• No results found

The effects of annoying advertisement on Brand Attitude and the moderating role of Exposure and Relevance

N/A
N/A
Protected

Academic year: 2021

Share "The effects of annoying advertisement on Brand Attitude and the moderating role of Exposure and Relevance"

Copied!
59
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

1

The effects of annoying advertisement on Brand

Attitude and the moderating role of Exposure and

Relevance

By:

Remco Kuilman

(2)

2

Master Thesis, Msc. Marketing Management

Faculty of Economics and Business

The effects of Annoying Advertisement on Brand

Attitude and the moderating role of Exposure and

Relevance

By:

Remco Kuilman

Student number: 2232065

Supervisor: Dr. D. Trampe

Second supervisor: Dr. L. Lobschat

January, 2013

| W.A. Scholtenstraat 26 // 9712 KW // Groningen | 06 20310085

(3)

3 Preface

In July 2011 I graduated for my Bachelor Sport Management at the University for Applied Science at the Hanzehogeschool of Groningen. My future was undecided and I received several job offers. At the same time I believed that there was more to learn. However, I had doubts whether or not to apply for University as I was wondering if I was qualified enough to pass the mathematics and statistics tests. Now, two and half year later, I am writing the last sentences of my Master’s thesis to complete the journey I started in September 2011. Applying at the University of Groningen resulted in a fantastic phase of my life. It was a choice which completed my student life, a choice which provided me with many resources, personally and intellectually, it was a choice I never will regret. This achievement and the moments of enjoyment were not possible without some people who are important to me. Therefore I would like to thank them in particular. From the beginning my parents motivated me to continue studying and during my study they were always there when I needed help. Second, my girlfriend Hester for her love, support and understanding, especially during exam periods when my attention for her was limited. Third, I would like to thank my friends that made studying more pleasant and of course for the other pleasant moments we enjoyed together during our free time. Lastly, I would like to thank Debra Trampe for her guidance and pleasant feedback during the process of writing this thesis.

(4)

4 Management summary

This study examined how different online types of advertising (annoying and non-annoying) influence consumers’ brand attitude and if this relation is moderated by the number of exposures and relevance of the ads. While previous studies focused on brand attitude enhancement by neutral advertisements, this study compared the differences in brand attitude development between exposure towards annoying and non-annoying ads. In this study, annoying advertisements are defined as advertisements that are perceived as intrusive by the respondent. Non-annoying advertisements are defined by the authors as not intrusive and should provoke a neutral feeling. The number of exposures and relevance of the ads are expected to influence this relationship. An advertisement is relevant when the ad is congruent with the website context. The opposite holds for irrelevant advertisements.

In a 2x2x2 between--group factorial experimental design four hypotheses are tested. The first hypothesis states that exposure towards annoying advertisements results in a lower brand attitude compared to exposure towards non-annoying ads. This hypothesis is based on the classical conditioning theory and confirmed by the data gathered for this study. When developing an advertisement, marketers should take into account how the advertisement will be perceived by the subjects.

Furthermore it is expected that when the frequency of exposure to an advertisement increases, consumers’ brand attitude will enhance for annoying, as well as non-annoying advertisements. These hypotheses are based on the mere exposure theory of Zajonc (1968). Zajonc argued that when the number of exposures increases the subject will process the stimuli more easily, which result in an enhancement of the brand attitude. For both, annoying and non-annoying internet advertisements the theory of Zajonc (1968) is confirmed. No significant difference in increase of brand attitude between the different types of advertisements is found. Concluding, marketers should keep in mind that brand attitude enhancement can be achieved by mere exposure regardless whether the advertisement is perceived as annoying or non-annoying.

(5)

5 Since subjects found advertisements less intrusive when they are relevant it is expected that the enhancement of brand attitude is greater for annoying compared to non-annoying advertisements when the ad becomes relevant. Because the enhancement of brand attitude due to a lower intrusiveness and a higher functionality (annoying) is equal to or greater than the enhancement of brand attitude due to a higher functionality (non-annoying). However, the results do not support the hypothesis. Even stronger, when the results are plotted in a figure, the opposite seems to be true. A possible explanation for this finding can be that intrusive advertisements are more difficult to process (Xia and Sudharshan 2002). A higher processing fluency ensures an enhancement of consumers’ brand attitude. Respondents may had difficulty with processing the annoying advertisements and became frustrated which resulted in a lower brand attitude.

Concluding, this study provides internet marketers with several new insights on how to manage and how not to manage their advertisement campaigns.

(6)

6

Table of Contents

1.0 Introduction ... 8 1.1 Relevance of research ... 10 1.2 Structure of thesis... 10 2.0 Literature review ... 11 2.1 Internet advertisement ... 11 2.2 Brand Attitude ... 13 2.3 Relevance ... 15 2.4 Conceptual model: ... 17 3.0 Methodology ... 18 3.1 Design ... 18 3.2 Participants ... 18 3.3 Pre-test ... 18

3. 5 Independent variable & moderators ... 20

3.5.1 Type of advertisement ... 20

3.5.2 Times of exposure ... 20

3.5.3 Relevance ... 21

3.5.4 Dependent variable Brand attitude ... 21

3.6 Manipulation check ... 21 4.0 Results ... 22 4.1 Demographics ... 22 4.2 Reliability analysis ... 23 4.3 Manipulation check ... 24 4.4 Multicollinearity check ... 25 4.5 Results hypotheses ... 26

5.0 Conclusions and Discussion ... 30

5.1 Conclusions ... 30

5.2 Limitations and further research ... 31

5.3 Managerial Implications ... 32

References ... 33

(7)
(8)

8

1.0 Introduction

You probably recognize the feeling: while examining the holiday options on the internet an advertisement pop-ups in front of your window. It distracts and interrupts you from your goal, booking your holiday. The advertisement ensures a slowdown in the whole process. In general, the intrusiveness of such an advertisement is perceived as annoying (Hairong 2002). In the past years researchers have explored which features of an advertisement provoke the annoyed feeling, but they have not yet examined how brand attitudes are affected by such specific advertisements. To get a better insight into the effects of annoying advertisements, this research tries to reveal how these advertisements affect consumers’ brand attitude.

According to Kotler (2001), advertisement can be defined as ‘any paid form of nonpersonal presentation and promotion of ideas, goods, or services by an identified sponsor’. Shortly after the introduction of internet, organizations started using online advertising. Internet advertisement became much more important over the years. According to the Interactive Advertising Bureau (IAB) and PwC US the expenditures on internet advertising have grown nearly 15% in 2012 compared to 2011. This results into a cumulative expenditure on internet advertising of almost $103 billion in 2012 (eMarketer 2012). Despite the growth in expenditure, the effectiveness of internet advertisement is a point of discussion in the literature (Korgaonkar 2002; Chatterjee 2003). Effectiveness of internet advertising can be measured in forms of click through rates, attitude towards an ad, attitude towards a web site and also brand attitude (Wang, 2013). In this research, effectiveness of internet advertising is measured in terms of brand attitude.

(9)

9 The mere exposure effect and classical conditioning are two theories which can explain these contradictions in the literature. The mere exposure effect underlies the findings of Metichell and Valenzuela (2005). Mere exposure is defined by Zajonc (1968) as follows; ’mere repeated exposure of the individual to a stimulus is a sufficient condition for enhancement of his attitude towards it. By ‘mere exposure’ is meant a condition which just makes the given stimulus accessible to the individual’s perception’. On the other hand, classical conditioning justifies the suggestion of Hairong et al (2012). Classical conditioning includes the repeated exposure towards two stimuli which results in one response. This includes an unconditioned- and a conditioned stimulus. The unconditioned stimulus is utilized to provoke a feeling. The conditioned stimulus is the promoted subject (Terence and Stuart, 1991). When the advertisement provokes a negative feeling it is thus unlikely that the advertisements enhance consumers brand attitude.

Relevance of the ad seems to be an important factor to determine how the advertisement is perceived by the consumer (Kim and Sundar 2012). Relevance is therefore a variable which is likely to influence the relationship between advertisement and brand attitude. Several researchers state that a higher degree of relevance results in a higher brand attitude (Russel 2002; Rodgers 2003) The consumer perceives an ad as relevant when there is congruity between with the context of the website and when the ad is based on user’s current actions (Moore, 2005) In other words, advertisements are seen as relevant when the content of the ad have a certain degree of fit with the content of the webpage and when the advertisement is personalized to current actions of the consumer. Edwards et al (2002) stated that when an ad is perceived as relevant the advertisement is perceived as less intrusive. Concluding, this statement suggest that relevance overrules the perceived annoyed feeling which is provoked by the advertisement.

Taking the different perspectives in current literature into account, the following research question can be formulated.

(10)

10

1.1 Relevance of research

All three variables (internet advertisement, brand attitude and relevance)are extensively studied in the past years. However, these variables are never linked to each other in such a way this research does. Especially the effects of annoying advertising on brand attitude are unknown. Therefore, this research will undoubtedly be useful for internet marketers to understand the effects of their online marketing campaigns. Furthermore, connecting these variables will fill the gap in the existing literature and will be fundamental for further research.

1.2 Structure of thesis

(11)

11

2.0 Literature review

In this chapter the theoretical framework about the individual variables is discussed. Hereafter, are the expected relationships displayed in a graphical conceptual model.

2.1 Internet advertisement

Expenditures on advertising have significantly increased in the past decades (eMarketer 2012). An interesting development is the shift from traditional advertising towards internet advertising. The increasing speed in which internet operates has led to an increase of 29% of expenditures in online video advertisement and even more astonishing is the rise in expenditures on mobile advertisements; an increase of 149% in 2012 compared to a year earlier. The way consumers perceive online advertisements differs, some advertisements are perceived in a positive sense, some negatively. Since this research is focused on the effects of the advertisements which are perceived as annoying, this chapter includes definitions of advertisements, an explanation about attitude towards the ad (Aad) and the effects of different types of attitudes toward the ad.

According to Puto et al (1984) advertisements include two concepts: information and transformation. High informational advertisement are advertisements which include factual and relevant brand data, which in turn improves the customers’ ability to evaluate the object. High transformational advertisements are advertisements with a unique set of characteristics to provoke a brand experience. The purpose of a high transformational advertisement is to make a product more exciting, warmer and more enjoyable compared to a solely objective description of the advertised brand. Advertisements are always a combination of these two concepts, however the degree of informational and transformational differs (high versus low)(Puto 1984).

(12)

12 (Kim,2012). In paragraph 2.3 more information is provided to gain deeper insights in the importance of personality of an advertisement.

Nowadays, a variety of forms of internet advertising exist. On the world wide web, advertisements can be delivered in forms of pop-up ads, paid text link, banner ads and short movies (Zeff 1999). Since this research is focused on internet advertisements it is important to thoroughly understand the different types of advertisements. Therefore, a short definition for each form is given. Pop-up ads are advertisements which appear in front of the existing window. Pop-ups are seen as an aggressive form of internet advertisements because of their high degree of intrusiveness (Edwards, 2002). Paid text links are sponsored links to website which are exposed based on algorithms and input from the surfer (Investopedia 2012). Banner ads are advertisements which are embedded within the website. In general, banner ads have click-through purposes. Despite the low click through rates on the internet, banner advertisement is still the most used form of internet advertisement (Investopedia 2012). Lastly, short movies are utilized to promote an object, often displayed before watching the preferred movie. Since the speed of internet connections has improved there is a large increase of use of this type of advertisement (Silverman 2013).

(13)

13 In contrast, ads which are perceived positively enhance the attitude towards the ad(Walker and Dubutsky 1994). Walker and Dubitsky (1994) conducted a study to acquire empirical evidence to see whether a likable ad is more effective or not. The authors found evidence that an ad which is liked, compared to a neutral advertisement, is indeed more effective in increasing post-exposure purchasing and that likable ads have a greater persuasive impact. In line with Walker and Dubitsky(1994), Youn et al (2001) conclude that likability of an ad has positive effects on company performance. Furthermore, especially the research of Haley and Baldinger (1991) showed a strong positive relation between likability of advertising and brand attitude. The following paragraph will elaborate on the concept brand attitude.

2.2 Brand Attitude

Brand attitude influences brand performance and therefore it is an important factor for marketers (Aaker 1991). Many empirical articles have tried to discover which factors influence consumers’ brand attitude (Aaker 2001; Lutz 1985). This paragraph includes definitions of brand attitude to create a better understanding of the concept. Furthermore, this paragraph will elaborate on relevant theories which explain how consumers’ brand attitude is affected.

In 1981, Mitchell and Olson defined brand attitude as ‘the individual’s internal evaluation of the brand’. According to Mitchell and Olson (1981) brand attitude is formed by evaluating an entity and creates a degree of (dis)favor. Attitudes are shaped over time and will be activated when the entity arouses in the environment of the consumer (Eagly 1993). The strength of the attitude depends on associative learning, the associative strength is determined by the accessibility of an attitude.

(14)

14 beautiful beach provokes a good feeling, nevertheless, at the same time the consumer is unaware that this exposure influence their attitude towards the liquor.

Prior research from Niedenthal (1990) and Razran (1938) concluded that classical conditioning works both ways, positively and negatively. When the conditioned stimulus is presented in combination with a negative unconditioned stimulus, consumers’ brand attitude shall be influenced in a negative way. For example, Niedenthal (1990) exposed her respondents to pictures which provoked negative feelings and found that the respondents developed a more negative attitude towards the subject. In contrast, when the conditioned stimulus is presented in combination with a positive unconditioned stimulus, consumers’ brand attitude will be influenced is a positive sense. An interesting example came from Razran (1938). Razran (1938) read during providing free meals, political slogans to people. Razrans’ research revealed a greater agreement with the political slogans when people received a free meal, compared to the group that did not receive a free meal. In other words, when the respondents received a positive stimulus (receiving a free meal) while listening to political slogans, they evaluated the political slogans more positive.

Taking classical conditioning into account, the following hypothesis can be formulated:

H1: Exposure to annoying advertisement will result in a more disfavorable brand attitude compared to exposure to non-annoying advertisement.

Another way of how consumers’ brand attitude is influenced is called ‘mere exposure’. Zajonc (1968) extensively studied the concept of mere exposure and defined mere exposure in the following way:

‘’mere repeated exposure of the individual to a stimulus is a sufficient conditions for enhancement of his attitude toward it. By ‘mere exposure’ is meant a condition which just makes the given stimulus accessible to the individual’s perception’’.

(15)

15 H2: When the frequency of exposures towards an annoying or non- annoying advertisement increases, the subject will develop a more favorable brand attitude.

As stated in the introduction it is likely that ‘relevance’ will influence the relationship between advertisement and brand attitude. Therefore, definitions of relevance are given in the following paragraph. Furthermore, an explanation is given for how relevance influences the relationship between annoying advertisement and brand attitude.

2.3 Relevance

Relevance is a concept which is also broadly studied in the literature. System oriented relevance is the most widely accepted definition of relevance. It is also known as the best match principle (Schamber 1990). In theory, an object seems to be relevant when there is a certain degree of congruency with its environment. The definition of Schamber (1990) is in line with definitions of other authors. Other authors most often used the term ‘conceptual relatedness’ to define relevance (Rees 1967; Saracevic 1970; 1975). Conceptual relatedness refers to the connection between two concepts. In perspective of this study, there is conceptual relatedness when the consumer experience a fit between the website and the online advertisement. Furthermore, Cooper (1971) used the term utility to define relevance. Within marketing context, utility is measured to which degree an advertisement is functional for the consumer. The higher the functionality the higher the advertisement is perceived as relevant.

(16)

16 marketers can make their ads more relevant by using a higher degree of personalization and congruency with the website.

Already in 1982, Mandler argued that advertisements with highly congruent information leads towards a slight favorable attitude towards the ad. In contrast, when advertisements are perceived as incongruent with its environment, negative evaluations are likely to occur. The same results are shown by many other authors(Kamins 1990; Russel 2002). For example, research showed enhancement of attitude towards an ad when a happy advertisement was shown within a happy television program. The same results were found for sad advertisements, when a sad commercial is placed within a sad program the brand attitude enhances (Lord 2001; Kamins 1990). More recently, Russel (2002) revealed the relationship between incongruent television advertisements and negative brand attitudes. Concluding from these studies, congruency indeed leads to more favorable attitudes towards the ad. These results were also determined for internet advertisements. Rodgers (2003) revealed that relevant sponsors on the internet provoke a more favorable brand attitude compared to irrelevant sponsors. Rodgers classified websites and sponsorship links with a high degree of congruency as relevant for the consumer. The opposite was classified as irrelevant. Following these studies, the following hypothesis can be stated.

H3: Exposure to an annoying or non-annoying relevant advertisement leads to a more favorable brand attitude compared to exposure to an annoying or non-annoying irrelevant advertisement.

It seems to be that relevance overrules the intrusiveness of an advertisement. A study of Edwards et al (2002) revealed a negative relationship between ad relevance and intrusiveness of the ad. In other words, the more the ad is perceived as relevant the lower the perceived feeling of intrusiveness. Hence, annoying ads which are relevant will be perceived as less annoying. A lower intrusiveness should lead to higher brand attitudes (Hairong 2012). Concluding, for annoying ads two reasons exist to suggest that brand attitude increases when the advertisement is perceived as relevant (lower intrusiveness and more functional). Only one reason (the higher degree of functionality) is applicable for non-annoying ads since non-annoying ads are not seen as intrusive.

Therefore the following hypothesis can be stated:

(17)

17

2.4 Conceptual model:

By means of the theory and the hypotheses that were stated in the literature section, the research question:

‘What is the influence of repeated exposure to annoying advertisement on brand attitude and is this relation moderated by relevance of the advertisement?’

can be answered. Below, the associated conceptual model is displayed. This model displays all the relationships. The independent variable, is expected to have a direct relationship on the dependent variable brand attitude (H1). This relationship is expected to be influenced by number of exposures (H2). Furthermore the direct relation is expected to be moderated by relevance (H3 and H4).

Figure 1 Ad type

Annoying versus non-annoying

Relevance

Brand attitude

(18)

18

3.0 Methodology

This chapter describes how the data was collected. In paragraph 3.1 the design of the study is discussed. Paragraph 3.2 contains a description of how the respondents are approached. Paragraph 3.3 includes a description of the pre-test which is performed. Furthermore, paragraph 3.4 explains the procedure of the experiment. Lastly, paragraph 3.5 and 3.6 includes a description how the different variables are manipulated and measured.

3.1 Design

In order to reveal the effects of types of advertisement, times of exposure and relevance on brand attitude an experiment was conducted. For this experiment a two (non-annoying and annoying advertisements) by two (one versus three exposures towards an advertisement) by two (relevant or irrelevant) between-subjects factorial design was utilized. To test these different conditions, eight questionnaires were developed.

3.2 Participants

Three hundred fifteen people participated in the experiment . Thirty seven people did not complete the whole questionnaire ( >1 questions) and were excluded for further analysis to overcome bias, leaving 278 participants. The experiment was conducted online, subjects could participate in their own time by opening a link to the online experiment. The respondents were randomly assigned to one of the eight versions of the questionnaire. The respondents were approached through social media and within the direct network of the researcher. The disadvantage of this method is that the respondents have in general common psychological and demographic characteristics. In order to confront this problem, a selection of participants was asked to spread the link within their network. This method of obtaining respondents is called convenience and snowball sampling (Malhotra 2010).

3.3 Pre-test

(19)

19 annoying- relevant or annoying-irrelevant. The respondents were informed that they wanted to purchase a new mobile phone. However, before purchasing the mobile phone, a short comparison research needs to be conducted on the website www.bellen.com. During their study the respondents were exposed to an annoying or non-annoying, relevant or irrelevant advertisement. The degree of consumers perceived the advertisement as annoying was measured on an eight item, 9 point bipolar scale based on the theory of Hairong et al (2002) (interfering; bothersome; invasive; intrusive; forced; obtrusive; distracting; disturbing). Relevance of the advertisement was measured on a nine-point bipolar scale as well. Participants were asked directly how relevant the advertisement was to them. For each condition five respondents completed the pre-test. The sample of the pre-test contained relatives of the researcher.

The results reveal a correct manipulation for relevance and annoyingness. The results are presented in the table below. Since the manipulation was successful, the exact same exposure situations were used later on in the actual experiment.

Table 3.1 Pre-test

T-Test t-value M SD Sig.

Relevance -4.705 6.095 1.976 .000

Annoyingness -9.322 .003 -1.704 .000

3.4 Experimental procedure

(20)

20 In the experiment unknown brands are used to minimize the effects of prior attitudes towards familiar brand names.

3. 5 Independent variable & moderators

This section discusses how the independent variable and the moderators are manipulated and measured during the experiment. The independent variable is ‘type of advertisement’ and the moderators are ‘times of exposure’ and ‘relevance’.

3.5.1 Type of advertisement

In the experiment two types of advertising are distinguished, non-annoying and annoying advertisements. Half of the respondents were exposed towards an annoying advertisement. Annoying advertisement is the condition where the advertisement which is shown evokes a feeling of irritation. Annoying advertisements are advertisements which are perceived as intrusive. Since Edwards et al (2002) concluded that pop-up ads are perceived as intrusive, pop-up ads are utilized in this study to provoke the annoyed feeling. The pop-up used in the experiment was tested in the pre-test and perceived as annoying. Non-annoying advertising refers to the situation in which the advertisement is perceived in a neutral sense by the consumer. An advertisement is perceived in a neutral way when the ad is not intrusive and includes cool colors (Metichell and Valenzuela 2005; Moore et al 2013). In context of this experiment, banner advertisements are seen as the least intrusive form of internet advertisement and therefore cool colored banner advertisements are used to arouse a non-annoyed feeling. Therefore the latter half of the participants was exposed to banner advertisement. An important remark here is that the advertisements are similar, only the way consumers are exposed to the ads differs (pop-up or banner). The banner used in the experiment was tested in the pre-test and perceived as non-annoying. The opposite accounts for pop-up advertisements.

3.5.2 Times of exposure

(21)

21

3.5.3 Relevance

Half of the respondents were exposed towards relevant advertisements, the other half were exposed towards irrelevant advertisements. Relevant refers to the situation in which the consumers perceive the advertisement as functional. Referring to chapter 2.3 an advertisement is perceived as functional when the advertisement is congruent with the context of its environment. Irrelevant refers to the situation when the advertisement is seen as unsuitable with the current situation. Therefore in this experiment, relevance was manipulated by exposing respondents towards advertisements which are congruent or incongruent with the respondents task. Within the relevant advertisements scenarios the advertisements were congruent with the task of the respondent. In contrast, the irrelevant advertisements had no connection at all with the respondents’ task. The respondents had the task to do a comparison study before buying a new smartphone. Half of the respondents were exposed towards smartphone advertisements (relevant) while the latter half of the respondents were exposed towards a trading website (irrelevant).

3.5.4 Dependent variable Brand attitude

After the participants had seen the advertisements during their shopping tasks, brand attitude of the respondent was measured using the five dimensions identified by Spears and Singh (2012) (unappealing/ appealing; bad/good; unpleasant/ pleasant; unfavorable/favorable; unlikeable/likeable). Participants had to rate these five dimensions on a nine point bipolar scale. To avoid that participants answer automatically, some dimensions were reversed in the questionnaire, such as (1) unpleasant and (9) pleasant, (1) appealing and (9) unappealing.

3.6 Manipulation check

(22)

22

4.0 Results

In this chapter the results of this study are discussed. First, the demographics of all the participants are presented. Since the study included eight different experiments it is tested whether demographical differences exist between the groups. Second, a reliability analysis was conducted to check whether the internal consistency of the applied scales were high enough to factor. Third, the results of the manipulation check are presented, these results show if the stimuli were perceived correctly. Hereafter, the results of the multicollinearity check are discussed. Lastly, the outcomes of the ANOVA analysis show whether the hypotheses can be accepted and the outcomes of a regression analysis show if other (non-hypothesized) variables had influence on the brand attitude.

4.1 Demographics

Three hundred fifteen people participated in the experiment . Thirty seven people did not complete the whole questionnaire ( >1 questions) and were excluded for further analysis to overcome bias, leaving 278 participants. From the remaining respondents, 51.1% were male and the mean age was 30.97 (SD=18.98). A major part of the respondents had an university degree (44.2%). Furthermore, a large part of the sample was still student (37.1%).

(23)

23 Employed, part time

Jobseeker Houseman/ housewife Retired 54 9 2 16 19.4% 3.2% 0.7% 5.8%

An ANOVA analysis is performed in order to test whether there are differences in age, education and occupation between the eight different groups. The null hypothesis is that there are no significance differences between the groups (p > 0.05). This hypothesis can be accepted since no differences exist between the groups; Gender (F=1.505 p=.165), education (F=.871, p=.530) and occupation (F=.289, p=.958). To test whether there is a difference in age between the groups a Chi-Square test was conducted. No significance difference are revealed (χ2= 220.286, p> .05), In conclusion, since there are no significance differences between the groups the sample can be used for further analysis.

4.2 Reliability analysis

To check the internal consistency a reliability test is conducted. The internal consistency is computed over the total sample (N = 278). The Cronbach’s Alpha is conducted for the brand attitude and manipulation scale. An overview of the reliability scores is presented in table 4.2.

Table 4.2 Reliability

Concept Scale Cronbach’s Alpha

Brand attitude Spears and Singh(2012) α = ,939

Intrusiveness Edwards et al (2002) α = ,960

Relevance - -

(24)

24 relevance since this was a one-item scale. The constructs are factored to one variable which is used in further analysis.

4.3 Manipulation check

After the reliability analysis a manipulation check was performed to test whether the advertisements were perceived correctly. The manipulation check is carried out by conducting two 2x2x2 ANOVA tests. The first ANOVA test exams whether the respondents perceived a difference in relevance of the advertisements. The second ANOVA test exams whether the respondents perceived a difference in annoyingness of the advertisements. It is expected to observe only a main effect in

relevance/annoyingness. The first ANOVA test reveals a main effect for relevance as expected. No further significant effects are revealed, see table 4.3.

Table 4.3 Manipulation

To check whether the difference is in the expected direction, means are compared. The lower the mean, the higher the perceived degree of relevance. The mean for relevant advertisements (M=5.055) is lower

ANOVA TABLE df Mean square F Sig.

(25)

25 compared to the mean of the irrelevant advertisements (M=6.965). Based on these findings, it can be concluded that relevance is manipulated correctly.

The second ANOVA test reveals the expected main effect for annoyingness. However, here an interaction effect between annoyingness and relevance is observed as well, see table 4.4 .

Table 4.4 Manipulation

To check whether the difference is in the expected direction, the means are compared. The lower the mean, the higher the perceived degree of annoyingness. The mean for annoying advertisements (M=-.506) is lower compared to the non-annoying advertisements (M=.502). Based on these findings, it can be concluded that annoyingness is manipulated correctly. The interaction effect revealed by the ANOVA test can influence the results, disadvantages of this effect are discussed in chapter 5.2.

4.4 Multicollinearity check

Lastly, the data is checked on multicollinearity. The multicollinearity check is to find whether the moderators and the independent variables are correlating with each other in the ANOVA. In order to check this, the individual parameters are evaluated regarding their (VIF) Variance Inflation Factor levels.

ANOVA TABLE df Mean square F Sig.

(26)

26 The outcomes are presented in table 4.5. In 1990 Myers stated that a variance inflation factor > 10 is a value to worry about multicollinearity. Here, all variance inflation factors show low levels of multicollinearity and should be no issue.

Table 4.5 Multicollinearity

Variable Sig. VIF

Annoyingness .000 1.017

Relevance .000 1.017

Exposures .002 1.001

4.5 Results hypotheses

This section describes the outcomes of the ANOVA conducted to test the hypotheses. Within the ANOVA the new variable brand attitude which is calculated by a factor analysis over the five items of the scale act as dependent variable. The ANOVA table shows the direct effects of the independent variables furthermore it also presents the interaction effects. Before the results are presented in table 4.6. the hypotheses are reiterated below.

H1: Exposure to annoying advertisement will result in a more disfavorable brand attitude. compared to exposure to non-annoying advertisement.

H2: When the frequency of exposures towards an annoying or non- annoying advertisement increases, the subject will develop a more favorable brand attitude.

H3: Exposure to an annoying or non-annoying relevant advertisement leads to a more favorable brand attitude compared to exposure to an annoying or non-annoying irrelevant advertisement.

(27)

27 Table 4.6 ANOVA

Hypothesis one tests whether annoyingness influences consumers’ brand attitude. From the ANOVA test can be concluded that annoyingness does influence consumers’ brand attitude (F=37.540, p=.000) By comparing the means, the results reveal that the advertisement perceived as annoying lowers the brand attitude. Concluding, H1 is supported by the data and therefore accepted. Table 4.7 presents the difference in brand attitude (M) and the standard deviation.

Table 4.7 H1

Annoyingness N M SD

Annoying 143 -.321 .074

Non-Annoying 135 .332 .076

H2 tests if an increase in numbers of exposure towards annoying and non-annoying advertisements results in a higher brand attitude. It can be concluded from the ANOVA analysis that the number of exposures has a significant influence on the consumers’ brand attitude (F= 14.432, p= .000). By comparing the brand attitude means for the different conditions, there can be concluded that an

ANOVA TABLE df Mean square F Sig.

(28)

28 increase in exposure results in an increase of the brand attitude (table 4.8). No significant difference between annoying and non-annoying advertisement are revealed by the ANOVA model (F=.945, p=.332)

Table 4.8 H2

Exposures N M SD

One exposure 131 -.197 .078

Three exposures 147 .208 .073

Hypotheses H3 tests whether relevance of the advertisement influences consumers’ brand attitude. The ANOVA analysis reveals that relevance of the advertisement has significant influence on consumers’ brand attitude (F=22.229, p=.000). Exposure to relevant advertisements leads to a more favorable brand attitude compared to exposure to irrelevant advertisements. Table 4.9 presents the results.

Table 4.9 H3

Relevance N M SD

Irrelevant 137 -.280 .075

Relevant 141 .291 .076

(29)

29 Relevant - Annoying - Non- Annoying 65 76 -.128 .710 .111 .101 Irrelevant - Annoying - Non-Annoying 78 59 -.514 -.047 .10 .114 Explorative results

To check whether other, non-hypothesized variables influence (in)direct consumers’ brand attitude explorative questions were included in the questionnaire. In this section the outcomes of these questions are discussed. During the questionnaire respondents’ mood and involvement of the product was measured. Other studies show that these variables influences product and brand evaluations therefore it seems to be important to include these (Chang 2004 ; Park and Young 1986). Since the variables are measured as continuous variables a regression analysis is performed to measure if these variables influence consumers’ brand attitude. The results are presented in table 4.11.

Table 4.11 Explorative variables Model Unstandardized Coefficients

(30)

30 There can be concluded that there are several significant effects. As known from the ANOVA analysis, the degree of annoyingness has a significant effect on consumers’ brand attitude. Furthermore, mood has a direct influence on consumers brand attitude (t=2.631, p=.009, B=.192). Meaning, a better mood means a higher brand rating. Next to the two main effects, two interaction effects are revealed by the regression. The interaction of mood * Involvement has significant influence on consumers’ brand attitude (t=2.187, p=.001). Lastly, an interaction effect between annoyingness * mood is revealed, however on a significance level of p < .10 (t=-1.920, p=.056).

5.0 Conclusions and Discussion

This section describes what can be concluded from the results that were presented in chapter four. First, the research results are discussed and evaluated. Thereafter, the limitations of this study, the managerial implications and suggestions for further research are dealt with.

5.1 Conclusions

The main objective of this study was to examine how consumers’ brand attitude is influenced by exposure to annoying advertisements and how this relation was moderated by number of exposures and relevance. Based on previous studies, it is hypothesized that exposure towards annoying advertisement compared to non-annoying advertisements results in a disfavorable brand attitude. In line with the theory, the results from the online experiment reveal that there is a direct relationship between the type of advertisement and brand attitude(H1). Likewise, Hairong et al (2012) and Reed (1999) both stated that annoying advertisements are likely to arouse a negative brand attitude. Hereby, the classical conditioning effect is confirmed and there is reason to believe that annoying advertisements provoke a negative feeling which is associated with the advertised brand.

However, as discussed earlier, there is an academic discrepancy while authors argue that exposure to annoying advertisements results in a disfavorable brand attitude. This is confirmed by the first hypothesis, it can be reasoned that mere exposures towards an annoying advertisement can enhance consumers’ brand attitude. Based on the mere exposure effect it is hypothesized in this research that if the frequency of exposures to an advertisement increases, the consumers’ brand attitude should increase, regardless of whether the advertisement is annoying or non-annoying.

(31)

31 annoying or non-annoying advertisements. While Bornstein (1989) suggested that the mere exposure effect maybe can be illuminated when the stimulus is liked or disliked, this study reveals that there is no difference. Mere exposure results in a higher brand attitude regardless of whether the advertisement is liked or disliked.

Thirdly, it is confirmed as hypothesized in this study that the increase of relevance of an advertisement results in an increase of brand attitude. This result is confirmed for both annoying and non-annoying advertisements(H3). Meaning, when a consumer is exposed to an advertisement that is personal and congruent within the site context, the consumers’ brand attitude of the advertised brand will enhance compared to exposure to irrelevant ads. While Rodgers (2003) revealed this effect for only one type of advertisements (sponsored links) this research reveals that the theory is also applicable for annoying and non-annoying advertisements.

In this study, it is stated that relevance should overrule the intrusiveness. Therefore it is expected that the increase of brand attitude due to relevance of the ad is greater for annoying advertisements compared to non-annoying advertisements. A significant effect is revealed on a significance level of p < .10 (H4). However, when the results are plotted, the opposite of what was expected seems to be true. The increase of brand attitude for non-annoying advertisements is greater than the increase of brand attitude for annoying advertisements. A possible explanation for this finding can be that intrusive advertisements are more difficult to process (Xia and Sudharshan 2002). A higher processing fluency ensures an enhancement of consumers’ brand attitude. Respondents may had difficulty with processing the annoying advertisements and became frustrated which resulted in a lower brand attitude. Another explanation can come from the main effect of mood on brand attitude which is revealed by the explorative questions. It might be possible that the mood of the respondents was negatively influenced by the annoying advertisements. An effect which could result in a lower brand rating.

5.2 Limitations and further research

(32)

32 check an interaction effect between annoyingness and relevance was revealed. This effect could bias the results, annoying advertisements could be seen as more or less relevant.

Additionally, this study is not generalizable over all product categories. For products that fall in a different pricing range, other results could be found. Further research is necessary to reveal if the finding are applicable for more categories.

5.3 Managerial Implications

Despite the limitations of this study, there are some relevant managerial implications for marketers, in particular for internet marketers. This study reveals that exposure to annoying internet advertisements results in a lower brand attitude compared to exposure to non-annoying internet advertisements. Therefore, marketers should take into account when developing an advertisement how it will be perceived by the consumer.

In this study, evidence is found that mere exposure towards annoying advertisements enhances consumers’ brand attitude. While the theory of Zajonc (1968) is confirmed by many studies the non-annoying advertisements. Here, evidence is revealed that there is a significant enhancement of consumers’ brand attitude when the frequency of exposures towards annoying advertisements increases. Therefore, marketers should keep in mind that brand attitude enhancement can be achieved by mere exposure, even when the advertisement is perceived as annoying.

Further, brand attitude enhancement can be achieved by making the internet advertisement more relevant. This finding accounts for both annoying and non-annoying advertisements. Hence, marketers should base their decisions concerning ad placement on functionality and context congruency. Lastly, while the last hypothesis is rejected, marketers should not focus on the hypothesized effect while developing an advertisement.

(33)

33

References

Aaker, D. A. (1996) Building Strong Brands. New York: The Free Press

Aaker, D.A. (1985) Causes of irritation in Advertising . Journal of Marketing, 47-57.

Aaker, D.A., & Jacobson, R (2001). The Value Relevance of Brand Attitude in High- Technology Markets. Journal of Marketing research, 485-493

Bayles, M. (2000). Just How ‘Blind’ are we to Advertising Banners on The Web? [Online]. Usability News, Available: http://psychology.wichita.edu/surl/usabilitynews/2S/banners.htm

Benway, J.P., & Lane, D.M. (1998). Banner Blindness: Web Searchers Often Miss Obvious Links [Online]. Available: www.internettg.org/newsletter/dec98/banner blindness.html.

Bornstein R.F. (1986) Exposure and Affect: Overview and Meta-Analysis of Research. Psychological Bulletin. (106): 265-289

Briggs, R., & Hollis, N. (1997). Advertising on the Web: Is There Response Before Click-Through. Journal of Advertising Research, 37 (2): 33-46.

Campbell, M.C., Keller, K. L. (2003) Brand familiarity and advertising repetition effects. Journal of Consumer Research. (30): 292-304

Chang, C. (2004). How mood and ad-self-congruency affect the relative influence of hedonic ad appeals and utilitarian ad appeals on product evaluations. Advances in Consumer Research. (31): 721-727 Chatterjeee, P., Hoffman, D. L., Novak, T.P. (2003) Modeling the clickstream: Implications for Web-based advertising efforts. Marketing Science. (22): 520-541

Cooper, W.S. (1971). A definition of relevance for information retrieval. Information Storage and Retrieval, 7(l), 19-37.

(34)

34 Edwards, S.M., Li, H., & Lee, J.H. (2002). Forced exposure and psychological reactance: Antecedents and consequences of the perceived intrusiveness of pop-up ads. Journal of Advertising, 31 (3): 83–95. eMarketer. (2012, Dev). eMarketer.com. Retrieved from www.emarketer.com.

Fang, X., Singh, S., Ahluwalia, R. (2007) An examination of different explanations for the mere exposure effect. Journal of Consumer Research. (34): 97-103

Field, A. (2005) Discovering statistics using SPSS. 2nd edition London: Sage

Goldfarb, A., & Tucker, C. (2011). Online Display Advertising: Targeting and Obtrusiveness. Marketing Science, 30 (3): 389-404.

Haley, R. L., Baldinger, A. L. (1991) The ARF Copy Research Validity Project, Journal of Advertising Research, Vol.31, p.11-32.

Hairong, M. S., Edwards, S. M., Lee, J. (2002). Measuring the Intrusiveness of Advertisements: Scale Development and Validation. Journal of Advertisement, 37-49.

Investopedia. (2012, March 15). Banner Advertising definition. Retrieved October 15, 2013, from Investopia: http://www.investopedia.com/terms/b/banneradvertising.asp

Internet Advertisisng Bureau (2013). Internet Ad Revenues A t $20.1 Billion Hit Historic High For

Half-Year 2013, Up 18% Over Same Time In 2012 (Online)

http://www.iab.net/about_the_iab/recent_press_releases/press_release_archive/press_release/pr-100913

Jeong, Y., & King, C.M. (2010). Impacts of Website Context Relevance on Banner Advertisement Effectiveness. Journal of Promotion Management, 16: 247-264.

Kamins, M.A. (199). An investigation into the ‘match-up’ hypothesis in celebrity advertising: When beauty may be only skin deep. Journal of Advertising. (19): 4-3

(35)

35 Kim, N. Y., & Sundar, S. S. (2012). Personal relevance vs. contextual relevance: The role of relevant ads in personalized websites. Journal of Media Psychology, 24 (3), 89-101

Kline, P. (1999). The handbook of psychological testing (2nd ed.). London: Routledge Kotler, P. (2001). Marketing Management. Englewood Cliffs: Prentice-Hall.

Korgaonkar, P & L. W. (2002). Web usage, advertising, and shopping: relationship patterns. Internet research, 191-204.

Lutz, R. J. (1985). Affective and Cognitive Antecedents of Attitude Toward the Ad: A Conceptual Framework. Hillsdale: Lawrence Erlbaum Associates.

Malhotra, N.K. (2010). Marketing Research: An Applied Orientation, 6th edition. Pearson Education India.

Mandler, G. (1982). The Structure of Value: Accounting for Taste. Affect and Cognition: The 17th Annual Carnegie Symposium, Hillsdale, NJ: Lawrence Erlbaum Associates, 203-230.

Metha, A. (2000). Advertising attitudes and advertising effectiveness. Journal of advertising research, 67-73.

Mitchell, A. A., Olson, J. C. (1981). Are product attribute beliefs the only mediator of advertising effect on brand attitude. Journal of Marketing Research (18): 318-332

Moore, R.S., Stammerjohan, C.A., & Coulter, R.A. (2005). Banner Advertiser-Website Context Congruity and Color Effects on Attention and Attitudes. Journal of Advertising, 34 (2): 71-84.

Myers, R. (1990) Classical and modern regression with applications (2nd ed.) Boston, MA: Duxbury Niedenthal, P. M. (1990) Implicit perception of affective information. Journal of Experimental Social Psychology, 26, 505-527

Park, C.W. Young, S.M. (1986) Consumer response to television commercials- the impact of involvement and background music on brand attitude formation. Journal of Marketing Research. (23): 11-24

(36)

36 Razran, G. S. (1938). Conditioning away social bias by the luncheon technique. Psychological Bulletin,

481.

Rees, A. M., & Schultz, D. G. (1967) A field experimental approach to the study of relevance assessments in relation to document searching. Cleveland, OH: Western Reserve University, School of Library Science,

Center for Documentation and Communication Research

Rodgers, S. (2003/2004). The Effects of Sponsor Relevance on Consumer Reactions to Internet Sponsorships. Journal of Advertising, 32: 67-76.

Russell, C. A. (2002). Investigating the Effectiveness of Product Placements in Television Shows: The Role of of Product Placements in Television Shows: The Role of Memory and Attitude. Journal of Consumer Research, 306-318.

Saracevic, T. (1970). The concept of “relevance” in information science: A historical review. Introduction to Information Science, 111-151.

Saracevic, T. (1975). Relevance: A review of and a framework for the thinking on the notion in information science. Journal of the American Society for Information Science, 321-343.

Schamber, L., Eisenberg, M. B., Nilan, M.S. (1990) A Reecamination of Relevance- Toward a Dynamic, Situational Definition. Information Processing & Management. (26): 755-776

Spears, N., Singh, S. N. (2012) Measuring Attitude toward the brand and Purchase Intentions. Journal of Current Issues & Research in Advertising. (26) 53-66

Terence, T. A. & Stuart, W. E. (1991). A Program of classical condintioning experiments testing variations in the conditioned stimulus and contents. Journal of Consumer Research, 1-12.

Valenzuela, A. M. (2005). How banner ads affect brand choice without clicking through. Online consumer Psychology, 125-142.

Walker, T. & Dubitsky, M. (1994). Why liking matters. Journal of advertising Research, 9-18.

(37)

37 Xia, L., Sudharshan, D. (2002). Effects of Interruptions on Consumer Online Decision Processes. Journal of Consumer Psychology. 12(3) 265-280

Youn, S., Sun, T., Wells, W. D., Zhao, X. (2001) Commercial Liking and Memory: Moderating Effects of Product Categories. Journal of Advertising research. 7-13

Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality, 1-27. Zajonc, R. B. (1998). Emotions. The handbook of social pshychology, 591-632.

Zajonc, R. B. (2001). Mere exposure: A Gateway to the subliminal. Current directions in pschychological science, 224-228.

(38)

38

Appendices

Questionnaire:

Version 1,2,6 and 8 included.

For more: r.kuilman.2@student.rug.nl Version 1:

Allereerst bedankt dat je deze enquête in wilt vullen. Het beantwoorden van de vragen zal afhankelijk van de versie twee tot vijf minuten in beslag nemen. Ik wil je vragen de gehele enquête in te vullen, niet volledig ingevulde vragenlijsten kunnen een probleem opleveren tijdens het analyseren van de data.

Succes!

(39)

39

(40)
(41)
(42)

42 Version 2

Allereerst bedankt dat je deze enquête in wilt vullen. Het beantwoorden van de vragen zal afhankelijk van de versie twee tot vijf minuten in beslag nemen. Ik wil je vragen de gehele enquête in te vullen, niet volledig ingevulde vragenlijsten kunnen een probleem opleveren tijdens het analyseren van de data.

Succes!

(43)
(44)
(45)

45 Version 6

Allereerst bedankt dat je deze enquête in wilt vullen. Het beantwoorden van de vragen zal afhankelijk van de versie twee tot vijf minuten in beslag nemen. Ik wil je vragen de gehele enquête in te vullen, niet volledig ingevulde vragenlijsten kunnen een probleem opleveren tijdens het analyseren van de data.

Succes!

(46)
(47)
(48)
(49)
(50)
(51)

51 Version 8

Allereerst bedankt dat je deze enquête in wilt vullen. Het beantwoorden van de vragen zal afhankelijk van de versie twee tot vijf minuten in beslag nemen. Ik wil je vragen de gehele enquête in te vullen, niet volledig ingevulde vragenlijsten kunnen een probleem opleveren tijdens het analyseren van de data.

Succes!

(52)
(53)
(54)
(55)
(56)
(57)
(58)
(59)

Referenties

GERELATEERDE DOCUMENTEN

Para ver de qué manera ‘Los niños de la furia’ se contrapone al discurso de seguridad de Felipe Calderón, se emplea un acercamiento discursivo de los estereotipos:

Table 3: Top URLs and Hashtags in User Groups By URL Bias Liberal URL Users Conservative URL Users Neutral URL Users.. Top

Overall, the baseline black and white cohort presented with similar ages, clinic and 24-hour blood pressures, but black adults had lower socioeconomic status and higher central

Comparison of DSM-5 criteria for persistent complex bereavement disorder and ICD-11 criteria for prolonged grief disorder in help-seeking bereaved children.. Boelen, Paul A.;

In this paper, our main contribution is that we present combinations of measurements for error modeling that can be used to estimate the quality of arbitrary GNSS receivers

[r]

This isn’t about global warming, where it might still just be possible to hold a principled sceptical position (although I very much doubt it); it’s about understanding how what you

(upper row 1), coiled-coil formation in the B-loop (blue) enables HA extension and insertion of the fusion peptide into the cell membrane (c1), followed by foldback of the hinge