, since all participants were sitting at a terrace enjoying the sun, beer was all around already; participants saw other people sipping a beer or were drinking beer themselves. This could also have impacted the effect of the prime on novelty seeking behaviour. Participants may have been already primed with beer consumption just by the

environment surrounding them when taking part in the experiment. This would imply that all participants in all conditions had been getting cues of beer consumption from their environment already, and thus priming them by means of the scrambled sentence test would not have affected that mind-set. Hence, in that scenario all participants in all conditions were primed with beer consumption and as a results, no differences in terms of the effect of priming on novelty seeking behaviour could have been found. Furthermore, the major part of the participants was student in the city of Groningen, and some of them actually recognized the scrambled sentence test for what it is – a means of priming people with certain behaviour. Bargh and Chartrand (2000, p. 11) argue that “if a person is aware of the relevance of the priming event to the later perception or judgment, there is an adjustment away from the presumed effect of that event. (…) But in the usual case, in which one is not aware of the potential influence, bias in the direction of the primed representation occurs.” So, if a significant part of the participants in the primed conditions was aware of the priming and the relevance of it to the M&M and candy bar test, it may have affected the effect of priming on novelty seeking behaviour. However, although some participants were familiar with the theory of priming, none of them seemed to understand exactly how all parts of the questionnaire were related to each other, and even if they did, it was only such a small part of the entire sample that it could not have caused major shifts in test outcomes.

to be rejected. The combined work of Oyserman and Lee (2008), Bargh and Chartrand (2000) and Harris, Bargh and Brownell (2009) led to the assumption that priming people with an emotion, behaviour, goal, or motivation will always have the same amplifying effect on the effect of an independent variable on a dependent variable. However, since in the present research no main effect of beer dosage on novelty seeking behaviour has been found, this may have caused the moderating effect of the prime to be cancelled out as well.

Moreover, gender is found to have a significant effect on novelty seeking behaviour, perhaps due to the fact that men and women are different in nature; the results show that men are more inclined to behave as novelty seekers, compared to women. Similarly, the control variable needstate also generated significant results; participants with a needstate of 2 (on a 7-point Likert scale ranging from not at all to a lot) scored a 5 on the M&M novelty seeking task, which is the highest possible score, since participants were asked to choose a combination of any five colours of M&M’s. However, needstate did not have a linear relationship with the outcome of the M&M task. As a matter of fact the exact order in which needstate affected novelty seeking behaviour was: 2, 4, 1, 7, 6, 5, 3. A reason for this might be that it can be challenging for participants to estimate whether their level of craving a beer on a 7-point Likert scale is, for example, 2 or 3. Although needstate

significantly affects novelty seeking behaviour, the exact way in which it does cannot be assessed with the data of the present experiment; it does provide a basis for future research however. Furthermore, gender and needstate only provided significant outcomes when the M&M task was implemented as the novelty seeking behaviour variable, using the candy bar task as the novelty seeking behaviour variable did not produce any significant outcomes. Following the work of Hong and Lee (2009), the M&M task and the candy bar task were expected to provide similar results (i.e. since gender and needstate were significant in the M&M task, they were also expected to be significant in the candy bar test). However, the positive relationship was only found when the M&M task was used to measure novelty seeking behaviour with. This could be due to the fact that people do not have as strong associations with differently coloured M&Ms as they do with different types of candy bars. Most consumers know what the different candy bars taste like, and probably have strong preferences towards the ones they like best; these preferences might gender and needstate to have less impact on novelty seeking behaviour when the construct is measured by means of the candy bar test.

In terms of the practical applicability of the results of this experiment, several other comments can be made. Since a significant effect of gender on novelty seeking behaviour has been found,

engage in higher levels of novelty seeking behaviour than women, and marketing outings should be adapted to this newfound knowledge, by offering more new and innovative products during commercial breaks of programs watched by primarily men. An example could be that the commercials during the Dutch TV-show ‘Voetbal International’ should focus more on raising awareness of the innovative aspects of products, and not just on creating brand awareness. Moreover, in chapter 1 several questions were asked: do consumers’ preferences change after drinking a beer or two? And how do these preferences change; do consumers want more novel brands or do they stick with the leading, well-known brands? These questions can be answered now; consumers’ preferences do not change after drinking a beer or two, but men and women do prefer different brands and products. As a result of the present research it can be stated that men prefer more novel brands, whereas women are more inclined to stick with the leading, well-known brands.

Limitations and suggestions for future research

One of the main limitations of this study is that the experiment took place in an individual context while group pressure and group homogeneity could have significant effects on the decisions

consumers make. Moreover, one of the main constructs of this experiment was beer dosage, and it is common knowledge that people often drink beer in groups. Also, prior research has identified group context (vs. individual context) to majorly influence the effects of priming. Future research is

therefore suggested to focus on the effects of beer dosage and priming people with beer

consumption not only in an individual context, but also in a group context. An example of this could be that all participants that are sitting at a table at the terrace together complete the same

scrambles sentence test and are actually allowed to discuss the sentences with each other; by doing this, the group aspect will come into play and results may be different. The novelty seeking tasks however, should be completed in isolation. Pilot tests are suggested to be used in future research to estimate what the perfect balance between working in groups and working individually is in this field of research.

Another basis for future research related to the priming task focuses on the priming task itself. In this research the scrambled sentence test was used, which in practice meant that participants in the primed conditions were presented with twelve different combinations of five words, of which five combinations included cues towards beer consumption. It might very well be that those five cues were not sufficient for the effect of the prime to be truly effective. Future research is proposed to be directed towards finding out what the perfect amount of sentences with and without beer

Furthermore, the present research focused solely on beer intake. Obviously, many other types of alcoholic beverages exist, but no registration of the intake of those other consumptions has been made. Even though participants may have had consumed zero glasses of beer at the time of the observations, their novelty seeking behaviour could have been affected by the consumption of other alcoholic beverages. Future research is proposed to address the effects of different types of alcohol on novelty seeking behaviour, or perhaps to address the influence of beer consumption but in a controlled environment. By controlling the environment, the researcher can make sure that

participants have really only consumed beer at the time of the observation. Of course, participants in the present study were randomly assigned to one of the conditions, which would generally cancel out such inconsistencies. The researchers, however, did observe a lot of participants drinking wine, or mixed beverages (e.g. Vodka-Cola). This provides an interesting basis for future research, as the precise effects of different types of alcoholic consumptions on novelty seeking behaviour are not well established at the moment.

Moreover, the control variable needstate produced significant results when the M&M task defined novelty seeking behaviour. Future research should be directed towards investigating the exact relationship between the two constructs; the present study did not find a linear relationship, which makes it difficult to assess the way in which needstate precisely affect novelty seeking behaviour. Another suggestion is related to the different outcomes of the candy bar task and the M&M task. Needstate and gender only produced significant results when the M&M task was used as a means of establishing novelty seeking behaviour, whereas the candy bar task did not produce and significant outcomes. Future research should focus on both test and assess the validity of the different tasks and the possible reasons behind the differences in outcomes.

In conclusion, beer dosage and priming participant with beer consumption do not significantly influence consumers’ novelty seeking behaviour, contradictory to prior research. Similarly, no evidence supporting the influence of priming people with consumption on either novelty seeking behaviour has been found, although gender and the level of craving a beer at the time of the

6

References

Abrams, D., Hopthrow, T., Hulbert, L., & Frings, D. (2006). “Groupdrink”? The effect of alcohol on risk attraction among groups versus individuals. Journal of studies on alcohol, 67(4), 628-636. Bargh, J. A. (2002), Losing Consciousness: Automatic Influences on Consumer Judgment, Behavior,

and Motivation. Journal of Consumer Research, 29(2), 280-285.

Bargh, J. A., & Chartrand, T. L. (2000). The mind in the middle. Handbook of research methods in social and personality psychology, 253-285.

Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait construct and stereotype activation on action. Journal of personality and social psychology, 71(2), 230.

Chuang, S. C., Cheng, Y. H., Wang, S. M., & Cheng, S. Y. (2013). The impact of the opinions of others on variety‐seeking behavior. Journal of Applied Social Psychology, 43(5), 917-927.

CNBC (2015), “Top 20 Beer Drinking Countries,” (accessed March 6, 2015), [available at http://www.cnbc.com/id/26789471].

Euser, A. S., van Meel, C. S., Snelleman, M., & Franken, I. A. (2011). Acute effects of alcohol on feedback processing and outcome evaluation during risky decision-making: An ERP study. Psychopharmacology, 217(1), 111-125.

George, S., Rogers, R. D., & Duka, T. (2005). The acute effect of alcohol on decision making in social drinkers. Psychopharmacology, 182(1), 160-169.

Harris, J. L., Bargh, J. A., & Brownell, K. D. (2009). Priming effects of television food advertising on eating behavior. Health Psychology, 28(4), 404.

Health Promotion Agency (2014), “Alcohol – The Body and Health Effects,” (accessed June 4, 2015), [available at http://www.alcohol.org.nz/sites/default/files/1.0%20AL802_Body%20and% 20health%20effects_Aug2014_web.pdf].

Hong, Jiewen, and Angela Y. Lee. Choosing with Crying Smiles and Laughing Tears: The Dual Effects of Mixed Emotions on Variety Seeking. Retrieved January 23 (2009): 2012.

Jellinek (2011), “Wat voelt men bij alcohol,” (accessed at March 7, 2015), [available at

Jellinek (2014), “Welk deel van alle alcohol wordt gedronken door de groep die het meeste drinkt?,” (accessed June 3, 2015), [available at http://www.jellinek.nl/vraag-antwoord/welk-deel-van-alle-alcohol-wordt-gedronken-door-de-groep-die-het-meeste-drinkt/].

Legohérel, P., Daucé, B., & Hsu, C. H. (2012). Divergence in Variety Seeking: An Exploratory Study Among International Travelers in Asia. Journal of Global Marketing, 25(4), 213-225.

Manning, K. C., Bearden, W. O., & Madden, T. J. (1995). Consumer innovativeness and the adoption process. Journal of Consumer Psychology,4(4), 329-345.

Netpanel (2014), “Nationaal Bieronderzoek 2014,” (accessed March 6,2015), [available at http://www.biernet.nl/nieuws/nationaal-bieronderzoek-2014-pils-blijft-favoriet]. McMillen, D. L., Smith, S. M., & Wells-Parker, E. (1989). The effects of alcohol, expectancy, and

sensation seeking on driving risk taking. Addictive behaviors, 14(4), 477-483.

Oyserman, D., & Lee, S. W. (2008). Does culture influence what and how we think? Effects of priming individualism and collectivism. Psychological bulletin, 134(2), 311.

Pedersen, E. R., Neighbors, C., & Larimer, M. E. (2010). Differential alcohol expectancies based on type of alcoholic beverage consumed. Journal of studies on alcohol and drugs, 71(6), 925. Ratner, R. K., & Kahn, B. E. (2002). The impact of private versus public consumption on variety‐

seeking behavior. Journal of Consumer Research, 29(2), 246-257.

Rijksoverheid (2015), “Alcohol in de wet,” (accessed March 7, 2015), [available at http://www.rijksoverheid.nl/onderwerpen/alcohol/alcohol-in-de-wet].

Sayette, M. A., Dimoff, J. D., Levine, J. M., Moreland, R. L., & Votruba-Drzal, E. (2012). The effects of alcohol and dosage-set on risk-seeking behavior in groups and individuals. Psychology of

addictive behaviors, 26(2), 194.

Sharma, P., Sivakumaran, B., & Marshall, R. (2010). Impulse buying and variety seeking: A trait-correlates perspective. Journal of Business Research, 63(3), 276-283.

Toledo, M., & Sandi, C. (2011). Stress during adolescence increases novelty seeking and risk-taking behavior in male and female rats. Frontiers in behavioral neuroscience, 5, 17.

Trimbos Instituut (2013), “Werking alcohol,” (accessed March 6, 2015), [available at

University of Rochester Medical Center (2015), “Understanding Alcohol's Effects,” (accessed June 3, 2015), [available at http://www.urmc.rochester.edu/encyclopedia/content.aspx?ContentTyp eID=1&ContentID=2860].

Van Koningsbruggen, G. M., & Stroebe, W. (2011). Lasting effects of alcohol: Subliminal alcohol cues, impairment expectancies, and math performance. European Journal Of Social Psychology, 41(7), 807-811.

7

Appendices

Appendix 1 – Conceptual model of the experiment.

Main effects

Level of beer consumption Novelty seeking behaviour (H1) Primed with consumption Novelty seeking behaviour (H2)

Interaction effect

Primed with consumption

Level of beer consumption Novelty seeking behaviour (H3)

Appendix 2 – Researcher’s introduction to potential participants.

Hallo, ik ben Lauren Weijers en ik ben momenteel bezig met een onderzoek voor mijn master thesis Marketing aan de Faculteit Economie en Bedrijfskunde hier aan de Rijksuniversiteit Groningen. Ik zou het enorm waarderen als je meedeed aan mijn onderzoek, het zal slechts ongeveer 5 minuten van je tijd kosten.

+ +

Appendix 3.1 – First part of the experiment; not priming participants with beer consumption. Het onderzoek bestaat uit enkele ongerelateerde deelstudies naar taal en redeneren. Deelname is vanzelfsprekend geheel anoniem en je gegevens zullen uitsluitend vertrouwelijk worden

gebruikt voor dit onderzoek en niet verder worden verspreid.

Het onderzoek zal slechts enkele minuten van je tijd vergen. Werk a.u.b. zelfstandig aan het onderzoek en overleg niet met anderen.

Je mag nu beginnen met de eerste deelstudie.

In deze taak wordt je gevraagd woorden te combineren zodat er een correcte grammaticale zin ontstaat. Je ziet telkens vijf woorden; van vier van deze woorden moet je een correcte zin maken. Het is dus de bedoeling dat je een correcte zin maakt van vier woorden. Deze zin schrijft je

vervolgens op de stippellijn onder de desbetreffende woorden. Hier is een voorbeeld: Stel, je ziet de volgende vijf woorden.

sappig zijn de sinaasappelen rijp Je zou dan de zin kunnen maken:

De sinaasappelen zijn rijp.

Een andere mogelijkheid is:

Zijn de sinaasappelen sappig?

Onthoud goed dat je maar vier van de vijf woorden gebruikt. Je mag GEEN andere woorden gebruiken of woorden veranderen.

1. bewolkt lijkt drank het behoorlijk

……….. 2. boek schrijver schreef het Marie

……….. 3. goed voorbereid was hij gezegd

……….. 4. appartement zij de ramen reinigden

……….. 5. zijn ook we zelfstandig hebben

6. riep hond de hij zwaaide

……….. 7. makkelijk papier winkel scheurde het

……….. 8. klas ik liep de koos

………..

9. mist Jan familie zijn zonlicht

……….. 10. laat is heeft bus de

……….. 11. de weg kapitein voer schip

……….. 12. koop ik morgen gehucht melk

………..

*** Je bent aan het eind gekomen van dit deelonderzoek.

Je mag nu verder gaan met de volgende taak op de volgende bladzijde. ***

Appendix 3.2 – First part of the experiment; priming participants with beer consumption. Het onderzoek bestaat uit enkele ongerelateerde deelstudies naar taal en redeneren. Deelname is vanzelfsprekend geheel anoniem en je gegevens zullen uitsluitend vertrouwelijk worden

gebruikt voor dit onderzoek en niet verder worden verspreid.

Het onderzoek zal slechts enkele minuten van je tijd vergen. Werk a.u.b. zelfstandig aan het onderzoek en overleg niet met anderen.

In deze taak wordt je gevraagd woorden te combineren zodat er een correcte grammaticale zin ontstaat. Je ziet telkens vijf woorden; van vier van deze woorden moet je een correcte zin maken. Het is dus de bedoeling dat je een correcte zin maakt van vier woorden. Deze zin schrijft je

vervolgens op de stippellijn onder de desbetreffende woorden. Hier is een voorbeeld: Stel, je ziet de volgende vijf woorden.

sappig zijn de sinaasappelen rijp Je zou dan de zin kunnen maken:

De sinaasappelen zijn rijp. Een andere mogelijkheid is:

Zijn de sinaasappelen sappig?

Onthoud goed dat je maar vier van de vijf woorden gebruikt. Je mag GEEN andere woorden gebruiken of woorden veranderen.

1. bewolkt lijkt regen het behoorlijk

……….. 2. dronk hij bier veel morgen

……….. 3. goed voorbereid was hij gezegd

……….. 4. appartement zij de ramen reinigden

……….. 5. smaakte biertje het goed zwaaide

……….. 6. makkelijk papier winkel scheurde het

……….. 7. bier me maakt ontspannen koos

8. mist Jan familie zijn zonlicht

……….. 9. geniet bier zij van president

……….. 10. de weg kapitein voer schip

……….. 11. alcohol plezier maar voor zorgt

……….. 12. kapot computer is mijn wordt

………..

*** Je bent aan het eind gekomen van dit deelonderzoek.

Je mag nu verder gaan met de volgende taak op de volgende bladzijde. ***

Appendix 4 – Second part of the experiment; measuring the DV with the M&M test.

In de afbeelding hieronder zie je 9 verschillende kleuren M&M’s, ze zijn allemaal genummerd. Je mag 5 M&M’s uitkiezen, in elke combinatie die je wilt.

Als je 5 M&M’s in dezelfde kleur wilt kan je 5x achter elkaar hetzelfde nummer invullen, als je verschillende kleuren wilt kan je 5x een anders nummer invullen.

*** Je bent aan het eind gekomen van dit deelonderzoek.

Je mag nu verder gaan met de volgende taak op de volgende bladzijde. ***

Appendix 5 – Third part of the experiment; measuring the DV with the candy bar test. In de afbeelding hieronder zie je 9 verschillende snoeprepen, ze zijn allemaal genummerd. Je mag 5 snoeprepen uitkiezen, in elke combinatie die je wilt.

Als je 5 dezelfde snoeprepen wilt kan je 5x achter elkaar hetzelfde nummer invullen, als je verschillende repen wilt kan je 5x een ander nummer invullen.

Er is geen goede of slechte combinatie, het gaat om jouw persoonlijke voorkeur.

*** Je bent aan het eind gekomen van dit deelonderzoek.

Appendix 6 – Fourth part of the experiment; measuring novelty seeking as a personality trait

using the Consumer Novelty Seeking (CNS) scale (Manning, Bearden and Madden 1995).

Hieronder staan 8 statements, door middel van de cijfers 1 (zeer mee oneens) tot en met 7 (zeer mee eens) kan je laten weten in hoeverre je het met deze statements eens bent.

Graag het cijfer van je voorkeur omcirkelen!

1 2 3 4 5 6 7

Zeer mee oneens Klein beetje mee oneens Klein beetje mee eens Zeer mee eens Mee oneens Neutraal Mee eens

Ik zoek vaak informatie op over nieuwe producten en merken

1 2 3 4 5 6 7

Ik ga graag naar plekken waar ik word blootgesteld aan informatie over nieuwe producten en merken

1 2 3 4 5 6 7

Ik vind tijdschriften die nieuwe merken introduceren leuk

1 2 3 4 5 6 7

Ik zoek vaak naar nieuwe producten en diensten

1 2 3 4 5 6 7

Ik zoek situaties op waar ik bloot word gesteld aan andere bronnen met product informatie

1 2 3 4 5 6 7

Ik ben voortdurend op zoek naar nieuwe product ervaringen

1 2 3 4 5 6 7

Als ik ga winkelen, besteed ik weinig tijd aan het bekijken van nieuwe/ongewone producten

1 2 3 4 5 6 7

Ik profiteer van de eerst mogelijke kans om nieuwe en ongewone producten te ontdekken

1 2 3 4 5 6 7

*** Je bent aan het eind gekomen van dit deelonderzoek.

Appendix 7 – Fifth part of the experiment; administering the control variables. Tot slot vragen we je de vragenlijst hieronder in te vullen!

Wil je hiernaast aangeven hoeveel glazen alcohol (dat mag alles zijn: bier, wijn, sterke drank, mix drank

etc.) je meestal drinkt op elk van de aangegeven dagen van de week?

Aantal glazen alcohol:

Maandag _………… Dinsdag _………… Woensdag _………… Donderdag _………… Vrijdag _………… Zaterdag _………… Zondag _…………

Hoeveel glazen bier heb je vandaag gedronken? Op welk tijdstip heb je je laatste glas bier gedronken?

Geef ook de grootte en soort bier aan (vaasje/amsterdammer/pint & speciaalbier/pils/etc).

Glazen bier

……… Hoe laat heb je je

laatste glas bier gedronken?

……….

Grootte (S/M/L) Soort

Op een schaal van 1 tot 7, hoe graag zou je op dit moment een biertje willen? (1= helemaal niet, 7 =

heel erg graag)

1 2 3 4 5 6 7

Als ik 2 glazen bier drink in 1 uur wordt mijn vaardigheid in keuze- en behendigheidsspelletjes:

(1 = Veel slechter; 5= Geen invloed; 9= Veel beter) 1 2 3 4 5 6 7 8 9

Geslacht

M V

Appendix 8 – Sixth part of the experiment; debriefing participants.

Participant number: ……… Time of observation: ………

Je bent nu aan het eind van het onderzoek gekomen. Hartelijk dank voor je medewerking. Ik zou je willen vragen in verband met de vertrouwelijkheid niet over het onderzoek met anderen te praten. Als je benieuwd bent naar de achtergronden van het onderzoek, schrijf dan hieronder je emailadres, dan stuur ik je zodra de resultaten bekend zijn een volledig overzicht van doel, opzet en resultaten van het onderzoek. Je emailadres zal voor geen ander doel dan dit gebruikt worden

Emailadres: _____________________________________________________________

Appendix 9 – The original CNS items; based on a 7 point Likert scale, ranging from strongly

disagree to strongly agree. Item 7 requires reverse scoring (Manning, Bearden and Madden 1995).

1. I often seek out information about new products and brands

2. I like to go places where I will be exposed to information about new products and brands 3. I like magazines that introduce new brands

4. I frequently look for new products and services

5. I seek out situations in which I will be exposed to new and different sources of product information. 6. I am continually seeking new product experiences