Effectiveness of movie trailers: Myth or reality

Effectiveness of movie trailers: Myth or

reality

The influence of movie trailers and cue recognition on

moviegoer’s uncertainty

by

Katya Damyanova

2

Effectiveness of movie trailers: Myth or

reality

The influence of movie trailers and cue recognition on

moviegoer’s uncertainty

Master Thesis

by

Katya Damyanova

University of Groningen

Faculty of Economics and Business

MSc Marketing Management and Marketing Intelligence

3

Management summary

Movie trailers are very often used promotional tool in the movie industry. However, compared to other methods to increase awareness for upcoming movies such as movie posters and websites, movie trailers are much bigger investment. Therefore, it is crucial to investigate how effective they are in decreasing viewer’s uncertainty about seeing a certain movie. In the following paper, the movie trailer effectiveness in decreasing movie uncertainty is tested against movie posters. Movie posters contain different cue recognitions such as actors, director and studio, which can be used to create various expectations in moviegoers. After conducting a survey by sending a questionnaire, where both upcoming movies’ trailers and posters were used it was found that movie trailers in fact increase movie uncertainty. However, cue recognitions were found more effective as they had a negative effect on movie uncertainty. Using genre preference was as moderator, showed that liking towards the genre can somewhat negate the increase of uncertainty caused by movie trailers.

Preface

For a very long time I have been passionate about advertising. I am talking about advertising when it comes to TV advertisements and promotional campaigns. Therefore, movie trailers as promotional tools were quite interesting for me to investigate, especially having in minded that there is not much research done in this field. To be honest I am not really a moviegoer and this is the first time in my life that I know all of the movies that will be coming on the big screen in the next six months. But I believe that because of this reason I managed to be impartial during the whole research and could accept all of the results with open mind.

4

Contents

Management summary ... 3

Preface ... 3

Introduction ... 5

Identifying the literature gap ... 7

Theoretical background ... 11

Cue recognition ... 11

Movie Genre Preferences ... 12

Conceptual Framework ... 13

Hypothesis building ... 13

Conceptual model ... 14

Methodology ... 15

Research design and sampling ... 15

Procedure ... 16

Data Processing ... 17

Data analysis method ... 17

Variables operationalization ... 18

Results: Analysis of Main and Moderating Effects ... 19

Discussion... 23

Limitations ... 25

Conclusion and Recommendations ... 26

References: ... 27

Appendix: ... 31

Appendix 2: Posters used in the survey ... 31

Appendix 2: Sample survey ... 37

5

Introduction

Imagine a typical Saturday night with some close friends. The atmosphere is rather relaxed, the drinks are cold and there is a sound from the kitchen of popcorns getting ready. The pleasant smell of butter is in the air. All of you are sitting on the couch in anticipation and there is only one question left to be answered: “Which movie are we watching tonight?”

Only in 2013 there were about 158 Hollywood movies released in various genres, targeting both children and adults. As a result of the variety of movies available, it is hard for moviegoers to make a choice, which one they want to watch as watching new unknown films is perceived as risky due to the high level of uncertainty and the lack of expectations (Somburanasin, 2010). Therefore, movie makers always have to engage in tough struggle for attention using various tools such as posters, websites, flayers and trailers to decrease the perceived risk and uncertainty while promoting the movie at the same time. Movie trailers are one of the most commonly used tool to reduce uncertainty as they can be used to shape customer expectations of what is going to happen in the movie itself based on the content and overall style of the trailers (Bridges, 1993). Trailers are a form of affective advertisement with an emotional nature where both visual and emotional factors are joined together (Belch and Belch, 2007). Moreover, they are not just a pure form of advertising but a hedonic experience as they can arouse various emotions in the viewer (Devlin et al., 2011). Most film trailers consist of three parts: in the first one the characters and the film environment are introduced. Then, the second part normally creates some tension in the storyline by introducing some life-changing events and lastly, the trailer’s pace escalates in the final part and hints some potential solution (Finsterwalder et al., 2012).

As the trailer usually shows parts of the movie, it creates certain expectations in the minds of the viewers. For example, if the film trailer has elements of horror, it will create expectations associated with scary movies and attract primary people interest in this genre. Moreover, if these expectations are not met, it will result in a disappointment for the movie goers (Austin, 1985). In other words, trailers introduce new movies to the audience by providing actual scene with the purpose of building

6

the budget for the whole advertising campaign is rather limited, it is very important to distribute the funds where they can be most effective.

Movie trailers are primary used to increase awareness and interest by revealing in most cases what is believed by the audience the most attractive scenes of the move (Iida et al., 2012). Around 71 % of the moviegoers revealed that they had seen the trailer when considering if they should watch the film (Takahashi et al., 2000). However, this number does not reveal how many of them decided to move to another movie as a result of disappointing trailer. In fact a lot of successful movies referred as a “classic” nowadays like Pulp Fiction did not have really good trailers but still the movie itself was a huge success (Lafrance, 2014). In this line of reasoning movie trailers effectiveness can be defined as the likelihood to see the movie after watching the trailer.

The purpose of this study is to shed more light on the topic about movie trailer effectiveness by looking how movie trailers compared to other promotional tools such as movie posters will affect uncertainty and reduce the perceived risk in watching a certain movie. Movie posters contain various cue

recognition such as well-known actors, directors or studio which can play important role in influencing customer’s desire to see a certain film (Sauer, 2013; Somburanasin, 2010; Kernan, 2004). For example, if the moviegoer has seen previously movie made by a certain director or featuring a favored actor there is a higher possibility to be interested in the movie. Therefore, cue recognitions, found in posters will play important role in decreasing uncertainty. Besides, people have their special preference what kind of genre wants to watch and what desired entertainment they want to get by doing so. Thus, the relationship between the cue recognition, movie trailers effectiveness and movie uncertainty will be moderated by the movie genre preference.

It is important to note that uncertainty should not be associated with only negative aspects when it comes to movies. In fact, some moviegoers enjoy the “nobody knows” factor leading to a box-office success for the particular movie (Walls, 2005). However, it should be kept in mind that uncertainty will make negative experience feel even more unpleasant while positive experience will be perceived as even more enjoyable (Wilson et al., 2009). In addition, uncertainty can be used to generate buzz as it keeps the attention the new coming movie and make the reactions more intense. Provoking curiosity in the viewer and make him more engaged (Wilson et al., 2005).

7

of the variables used in this study will be given and hypothesis will be formed concerning them in order to build a conceptual model. Then in the methodological section the data collection method and the operationalization of the variables will be explained. Then, there will be a section concerning the data analysis results and based on them discussion section will be provided. Next, the limitations of the study will be outlined. Finally, the paper concludes with a summary of the main findings and

recommendations.

Identifying the literature gap

The investigation on movie trailers has been relatively recent one yet in the last couple years there are more and more studies on it. Table 1 contains a review of papers illustrating the current knowledge on the topic of movie trailer effectiveness. Most of the studies in the review start from the point where the movie trailers are assumed to be effective and try to improve this effectiveness by using different factors such as actors, directors, reviews and other cue recognitions (Finsterwalder et al. 2012; Yi et al., 2013; Sauer, 2014).

Other studies are trying to see how the effectiveness of movie trailers will change depending on external factors such as the mood of the viewer, his preference towards genre and gender (Hixson, 2006; Delvin et al., 2011; Jerrick, 2013). There are even studies searching for the best channels for trailers (Jerrick, 2013) and recommend tailoring the trailer towards certain target group in order to increase its effectiveness (Hixson, 2006).

New technologies open opportunities to not only enhance the trailers but also give completely new ways of investigating the effect of trailers. For example, the rising popularity of mobile devices such as smartphones and tablets make movie trailers accessible anywhere and anytime. As a result, it is easier to reach certain target group and spread greater awareness among it (Johnson, 2008). Movie trailer can take this advantage to improve their effectiveness.

8

are not an exception. Using automatically selecting shots it can be possible that in the future people can be shown trailers not tailored for specific segment but specific individual (Smeaton et al., 2006).

In contrast of other researches the main focus of this study is to examine if the movie trailers are indeed as effective as presumed. This is vital as before trying to enhance their effectiveness as the other studies try to do, it should be confirmed that movie trailers in reality influence viewers consumption choice. Moreover, even if movie trailers do have significant effect on movie uncertainty, it should be examined if trailers encourage positive movie consumption related to increasing desire of seeing the movie or they discourage it, making the viewer less likely to see the movie because of the trailer.

9

Study

Main Finding

Hixson, T. K., 2006  Moviegoers with similar genre preferences have similar

reactions to trailers. Hence, the effectiveness of a movie trailer can be boosted by tailoring it for a particular segment

Septak, D., 2008  Trailers that are perceived as interesting have a persuasive effect to see the movie on the viewer on individual

psychological level

 The greater investment in teasers and trailers is done on economical level, a greater profitability and revenue will be achieved

Johnson, K. M., 2008  The new technology brings new ways of watching movie trailers. Thus, factors like mobility and interaction with the audience can enhance the effectiveness of movie trailers as a promotional tool

Liu, A., Zhang, Y. and Li, J., 2009  Providing personalized movie synopsis recommendations can be more effective way for advertising movies that using movie trailers

Delvin, M. B., Chambers, L. T. and Callison, C., 2011  Interaction between mood, trailer type and gender

 Men find more attractive trailers that portray genres different from their mood (positive or negative)

 Women show congruency. When being in a good mood they prefer comedy trailers while in negative mood prefer more serious trailers

Finsterwalder, J., KUppelwieser, V. G. and Villiers, M., 2012  actors influence the perceived film quality, thus in a trailer should be selected episodes that best portray the performance of a certain actor

 genre influences the content expectations and can influence the liking of a movie trailer

 Using a favorable music in a trailer can influence the perception of the whole movie positively

10

 The trailer needs to provide enough information for the viewer to build judgment for the movie in order to be effective Iida, T., Goto, A., Fukuchi, S. and Amasaka, K., 2012  Most appealing factors in movie trailers are picture, content

and role

 Changing the brightness of the picture in the trailer can impact viewers

 Skillful changes in the soundtrack during the trailer can impact viewers

 Sound effects can capture viewers’ attention

 Long action scenes included in trailer may result in viewer fatigue

 Highest viewer stimulation in the beginning and at the end of the trailer and lowest at the middle point

Jerrick, D., 2013  Movie trailers affect viewer’s decisions to see the movie to some extend

 Trailer placements on the television was found the most effective

 The mood was found to only slightly affect how the movie trailer was perceived

Yi, H. W., Eileen, H., Ling, L. Y. and Wen, W. J., 2013  Positive relationship between movie trailers and decision about watching the movie

 Positive influence of critics review on decision about watching the movie

 Positive impact of star power on decision about watching the movie

 Positive relationship of WOM and movie consumption Sauer, M., 2014  Including certain cue recognition in the trailer can boost its

effectiveness

 Using faded in titles can affect the trailer positively only if the style of the titles is perceived as suitable

 Participant that liked animated movie made their decision mostly based on the cues and not the trailer

11

Theoretical background

Cue recognition

As often comparing various products in order to make the best decision can be very challenging task especially if you do not have sufficient information and time, people tend to simplify judgments by using subconsciously heuristic methods (Pachur and Hertwig, 2006). In other words, many decisions are made by using mental short cut-ups such as previous experience and knowledge. Because people in general are cognitively limited, they rely on association and cue recognition in choice making (Heijden, 2004). When it comes to movies, recognition of studio, actors and/or a director can serve as signs for cue recognition as there is a high possibility that the moviegoers have seen other movie from the same director or well-known actors (Somburanasin, 2010). Consequently, based on their previous experience and knowledge, they will form certain expectations about the movie. If there is positive expectation the possibility to see the newly released movie increases.

Movie posters are tools that are often used to present different cues of recognitions to the audience. They have a commercial purpose as medium to advertise a movie. Although there are different

variations of the posters’ design the, most commonly used elements in movie posters are the names of the leading actors, the name of the movie, the upcoming date and sometimes the name of the directors (Burty, 2013). All of these elements are included in order to create certain expectations and familiarity. One of the most effective cue recognition that can be found on a poster is the actors starring in the movie (Srinivas, 2009). This cue can be presented either with the names of the actors, only their pictures or a combination of both approaches. Besides, the effectiveness of the cue often depends on its

accessibility in the consumer mind. The faster the one can retreat a certain cue, the stronger its influence on consumer choice is (Biehal and Chakravarti, 1983). Hence, famous movie stars, who are expected to have high accessibility in the consumer mind will have string influence on the consumers. According to Wallace et al. (1993), the participation of certain actors depending mostly on their

popularity at the moment can strongly influence the success of the movie. For example, if there are stars in the cast it is expected roughly around $3 000 million boost of box office revenue (Elberse, 2007) just because of her/his presence.

12

one negative review the uncertainty about the movie raises (Sridhar and Srinivasan, 2012). Therefore, the presence of well-known actors can negate the negative influence of critics’ comments. A possible expiation for this effect is that the movie is perceived as more familiar and trustworthy (Levin et al., 1997).

Yet, it should be kept in mind that similar to movie trailers and even online reviews cue recognitions can have a dual effect. Their credibility can be affected by the self-selection biases, meaning that viewers may be attracted to watch the movie by an actor that they like but at the same time they can decide not to see the move because of a disliked actor (Hu et al., 2006). However, a further research is needed to estimate the real strength of this phenomenon.

Kindem (1982) states that the “star value” as recognition cue is related to the fact that “in an industry where multi-million dollar films are released to compete with one another head-to-head, the presence of a star act as a beacon that makes the movie stands out and entices larger audiences.” Thus, the value of the movie star and stories is related to the fact that they serve as a ‘publicity machines’ that optimize advertising effectiveness by increasing the levels of the awareness and creating a buzz around the movie (Bakker, 2001), resulting in lower uncertainty.

Movie Genre Preferences

Some people prefer exciting, action movies, others the thrill of a blood freezing horror or just a comedy that will make them laugh out loudly. In any cases everyone has a specific desire to be fulfilled when choosing a movie which is strongly reflected in the genre of the chosen film. Genre can be defined as the similarities in the narrative elements used in a film and shared with other movies (Kernan, 2004). Due to these similarities genre can create certain exceptions in the mind of the consumer and affect the

perceived risk and uncertainty related to the new movie.

Moreover, some moviegoers have a genre preference towards particular types as they believe that can reach their expectations the best. As a result knowledge and liking of the films plot and genre are often the strongest reasons why people want to watch certain movies. People pay more attention to

something that is relevant and familiar to them as find more entertainment value in it compared to something that is not (Hixson, 2006).

13

trailers as they will have similar expectations, forming a segment that can be targeted by marketers. In reaction, the marketers will adjust their promotion strategy when it comes to the positioning of the movie trailer to reach these segments most efficiently (Jerrick, 2013).

The moviegoers’ genre preference will play an important role, how effective will be the movie trailer for a certain individual or segment. It will be expected that if the genre is to their preferences they will be more positive towards the trailer as well and will affect them more. Furthermore, movie genre

preferences will influence how we perceive previously acquired knowledge like cue recognition (Sauer, 2014). Thus, genre preference would be an appropriate moderator of the relationship between the dependent and independent variables discussed in the paper.

Conceptual Framework

Hypothesis building

According to Somburanasin (2010), when there are some recognition factors familiar to the viewer, then the uncertainty will be reduced while the probability to watch the movie will increase. The reason is that the perceived risks will decrease, if the viewer believes that the new movie has familiar attributes such as adaptation, famous actors and directors or sequels (Sedgwick and Pokomy 2005). Therefore, cue recognition can as well affect uncertainty.

For example, using famous actors can be very influential as they have a track of records, formed via the previous movies that they started in that helps the viewer to form some expectations about the movie based on the actor’s ability and previous type of roles (Vany and Walls, 1999).

A recognizable actor or a director can be used to identify quality of the production, the same way that luxury brand names are associated in the mind of the buyer with high quality (Levin, et al., 1997). Moreover, they influence the viewer’s decision to watch the movie as they have brand equity, thus a film that have more famous stars will be perceives as attractive. As a result, studies have shown that famous actors and directors can bring several millions of dollars in revenues (Elberse, 2007).

14

H1: Cue recognition will influence movie uncertainty negatively.

The music theme, used in the trailer can also be used to reduce uncertainty and create expectations in the customers as they often associate particular type of music with a specific feeling like soft music for example is associated with romance (Flanagan, 2012). Furthermore, by just switching the music theme and the pace of the commenter if the trailer is featuring a voice over, a horror movie trailer can create expectations of comedy and the other way round (Lafrance, 2014).

Although the movie goers do not solely base their expectations of the trailer but also relay on other sources like word of mouth and critic reviews (Liu, 2006), film trailers still have a strong influence on reducing risk and uncertainty by creating expectations in the viewers as it allows the viewer to “window-shop” and let him deciding whether the film is what they would want to watch by portraying the tone and genre, story and actors in the movie (Kernan, 2004).

H2: Movie trailers effectiveness compared to posters has a negative effect on movie uncertainty.

If there is a liking towards the movie genre, then the cue recognition will play much more important role in uncertainty avoidance. A reason for that is that favorable genres are associated with the presence of certain cues (Source, 2014). Therefore the viewer will be searching for the presence of these cues more attentively when s/he is interested in the genre and cues will play a much more important role as tool for decreasing movie uncertainty. Hence:

H3: The relationship between cue recognition and movie uncertainty is stronger when there is a

preference towards the movie genre, compared to when there is no such a preference.

Yet, if there is no other recognized cues, the genre portrayed by the movie trailers and the preference towards it will be crucial factor for moderating the relationship between movie trailer and movie uncertainty as the displayed genre will create some expectations towards the movie by relying on previous experience with movie from the same genre (Bridges, 1993).

H4: The relationship between movie trailers and movie uncertainty is stronger when there is a preference

towards the movie genre, compared to when there is no such a preference.

Conceptual model

15 Figure 1: Conceptual Model

Methodology

Research design and sampling

The research design used to investigate the effectiveness in movie trailers in reducing the viewer uncertainty is causal casual. Only primary data was used collected though a questionnaire, developed specially to investigate the problems and objectives of the current study. Furthermore, survey was chosen as data-collecting method as it enables gathering the data inn fast and cost-efficient way (Sekaran, 2010).

The sample consists of men and women between 15 and 33 years old. The participants were

16

Procedure

The questionnaire used primary close-ended questions and scale questions, easily to be interpreted and answered by the respondent. The questionnaire consisted of three general parts. The first part gives information about the demographics of the respondent such as gender, age and nationality. The second part investigates general movie consumption and preferences. For example, one of the questions is using a 5-point Likert scale to measure the respondent preference towards certain movie genre. The third part tries to compare the effectiveness of movie posters and movie trailers by asking the

participant how likely is to see the movie after first seeing the poster. Then the participant is asked how likely is to see the movie again seeing the trailer of the same movie as the poster.

The first two parts are compulsory for everyone taking the survey. However, the third one is randomize depending on what answer the respondent has given on the question about genre preference. Question 5 in the survey investigates genre preferences towards action, comedy, drama, horror, fantasy and crime. Each preference is measured by 5-point Likert scale and depending on the answers the participant will randomly get 2 different genres in the third part. Only genres that were access as “neutral”, “agree” (I like watching genre) or “strongly agree” (I like watching genre), will be appointed in order to avoid observant disliking the genre and bias the results of the study by not watching the trailer for example.

In third part 6 movies are used (Table 2). Each of the movies used was up-coming at the time of the data collection. Hence, none of the participant could have seen the movie beforehand and they had to solely rely on the posters and trailers to make their choice. The participants are first shown the official poster of the movie and then asked if they like the genre of the movie given (see Appendix 1). Next, they are asked if they recognize some of the actors on the poster. If they do, a question about how much the respondent likes the actor is asked. If they do not recognize any of the actors this question is skipped. The next question measures using a 5-point Likert scale, how like is for the participant to see the movie after seeing the poster. In the next part the trailer of the same movie as the poster is shown and the respondent is asked how likely is to see the movie based on the trailer this time.

17

not like watching particular genre of movie, s/he will not get this genre in the third part. Therefore, there were participants that had to answer questions about one movie only or did not get any movies at all. A sample questionnaire is provided in the Appendix (see Appendix 2)

Drama Action Comedy

The Fault in our Star Edge of tomorrow Chef

The Immigrant Hercules The Grand Seduction Table 2: Movies investigated

Data Processing

After collecting the initial data from the survey, it had to go through several process stages before it could be preceded with further analyses. First of all the data was edited in order to be clean from all incomplete or responses through the questionnaire. Consequently, the data was reduced from 155 to 118 respondents that had filled in the survey fully.

In the next step the data for the two randomly assigned movies was pooled together. Meaning that each respondent’s answers were taken twice: once with the response of the questions concerning the first movie and the second time for the second movie. Thus, the data set was doubled from the original 118 respondents to 236. After further processing the data it was discovered that for 6 of the respondents only one movie was assigned. Therefore, the dataset needed to be reduced to a total of 230

observations.

Afterwards the data was checked for missing values and outliers. Only missing data was found the question: “How much do you like in general the actors that you recognize”, around 57.8%. This result can be explained by the fact that participants who answered that they do not recognize any of the actors on the given poster skipped this question. As there was not enough data collected on this question, it was decided not to be used in further research. Using the Z-scores of the variables, the data was checked for outliers. None of the Z-scores were higher than 3.29 or smaller than -3.29. Hence, no outliers were found and the analyses could proceed with all 230 observations.

Data analysis method

In order to check the movie trailer effectiveness and the cue recondition influence on movie

18

variable. The whole model was moderated by the movie genre preference. In the next section the operationalization of the variables in order to run the regression will be discussed.

Variables operationalization

In order to fit the conceptual framework investigated in this paper of the questions needed to be operationalized into the needed variables. The dependent variable “movie uncertainty” was created by subtracting the likeliness to see the movie after seeing the poster from the likeliness of seeing the movie after seeing the trailer. Both the likeness of seeing the movie after seeing the poster and seeing the movie after seeing the trailer were measured using a 5-point Likert scale. For the moderator “movie genre preference”, a nominal dichotomously coded scale was used, where 0 means “No I do not like the genre” and 1 means “Yes I do like the genre”. This question initially designed as probing for question 5, which already investigates the general preferences towards the genre. However, after further analysis it was discovered that question with using a dichotomous scale, captured the moderating effect better. As a result, the analysis proceeded using this question even though it does not capture the effect of “genre preference” as good as questions using Likert scale.

19

made it less likely to see the movie compared to the poster only. The situation where there was no changed in the desire to see the movie was taken as the base category and coded as 0.

Results: Analysis of Main and Moderating Effects

The results analyzed and described in the following section are based on the regression analysis of the initial conceptual model. The general regression equation is presented below:

The estimated model managed to explain 37.9% of the common variance between the variables (adjR2=.37). Table 3 shows the estimated main effects of cue recognition and trailer effectiveness on Movie uncertainty. It also shows the moderating effects influencing these main effects. Cue recognition was found to have a significant negative impact (β=-.466, p=.021), supporting H1. Thus, indeed was found that if there is a certain cue of recognitions the viewer’s uncertainty if s/he wants to watch the movie will decrease.

On the hand, there were mixed results about the Movie trailer effectiveness. Movie poster effectiveness was found insignificant (β=-.124, p=.675), while Movie trailer effectiveness was found to have a

20

Variables

Dependable variable: Movie Uncertainty

B

SE

p-value

VIF

Main Effect

Cue Recognition

-.466

.201

.021

3.618

Movie trailer effectiveness

1.407

.226

.000

4.676

Movie poster effectiveness

.124

.295

.675

4.770

Constant

.070

Interaction effect

Cue Recognition*Genre

Preference

.454

.237

.057

4.464

Movie trailer

effectiveness*Genre

Preference

-.569

.268

.035

6.074

Movie poster

effectiveness*Genre

Preference

-.666

.348

.057

5.038

21

The moderation effect of genre preference on the relationship between cue recognition and movie uncertainty was found marginally significant (β=.454 p=.057). The positive moderating effect of genre preference (β= .454) almost equals the negative effect of cue recognition (β=-.466) on movie

uncertainty, meaning that when there is a strong genre preferences cue recognition will not be

effective. Therefore, H3 is only partially supported as even though genre preference does moderate the relationship between cue recognition and movie uncertainty, it does not strengthen it but weakens it. Furthermore, as figure 2, using the unstandardized regression coefficient of the interaction effect and the Cue recognition to be built, shows that Genre preference has a stronger moderating effect in a condition where there is high cue recognition while when there is low cue recognition there is no real difference between low and high genre preference. When there is a high amount of cue recognition, high genre preference will decrease movie uncertainty even more compared to when there is low genre preference.

Figure 2: Moderation effect of Genre Preference on Cue recognition and Movie Uncertainty

22

and low genre preference. On the other hand, when Movie trailer effectiveness is high, there is a need of high genre preference to decrease the Movie uncertainty.

Figure 3: Moderation effect of Genre Preference on Movie trailer effectiveness and Movie Uncertainty Even in conditions where posters were found more effective than movie trailers as a tool to decrease uncertainty, genre preference has a negative marginally significant moderating effect (β=-.223, p=.057). Figure 4 shows that in cases where Movie poster effectiveness is low having low genre preferences will result in lower movie uncertainty. Otherwise having high genre preference will play an important role in decreasing movie uncertainty.

23 A summary of the main findings can be found in Table 4.

Hypothesis

Findings

Main Effect

H1: Cue recognition will influence movie uncertainty negatively.

Supported

H2: Movie trailers effectiveness has a negative effect on movie uncertainty.

Not Supported

Interaction Effect

H3: The relationship between cue recognition and movie uncertainty is stronger when there is a preference towards the movie genre,

compared to when there is no such a preference.

Partially Supported

H4: The relationship between movie trailers and movie uncertainty is stronger when there is a preference towards the movie genre, compared to when there is no such a preference.

Partially Supported

Table 4: Summary of the main findings

Discussion

The results after analyzing the obtained respondents’ data leave several interesting points for discussion. First of all, cue recognition in the form of actors portrayed on movie posters indeed can decrease uncertainty about seeing the movie supporting the literature and finding from previous studies. However, the findings about movie trailers effectiveness compared to movie posters were very surprising.

24

as much attention of the viewers as possible. This results in false expectations and big disappointment for the viewer after seeing the actual movie. It becomes even worse once when expectations and disappointment become a regular occurrence, turning the movie trailers as another perceived deceiving marketing tool.

In fact close to 47% of the participants in the survey felt that after seeing the trailer they were either less willing to see the movie or the desire was the same as when they saw the poster only (see appendix 3). Therefore, the effectiveness of movie trailers should be reconsidered and instead of trying to make trailers more and more attractive to viewers, they should be made more believable. Using more methods to increase the visual attractiveness of the trailer can increase the skepticism among the viewers how honest is the really the trailer. Thus a first step is to invest in ways to decrees the move trailers skepticism. Once, this hurdle is overcome it will be easier for viewers to be influenced by more attractive trailers.

The moderating effect of genre preference was proven to weaken on the relationship between cue recognitions and movie uncertainty. In fact, the moderating effect almost chancels out the main effect. A possible reason is that when there is high preference towards certain genre the viewer knows what expect to a certain degree and does not need other cues to guide his choice. Following, this reasoning it is possible that if one has a strong genre preference, the genre of the movie becomes the greatest cue for decreasing the uncertainty and any other cues turned into supporting.

In some cases when the participants were asked about their general liking towards a particular genre they gave in some cases different answers compared to when they were asked if they liked the genre after seeing the poster. For example, some participants indicated that they really enjoy watching comedies yet after seeing the poster of “The grand seduction” and were asked “do you like the genre” the answer was “no”. A possible explanation may be that the poster does not contain sufficient cues that will be identified as a comedy in the mind of the respondent and therefore cannot it cannot be identify as a comedy.

Contrary on the expectations the relationship between movie trailer effectiveness and movie

25

the artistic elements in the trailer and they had the most influence on the participants’ desire to watch the movie or not (Sauer, 2013).

Furthermore, the more knowledge the viewer has about actors, director etc., the more he will rely on his previous experience to make the decision if he wants to watch the movie (Finsterwalder et al., 2012). Accordingly, movie trailers will be found more likable to the audience if there are some cue recognition elements and preferable genre compared to the ones with none (Sauer, 2013). This can again be connected with the skepticism towards movie trailers. Though one can doubt how much the trailer represents the reality s/he is willing to trust previous experience.

Limitations

Although the current research did bring some further light on the topic about movie trailers and

understanding their effectiveness, it is important to mention some of the limitation of the study. First of all the sample can be viewed as limited as it contains people from 15 to 33 years old without covering more age groups that will probably have different movie consumption behavior. Moreover, as pointed before the sample consists mainly of students. Hence, it can be argued that due to smaller income they go to the cinema less often compared to people with regular full time jobs. Moreover, the sample was collected via Facebook and personal contacts and may be argued that is not fully representative. Another limitation is that only one type of cue recognition was used, namely the actors. However, the director and the film studio can be as well a valuable source of cue recognitions. However, as posters were used as a tool for showing cue recognitions in order to be comparable to each other they needed to contain the same cues. As a result only the actors could be used as cues as no sufficient number of posters that contained all cues like actors, director and studio to be compared was found.

Furthermore, finding suitable posters to represent all the investigated genres (action, comedy and drama) and cover all the selection criteria was proven to be challenging. Thus, there are posters that contain a lot of actors and others with only one or two actors. As a results there could not be made any significant difference between recognizing all the cues and only some of them. Both conditions were treated as the same resulting in possible inaccuracies.

26

already had an opinion about his/her desire to see the movie. Moreover, it is possible that they saw some kind of previews or discussions about some of the movies in the past. Therefore, they had more information available to make their decision if they want to see the movie than having just the trailer and the poster only.

Last and possibly the greatest limitation of the study is that the constructs investigated were measured by using only one or two questions. There are no scales up to date created and validated to measure movie uncertainty or trailer effectiveness unlike other constructs. Consequently, this study could use only limited amount of questions to operationalize the variables. Conducting the same study using validated scales for the variables in the future can come to different and even more accurate results.

Conclusion and Recommendations

27

References:

Austin, B. (1985). Current research in film. 1st ed. Norwood, N.J.: Ablex.

Bakker, G. (2001). Star and Stories: How Films Became Branded Products. 1st ed. Routledge.

Bar-Anan, Y., Wilson, T. and Gilbert, D. (2009). The feeling of uncertainty intensifies affective reactions.

Emotion, 9(1), p.123.

Belch, G. and Belch, M. (2007). Advertisingand Promotion: AnIntegrated Marketing Communications

Perspectives. 1st ed. New York: McGraw-Hill/Irwin.

Biehal, G. and Chakravarti, D. (1983). Information accessibility as a moderator of consumer choice.

Journal of Consumer Research, pp.1--14.

Bridges, E. (1993). Service attributes: Expectations and judgments. Psychology & Marketing, 10(3), pp.185--197.

Burty, D. (2013). A Semiotic Analysis on Movie Posters of Harry Potter and the Deathly Hallows.. Universitas Brawijaya.

De Vany, A. and Walls, W. (1999). Uncertainty in the movie industry: Does star power reduce the terror of the box office?. Journal of Cultural Economics, 23(4), pp.285--318.

Devlin, M., Chambers, L. and Callison, C. (2011). Targeting mood: Using comedy or serious movie trailers.

Journal of Broadcasting & Electronic Media, 55(4), pp.581--595.

Elberse, A. (2007). The power of stars: Do star actors drive the success of movies?. Journal of Marketing, 71(4), pp.102--120.

Finsterwalder, J., Kuppelwieser, V. and de Villiers, M. (2012). The effectsof film trailers on shaping consumer expectations in the entertainment industry: a qualitative analysis. Journal of Retail

Consumer Service, 19(6), pp.589–595.

Flanagan, M. (2012). Microfilmmaker Magazine - Tips & Tricks - How to Edit a Trailer That Will Get Your

Film Noticed, Pg. 1 of 3. [online] Microfilmmaker.com. Available at:

28

Film Noticed, Pg. 1 of 3. [online] Microfilmmaker.com. Available at:

http://www.microfilmmaker.com/tipstrick/Issue14/Edit_Trl.html [Accessed 23 Jun. 2014]. Hixson, T. (2006). Mission possible: Targeting trailers to movie audiences. Journal of Targeting,

Measurement and Analysis for Marketing, 14(3), pp.210--224.

Hu, N., Pavlou, P. and Zhang, J. (2006). Can online reviews reveal a product's true quality?: empirical findings and analytical modeling of Online word-of-mouth communication. pp.324--330. Iida, T., Goto, A., Fukuchi, S. and Amasaka, K. (2012). A Study On Effectiveness Of Movie Trailers

Boosting Customers’ Appreciation Desire: A Customer Science Approach Using Statistics And GSR.

Journal of Business & Economics Research (JBER), 10(6), pp.375--384.

Jerrick, D. (2013). The Effectiveness of Film Trailers: Evidence from the College Student Market. UW-L

Journal of Undergraduate Research, XVI, 13.

Johnston, K. (2008). 'The Coolest Way to Watch Movie Trailers in the World'Trailers in the Digital Age.

Convergence: The International Journal of Research into New Media Technologies, 14(2),

pp.145--160.

Kernan, L. (2004). Coming attractions. 1st ed. Austin: University of Texas Press.

Kindem, G. (1982). The American movie industry. 1st ed. Carbondale: Southern Illinois University Press. Lafrance, A. (2014). Why Classic Movies Have Terrible Trailers. [online] The Atlantic. Available at:

http://www.theatlantic.com/entertainment/archive/2014/01/why-classic-movies-have-terrible-trailers/283332/ [Accessed 23 Jun. 2014].

Levin, A., Levin, I. and Heath, C. (1997). Movie stars and authors as brand names: Measuring brand equity in experiential products. Advances in consumer research, 24, pp.175--181.

Liu, Y. (2006). Word of mouth for movies: Its dynamics and impact on box office revenue. Journal of

marketing, 70(3), pp.74--89.

29

Sauer, M. (2014). Cue-recognition effects in the assessment of movie trailers. Journal of Retailing and

Consumer Services, 21(3), pp.376--382.

Sekaran, U. (2010). Research method for business : A skill approach. 5th ed. New York: John Wiley & Sons.

Septak, D. (2008). Understanding the Effectiveness of Trailers, Teasers, and Television Spots in Marketing Movies.

Smeaton, A., Lehane, B., O'Connor, N., Brady, C. and Craig, G. (2006). Automatically selecting shots for action movie trailers. pp.231--238.

Somburanasin, M. (2010). Risky Busines: Does recognition reduce uncertainty of the movie industry

global box office revenue?.

Sridhar, S. and Srinivasan, R. (2012). Social influence effects in online product ratings. Journal of

Marketing, 76(5), pp.70--88.

Srinivas, L. (2009). Cinema in the City: Tangible Forms, Transformations and the Punctuation of Everyday Life. Visual Anthropology, 23(1), pp.1--12.

Takahashi, Y., Hasegawa, K., Sugiyama,, K. and Watanabe, M. (2000). Takahashi, Y., Hasegawa, K., Sugiyama, K. and Watanabe, M., (2000), A new movie summarization algorithm for effective movie selection: Toward human content interface design (4), Bulletin of Japanese Society for Science of Design, Vol.142, No.1, 50. Bulletin of Japanese Society for Science of Design, 142(1), pp.41-50.

Van der Heijden, H. (2004). User acceptance of hedonic information systems. MIS quarterly, pp.695--704.

Wallace, W., Seigerman, A. and Holbrook, M. (1993). The role of actors and actresses in the success of films: How much is a movie star worth?. Journal of cultural economics, 17(1), pp.1--27.

30

Wilson, T., Centerbar, D., Kermer, D. and Gilbert, D. (2005). The pleasures of uncertainty: prolonging positive moods in ways people do not anticipate. Journal of personality and social psychology, 88(1), p.5.

Yi, H., EileenI, H., Ling, L. and Wen, W. (2013). The effect of tarilers, critics reviews, star power and word

of mouth towards decision making on movie consumption. University of Tunku Abdul Rahman.

31

Appendix:

Appendix 2: Posters used in the survey

32

33

34

36

37

Appendix 2: Sample survey

Hello! I am a Master student writing my master thesis, focusing on movie trailers. I am investigating the role of movie trailers when choosing a movie to watch. The following survey will take only 10 to 15 minutes of your time but your contribution can be a big help in better understanding the real effect of movie trailers.

Thank you very much for your participation and time!

1. What is your gender? Male

Female

2. What is your age?

3. What is your nationality?

The following set of questions is concerning some general information about your preferences when it comes to choosing movies:

4. How often do you watch movies? Less than Once a Month Once a Month

38

5. To what extent do you agree with the following statement?

Strongly disagree

Disagree Neither Agree nor Disagree

Agree Strongly agree I like watching Action

as a movie genre

    

I like watching Comedy as a movie genre

    

I like watching Drama as a movie genre

    

I like watching Horror as a movie genre

    

I like watching Fantasy as a movie genre

    

I like watching Crime as a movie genre

    

6. Do you search for some kind of additional information like actors, director, rating, etc. before making a decision on watching a movie?

Yes No

7. To what extent do you agree with the following statement?

Strongly disagree

Disagree Neither Agree nor Disagree

Agree Strongly agree Before watching the

movie, it is important to know the people or star that acted in the movie

    

The trailer makes me excited to watch the movie

39

8. On the picture below, you are seeing the poster of an upcoming Genre movie, called "Movie 1". Please, have a look at the poster and answer the questions below.

9. Do you like the genre of the movie? Yes

No

10. Do you recognize some of the actors staring in the movie? I do not recognize them

I recognize some of them I recognize all of them

11. How much do you like in general the actors that you recognize? (Skipped if the answer of the

previous question is “I do not recognize them”)

Not at all

Somewhat dislike them Neutral

Somewhat like them Like them a lot

40

12. After seeing the poster how likely is that you will consider seeing the movie? Extremely unlikely

Unlikely Neutral Likely

Extremely likely

After seeing the poster of the movie, you will see the movie trailer of the same movie as well. Please, answer the questions below.

13. After seeing the movie trailer how likely is that you will consider seeing the movie? Extremely unlikely

Unlikely Neutral Likely

Extremely likely

14. Did your uncertainty about seeing the movie changed when you saw the trailer compared to seeing the poster only?

No, it is the same level

Yes, it is less likely to see the movie after seeing the trailer Yes, it is more likely to see the movie after seeing the trailer

Questions from 8 to 14 are repeated for “Movie 2”

Thank you for your participation!

41

Appendix 3: Change of Uncertainty after seeing the trailer compared to the

poster

Less Likely Same More Likely

53,4%

18,3%