In-stream video advertisements : to fit or not to fit

In-stream video advertisements:

To fit or not to fit.

Master Thesis

Joke Oppenhuisen

peter.jonker@student.uva.nl Word count: 5075

Summary

Advertisers are constantly seeking for new techniques to counter advertising avoidance. One strategy for doing this is to use forced exposure advertising; a technique that is often used with in-stream video advertisements on YouTube. These ads are often found to be positive in influencing brand recall, but negatively in

influencing brand evaluations. This study investigated if there are ways in which these negative effects on brand attitude could be prevented. More specifically, this research investigated what the effects of congruency (congruent vs. Incongruent) and ad-position (pre-roll vs. mid roll) of in-stream video ads are on brand attitude, and if there are moderating effects of intrusiveness and ad-position. An online experiment was conducted in order to test the proposed hypotheses. There was not found any support for the hypotheses, but theoretical implications are given.

1. Introduction

Since the introduction of online advertising in the mid-1990s, marketers and researchers speculated about a new era in advertising (Rust & Varki, 1996). From 2006 to 2012, only digital video advertising showed continuous growth as a percentage of total internet advertising revenues in the United States (Interactive Advertising Bureau; PricewaterhouseCoopers, 2013). Global total internet advertising revenue is forecast to grow from $135,42 billion in 2014 to $239,87 billion in 2019. Therefore, internet advertising is expected to exceed TV to become the largest single advertising category in 2019 (PricewaterhouseCoopers, 2015). Although these trends show that marketers have constantly increased online video advertising expenses, research has shown that consumers are increasingly avoiding any source of annoyance that disrupt their access to content of the internet (Baek & Morimoto, 2012; Edwards, Li, Lee, & Taylor, 2002). Together with the increase in popularity of rich media ads, new delivery techniques have been developed to reduce avoidance to online ads (Edwards et al., 2002). However, it is still unclear in what way these different techniques should be used in order to optimize advertising effects.

One obvious strategy for preventing avoidance is the use of Forced Exposure advertisements (Fransen, Verlegh & Kirmani, 2015). For example, visitors of online video-sharing website YouTube often are forced to view video advertisements before they are able to watch the desired video content. Research about forced exposure in-stream video ads has shown mixed results. For instance, Hegner, Kusse, and Pruyn (2014) found that these types of advertisements could brand evaluations negatively because consumers perceive forced exposure ads to be as more intrusive than traditional forms of advertisements. According to Edwards et al. (2002), feelings of

irritation can be elicited and advertisements are often avoided when they are

perceived as intrusive by recipients. Subsequently, this could result in negative brand and advertisement evaluations. On the other hand, forced exposure ads are found to be positively affecting brand recall and brand recognition (Li & Lo, 2015). Li and Lo (2015) investigated whether brand recall and recognition were affected by ad characteristics such as format, timing, and context. Their results showed that ad-context incongruity enhanced brand name recognition when the ad was inserted after the content was completed. Their findings also provided empirical evidence on the effects of different ad positions on brand recognition; when the context of

commercials is thematically congruent with the video program, a mid-roll ad

enhances brand name recognition. But when the ad content is incongruent, a post-roll format generates better brand name recognition.

In brief, previous research about in-stream video advertisements shows that these ads could improve brand recall and recognition, but they influence brand and advertisement evaluations negatively because this advertising format often interrupts one’s viewing of a video. However, there are still some uncertainties and

shortcomings when we look closer at the literature about online in-stream video advertisements. Firstly, it is still unclear what the effects on brand attitudes are of in-stream video ads characteristics. Several characteristics of interruption, such as timing and context, can affect an individual’s response to the interruption (Kirmeyer, 1988). Consequently, it is should be investigated what the effects of congruency and ad position of online video advertisements on brand attitudes are. This could result in an important theoretical and practical contribution, as it will be clearer how to minimize the negative perceptions of in-stream video advertisements on brand attitude, while taking advantage of the potential effective benefits of these ads on brand recall and

recognition. Secondly, perceived ad value as a potential moderating variable has not been taken into account in previous research about the effects of online in-stream video advertisements. Edwards et al. (2002) found that perceived value of pop-up advertisements is a possible positive social influence that may affect the perception of intrusiveness. Because pop-up ads are somehow similar to in-stream video

advertisements, as in that they are both seen as forced exposure advertisements (Hegner et al., 2014), it is relevant to test whether similar effects are found for in-stream video ads. This research will focus on investigating the effects of ad

characteristics (e.g., context and position) of online in-stream video advertisements on brand attitude, and whether these effects are moderated by intrusiveness and ad value. Besides the theoretical contribution this research will provide, results of this research will help practitioners to decide where (congruent vs. incongruent / pre-roll vs. mid-roll) to place their advertisements on YouTube, considering what the effects will be on the evaluations of the advertised brand. The research question of this study sounds as follows:

Research Question: What are the effects of ad-context congruity and ad-position of

in-stream video ads on YouTube on brand attitude, and what are the moderating roles of perceived intrusiveness and perceived ad-value?

2. Theoretical background

In this section theory will be discussed to develop hypotheses which will be tested in the current research. In the first part, theories concerning the effects of ad-context congruity on brand attitude will be discussed. Subsequently, in relation to these

theories, expectations about the moderating role of intrusiveness will be mentioned. In the second part, theories which will explain expectations about the effect that

ad-position has on brand attitude will be discussed, and the moderating role of perceived ad value will be argued.

2.1 The effect of ad-context congruity on brand attitude

Consumers develop knowledge about brands when encountering brand-related information such as advertising, packaging, and trying out products. Over time, as brands become increasingly familiar, consumers’ brand knowledge and perceptions about a brand together build a brand schema which can be defined as ‘all their previous knowledge of a brand’ (Dahlén, 2005). An individual’s brand schema is the end result of all previous processing that he or she has dedicated for a particular brand, and it pays off by reducing the effort that is required on future encounters with the brand. When consumers face new information about a brand, the brand schema about this brand is activated and the new information will be interpreted in light of this schema. The brand schema also forms expectations about where and how the brand will be processed in the future (Alden, Mukherjee, & Hoyer, 2000; Dahlen et al. 2005). Halkias and Kokkinaki (2014) used the term schema (in)congruity to refer to information presented in persuasive messages that conforms (or does not conform) to consumers’ expectations and their perceptions of what is relevant, according to their brand schema.

Ad-context congruity is described as the matching of content, or the matching of an affective or emotional execution style or tone, in ads and media vehicles

(Feltham & Arnold, 1994; Kim & Sundar, 2010; Moore, Stammerjohan, & Coulter, 2005; Sharma, 2000). Scientists have, among other things, investigated the effects of advertisement (in)congruity on information processing (Dahlén, Rosengren, Törn, & Öhman, 2008; Kent & Allen, 1994; Meyers-Levy & Tybout, 1989) and attitudes toward the brand and advertisement (Lee, 2000; Peracchio & Meyers-Levy ,1994;

Phillips,2002). When information about a brand conforms to expectations (such as when a lifestyle brand advertises in a lifestyle magazine), there is little need to process this information in the ad in-depth to relate it to the existing brand schema (Kent & Allen, 1994). These ads are more predictable and do not stimulate additional elaboration (Halkias & Fokkinaki, 2014). Contrary, when information does not

conform expectations (such as when a car brand advertises in a beauty magazine), this will encourage people to pay attention to the information presented in the ad and motivates them to think about it (Fiske, Kinder, & Larter, 1983). Experimental research of Moore et al. (2005) has shown that ads which are incongruent elicit more processing than ads which are congruent. Their explanation of these results was based on their notion that ads which are placed in a thematically incongruent medium often do not fit with the brand’s existing associations or previous media placements. In other words, incongruent ads conflicts with the brand schema, which will lead

consumers to engage in more extensive processing to resolve the incongruence. While congruent ads do the opposite; they will lead consumers to engage in less extensive processing of brand information (Meyers-Levy & Tybout, 1989; Dahlén et al., 2008). Subsequently, it has to be discussed, in what ways information processing could affect brand attitude. According to the literature, information processing affects brand attitude in two ways. First, more in-depth processing leads to higher confidence in resolution judgments, which is usually interpreted as a greater liking of the ad and the brand (Lee, 2000). Secondly, it seems that incongruent information can produce some kind of entertainment value. That is, it challenges the consumer and presents him or her with some kind of puzzle to solve. Solving the puzzle, caused by the thematically incongruent ad placement could in turn improve brand evaluations. Peracchio and Meyers-Levy (1994) conducted an experiment where they compared

the effects of ambiguous advertisements pictures versus non-ambiguous advertisements. Their findings indicated that ambiguity could enhance brand evaluations if ad recipients are sufficiently motivated to mentally complete the ambiguous advertisements, or in other words, if they were able to complete the puzzle. Phillips (2002) conducted a similar experiment, where he compared effects of complex ads with an explanatory headline versus effects of complex ads without an explanatory headline. Results indicated that that adding an explanatory headline that offers a clue to the image’s meaning decreased ad and brand liking by giving away the ads’ message and decreasing subjects’ enjoyment of interpreting the ad by

themselves. Even though these experiments used complex advertisements as stimulus material, Dahlen et al. (2008) implicated that a similar process occurs when recipients are presented with context incongruent advertisements. Recipients seem to process advertisements more (less) when advertising is placed in a thematically incongruent (congruent) medium, and therefore brand associations will be perceived more (less) strongly when advertising is placed in a thematically incongruent (congruent) medium. When we apply these findings to the current research, the first hypothesis sounds as follows:

H1a: Brand attitudes will be more positive when the ad is placed in a thematically incongruent context than when the ad is placed in a thematically congruent context.

However, perceptions of intrusiveness evoked by in-stream video ads could affect the effect that context (in)congruity has on brand attitude. Ha (1996) describes

intrusiveness as ‘the interruption of editorial content, which could lead to negative feelings and reactions towards the content’. Intrusiveness can be thought of as a psychological consequence that arises when an individuals’ cognitive process is activated (Edwards et al., 2002). According to Edwards et al. (2002), intrusiveness

could affect advertising effectiveness negatively because it could result in reactance towards the advertisement. Brehm and Brehm (1981) describe reactance as a

psychological process in which recipients perceive that a certain persuasion attempts as intrudes their freedom, which results in ignorance or resistance towards the

persuasion attempt. Consequently, it can be expected that intrusiveness could serve as a moderator on the effect of congruency on brand attitude. More specifically, the negative effect of congruency on brand attitude can be expected to be strengthened when individuals perceive these ads to be as intrusive, or weakened when they perceive these ads to be less intrusive. Therefore, it could be expected that:

H1b: The negative effect of ad-context congruity on brand attitude will be moderated by perceived intrusiveness; the negative effect of ad-context congruity on brand attitude will be stronger when perceived intrusiveness is high, and less stronger when perceived intrusiveness is low.

2.2 The effect of ad-position on brand attitude

Expected variations in brand attitude caused by the position of in-stream video advertisements (pre-roll vs. mid-roll) will be discussed with theories explaining the influence that ad-position has on perceived intrusiveness. In a survey among

consumers from the U.S.A, Krugman, Bauer and Greyser (1968) identified some of the main motivations of consumers for criticizing ads. The most salient motivation which the authors found among their target group was the fact that advertisements are often causing irritation because it is interrupting editorial content. Advertisers seek to interrupt editorial content often with their ads, because the main objective of

persuasive communication is to get noticed. However, advertising limits the numbers of actions that consumers can take to attain their goals, since advertising often

interferes with goals of consumers. As discussed earlier, when advertisements are interrupting editorial content, they can be seen as intrusive (Ha, 1996). This can cause recipients to feel annoyed and irritated, which could spill over to brand information and thus lead to a more negative brand evaluations (Aaker & Bruzzone, 1985).

The only question which has to be answered is for which in-stream ad placement position (pre-roll vs. mid roll) intrusiveness will be expected to be higher (lower). This will be discussed with literature that focuses on a recipients’ attention while consuming online video content. As a result of limited and selective attention of recipients, messages in video ads may be processed and understood differently by them (Basil, 1994). Attention is the degree to which processing activity is devoted to a specific stimulus and will vary according to the stimulus and the recipient

(Solomon, 2013). When we take a look at this from the perspective of information processing, a larger amount of attention encourages deeper processing of a persuasive message (Yun & Kim, 2005). Nelson and Meyvis (2008) stated that, even though online media is relatively high attention-demanding, an adaption effect can occur when individuals are watching online videos. This means that, as viewers continue to receive a certain stimulus, their responses to the same stimulus will decrease. This leads to an increase of a viewers’ attention to the video as the video progresses and eventually, after reaching a peak, a decrease in attention. Therefore, viewers are expected to be more mentally engaged in the middle of the video than before or after it (Li & Lo, 2015). When we relate these findings to the current research, a viewers’ attention will be expected to be higher when the mid-roll in-stream advertisement will be shown, and is expected to be lower when the pre-roll in-stream advertisement is shown. It can be expected that when recipients’ attention is higher, advertisements are seen as more interruptive than when recipients’ attention is lower. Consequently, it

mid-roll advertisement should be seen as more interrupting and intrusive as pre-roll advertisements. Because intrusiveness will have a negative effect on brand attitude, the second hypothesis of this research sounds as follows:

H2A: Ad-position will have a negative influence on brand attitude; brand attitude will be higher when participants are exposed to a pre-roll in-stream video advertisement than when participants are exposed to a mid-roll in-stream video advertisement.

However, Paasedeos (1990) found out that, when ads contain useful information, and consequently are perceived as valuable, they elicit less irritation and avoidance and are therefore perceived as less intrusive. According to Ducoffe (1995), ad value could be defined best as ‘an overall representation of the worth of advertising to

consumers’. Ducoffe’s (1995, 1996) studies indicate that ad value is positively

correlated with the entertainment value, and perceived informativeness of an ad. Both information and entertainment value are essential for communication exchanges between advertisers and consumers. Therefore, perceived intrusiveness could be affected when an ad is perceived as valuable by recipients, that is perceived as informative and entertaining. When ads are not perceived as valuable by recipients, they could be perceived as unwelcome and therefore perceived as more intrusive. (Ducoffe, 1996). Hence, it could be expected that advertising value works as a

moderator on the effect that ad-position has on brand attitude. The final hypothesis of this research sounds as follows:

H2B: The negative effect of ad position on brand attitude will be moderated by ad value; the effect of ad position on brand attitude will be less stronger when perceived ad-value is high and less stronger when perceived ad-value is low.

Figure 1.

Conceptual model of the theoretical framework

3. Method

3.1 Design, procedure and sample.

In order to test the proposed hypotheses a 2 x 2 factorial between-subjects design was used. An experimental design was used in order to manipulate independent variables congruency (congruent vs. incongruent) and ad-position (pre-roll vs. mid-roll). The experiment was placed in online survey software Qualtrics. A convenience sample and snowball method were used to gather participants; an invitation was send by e-mail and via social media, containing a short introduction to the experiment and a link to enter the experiment. Participants were also asked if they could send this invitation to potential participants in their surroundings. When participants clicked on the link, a small introduction to the experiment was presented; they were told that the

experiment was about the effects of online videos. Participants were not told that the experiment was actually about the effects of the advertisements shown in the videos, because this could create a setting wherein participants would focus too much on the

Perceived intrusiveness -( Ad-context congruity Brand Attitude + Ad-position (pre-roll vs. mid-roll) -( Perceived ad-value -

advertisements presented in the video. This could in turn affect the ecological validity negatively, because people are usually not watching YouTube videos for the ads but for the actual content. After the introduction, participants were exposed to one of the four YouTube videos in which the independent variables were manipulated.

Subsequently, participants were asked to fill in a questionnaire related to the video they had seen, in which brand attitudes were measured as well as demographic variables.

In total 134 participants participated in the online experiment. From these 134 participants, 115 were selected to be taken into analysis because there were 19 people who did not finish the experiment sufficiently. The average age of the participants was 33 (SD=14,19) and varied from 11 years old to 84 years old. Of all participants, 52,6% was male and 47,4% was female. The majority of participants indicated that university was their highest level of education (43,9%), followed by higher

professional education (34,1%). Participants were randomly assigned into one of the four conditions.

3.2 Stimulus material

The advertisement that was used to create the stimulus material was an existing Dutch video commercial about the Ford C-Max Ecoboost car, which had a total length of 30 seconds. An existing brand and commercial were chosen to create a more realistic setting and therefore improve ecological validity. In order to manipulate congruency of the ad, two videos were selected which were shown alongside the commercial. The video that was used for the congruent condition was a fragment of the Dutch

television show ‘De Slechtste Chauffeur van Nederland’, which is an entertainment show about the worst car drivers of the Netherlands and broadcasted by the Dutch

the content was expected to be congruent with the advertisement about the Ford C-Max Ecoboost. The show ‘De Slechtste Chauffeur van Nederland’ can be labeled as entertainment television. In order to keep the two videos as similar as possible, a similar kind of video needed to be found for the incongruent condition; the only thing that had to be different about this video was that it should incongruent with the Ford advertisement. The video that was used for the incongruent condition was a fragment of ‘Wie is de Sjaak’, which also is an entertainment program broadcasted by RTL 5. In the fragment a woman had to perform different tests on the street in order to win money prizes. The fragment was not about cars (and there were no cars shown nor mentioned during the fragment) and therefore not expected to be incongruent with the Ford commercial. Both videos, including the ads, had a similar length of 2 minutes and 48 seconds.

In order to manipulate ad-position, the Ford commercial was either placed in front of the video (pre-roll) or in the middle of the video (mid-roll). When the

commercial was shown, the text ‘advertisement’ with a timer counting down until the ad was finished was shown on the bottom left. This was implemented because

YouTube implements a similar timer during the presentation of their ads. This made the YouTube content seem more realistic and participants were able to recognize the advertisement during the video.

To summarize, four videos were created in order to manipulate the

independent variables congruency and ad-position. The first was a congruent video with a pre-roll ad. The second video was a congruent video with a mid-roll ad. The third video was an incongruent video with a pre-roll ad. And the final video was an incongruent video with a mid-roll ad. Each of these videos were published on YouTube and these videos were embedded in the online questionnaire in Qualtrics.

Because an online experiment was used, it was possible to show the videos and advertisements in a YouTube video player, which improved ecological validity; using a different video player would make the experiment seem less realistic.

Figure 2.

Examples of the Ford advertisement (left), congruent video (middle) and incongruent video (right)

3.4 Measures

The definition of brand attitude used in current research is: ‘the extent to which participants think the brand is good and attractive’ Bruner (1998). In order to measure brand attitude, two 7-point bipolar items (totally not agree – totally agree) were used, with which participants could report their attitude towards Ford. The two items together proved to be a reliable scale to measure brand attitude (α = 0,94). Perceived intrusiveness of the advertisement was measured by the scale used by Edwards et al. (2002). Participants were able to report perceived level of intrusiveness of the ad by answering three 7-point bipolar items (totally not agree – totally agree); they could indicate whether they thought the advertisement presented in the video was intrusive, distractive and disturbing. These three items together proved to be a reliable scale to measure perceived intrusiveness (α = 0,91). Finally, ad-value was measured with two 7-point bipolar scales (totally not agree – totally agree), which was adopted from Ducoffe (1995); participants could indicate whether they thought the advertisement was useful and informative. These two items also proved to be a reliable scale (α = 0,84). Since all items proved to be reliable indicators in order to measure the

dependent variables and/or moderators, scale variables were constructed by taking the weighted average of all items. At the end of the questionnaire, demographic variables were measures indicating participants’ gender, age and educational level.

4. Results

First, preparatory analysis will test whether the control variables ( age, gender, education) are equally distributed among the conditions of the independent variables (congruency, ad-position). Two ANOVA’s with congruency and ad-position as dependent variable shows that there is no significant difference in age (M=32,66, SD=14,19), (F(1,113)=0,581, p=0,448) among the congruency conditions and ad-position conditions (F(1,113)=0,303, p=0,583), no significant difference in gender (M=1,47, SD=0,5), (F(1,113)=1,694, p=0,196) among the congruency and ad-position conditions (F(1,113)=4,39, p=0,38) and no significant difference in

education level (M=7,11, SD=1,16), (F(1,113)=0,031, p=0,86) among the congruency and ad-position conditions (F(1,113)=1,315, p=0,0,254). This implies that

randomization worked out well and that these variables will not be added into the main analysis as control variables.

Second, in order to test the main hypothesis (H1A and H2A), a two-way ANOVA was conducted to check whether average brand attitude differed

significantly among the congruency conditions (congruent vs. incongruent) and the ad-position conditions (pre-roll vs. mid-roll). The results of the ANOVA showed that average brand attitude of participants in the congruent conditions (M=4,08, SD=1,36) did not differ significantly of participants in the incongruent conditions (M=4,07, SD=1,24), (F(1,113)=0,002, p=0,967). The average brand attitude of participants in the pre-roll conditions (M=4,09, SD=1,11) did not differ significantly of participants in the mid-roll conditions (M=4,05, SD=1,47), (F(1,113)=0,029, p=0,865). Therefore,

hypothesis H1A and H2A can be rejected.

Finally, the hypotheses with regard to the moderation effects (H1B and H2B) of perceived intrusiveness and ad value have to be tested. A multiple regression analysis with brand attitude as dependent variable and congruency, intrusiveness and the interaction term between congruency and intrusiveness as independent variables will be conducted. The multiple regression analysis shows that the main effect of congruency on brand attitude is not significant (F(-0,43, 0,42)=-0,002, p=0,983), the main effect of intrusiveness on brand attitude is significant (F(-1,64, -0,27)=-0,74, p=0,007) and the interaction effect of congruency and intrusiveness on brand attitude is not significant (F(-0,19, -0,66)=0,29, p=0,281). Since there was not found a

significant interaction effect, it can be concluded that perceived intrusiveness does not serve as a moderator on the effect that congruency has on brand attitude; hypothesis 1B is therefore rejected. Second, a multiple regression analysis with brand attitude as dependent variable and position, value and the interaction term between ad-position and ad-value as independent variables will be conducted. The multiple regression analysis shows that the main effect of ad-position on brand attitude is not significant (F(-0,57, 0,35)=-0,04, p=0,636), the main effect of ad-value on brand attitude is not significant (F(-0,28, 1,18)=0,35, p=0,221) and the interaction effect of ad-position and ad-value on brand attitude is not significant (F(-0,51, 0,43)=0,47, p=0,869). Since there was not found a significant interaction effect, it can be concluded that ad-value does not serve as a moderator on the effect that ad-position has on brand attitude; hypothesis 2B is therefore rejected.

Discussion

found a moderation effect of perceived intrusiveness. Neither did brand attitude differed significantly for participants who were exposed to ads varying in position, ad-value did not serve as a moderator. However, this does not mean that the current research was useless; in this section, shortcomings of current research will be discussed in order to give implications for future research on the effectiveness of in-stream video advertisements. Since there was not found support for the hypotheses, implications of this study will be mainly theoretical.

The first shortcoming that will be discussed is one that could have affected the internal validity of the results negatively. More specifically, the manipulation material that was used probably did not manipulated congruency not extremely enough; the two fragments that were used might have been too similar to each other. The manipulation material was not pre-tested, while a pre-test could have shown that congruency was not manipulated good enough. This would have made it possible to make adjustments to the stimulus material before the data was collected. The lack of manipulation of congruency could be an explanation for the fact that congruency did not affect brand attitude significantly. Therefore, future research should be paying more attention to the creation/selection process of the manipulation material and also the pre-testing process of the stimulus material to manipulate congruency.

Second, because the experiment was conducted online, it was relatively hard to keep experimental control over confounding variables. For instance, it was impossible to check whether participants had comparable attention for the stimulus material. Participants could have been distracted easily when they were taking part in the experiment online; the doorbell could have rung, a friend could have called or they were distracted by the television, radio or internet. These are only a few of examples which could have influenced attention to the stimulus material negatively.

Some of these variables should be kept under control in future research. Keeping control over these variables is much easier when the experiment will take place in a laboratory setting. One of the negative aspects of laboratory experiments, however, is that the ecological validity of the results is relatively low; it is very hard to recreate a real-life setting in laboratories. Even though, this can be prevented by, for instance, presenting participants with a good cover story, future research on this topic has to keep in mind that choosing the experimental setting has big influence on the validity of the results. One possible suggestion is that research about the effects of in-stream video advertisements first have to focus on optimizing the internal validity (real-life setting), and subsequently could focus on optimizing the external validity (laboratory setting).

Finally, some implications for future research will be given when we look closer at theories about the effectiveness of in-stream video advertisements. Ewards et al. (2002) mentioned that there could be other things that influence perceived levels of intrusiveness. When viewers are focused, for instance, they perceive interruptions as more severe than when not focused. In current research, it was expected that

participants were more focused in the middle of the video, and less focused in front of the video (Li & Lo, 2015). Because of this, brand attitude was expected to be higher when a mid-roll ad was displayed than when a pre-roll ad was displayed. However, it could be that participants’ focus was not more (or less) in the beginning (in the middle) of the video. This might as well be a cause of the fact that participants were aware that they were participating in an experiment; they knew that questions would be asked after watching the video, which could make them more focused on the video than they usually would be. Future research could therefore focus more on

in-stream video advertisements on YouTube. Besides cognitive effort Edwards et al. (2002) mentioned that intrusiveness can be moderated by the use of creative advertisement placement strategies. More specifically, when advertisements are perceived as more creative they are perceived as less intrusive. Future research should therefore also test whether the use of creative strategies also works for online in-stream video advertisements to generate less perceptions of intrusiveness. All together, current research builds a good foundation on which future research can continue; a compromise should still be found in the future how to minimize the

negative perceptions of in-stream video advertisements on brand attitude, while taking advantage of the potential effective benefits of these ads on brand recall and

References

Aaker, D. A., & Bruzzone, D. E. (1985). Causes of irritation in advertising. Journal of Marketing 3(2), 47-57. doi:10.2307/1251564

Alden, D. L., Mukherjee. A., & Hoyer, D. W. (2000). The effects of

incongruity, surprise and positive moderators on perceived humor in

television advertising. Journal of Advertising, 29(2), 1-15.doi:10.1080/00913 367.2000.10673605

Baek, T. H., & Morimoto, M. (2012). Stay away from me. Journal of Advertising,

41(1), 59-76. doi:10.2753/joa0091-3367410105

Basil, M. D. (1994). Multiple resource theory 1: Application to television viewing.

Communication Research, 21(2), 177-207. Doi:10.1177/0093650940210020

03

Brehm, S. S., & Brehm, J. W. (1981). Introduction: Freedom, control, and reactance theory. Psychological Reactance, 1-7. doi:10.1016/b978-0-12129840-1.500- 05-3.

Dahlén, M. (2005). The medium as a contextual cue: Effects of creative media choice. Journal of Advertising, 34(3), 89–98. doi:10.1080/00913367.2005.10- 639197

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. doi:10.1080/00913-

367.2002.10673678

Feltham, T. S., & Stephen J. A. (1994). Program involvement and ad/program consistency as moderators of program context effects. Journal of

Fiske, S. T., Kinder, D. R., & Larter, W. M. (1983). The novice and expert:

Knowledge-based strategies in political cognition. Journal of Experimental

Social Psychology, 19(4), 381–400. doi:10.1016/0022-1031(83)90029-x

Fransen, M. L., Verlegh, P. W. J., & Kirmani, A. (2015). The review of marketing communications a typology of consumer strategies for resisting

advertising, and a review of mechanisms for countering them.

International Journal of Advertising, 487(1), 37–41. doi:10.1016/0022-10- 31(93)90029-x

Ha, L. (1996). Advertising clutter in consumer magazines: Dimensions and effects, Journal of Advertising Research, 36(6), 76-83.

Halkias, G., & Kokkinaki, F. (2014). The degree of ad–brand incongruity and the distinction between schema-driven and stimulus-driven attitudes. Journal of Advertising, 43(4), 397–409. doi:10.1080/00913367.2014.891087 Hegner, S. M., Kusse, M. D. C., & Pruyn, A. T. (2016). Watch it! The influence of forced pre-roll video ads on consumer perceptions. In P.W.J. Verlegh, H.A.M. Voorveld, & M. Eisend, (Eds.). Advances in advertising research series (Vol. VI), 63-73, Wiesbaden: Springer Gabler. doi:10.1007/978-3-658-10558-7_6 Kim, N. Y., & Shyam S. S. (2010). Relevance to the rescue: Can smart ads reduce negative response to online ad clutter? Journalism and Mass

Communication Quarterly, 87(2), 346–62. doi:10.1177/107769901008700208 Krugman, H. E., Bauer, R. A., & Greyser, S. A. (1969. Advertising in america: The consumer view. Journal of Marketing Research, 6(2), 240. doi:10.2307/3149-

683

Lee, Y. H. (2000). Manipulating ad message involvement through information expectancy: Effects on attitude evaluation and confidence. Journal of

Advertising, 29(2), 29–43. doi:10.1080/00913367.2000.10673607 Li, H., & Lo, H.-Y. (2015). Do you recognize its brand? The effectiveness of online in-stream video advertisements. Journal of Advertising, 44(3), 208 -218. doi:10.1080/00913367.2014.956376

McCoy, S., Everard, A., Polak, P., & Galletta, D. F. (2008). An experimental study of antecedents and consequences of online ad intrusiveness. International

Journal of Human-Computer Interaction, 24(7), 672–99. doi:10.1080/10447- 310802335664

Meyers-Levy, J., & Tybout, M. J. (1989). Schema congruity as a basis for product evaluation. Journal of Consumer Research, 16(6), 39–54. doi:10.1086/209192 Nelson, L. D., & Meyvis, T. (2008). Interrupted consumption: Disrupting

adaptation to hedonic experiences. Journal of Marketing Research, 45(6), 654-664. doi:10.1509/jmkr.45.6.654

Dahlén, M., Rosengren, S., Törn, F., & Öhman, N. (2008). Could placing ads wrong be right?: Advertising effects of thematic incongruence. Journal of

Advertising. 37(3), 57-67. doi:10.2753/joa0091-3367370305

Moore, R. S., Stammerjohan, C. A., & Coulter, R. A. (2005). Banner advertiser-web site context congruity and color effects on attention and attitudes. Journal of Advertising, 34(2), 71–84. doi:10.1080/00913367.2005.10639189 Phillips, B. J. (2000). The impact of verbal anchoring on consume response to image ads. Journal of Advertising, 29(1), 15–24. doi:10.1080/00913367.200- 0.10673600

Peracchio, L. A., & Meyers-Levy, J. (1994). How ambiguous cropped objects in ad photos can affect product evaluations. Journal of Consumer Research, 21(6), 190–204. doi:10.1086/209392

Rust, R. T., & Varki, S. (1996). Rising from the ashes of advertising. Journal of

Business Research, 37(3), 173-181. doi:10.1016/s0148-2963(96)00067-7

Solomon, M. R., Dahl, D. W., White, K., Zaichkowsky, J. L., & Polegato, R.

(2014). Consumer behavior: Buying, having, and being. Upper Saddle River, NJ: Prentice Hall. doi:10.1108/0025174091096019

Sharma, A. (2000). Recall of television commercials as a function of viewing context: The impact of program-commercial congruity on commercial

messages. Journal of General Psychology, 127(4), 383–96. doi:10.1080/0022- 1300009598592

Wedel, M., & Pieters, R. (2008). A review of eye-tracking research in

marketing. Review of marketing research, 4(2008), 123-147. doi:10.1108/s15-

48-6435(2008)0000004009

Yun, Y, C., & Kim, K. (2005). Processing of animation in online banner advertising: The roles of cognitive and emotional responses. Journal of

Appendix 1: Questionnaire

Q01

Geachte heer/mevrouw,

U bent uitgenodigd om deel te nemen aan een onderzoek dat wordt uitgevoerd onder verantwoordelijkheid van de Graduate School of Communication, onderdeel van de Universiteit van Amsterdam. Dit onderzoek is niet geschikt om uit te voeren op een mobiele telefoon of tablet. Daarom verzoek ik u om het onderzoek uit te voeren op een pc of laptop. Het doel van het onderzoek is meer inzicht te krijgen in de effecten van online video's. Het onderzoek duurt ongeveer 6 minuten. Omdat dit onderzoek wordt uitgevoerd onder de verantwoordelijkheid van de ASCoR, Universiteit van Amsterdam, heeft u de garantie dat uw anonimiteit is gewaarborgd en dat uw

antwoorden of gegevens onder geen enkele voorwaarde aan derden worden verstrekt, tenzij u hiervoor van te voren uitdrukkelijke toestemming hebt verleend. Voor meer informatie over dit onderzoek en de uitnodiging tot deelname kunt u te allen tijde contact opnemen met de projectleider Peter Jonker, te bereiken via

peter.jonker@live.nl.

Mochten er naar aanleiding van je deelname aan dit onderzoek klachten of

opmerkingen zijn, dan kunt u contact opnemen met het lid van de Comissie Ethiek van de afdeling communicatiewetenschap, per adres: ASCoR secretariaat, Comissie Ethiek, Universiteit van Amsterdam, Postbus 154793, 1001 NG Amsterdam: 020-525 3680; ascor-secr-fmg@uva.nl. Een vertrouwelijke behandeling van je klacht of opmerking is daarbij gewaarborgd. Ik hoop u hiermee voldoende te hebben

geïnformeerd en dank u bij voorbaat hartelijk voor uw deelname aan dit onderzoek dat voor mij van grote waarde is. Met vriendelijke groet, Peter JonkerKlik op >> om verder te gaan

Q02

Ik verklaar hierbij op voor mij duidelijke wijze te zijn ingelicht over de aard en methode van het onderzoek, zoals uiteengezet in de uitnodigingsmail voor dit onderzoek.

Ik stem geheel vrijwillig in met deelname aan dit onderzoek. Ik behoud daarbij het recht deze instemming weer in te trekken zonder dat ik daarvoor een rede hoef op te geven. Ik besef dat ik op elk moment mag stoppen met het onderzoek.

Als mijn onderzoeksresultaten worden gebruikt in wetenschappelijke publicaties, of op een andere manier openbaar worden gemaakt, dan zal dit volledig geanonimiseerd gebeuren. Mijn persoonsgegevens worden niet door derden ingezien zonder mijn uitdrukkelijke toestemming.

Als ik meer informatie wil, nu of in de toekomst, dan kan ik me wenden tot Peter Jonker, te bereiken via peter.jonker@live.nl. Voor eventuele klachten over dit onderzoek kan ik me wenden tot het lid van de Commissie Ethiek namens ASCoR,

per adres: ASCoR secretariaat, Commissie Ethiek, Universiteit van Amsterdam, Kloverniersburgwal 48, 1012 CX Amsterdam; 020-525 3680; ascor-secr-fmg@uva.nl  Ik begrijp de bovenstaande tekst en ga akkoord met deelname aan het onderzoek.  Ik ga niet akkoord, en doe daarom niet mee aan het onderzoek.

Q03

Het onderzoek gaat bijna beginnen. U krijgt zometeen een YouTube video te zien waarover later in het onderzoek een aantal vragen zullen worden gesteld. De video bevat een kort fragment van een televisie programma. Zorg ervoor dat het geluid aan staat, ga rustig zitten en bekijk de video in zijn geheel af. Het is de bedoeling dat u tijdens het kijken van de video op de onderzoekspagina blijft.

Wanneer u hier klaar voor bent, klik dan op >> om verder te gaan, de video zal dan automatisch starten. Wanneer u de video in zijn geheel heeft afgespeeld, krijgt u de mogelijkheid om door te klikken naar het vervolg van het onderzoek.

Klik op >> om de video te starten. Q04 Video condition 1 Q05 Video condition 2 Q06 Video condition 3 Q07 Video condition 4 Q08

Q09

Ik vond de advertentie van Ford in de video:

1 2 3 4 5 6 7 Zeer opdringerig:Helemaal niet opdringerig        Zeer afleidend:Helemaal niet afleidend        Zeer storend:Helemaal niet storend        Q10

Ik vond de advertentie van Ford in de video:

1 2 3 4 5 6 7 Helemaal niet nuttig:Zeer nuttig        Helemaal niet informatief:Zeer informatief        Q11

Mijn algemene gevoel over Ford is:

1 2 3 4 5 6 7 Zeer slecht:Zeer goed        Zeer onaantrekkelijk:Zeer aantrekkelijk        Q12

Het onderzoek is bijna klaar. Op de volgende pagina dient u enkel nog wat

persoonlijke gegevens in te vullen. Klik op >> om verder te gaan en bovenstaande antwoorden op te slaan.

Q13

Wat is uw leeftijd? ______ Mijn leeftijd is: Q14

Wat is uw geslacht?  Man

 Vrouw Q15

Wat is uw hoogst genoten opleiding?  geen

 basisonderwijs

 lager beroepsonderwijs (LBO, VMBO)

 middelbaar algemeen voorbereidend onderwijs (MAVO)  middelbaar beroepsonderwijs (MBO)

 hoger algemeen voorbereidend, wetenschappelijk onderwijs (HAVO, VWO)  hoger beroepsonderwijs (HBO), kandidaats/propedeuse wetenschappelijk

onderwijs

 (post) wetenschappelijk onderwijs  weet niet / wil ik niet zeggen Q16

Klik op >> om bovenstaande antwoorden op te slaan. Q17

Dit was het einde van het onderzoek. Enorm bedankt voor uw deelname! Klik op >> om het onderzoek af te sluiten.