PROCESSING DEPTH AS A POSSIBLE
EXPLANATION FOR THE EFFECTIVENESS OF
IRRITATING ADVERTISING
LONNEKE GROOTE HAAR
Student number: 2029308
Akkerstraat 4
9717KH Groningen
tel: +31 (0)6-54353013
e-mail: l.e.groote.haar@student.rug.nl
Supervisor: Dr. D. Trampe
Second supervisor: Dr. M.C. Leliveld
Faculty of Economics and Business
Marketing department
TABLE OF CONTENTS
Abstract 3
Introduction 4
Literature review 7
Irritation and ad effectiveness 7
Involvement and processing depth 9
Attitude formation 10
The moderating role of involvement and processing depth 11
Methodology 13
Participants and research design 13
Procedure 14 Measures 15 Commercial selection 16 Pre-test 16 Results 18 Manipulation checks 21
The effect of ad type and processing depth on Aad 22
Mediator effect of Aad between ad type and brand attitude 22
Discussion 24
The effect of ad type on Aad and brand attitude 24
Processing depth as the explanation for the effectiveness of irritating ads 25 Other explanations for effective irritating advertising 27
General conclusion 28
References 29
Appendix 1: Survey questions 32
ABSTRACT
The effectiveness of perceived irritation in advertising is researched in terms of attitude towards the ad and brand attitude. An irritating ad is assumed to lead to a more negative brand attitude than a neutral ad, but this effect is mediated by the attitude towards the ad. An
irritating ad thus leads to a more negative attitude toward the ad, which results in a more negative brand attitude. Besides this mediating effect of attitude towards the ad on the
relationship between ad type and brand attitude, processing depth is assumed to moderate the relationship between ad type and attitude toward the ad. The negative effect of irritation on attitude toward the ad will be reduced, when the arguments in the ad are deeply processed, resulting in a less negative attitude towards the ad. However, results show that ad type indeed leads to brand attitude through attitude towards the ad, but processing depth does not
INTRODUCTION
Commercial breaks are a source of irritation for almost everyone watching television. Irritation is seen as a negative emotion resulting in disliking an ad (De Pelsmacker & Van den Bergh, 1998). Several researchers investigated irritation in advertising, often resulting in outcomes as ad ineffectiveness and a negative brand attitude (Aaker & Bruzzone, 1985; Aaker & Stayman, 1990; De Pelsmacker & Van den Bergh, 1998). But if the consequences of an irritating ad are that negative, then why do consumers still buy products that are advertised in an irritating way? I suggest the Carglass commercial is seen as annoying by practically everyone receiving it, but in case of a broken windshield, Carglass is the first company popping up in consumers’ minds: Carglass is Europe’s leader in brand awareness in their industry (Kapferer, 2012). In the commercial, an actor acts as a Carglass employee who is on his way to a customer with a broken windshield. He talks in a dialect, possibly to show that Carglass’ service is excellent: conversations can even be held in dialect. The acting is very overdramatized, which is, I assume, the biggest irritation. Aaker & Bruzzone (1985) also claim that over dramatization of the situation will increase irritation, which will explain the irritation that this commercial evokes.
Bartos (1981: 138) states, “a certain amount of irritation helps to make advertising effective”. Irritation can actually be helpful for the ad to stand out of the clutter. Chakrabarty & Yelkur (2005) agree that irritation stimulates attention and say that it can be possible that consumers process the information without transferring the negative reaction to the brand. On the other hand, irritation towards an ad might inhibit counter argumentation, which may actually also enhance persuasion (Aaker & Bruzzone, 1985; Chakrabarty & Yelkur, 2005). The mind is then already occupied by the irritation, reducing space for elaborating on the actual arguments that are given.
like the commercial. However, the same study also shows that no specific difference in effectiveness exists for extremely liking or extremely disliking the ad. This implicates that irritation resulting in disliking the ad is not per se harmful for ad effectiveness, which is in line with the ‘law of extremes’ hypothesis by Silk & Vara (1974) in which a J-shaped relationship is predicted between ad attitudes and brand attitudes: both liked and disliked ads are more effective than neutral ads, but liked ads are in turn still more effective than disliked ads. Although irritating ads can be more effective than neutral ads, well-liked ads are probably even more effective than irritating ads (Chakrabarty & Yelkur, 2005).
De Pelsmacker & Van den Bergh (1998) take this even further in stating that irritation always leads to a negative attitude towards the ad (Aad), resulting in negative communication
and brand effects. Edell & Burke (1987) confirm this effect in saying that negative evoked feelings directly or indirectly lead to more negative brand effects. The impact of these negative feelings is mediated by the Aad, suggesting that an irritated (negative) reaction to an
ad result in a negative attitude towards the brand.
If it is really true that negative affective reactions to a commercial are that harmful for effectiveness, brand attitude and brand choice, then why does Carglass seem to be immune to these effects? Can it be that people really interested in Carglass’ services are not interested in the way these are advertised? Or can irritating ads indeed be effective, if specific conditions are taken into account? In this paper I will investigate the role of processing depth of the advertised arguments in this issue.
The depth of processing of the information in an ad depends on how much the consumer is involved in the advertised product. As a consequence of involvement, a message is possibly deeper processed than in case of no involvement at all. A lot is known about the role of involvement in creating attitudes based on advertisements, but this knowledge is mainly about the different ways of processing information as a result of the level of involvement, i.e. peripheral processing in case of low involvement or central processing in case of high involvement (Petty et al., 1983). In these cases, processing depth is seen as a dependent variable, depending on involvement in the advertised product. However, nothing is written about processing depth as a predictor for attitudes toward advertisements.
This raises the following research question: “Does attitude toward the ad mediate the
relationship between ad type and brand attitude and does processing depth moderate the negative effect of an irritating ad on attitude towards the ad?”. I predict that this can be
emotion resulting in a negative attitude towards the ad, since the emotion serves as the foundation for the attitude towards the ad, rather than deeply processed arguments. I expect attitude towards the ad will in turn predict brand attitude. For high processing depth I expect a person to be that much involved in the product that the evoked irritation is overruled by deeply processing of the arguments, for example about the quality of the product, resulting in an attitude in which irritation is ignored. Attitude toward the ad and brand attitude are then based on the processing depth of the arguments in the ad, rather than on the irritation that is possibly evoked by the ad.
Extensive research on irritation in advertising shows contradicting results, namely that irritating advertising can be effective and ineffective. However, the role of processing depth in the effectiveness of ads that are specifically perceived as irritating is not researched yet, in spite of its important predicting role in both brand attitude and ad attitude formation. Negative emotions as a result of an ad are mainly assumed to be ineffective, but still several researchers claim for the effectiveness of these. This paper will contribute academically to the explanation of the effectiveness of irritating ads since processing depth is researched as an explanation for irritating ads to be effective.
The practical relevance of this paper consists of the consequences that the hypothesized moderation effect has. If it is indeed true that processing depth moderates the negative effects of irritation in advertising, marketing managers can use this knowledge to make their advertising more effective. This might be helpful in situations of advertising that are already experienced as ineffective, explaining the effect of processing depth and the consequences of irritation in advertising in case of low processing depth. Besides that, it might contribute to the knowledge about how to advantage from irritation in advertising in answering the question on whether or not to use irritating advertising.
LITERATURE REVIEW
Irritation can be defined as provoking, annoying, causing displeasure and momentary impatience (Aaker & Bruzzone, 1985: 48, De Pelsmacker & Van den Bergh, 1998: 6) and can thus be seen as a negative emotion. In an advertising context, an irritating ad can be seen as negative affective-based, as the ad is based on a negative emotion (De Pelsmacker & Van den Bergh, 1998). Irritation does not always follow from content only. The kind of advertisement medium plays a role as well, as television seems the most irritating medium for advertising. The level of irritation consumers perceive also depends on product category: for example, toothpaste is a product with a higher than average irritation level. The frequency of exposure to the advertisement is another determinant for irritation and consumer characteristics play a role as well, as some consumers are more irritated than other consumers (De Pelsmacker & Van den Berg, 1998). It seems however reasonable for this research to narrow irritation causes to following from advertising content only.
Irritation and ad effectiveness
Contradicting statements exist about the relationship between irritating ads and ad
In other articles, however, irritation is seen as a distraction from counter arguing and an enhancer for advertising effectiveness and persuasion (Aaker & Stayman, 1990; Aaker & Bruzzone, 1985): the mind might be already occupied by the emotion, leaving no space for deeply processing the arguments that are given in the ad. In case of an irritating ad with a weak message that a consumer would reject when he/she should process it, this distraction and inhibition from counter arguing might prevent a negative brand attitude due to the weak message, and instead enhances persuasion. This reasoning is an argument for an irritating advertising being effective.
According to Bartos (1981), irritation can enhance ad effectiveness by helping the ad to stand out of the crowd of advertisements. Kwon (2014: 17) takes this paradox even further, stating “the experience of negative affect indicates to the individual a threat to the achievement of desired goals; in those experiences, the situation calls for systematic and attentive processing.” Negative affect thus leads to deeper processing of the advertisement in this point of view, using affect as the reason to centrally process the arguments given in the ad. This prediction is supported by Chakrabarty & Yelkur (2005: 40), who state “an irritating ad might strengthen brand attitude by promoting cognitive processing of the ad due to increased attention towards it.” In this research, 125 students were exposed to five non-comparative commercials with varying degrees of irritation. After showing the commercial, the participants had to fill out a questionnaire in which their brand attitudes, feelings evoked by the commercial, Aad, attitude towards advertising in general and perceptions of credibility
of the ad were measured. Results showed that no significant effect existed of ad irritation on brand attitude. This cognitive processing has to be in positive direction towards the arguments in the commercial. It is not taken for granted that processing leads towards persuasion, as processing can also increase possible aversion towards the brand when it leads to confirmation of already existing counterarguments towards the brand.
ad. Then, if the processed arguments are strong, the irritating ad might lead to a positive attitude towards the advertised brand.
Involvement and processing depth
“Individuals are involved in an issue when the topic has personal relevance, intrinsic importance, or significant consequences in their lives” (Yoon & Tinkham, 2013: 31, 32). In case of high involvement, one is more likely to process an advertised message centrally. According to Hoyer et al. (2013), the depth of processing of the message depends on consumers’ motivation, ability and opportunity to process information. If motivation, ability and opportunity are present, consumers will process the information in the message in a central way, meaning that they really think about the message.
Thus as a consequence of involvement, a message is deeper processed than in case of no involvement at all. Brown et al. (1998) state that product involvement stimulates central processing. This means a person diligently considers the information that is, in the person’s mind, central to the pros and cons of the product that the message is about. If involvement is low, processing occurs via the peripheral route of processing instead of the central route of processing. In processing a message peripherally, the pros and cons of a message are not considered deeply, but they are associated with simple cues that can be positive or negative in a person’s mind. For example, a person might accept the commercial, not because he/she diligently thought about the arguments, but because he/she very much liked the music played in the commercial, or because he/she saw the commercial during a pleasant meeting with a good friend. The other way around is also possible: a person might reject the commercial because he/she simply doesn’t like the music, without even thinking about the given arguments (Petty et al., 1983).
about the role the depth of processing plays within negative affect-based ads and the effectiveness of these for varying levels of processing depth.
Attitude formation
A lot of researchers concluded that the way of processing determines whether the attitude resulting from this thinking will be strong or weak: deep processing results in strong attitudes that are resistant to change (Hoyer et al., 2013, Barden & Tormola, 2014), but when the attitude change is based on peripherally processing the message, under low elaboration, these attitudes are relatively temporary and less predictive of behavior (Petty et al., 1983).
Attitudes are based on thought and on emotions (Hoyer et al., 2013: 129) and can be changed by the extent of processing of the message. This processing depth is mainly seen as a result of how much the consumer is involved in the advertised product. It is almost generally assumed that affect-based advertising is more effective for peripheral processing and argument-based advertising is more effective for central processing, since highly involved consumers are more likely to deeply (centrally) process a message and are less likely to experience affective responses (Dens & De Pelsmacker, 2010). It is however not per se true that central route processed messages are purely based on cognition, since I assume consumers can also be highly involved in an affect-based message and base their attitude on the emotion they feel as a result of the message. The emotion then acts as a source of information on which consumers will rely to evaluate and process the message as a consequence of the involvement in the messaged product (Hoyer et al. 2013). The formed attitude as a result of this centrally processing is then affect-based instead of cognitive based. When the affect-based message is processed peripherally, this can be based on, for example, the mental mood the message causes, but this mood is then not further elaborated on.
An irritating ad results in a negative perception of the ad (De Pelsmacker & Van den Bergh, 1998) and a general reduction in advertising effectiveness. Burke and Edell (1989) state that negative emotions always have a negative effect on advertising effectiveness. Since emotion caused by the ad can serve as the information to process in order to form an attitude, I expect an irritating ad will result in a negative attitude. I thus hypothesize that an irritating ad will result in a negative attitude towards the ad:
Advertising attitude can be defined as “a predisposition to respond in a favorable or unfavorable manner to a particular advertising stimulus during a particular exposure occasion" (MacKenzie et al., 1986: 131, 132). Strong attitudes predict brand consideration, intention to purchase, purchase behavior, and brand choice (Whan Park et al. 2010). A consumer will respond in a certain way to a commercial: he/she likes or dislikes the exposure to it. This liking or disliking will result in an attitude this consumer will hold towards the ad. According to MacKenzie et al. (1986), this Aad is a predictor for brand attitude formation.
Based on this reasoning, I expect attitude towards the ad to be a mediator in the relationship between extent of perceived irritation in a commercial and negativity of brand attitude:
H2: Attitude towards the ad is a mediator between extent of irritation in the ad and negativity of brand attitude.
The moderating role of involvement and processing depth
Rossiter (2014: 537) defines brand attitude as “the buyer’s evaluation of the brand with respect to its expected capacity to deliver on a currently relevant buying motive. This motive-anchored definition means that the prospective buyer can hold different overall attitudes toward the same brand depending on his or her main reason for buying it on a particular purchase occasion.” Involvement in the product plays an important role in this definition, since it determines the relevancy of the product for the consumer that might turn the negative attitude into a more positive one for the same ad. As stated above, in case of high involvement, it is more likely that the consumer will process the advertised arguments deeper than in case of lower involvement. In other words, involvement might reduce irritation from an ad, because consumers receiving the irritating ad might need the product and in doing so, pay less attention to the way the product is advertised. Instead of this, they deeply process the message about the advertised product, resulting in an attitude that is based on cognitive processing of the actual message instead of an attitude that is based on simple (negative, in case of irritation) cues.
deeply and the change of attitude might be positive -if the quality of the message is high and the arguments are strong-, instead of negative based on the emotion purely.
For disliked ads it is thus important that the message is processed deeply. If, like stated above, the message is processed deeply and arguments are strong, the ad might be effective in terms of creating a positive attitude towards the ad, even if the ad is perceived as irritating at first instance. Processing depth is then seen as an explanation for the effectiveness of irritating ads. I thus expect processing depth as a result of involvement to moderate the relationship between irritation as a result of the ad and the attitude towards the ad (in a negative way) as a consequence of this irritation.
H3: The deeper the advertised message is processed, the less negative influence the irritation evoked by the ad has on the attitude towards the ad
In the following conceptual model, the described relationships are made visual. The more irritation is evoked by the ad, the more negative the Aad will be. The more negative the Aad,
the more positive brand attitude will be in turn. However, if the ad is centrally processed as a result involvement with the advertised product, this relationship will be moderated negatively: the perceived amount of irritation that is evoked by the ad has less negative influence on the Aad.
METHODOLOGY
The aim of this research design is to collect the data to answer the research question stated above: “Does attitude toward the ad mediate the relationship between ad type and brand
attitude and does processing depth moderate the negative effect of an irritating ad on attitude towards the ad?” Does Aad mediate the relationship between irritation and brand attitude and
does the depth of processing of the advertised message moderate the effect of irritation on Aad? The research will take form of an experiment, in which ad type and processing depth will
be manipulated.
Participants and research design
Data collection took the form of an online experiment. The proposed research consisted of two variables: an irritating or a neutral ad and processing or not processing of the advertised message. The former was in the form of a real video of the ad that the respondent was
exposed to; the latter was in the form of an instruction to process the information in the video deeply while watching the video, or to watch the video without instruction.
Research took the form of a between-subject design (Aronson et al., 1998). 82
respondents were randomly assigned to different ad types and randomly assigned to different conditions of processing depth. This resulted in a 2 (ad type: neutral, irritating) x2 (processing depth: processing, no processing) between participants research design. Respondents were picked from the complete Dutch consumer market, since processing depth and irritation can be the case for every consumer; no restrictions or criteria were relevant. Table 1 shows the assignment of the different groups to variables and their conditions. In order to select appropriate advertisements, at least 20 participants were necessary for each advertisement condition, resulting in a need for at least 80 participants.
Irritating ad Neutral ad
Processing of the message Condition 1 Condition 3 No processing of the message Condition 2 Condition 4
Procedure
Since the research took place online, a link was posted on social media and friends and family were asked to share this link with a request to fill in the research. The online survey tool randomly assigned the individual respondents to the different conditions beforehand.
In the survey, participants first got to see a commercial of the irritating or the neutral ad. Before the participants got to see the commercial, they had to read a small instruction text, in which was written that they had to watch a small video and answer some questions after watching. Manipulation of the IV ‘ad type’ occurred by randomly assigning participants to the irritating ‘HoyHoy.nl’ commercial or the neutral ‘Consumentenbond’ commercial. Besides that, participants were randomly assigned to the processing or the not-processing condition for the manipulation of IV ‘processing depth’. The group that was assigned to the processing condition got a small instruction that they carefully had to take the arguments and information in the commercial into account, while the group that was assigned to the not-processing condition was only instructed that they just had to watch the video.
After watching the commercial, participants had to fill in the survey, in which their Aad and brand attitude was measured. At the end of the survey, participants had to indicate to
what extent they were actually irritated about the ad and to what extent they actually processed the information given in the commercial in order to check the manipulations for irritation and processing depth.
To be able to check for covariation of the control variables, questions about age, gender and education level were asked. Whether the participant makes his/her own decisions about health insurance was also an important control variable, since a participant whose parents arrange the health insurance would possibly be less concerned on this issue, which could have influenced the results. Lastly, the amount of switches in insurance company in the last years was a control variable, since a regular switcher might be more focused on the information given, which could also have influenced the results.
Concluding, ad type and processing depth represented the independent variables that were manipulated, Aad and brand attitude represented the dependent variables that were
Measures
Measurement of the DV’s
Aad was measured using a scale created by Burke & Edell (1989), to judge advertisement
using 24 adjectives about the commercial of which a complete overview can be found in appendix 1. This broad scale is chosen in order to avoid unilateral answering, since adjectives were presented both in negative and positive direction. During the actual study, the
participants were asked to judge the commercial that they were randomly assigned to on a scale from 1 to 7, with 1 being “I do not agree at all” and 7 being “I absolutely agree”. They were asked: “Some statements will follow about the commercial that you just saw. Please indicate how much you agree with each of the statements.”
Brand attitude was measured in terms of intention to use, since a website to compare health insurances is a service that is used instead of a product that is bought. Brand attitude was measured in three different dimensions with respect to user intention, based on scale that is created by Lin & Lu (2000): worthiness of reading, intention to use in the future, and bookmarking the site for future reference. Three statements are given: “The website is worth remembering”, “I will use this website in the future” and “I will remember this website for use in the future”. For all of these statements, participants have to indicate how much they agree to the statement on a scale from 1 to 7, with 1 being “I do not agree at all” and 7 being “I absolutely agree”. They were asked: “Some statements will follow about the commercial that you just saw. Please indicate how much you agree with each of the statements.”
Measurement of the manipulation checks of the IV’s
In order to check manipulation of the variables ‘irritation’ and ‘processing depth’, two questions are asked at the end of the survey. For irritation, participants are asked “How
irritating was the commercial that you saw on a scale from 1 to 7?”. The end poles of the scale were for 1 “not irritating at all” and for 7 “very irritating”. A manipulation check was done for processing depth as well, participants were asked: “How much attention did you actually pay to the arguments given in the commercial?”. They also had to indicate this question on a scale from 1 to 5, with 1 meaning “I did not listen at all to the arguments” and 5 “I paid a lot of attention to the arguments”.
Measurement of the control variables
effects of this issue. In order to do this, participants were asked whether or not they decide their selves about their health insurance. If they do, they also had to answer how many times they have been switched their health insurance company in the past two years.
Commercial selection
Irritating ad – Hoyhoy.nl
A commercial is irritating when a situation is overdramatized, when a person is put down with respect to knowledge or when an unsympathetic character is portrayed (Aaker & Bruzzone, 1985). In the Hoyhoy commercial, a sloth is hanging on a ceiling of a hospital, explaining the way he saved money on his health insurance. In 2015, this commercial won the ‘Loden Leeuw’, an award for the most annoying Dutch commercial, which is elected each year. According to a Dutch television program called ‘Radar’, consumers were especially irritated by this commercial because of the image that was created of the Dutch consumers: they feel compared to sloths that only want to save on their health insurance in order to spend the saved money on useless cosmetic surgeries like a vocal chord correction (Barbier, 2015). This is in line with Aaker & Bruzzone (1985) stating that putting a person down or portraying an unsympathetic character increases irritation in an ad.
Neutral ad – Zorgvergelijker Consumentenbond
If the commercial is perceived as being informative, irritation will be reduced (Aaker & Bruzzone, 1985). In the health insurance comparator of the ‘Consumentenbond’, only the possibilities of the website are announced in a very informative way. No story is created to bring the message, the message is only supported by a kind of a demonstration of the website. This informative way of advertising will be less irritating, according to Aaker & Bruzzone’s reasoning above.
Pre-test
In order to test whether these advertisements are indeed perceived as irritating and neutral, a pre-test was done. In this pre-test, 44 respondents participated, of which 17 were male and 27 were female. An assumed irritating commercial (Hoyhoy.nl) and an assumed neutral commercial (Zorgvergelijker Consumentenbond) were randomly assigned to
RESULTS
82 participants took part in the study. Of these, answers of one participant were deleted, since this participant did not answer any of the questions, which made the results of this survey unusable. Of the rest of the participants, 40 were male and 41 female.
Participants were randomly assigned to the different commercials (e.g. the irritating and the no-irritating commercial), which resulted in a division of 35 participants exposed to the irritating ad and 46 participants exposed to the not-irritating ad. An overview of these numbers can be found in table 1. A relatively major part of the participants joined higher education. Not a single participant joined VMBO, only 1 joined HAVO and only 4 MBO. Besides that, the average age (29) was quite low with respect to the overall society. In tables 3 and 4 an overview of means and standard deviations for the dependent variables brand
attitude and Aad can be found. Except for processing depth, all variables were measured on a
7-point scale; processing depth was measured on a 5-point scale. The mean of processing depth was moderate in relation to the overall scale.
Irritating ad - processing condition Irritating ad - non-processing condition Neutral ad - processing condition Neutral ad - non-processing condition Overall Deciding health insurance Yes No 11 5 14 5 22 3 15 6 62 19 Switches in last two years 0 1 > 2 5 6 5 8 6 5 13 7 2 8 6 1 34 25 13 Total 16 19 25 21 81
Table 2: general descriptives
Irritating ad Neutral ad Total effect of processing depth Processing of the message 3.10** (0.85) 3.73** (0.88) 3.32*** (0.95) No processing of the message 2.95** (0.95) 3.65** (0.85) 3.49*** (0.91) Total effect of ad type 3.02* (0.89) 3.70* (0.85) 3.40 (0.93) * p=0.00
** p=0.86 *** p=0.56
Table 3: means and (SD)’s for DV Aad
Irritating ad Neutral ad Total effect of processing depth Processing of the message 3.45** (1.65) 4.08** (2.03) 3.84*** (1.89) No processing of the message 3.00** (1.37) 4.02** (1.70) 3.53*** (1.62) Total effect of ad type 3.21* (1.50) 4.05* (1.86) 3.69 (1.76) * p=0.04
** p= 0.62 *** p= 0.51
In order to check for covariation between de control variables (age, gender, education level, deciding about health insurance, switch between health insurance) and the DV’s (Aad and
brand attitude) and to check whether the control variables undesirable influence the IV’s (perceived irritation and processing depth), several univariate analyses of variance and correlation checks were conducted. It might indeed be possible that one of the control variables influence to what extent one might be irritated by the ad or to what extent one will process the information in the ad. If one of these factors is influenced by the control variables, this needs to be taken into account in the subsequent analyses. For the univariate analyses of variance, gender and whether or not participants decided about their health insurance were IV’s, and the manipulation checks on irritation level and processing depth and Aad and brand
attitude were DV’s. No significant effects of the control variables existed. A correlation check was done between age, education level, the manipulation checks on perceived irritation and processing depth, number of switches between health insurance, Aad and brand attitude. No
significant correlations existed between control variables and the DV’s or between control variables and the manipulation checks for the IV’s. A correlation table is shown in table 5.
In order to be able to measure Aad, all 24 items that measured this variable were
1 2 3 4 5 6 7 8 Gender (1) - - - - Age (2) -.13** - - - - Education level (3) .37* -.05** - - - - # switches (4) -1.92** -1.19** .19** - - - - - Irritation (5) 0.07** 0.05** 0.02** -.12** - - - - Processing depth (6) -.17** -.17** .11** .11** -.28* - - - Aad (7) .06** -.09** .11** .12** -.81* .38* - - Brand attitude (8) .16** -.13** .14** .15** -.58* .48* .69* - * p<0.05 ** p>0.05
Table 5: correlations between control variables, manipulation checks on extent of perceived irritation and processing depth en DV’s Aad and Ab
Manipulation checks
An analysis of variance showed that the effect of ad type on extent of perceived irritation was significant. The condition ‘ad type’ was the IV and the extent of perceived irritation in the ad was the DV. The model showed that a significant difference existed in perceived irritation in the ad between participants that were exposed to the irritating ad (M=5.49, SD=1.52)
compared to the participants that were exposed to the neutral ad (M=3.11, SD=1.51),
F(1,80)=48.96, p=0.00. This indicates that participants that were exposed to the irritating ad
indeed judged the ad more irritating than participants that were exposed to the neutral ad did, so manipulation of ad type worked well.
An analysis of variance showed that the effect of processing depth on the extent to which participants processed the information in the ad was significant. The condition ‘processing depth’ was the IV and the extent of processing of the information in the ad was the DV. The model showed that a significant difference existed in processing depth between participants that were instructed to process the information (M=3.80, SD=0.95) compared to the participants that were not instructed to process the information (M=3.20, SD=1.13),
F(1,80)=6.74, p=0.01. This indicates that participants that were instructed to deeply process
The effect of ad type and processing depth on Aad
In order to test the main effect of ad type (IV) on Aad (DV) and the interaction effect of
processing depth (moderator), a two-way analysis of variance showed a significant main effect of ad type on Aad, F(1,81)=11.16, p=0.00, such that Aad was significantly more positive
for participants that were exposed to the neutral ad (M=3.70, SD=0.86) than for participants that were exposed to the irritating ad (M=3.02, SD=0.89). This indicates that exposure to an irritating ad evokes a more negative Aad than exposure to a neutral ad, which is in line with
hypothesis 1.
The main effect of processing depth on Aad was not significant, F(1,81)=0.35, p=0.56,
such that no significant difference existed between participants that were instructed to deeply process the information in the ad had no significant more positive Aad than participants that
were not instructed so.
The interaction effect of processing depth and ad type was also not significant,
F(1,81)=0.03, p=0.86. This means that the effect of ad type on Aad was not moderated by
processing depth of participants, which indicates that hypothesis 3 must be rejected: the effect of an irritating ad is not less negative in the processing condition than in the not-processing condition. Statistics can be found in table 3.
Mediator effect of Aad between ad type and brand attitude
In order to test whether a mediating effect of Aad exists between ad type (IV) and brand
attitude (DV), the Baron & Kenny four-step procedure was performed, followed by the bootstrap procedure. The model of the Baron & Kenny four-step procedure (R2=0.48) showed that the direct effect of ad type on brand attitude (path C) was significant, B=-0.84, p=0.03. The direct effect of ad type on Aad (path A) was significant, B=-0.67, p=0.00 and the direct
effect of Aad on brand attitude (path B) was also significant, B=1.31, p=0.00. Controlling for
the effect of Aad eliminated the significance (p=0.88) of the direct effect of ad type on brand
attitude (path C’), meaning that, taking Aad into account, no significant direct effect exists
between ad type and brand attitude. This indicates that Aad fully mediates the effect of ad type
on brand attitude, which supports hypothesis 2. These results are made visible in figure 2. To confirm the mediation effect, the Bootstrap method was performed to test the indirect effect of Aad as a mediator between ad type (IV) and brand attitude (DV). 5000
bootstrap samples were included in this test. The total indirect effect of Aad is 95% likely to
the indirect effect is significant, indicating that mediation exists which is in line with hypothesis 2.
DISCUSSION
The purpose of this study was to research whether perceived irritation in an ad would lead to a negative brand attitude via a negative attitude towards the ad, and whether processing depth moderates the relationship between perceived irritation and Aad. Several researchers state that
irritation can increase ad effectiveness (Aaker & Bruzzone, 1985; Aaker & Stayman, 1990; Bartos, 1981; Kwon, 2014; Chakrabarty & Yelkur, 2005; Holbrook & Barta, 1987; Silk & Vara, 1974), by helping the ad to stand out of clutter or by inhibiting counter argumentation in occupying the mind. Irritation also would stimulate attention towards the ad. Other research concluded that irritation is only harmful for ad effectiveness (Aaker & Stayman, 1990; De Pelsmacker & Van den Bergh, 1999; Edell & Burke, 1987; Greyser, 1973). According to this research, irritation can be labeled as a negative emotion that leads to a negative attitude towards the ad and the brand, reducing ad effectiveness.
This study had two aims, firstly to investigate whether the perceived irritation in the ad indeed led to a negative Aad and whether this negative Aad indeed led to a negative brand
attitude. Besides that, if irritation increases ad effectiveness, could the depth of processing of the information in the commercial be an explanation for this effectiveness?
The effect of ad type on Aad and brand attitude
The relation between perceived irritation and Aad was indeed supported by the results,
participants that rated the ad as irritating had a more negative Aad than participants that rated
that ad less irritating. At first instance, this means that an ad that is perceived as irritating indeed leads to a more negative Aad, which is in support of hypothesis 1. However, attitude
towards the ad does not immediately explain effectiveness, since this effectiveness is a result of the consequence the Aad has on brand attitude. The mediating effect of Aad on brand
attitude was indeed supported by the results, which is in line with hypothesis 2. Irritation led to a more negative attitude towards the ad and the brand, meaning that participants had a negative attitude towards both when perceived irritation was high. But when the relation between ad type and brand attitude was tested, while controlling for Aad, the direct relation
between ad type and brand attitude was not significant anymore. Participants that perceived the ad as irritating thus developed a negative brand attitude only if they also developed a negative Aad. These findings indicate that when a consumer develops a negative attitude
implicating that it is important to carefully considerate what makes an advertisement irritating when developing it. Aaker & Bruzzone (1985) already researched what increases irritation in advertising. Among others, an unbelievable situation, overdramatization, disrespect,
threatening of an important relationship or an unattractive character in the advertisement are factors that increase irritation and that have to be considered when developing an ad. Processing depth as the explanation for the effectiveness of irritating ads
How did processing depth play a role in these effects? Processing depth did not significantly moderate the effect of perceived irritation on Aad. Processing depth thus did not reduce the
negative effect of perceived irritation on Aad, indicating that it did not matter whether the
arguments in the commercial were deeply processed or not, Aad was negative in both cases.
Hypothesis 3 thus is rejected. The lack of support for this moderating effect can be found in the measurement scale. Processing depth was measured on a 5-point Likert scale: 1 = “I did not listen at all”, 2 = “I listened not that good”, 3 = “Neutral”, 4 = “I listened a bit”, 5 = “ I listened very well”. Afterwards, this scale was not accurate enough, since poles 2 and 4 actually say the same thing, namely that the participant did at least listen a bit to the information. Besides that, only 1 participant did not listen at all, which is understandable, since it is assumed that people always listen a bit to what is told, even if they are not interested. Although manipulation seemed to work marginally, with a significant, but marginal, difference between the two processing conditions (no processing: M=3.20; processing: M=3.80, p=0.01), it might be possible that processing depth was not measured accurately. This causes a bias in the results and makes it still impossible to create a clear frontier between participants that did listen and participants that did not listen even though manipulation seemed to work. With these results, it cannot be concluded that processing depth works as an explanation for the positive effects that perceived irritation on ad effectiveness could have according to the statements in the literature review.
in these products. Products for which advertisements will always be irritating can possibly include health insurances: consumers buy a health insurance because they are obliged to do so, not because a health insurance brings a lot of pleasure. Motivation for buying these products is not intrinsic but extrinsic, resulting in a lower motivation to really pay attention to the arguments given in the advertisement. If it is true that for certain product groups, motivation to process the information is low, then processing depth will never moderate the effect of the irritation that is evoked by the ad on the attitude toward that ad and consequently, brand attitude. This reasoning is however contradicting to reasoning of De Pelsmacker & Van den Bergh (1998), who stated that for approach products; products that bring pleasure, irritation in the ad lead to more negative effects than for products that are necessary; so called avoidance products. This dimension establishes the way consumers are involved in the product category that is (irritatingly) advertised: do they need the product or simply just want it? According to this research, involvement an avoidance product will be higher since the necessity for this product is high. Because of the high involvement, the advertised message probably will be processed centrally, reducing the influence of the irritation in judging the advertisement. The message is then thus processed deeply, rather than the irritation is emphasized in the consumers’ minds.
Thus, even if processing depth reduces the negative effect of irritation for other product groups than the product tested in this research, for certain product groups it might be possible that processing depth does not. For these ‘necessary’ products, irritation can cause damage of brand image if consumers do indeed not process the information in the ad. The attitude about the ad might then be formed by peripherally processing the cues in the ad. The irritation that is then caused by exposure to the ad might be a possible cue, resulting in a negative attitude toward the ad and consequently a negative attitude toward the brand.
Further research has to be done in order to find out what products are ‘necessary’ products for which processing does not reduce the negative effect of irritation on Aad.
Since no significant direct effect of processing depth on Aad existed, the measurement scale
Other explanations for effective irritating advertising
Then how can the Carglass commercial still be effective? Mittal (1990) states that brand attitude is only partly caused by Aad, since existing beliefs about the brand play an important
role in brand attitude formation as well. Beliefs about the functioning of the brand and the usability of the brand explain a part of the variance in brand attitude and Aad explains another
part. This means that a negative Aad not per se immediately leads to a negative brand attitude.
It thus is possible that consumers have positive beliefs about Carglass, for example because they heard positive stories about Carglass or they have a positive experience in using
Carglass’ services. If these beliefs partly influence brand attitude, Aad is not that meaningful
for brand attitude compared to when these consumers did not have any additional belief about Carglass.
However, Aad can also confirm already existing beliefs about the brand, reinforcing
the brand attitude that a consumer already holds. For an advertisement that is perceived as irritating, this can be harmful for the brand if the consumer already held a negative attitude towards that brand, since the ad in that case confirmed this negative attitude. But if one already holds a positive belief about the brand and a positive attitude towards the ad confirms this belief, this would strengthen brand attitude. For Carglass, the negative Aad might not be
that harmful, since a consumer might already have positive beliefs about Carglass’ service. However, if a consumer has a negative experience with Carglass, the negative Aad might
reinforce this negative attitude. MacKenzie et al. (1986) concluded on the other hand that Aad
both influences brand attitude and cognitive reactions to the brand, but that these cognitive reactions evoked by the ad do not influence brand attitude, meaning that brand attitude is simply caused by Aad. According to this statement, beliefs about the quality of the brand,
before or after exposure to the ad, do not influence brand attitude. Further research is
necessary in order to discover to what extent Aad solely influences brand attitude, independent
from other factors that might possibly influence brand attitude together with Aad.
is however tried to distribute the survey in other networks as well. Besides this, questioning of education level was confusing: participants might have understood the question as ‘what is your education level, including the education that you are currently joining’ or ‘what is your highest finalized education level’. Answering the question using the first explanation might give a higher general education level than the second explanation. Nevertheless, for
managerial implications, young and highly educated people are an interesting target group, since these have relatively more future potential, which makes it still interesting to take the results of this research into account.
General conclusion
Purpose of this study was to answer the following research question: “Does attitude toward
the ad mediate the relationship between ad type and brand attitude and does processing depth moderate the negative effect of an irritating ad on attitude towards the ad?”. Results showed
that participants that were exposed to the irritating ad indeed had a more negative Aad than
participants that were exposed to the neutral ad. Besides that, participants that had a were exposed to the irritating ad indeed had a more negative brand attitude, however, this negative brand attitude was a consequence of the negative Aad participants developed because of the
irritating ad. This means that the relationship between ad type and brand attitude is indeed mediated by Aad. Processing depth was hypothesized as an explanation for the effectiveness
of irritating advertisements. However, results showed that no moderation effect of processing depth existed on the relation between ad type and Aad. The negative effect of irritation in the
ad on Aad thus was not reduced by deeply processing the information in the ad.
This practically indicates that, according to the results of this research, irritating advertisement is not effective, even when the information is deeply processed. This means that in developing advertisements, factors that increase irritation should be taken into account very carefully, since irritating ads are harmful for brand attitude, even if the consumers pays careful attention to the arguments in the ad.
REFERENCES
Aaker, D.A. & Bruzzone, D.E. 1985. Causes of irritation in advertising. Journal of
marketing. 49: 47-57.
Aaker, D. & Stayman, D. 1990. Measuring audience perceptions of commercials and relating them to ad impact. Journal of advertising research. 30: 7-18.
Andrews, J., Durvasula, S. & Akhter, S. 1990. A framework for conceptualizing and measuring the involvement construct in advertising research. Journal of advertising. 19(4): 27-40.
Aronson, E., Wilson, T.D. & Brewer, M.B. 1998. Experimentation in social psychology. The Handbook of Social Psychology. 1: 99-142.
Barbier, J. 2015. HoyHoy.nl wint Loden Leeuw voor meest irritante reclame.
Volkskrant. [Online]. 15-01-2015. [Date accessed: 18-04-2015]. Available from:
http://www.volkskrant.nl/binnenland/hoyhoynl-wint-loden-leeuw-voor-meest-irritante-reclame~a3823505/
Barden, J. & Tormola, Z. 2014. Elaboration and attitude strength: the new meta-cognitive perspective. Social and Personality Psychology Compass. 8: 17-29.
Bartos, R. 1981. Ads That Irritate May Erode Trust in Advertised Brands. Harvard
Business Review. 59(4): 138-140.
Brown, S.P., Homer, P.M. & Inman, J.J. (1998) A meta-analysis of relationships between ad-evoked feelings and advertising responses. Journal of Marketing Research. 35(1): 114-126.
Burke, M. & Edell, J. 1989. The impact of feelings on ad-based affect and cognition.
Chakrabarty, S. & Yelkur, R. 2005. The Effects of Ad Irritation on Brand Attitudes.
Journal of Promotion Management. 11(2-3): 37-48.
Dens, N. & De Pelsmacker, P. 2010. Consumer response to different advertising appeals for new products: the moderating influence of branding strategy and product category involvement. Journal of brand management. 18: 50-65.
De Pelsmacker, P. & Van den Bergh, J. 1998. Advertising content and irritation: A study of 226 TV commercials. Journal of international consumer marketing. 10(4): 5-27.
Edell, J.A. & Burke, M. 1987. The power of feelings in understanding advertising effects. Journal of consumer research. 14(3): 421-433.
Greyser, S.A. 1973. Irritation in Advertising. Journal of Advertising Research, 13(1): 3-10.
Holbrook, M. & Batra, R. 1987. Assessing the role of emotions as mediators of consumer responses to advertising. Journal of consumer research. 14(December): 404-420.
Hoyer, W.D., MacInnis, D.J. & Pieters, R. 2013. Consumer Behavior (6th ed., international ed.). South-Western, Cengage Learning.
Kapferer, J. 2012. The new strategic brand management: advanced insights and strategic thinking. [E-reader version] Retrieved from
https://books.google.nl/books?id=JnK8pRjSGZUC&pg=PA188&lpg=PA188&dq=carglass+b rand+awareness&source=bl&ots=BW5gBc78KZ&sig=5GMCXZBAbHxGQjXueLVGvcdZP
z4&hl=nl&sa=X&ei=z5ESVY-EHImwPPrqgdgH&ved=0CDEQ6AEwAA#v=onepage&q=carglass%20brand%20awareness &f=false .
Lin, J. & Lu. H. 2000. Towards an understanding of the behavioral intention to use a web site. International journal of information management. 20: 197-208.
MacKenzie, S.B., Lutz, R. & Belch, G.E. 1986. The role of attitude toward the ad as a mediator of advertising effectiveness: a test of competing explanations. Journal of marketing research. 23(2): 130-143.
Mittal, B. 1990. The relative roles of brand beliefs and attitude towards the ad as mediators of brand attitude: a second look. Journal of marketing research. 27(2): 209-219.
Petty, R.E., Cacioppo, J.T. & Schumann, D. 1983. Central and peripheral routes to advertising effectiveness: the moderating role of involvement. Journal of consumer research. 10(2): 35-146.
Rossiter, J.R. 2014. ‘Branding’ explained: defining and measuring brand awareness and brand attitude. Journal of brand management. 21: 533-540.
Silk, A. J. & Vara, T. G. 1974. The influence of advertising’s affective qualities on consumer response. In G. D. Hughes & M.L. Ray (Eds.), Buyer/Consumer Information
Processing. Chapel Hill, NC: University of North Carolina Press: 157-186.
Whan Park, C., MacInnis, D. & Priester, J. 2010. Brand attachment and brand attitude strength: conceptual and empirical differentiation of two critical brand equity drivers. Journal
of marketing. 74: 1-17.
Yoon, H. & Tinkham, S. 2013. Humorous threat persuasion in advertising: the effects of humor, threat intensity, and issue involvement. Journal of advertising. 42(1): 30-41.
APPENDIX 1: Survey questions 1. What is your gender?
o Male o Female
2. What is your age? …. years
3. What is your education level? o VMBO o HAVO o VWO o MBO o HBO o WO o Other, … Processing condition
A short video will be shown about a website for comparing health insurances. Watch the video and make sure you pay careful attention to the information that is given in the video.
Not-processing condition
A short video will be shown about a website for comparing health insurances. Watch the video and click ‘next’.
4:
Totally disagree Totally agree
Believable 1 2 3 4 5 6 7 Energetic 1 2 3 4 5 6 7 Exciting 1 2 3 4 5 6 7 For me 1 2 3 4 5 6 7 Gentle 1 2 3 4 5 6 7 Humorous 1 2 3 4 5 6 7 Imaginative 1 2 3 4 5 6 7 Informative 1 2 3 4 5 6 7 Ingenious 1 2 3 4 5 6 7 Irritating 1 2 3 4 5 6 7 Meaningful to me 1 2 3 4 5 6 7 Merry 1 2 3 4 5 6 7 Novel 1 2 3 4 5 6 7 Playful 1 2 3 4 5 6 7 Phony 1 2 3 4 5 6 7 Ridiculous 1 2 3 4 5 6 7 Serene 1 2 3 4 5 6 7 Soothing 1 2 3 4 5 6 7 Tender 1 2 3 4 5 6 7 Terrible 1 2 3 4 5 6 7 Unique 1 2 3 4 5 6 7 Valuable 1 2 3 4 5 6 7 Vigorous 1 2 3 4 5 6 7 Worth remembering 1 2 3 4 5 6 7
Totally agree Totally disagree 5. The website is worth remembering
1 2 3 4 5
6. I will use this website in the future
1 2 3 4 5
7. I will remember this website in order to use it in the future
1 2 3 4 5
8. Do you decide about your health insurance yourself? (n.b. choose ‘no’ if, for example, one of your parents decide about your health insurance for you)
o yes o no
9. How often did you switch health insurance company in the last two years? o 0 times
o 1 time
o 2 times or more often
10. How irritating did you find the commercial on a scale from 1 to 7?
Not irritating at all 1 2 3 4 5 6 7 Very irritating 11. To what extent did you really listen to the information given in the commercial? I did not listen at all 1 2 3 4 5 I paid a lot of attention 12. What was the commercial about?
o Hoyhoy.nl
APPENDIX 2: Management summary
A lot of different conclusions are drawn about the effect of irritating advertisement on the consumers’ attitude towards the ad and the advertised brand. In this thesis is researched if it matters whether the consumer processes the information in the ad deeply or not. In other words: does the consumer judge the irritating ad less negative if he/she is focused on the arguments and information given in the ad?
Irritation in the ad is caused by several factors: the medium, the content of the ad, the repetition of exposure to the ad and the kind of product. Contradicting conclusions are drawn with respect to the effectiveness of irritating advertisement: on the one hand, an irritating ad may lead to a negative attitude towards the brand. Irritation is labeled as a negative emotion and negative emotions towards a brand always lead to negative feelings towards the brand that is advertised. On the other hand, several researchers claim irritating ads to stand out of the clutter and that irritation in the advertisement causes processing of the information and arguments that are given in the ad. If these arguments are strong, this may lead to a more positive attitude towards the brand than in case of no processing of the information.
In this thesis is researched whether the focus on the information in the irritating ad influences the negative effect of the irritation on the attitude towards the ad and the brand. Results showed that it does not matter whether the information in the irritating ad is deeply processed; the attitude towards the ad remains negative. This means that an irritating ad almost always leads to a negative attitude towards the ad. However, an ad that is perceived as irritating does not directly lead to a negative attitude towards the brand. No direct relation exists between the irritation in the ad and the attitude towards the ad; while there is an indirect effect. This means that when the attitude towards the ad is negative as a consequence of the irritation, the attitude towards the brand will be negative as a consequence of the negative attitude towards the ad. Irritation thus leads to a negative brand attitude via the attitude towards the ad. The depth of processing of the information in the ad does not moderate the relation between the perceived amount of irritation in the ad and the attitude towards the ad.
