• No results found

The impact of multiple stars on movie performance in terms of box office revenues and online movie ratings

N/A
N/A
Protected

Academic year: 2021

Share "The impact of multiple stars on movie performance in terms of box office revenues and online movie ratings"

Copied!
42
0
0

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

Hele tekst

(1)

MSc in Business Administration – Marketing track

Master Thesis

The impact of multiple stars on movie performance in terms

of box office revenues and online movie ratings

Author: Dominika Pastoreková Student ID: 10828826

Date of submission: 29th June 2015 Supervisor: Frederik Situmeang

(2)

2

Statement of originality

This document is written by Dominika Pastoreková, who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

Abstract

Extensive research has been conducted about star power. Within this topic there is an issue that hasn’t been investigated properly and is relatively unexplored. We are speaking about casting of multiple stars in one movie production. Recent findings suggest that star power has a positive impact on the movie performance, mainly the box office revenues. This thesis analyses the impact of multiple stars on the total box office revenues, using the sample of 1655 movies released between years 2000 and 2013. The sample consists of movies, in which at least one award winning or nominated actor or actress was cast. The considered movie awards are the Academy Awards, the Golden Globes and People’s Choice Awards. The main hypothesis of this thesis suggest that with increasing number of stars in one movie production, the box office revenues will start to decrease at some point. The suggestion is argued by the possible conflicts occurring between the performers, what could lead to movies with lower quality. The empirical findings show a significant positive relationship between the number of stars in a movie and the total box office revenues. It is also found that this relationship is positively moderated by sequels, but director power doesn’t have a moderating effect on the relationship. The results show that the relationship between the number of stars in a movie and the total box office revenues is not mediated by the online movie reviews.

(4)

4

Table of contents

1 Introduction ... 6 2 Literature Review ... 8 2.1 Genre of movie ... 8 2.2 Reviews ... 9 2.3 Star power ... 10 2.4 Director power ... 12 2.5 Awards ... 13 2.6 Sequels ... 14 2.7 Signal theory ... 15 2.8 Research question ... 16

3 Theoretical framework and hypothesis ... 17

4 Research design and methodology ... 21

4.1 Sample ... 21 4.2 Data collection ... 21 4.3 Variables ... 23 4.3.1 Independent variable ... 23 4.3.2 Dependent variables ... 23 4.3.3 Moderating variables ... 23 4.3.4 Mediating variables ... 23 4.4 Analysis ... 24 5 Results ... 25 5.1 Descriptive statistics ... 25 5.2 Hypothesis testing ... 27

5.2.1 H1: The impact star power on the opening week box office revenues ... 27

5.2.2 H2: The impact of star power on the total box office revenues ... 28

5.2.3 H3: The moderating effects of the director power and sequels ... 29

5.2.4 H4: The mediating effects of online movie reviews ... 30

6 Discussion ... 33

6.1 Academic relevance ... 34

6.2 Managerial implications ... 35

6.3 Limitations and future research ... 35

7 Conclusion ... 37

(5)

5

List of figures

Figure 1: Positive impact of number of stars on the opening week revenues ... 17

Figure 2: Expected inverted U-shaped curve ... 18

Figure 3: moderating effect of sequels and director power ... 19

Figure 4: mediating effect of online movie reviews ... 20

List of tables

Table 1: Descriptive statistics and correlations ... 26

Table 2: Regression analysis H1 ... 27

Table 3: Regression analysis: H2 ... 29

Table 4: Moderating effect of sequels and director power ... 30

(6)

6

1 Introduction

Every year more and more movies are being released with big and costly marketing campaigns (MPAA report, 2014). It is important to know where to invest in movie production to achieve a high quality movie that will generate high total revenues. One of the factors, worth considering, is casting of top star actors and actresses. Movies with famous movie stars are generally perceived as of high quality. However, it is important to know how many is enough. A romantic comedy Valentine’s Day, which was released in 2010, featured award winning stars like Julia Roberts, Anne Hathaway, Jessica Alba, Ashton Kutcher, Bradly Cooper, and many others. The consumers had big expectations, but the result of this movie production was big disappointment. The movie was rated just by 5.8 points (out of 10 point scale) on the Internet Movie Database (IMDb), the most popular online movie database. Similarly the another movie called Movie 43, released in 2013, which featured the total off twelve award winning or nominated stars like Kate Winslet, Hugh Jackman, Emma Stone, Richard Gere and Uma Thurman, etc. scored only 4.4 points on IMDb. And on the other hand, a movie Avatar, from the year 2009, leading in terms of the total box office revenues, featured only one award winning and one award nominated actress. The question stands: Is the involvement of stars a key to success in the movie industry? And if so, is it possible to say, the more top stars, the better? It would be interesting to look at the influence of top star actors and actresses on the overall performance of the movie, as casting famous stars can be very expensive for the production and it does not necessary lead to movies with higher quality. The number of studies that investigate the impact of star power on movie performance is growing. This is due to a fact that previous research found strong connection between top stars and box office performance (Sochay, 1994; Nelson et al., 2001; Elberse, 2007; Elliott & Simmons, 2008). On the contrary some studies show that the impact of stars on box office revenues is not significant (Litman &

(7)

7

Ahn, 1998; Ravid, 1999). For movie producers it is important to understand the power of movie stars, as it can affect the success of a movie and thus total box office revenues.

The main goal of this study is to investigate the impact of casting a number of top star actors and actresses on the overall movie performance, in terms of movie revenues and ratings. Casting multiple stars may lead to conflicts between the performers, as each one considers him/herself important and wants to do the movie his/her own way. Although from award winning actors and actresses we expect high quality movies, due to these conflicts the quality of the movie is lower and thus the revenues smaller.

Regarding the stars, celebrity media exposure is an increasingly strong source of actors’ popularity and it is possible that the acting skills are not reflected. This study will use wins and nominations for various movie awards, to calculate star power to make sure actors’ real talent is taken into consideration. The chosen awards are specifically Academy Awards as peer selected movie awards, Golden Globes as expert selected movie awards and People’s Choice Awards as consumer selected movie awards.

This thesis is structured as follows. The first section reviews relevant literature, and ends with the research question. Second section describes theoretical framework and sets hypothesis that are going to be tested in the research. Methodology follows with explanations about sample, data collection and how the research is going to be conducted, which research design will be used, timeframe in which data are going to be collected, dependent and independent variables and expected results. In the fourth section the results of the research will be presented. Next section focuses on discussion about these results and the final one is a conclusion that will sum up the findings and implications resulting from them.

(8)

8

2 Literature Review

It is known that motion pictures are uncertain products. Movie makers are struggling to produce high quality movies that will attract the audience and therefore ensure high total revenues. There are several factors influencing movie performance, movie revenues and ratings. Key factors that should be taken into consideration are movie reviews (professional and amateur), genre of the movie, movie awards and top star actors and directors working on a movie, in other words star power (Desai, Basuroy, 2005).

2.1 Genre of movie

Prior research suggests that movie genre is the first factor consumers look at, when choosing a specific movie to watch (De Silva, 1998). Based on previous study (Litman & Ahn, 1998), the movies are categorized into seven genres: action/adventure, children/family, comedy, drama, horror, mystery/suspense, and sci-fi/fantasy. De Silva’s study (1998) shows that most popular genres are comedies and dramas. Sochay (1994) found that especially these two genres have significant impact on the box office revenues. Thus, dramas and comedies are favourite among both consumers, as the highest revenues are generated by these genres, and also among directors and producers, as the largest number of releases belongs particularly to dramas and comedies (Desai, Basuroy, 2005).

However, there are significant differences in studies about movie genre, Chang & Ki (2005), in their study, showed concerns about multi-genre trend. Recent study by Shon, Kim and Yim (2014) focuses on this trend and claims that movie can’t be categorized into just one genre. They came up with different type of categorization of movies based on how the audience feels and reacts to the movie. The new nine categories are roller-coaster – described as fast, dynamic, powerful, eye-catching and wild; playground – fun movies targeted for families and kids; mosaic – various characteristics mixed; odd- ball – novel, new, unique, fresh; soul-trigger

(9)

9

– pitiful, sad, moving, and sentimental; lone wolf – difficult, complicating, solitary, and strange; nerve-wrecker – shocking, odious, cruel, brutal; deja-vu – relying too much on the typical plots and techniques; and marshmallow – cozy, soft, warm; (whereas roller-coaster provides biggest total revenues and marshmallow lowest).

After selecting the genre of a movie that the consumers want to watch, they tend to look at the reviews of movies, to find out more information and know, more or less, what to expect from it.

2.2 Reviews

Movies belong to experience goods. The main characteristics of these goods is that consumers can evaluate their quality only after the purchase and consumption, which makes the decision making very difficult. Movies are typical example of experience goods since consumers can’t judge them before seeing (Neelamegham, Jain, 1999). That’s why reviews are so powerful. Consumers are forced to rely on word of mouth (WoM) as it is the only information that can help them decide, whether or not to see a movie and it makes reviews pretty influential.

Consumer and expert critics generate electronic word of mouth that can be seen in many forms, e.g. online forums, discussion boards, chat rooms, blogs etc. (Duan et al., 2008). This way it is easily spread around and it has impact on a large number of consumers. Previous studies distinguish between expert and consumer reviews. Some claim that expert reviews are more persuasive in consumers’ purchase decisions, due to higher expertise of the source, but on the other hand, consumer reviews were found to be more trustworthy, which makes them more effective (Willemsen et al., 2010). Another study by (Flanagin & Metzger, 2013), based on an experiment, conducting more than one thousand adults from the US, using online movie ratings, found that people prefer expert reviews where there is a low-information volume and they tend to favour user-generated reviews under condition of high-information volume. Jong

(10)

10

and Burgers (2013) conducted a study in order to compare consumer and professional reviews and found differences in numeral points. Most importantly, consumer critics speak in first person and give their own opinion on a particular movie and evaluate the movie more explicitly. Expert critics, on the other hand, try to remain neutral and objective. They speak usually in third person and just want to inform the readers about the movie. Furthermore, amateur reviewers don’t go into details about the movie, compared to professional critics who focus also on the history of, and the news about the movie.

During the decision making process about watching a movie, the consumers also look at another important factor – the cast of the movie.

2.3 Star power

In the movie industry, especially in Hollywood, it is generally proven that top stars are key to a film’s success (Desai, Basuroy, 2005). Highly popular star in a movie is a predictor that consumers can expect high quality, or highly entertaining movie.

Nevertheless, academic literature investigated the impact of movie stars on movies’ financial performance many times and the results of these empirical studies are conflicting. According to some researchers (Sochay, 1994; Nelson et al., 2001; Elberse, 2007; Elliott & Simmons, 2008), the presence of stars in the cast of a movie has a significant impact on box office revenues. On the contrary, others (Litman. 1983; Litman & Ahn, 1998; Ravid, 1999), found no significant relationship between star power and market performance. However, in this case, it is necessary to look at the period in which the studies were conducted. The influence of star power may differ in time. More recent studies found a positive effect of using stars in movie production. It has major impact on ratings and sales of movies. Treme (2010) described powerful stars as key predictor of a movie performance.

(11)

11

The studies define the term powerful star, or star power, differently. In some popularity of the stars was measured based on how many times has this actor or actress appeared in a popular magazine. Other studies say the stars are considered powerful, when revenues of the movies, they’ve performed prior the current movie, where significant, compared to other movies. This is the most common way to identify star power. Some authors focus more on the popularity among the consumers, by looking at the number of page visits, and rating of stars on the Internet Movie Database (IMDb), specifically the STARmeter. Other authors determine star power based on number of ‘’likes’’ of actors’ and actresses’ Facebook page, number of tweets, or favourability on other social media websites. Another way is to look at the movie awards wins and nominations of the stars.

Akdeniz and Talay in their study (2012) gave two main explanations to why star power can be signal of quality. First, famous movie stars bring more publicity to the movie, so it is more likely to attract attention of the audience. Second, top movie stars are concerned with their reputation, thus they only choose movies that are expected to be successful and of good quality. The authors also found that the strength of star power differs in many countries. It is the strongest in countries with high uncertainty avoidance and cultural openness.

The impact of star power can be influenced by the genre of movie. Desai’s and Basuroy’s study (2005) showed that star power has no impact on total movie revenues in case of familiar movie genre (comedies, dramas). On the other hand, in movie genres that are not so popular, greater star power ensures positive impact on movie performance. But generally, the study found out that movies with less star power are perceived to be of poor quality, in the eyes of consumers.

Previous research also focused on how different characteristics of top star actors affect movie performance. Treme’s study (2010) investigated how stars’ age and gender influences box office revenues. They found out that in case of gender, male actors generate higher total

(12)

12

revenues, than female actresses. Regarding the age of movie stars, male actors at the age 42 and more will ensure high revenues, whereas in case of actresses the age should be up to 32 to generate high revenues.

Most of the studies that tested impact of star power, focused only on presence of one top star in a movie. They failed to take into account presence of multiple stars in one movie production. This fact could lead to overestimated impact of single star’s power on a movie performance. However, there was a study from Nelson and Glotfelty (2012) that investigated what would happen, if three regular actors would be replaced by top star actors. The star power in this study was calculated based on the number of page visits on the Internet Movie Database website. Authors could use a continuous measure of star power thanks to STARmeter ranking on the IMDb website. This research found out that replacing three regular actors by top star actors would generate almost 4 times higher box office revenues, than replacing just one actor by a top star. But, the question stands, is it possible to assume that this number will still increase with increasing number of replaced stars? I don’t think so. There might be a point in which the curve will start to sink again.

2.4 Director power

It is also important to look at the role the director plays in the movie production. Although working behind the scenes, the director impacts whether or not the movie becomes successful (Kim, 2013). Several studies found that director may be one of the factors consumers look at when deciding whether or not to see a movie. Top star actors and also top star directors are associated with a higher hit probability (Jung & Kim, 2010). Later (Kim, 2013) proved that both these powers have significant impact on the hit potential of a movie, however, he claims that star power is more important than director power. It is more influential in decision making,

(13)

13

and more important in determining whether or not the movie will generate high box office revenues.

As already mentioned, star power and director power can be measured in more ways. One of the methods is to look at number of various movie awards nominations and wins. The most famous award in motion picture industry is Oscar Academy award, which is awarded annually by the Academy of Motion Picture Arts and Sciences.

2.5 Awards

The most important function of any award is signal of quality. Awards are especially essential in the motion picture industry, as consumers don’t know if they will like the movie before they see it. Here, awards play role as a signal of quality and help consumers decide.

Deuchert et al. (2005) claim that box office revenues are generated primarily by the nominations of movie stars, not actual wins. This can be explained by the fact that movie stars first compete to be nominated, which already is a huge success, and after that they compete for the actual award. The study found out that for the potential audience, being nominated for the Academy Awards is a signal of quality sufficient enough, in their decision to watch a movie.

The study from Nelson et al. (2001) focuses on the impact of different categories in the Oscar Academy Awards. The most influential categories are best picture, best actor in leading role and best actress in leading role. Nominations in these categories generate positive impact on movie’s probability of survival and box office revenues. Other categories, like best supporting actor/actress, have only a little impact on these variables. Furthermore, the study shows that release date of the movie has big impact on the box office revenues, mainly opening week revenues. The closer is the movie release to the process of evaluation for Oscar nominations, the bigger revenues are generated in the opening week. As the Oscar Academy

(14)

14

Awards are carried out in February and it is common practise to delay movies from the first to fourth quarter to ensure higher box office revenues.

There are many awards within the motion picture industry. The most studies focused on the most famous one – The Academy Awards – Oscars. Research on one specific award is extensive, but there are not many studies that would compare more kinds of awards. And as it is important to consider also another movie awards, beside the Oscars, Gemser, Leender and Wijnberg (2008) examine, which awards have the biggest impact on the competitive performance of the award winners. The study distinguishes three types of awards, based on the jury that selects the winners. These types of awards are – peer selected, expert selected and consumer selected. Authors tested, if for mainstream, big Hollywood movies, awards, chosen by the end consumers, will have greater effect on the movie revenues, than peer and expert selected awards. This hypothesis wasn’t supported, as effectiveness of consumer selected awards wasn’t found to be significantly different from other types of awards. In case of mainstream movies, the credibility and trustworthiness of the different types of awards seems to be the alike.

2.6 Sequels

Eight out of the top ten most revenues generating movies, released in the last ten years, were sequels. It is clear that there is an increasing phenomenon of sequels in the movie industry. We can see this phenomenon not only in the movie industry, but also in video game and book industry. The importance of sequels is rising, and they are becoming more popular. They are usually big blockbuster movies that rarely win Academy Awards, despite that we can say that they are successful.

Most of the previous research focused on the impact of parent movie on sequel. (Basuroy & Chatterjee, 2008) found that the time difference between the release of the sequel

(15)

15

and parent movie is crucial. The sooner after the parent movie is the sequel released, the better do the sequels perform. They also say the sequels perform worse than the parent movie in terms of box office revenues. Another finding is that increasing number of sequels has positive impact on the whole franchise, due to the fact that creating buzz around the previous movies helps the current sequel succeed. Moon et al. (2010), on the other hand, say that with number of sequels, the fan base is diminishing, so studios have to be very thoughtful when deciding whether or not to extend the franchise even more. Recent research supports this claim, and shows that although user ratings have an impact on the first sequel, neither expert, nor user ratings of parent movie have an effect to large number of sequels. However, there was an interesting finding that says, the attendance of predecessor movie positively influences attendance of large number of sequels (Hou Yong, Wang Tie-nan, & Li Xiang-yang, 2013). We can conclude that, what is important in success of the sequel, is not the parent movie, but the previous sequel movie (in case of the first sequel, it is of course the parent movie).

2.7 Signal theory

As it is hard to evaluate quality of the movie before watching it, consumers use quality signals that form their expectations. All the influencing factors mentioned above are signals of quality that help consumers with their decision making.

Marketing is a tool for signalling, as it provides information to customers, and this way helps them with pre-purchase assessment of product quality, and purchase decision making. Marketing signals are very important in evaluating the actual quality of experience goods, as these are not tangible and we can’t see any physical features of the products (Akdeniz M. B. & Talay M. B, 2013).

In case of motion picture industry, moviegoers look for credible information to evaluate the quality of a movie. The studios invest in the movie cast and advertisement, in order to send

(16)

16

positive signals about the movie quality. These are marketing signals to positively influence moviegoers, and their perception of quality. Hennig-Thurau et al. (2006) claim that consumers are strongly influenced by the marketing signals, released by studios. Akdeniz and Talay state that popular movie-related signals are sequel movies, production budget, star power and expert rating. In this study I’m going to focus mainly on the star power as a signal of quality.

2.8 Research question

Previous studies clearly show that star power generates higher revenues and ratings of movies. An important research gap within this topic is that these studies focused on the impact of only one lead star on the movie performance. But what happens if there are many top stars in one movie? It would be interesting to investigate how experience of number of top star actors and directors influence the quality of a movie and thus its performance. Lot of experience can be highly beneficial for the movie. Every star brings some knowledge and ideas that may lead to overall higher movie performance. However, it can come to situations in which the ideas of the director and the cast are conflicting. With number of nominations and awards, the strength of stars increases. And with the stronger power of the actor, actress or director increases also their bargaining power. In some cases of multiple stars in one movie production, the differences between ideas of the involved parties, can cause problems and conflicts, as every one of them might want to make the movie their own way. Another issue here could be that if the stars in the movie are all male, it could mean that too much testosterone causes the conflicts, or in the case of female presence, the affection between performers can lead to drama. This may entail problems and conflicts in case of multiple stars in one movie production. Dissonance between visions of each performer and trying to take into consideration conditions of all the stars may lead to movies with lower quality and thus lower revenues, ratings, it may cause bad reviews and lower the overall movie performance, even though we could have expected high quality movies from award rewarded actors, actresses and directors.

(17)

17

Thus, the research question of this study is: ‘’What is the impact of multiple stars, compared to just one top star, on movie performance, in terms of box office revenues and online movie ratings?’’ It will answer an important literature gap that is worth investigating.

3 Theoretical framework and hypothesis

In order to investigate the research question, I’ve set several hypothesis. This section will wrap up the discussed literature from previous sections. Thereafter hypotheses will be drawn, while guided by the theoretical framework.

With number of top stars featured in the movie, increases the buzz around the movie and WoM. This can lead to attracting more people to go see the movie, therefore larger opening week box office revenues.

Figure 1: Positive impact of number of stars on the opening week revenues

H1: The more award winning movie stars are cast in the movie, the higher the opening week box office revenues.

After some time the movie is running in the cinemas, more and more reviews are appearing on the websites. As explained before, reviews are a powerful tool in the decision making for experimental goods like motion pictures. Due to conflicts that may occur between featured stars, I predict that the quality of the movie will decrease with rising number of top

(18)

18

stars featured in it, which means that consequently the movie will generate lower total revenues, than it would generate, if it featured less stars. An inverted U-shaped curve can be expected.

Figure 2: Expected inverted U-shaped curve

H2: With increasing number of award winning movie stars featured in the movie, the total box office revenues will start to decrease at some point.

There are lot of factors moderating the relation between number of stars in the movie and total box office revenues. As demonstrated in the theoretical framework, I’m going to invesigate two of them. First, it’s the director, who can serve as a strong signal of quality. The power of director is very important in the movie production. With a strong director, who is also popular among the peers, experts and consumers (based on the achieved movie awards wins and nominations), the quality of the movie can be expected to be high and subsequently the revenues large. Second, as found in the previous research sequels are very popular and important in the nowadays motion picture industry, generating larger total revenues than normal

(19)

19

movies. In a combonation with lots of top stars featured in the movie, the sequels can generate even higher, eventually lower revenues. It is interseting to test the moderting effect of sequels.

Figure 3: moderating effect of sequels and director power

H3a: Sequels, moderate the impact of star power on the total box office revenues.

H3b: Award winning or nominated director of the movie moderates the impact of star power on the total box office revenues.

It is expected that the more top stars are featured in the movie, the bigger number of critics will review the movie. This conclusion is based simply on the fact that more stars attract bigger audience, as every one star has their own fans, admirers and critics. I assume, this effect can be predicted more among user reviewers, than expert reviewer. Expert reviewers focus on quality of the performance and have a greater general knowledge about all the actors and actresses. User reviewers are driven more by the emotions and affection towards an actor or actress. The subsequent ratings may also be based on the number of stars featured in the movie and they may strongly influence the decision making of movie goers and thus the total box office revenues.

(20)

20

Figure 4: mediating effect of online movie reviews

H4a: The impact of star power on the box office revenues is mediated by the count of expert reviewers.

H4b: The impact of star power on the box office revenues is mediated by the score from the expert reviewers (metascore).

H4c: The impact of star power on the box office revenues is mediated by the count of consumer reviewers.

H4d: The impact of star power on the box office revenues is mediated by the score from the user reviewers.

(21)

21

4 Research design and methodology

4.1 Sample

The empirical context for testing the thesis is movie industry, especially big Hollywood movies, as the US motion picture industry provides an ideal setting to investigate star power and due to public availability of Hollywood movies. The study will analyse a sample of 1655 movies, released between years 2000 and 2013, whereas the data are retrieved from the internet. For the purpose of this research, data about movies – revenues, ratings, directors and cast, and movie awards – winners and nominees of specific categories, need to be collected.

4.2 Data collection

To collect data about the movie revenues, webcrawling is used on a website called thenumbers.com. The information needed for this research are opening week box office revenues and total box office revenues. The data about movie reviews – the Metascore, User score and count of both expert and consumer reviewers, and data about directors and cast are collected from website metacritic.com. The data retrieved from internet were very comprehensive and included information about the movies, approximately from the year 1915. Most of these data needed to be filtered out. For clearing and filtering of data, multiple functions in Microsoft Excel and Access were used. From all the movies between years 2000 and 2013, just the ones, in which powerful stars are featured, will be selected for analysis. The powerful stars are determined by the number of movie award nominations and wins they gained. The star will be considered powerful, if he/she has won, or has been nominated, for at least one of the awards, at least once.

There are lots of different awards in the movie industry. To ensure, that all different selection processes are covered, for this study are carefully chosen the most salient ones in three categories – peer selected, expert selected and end-consumer selected. First is Oscar Academy

(22)

22

Awards. This award was picked as it is the most famous peer selected one. The data were retrieved about all Oscar winners and nominees from the year 1950 onwards, in categories best actor in leading role, best actress in leading role, best actor in supporting role, best actress in supporting role and best director. The reason to collect data from year 1950, even though the study will analyse only movies released after year 2000, is that star featuring in these movies could have won some awards in years prior to year 2000. This could influence the box office revenues and overall movie performance. These data were gathered from the official website. Second focus was on the most important expert selected awards, which are the Golden Globes. Again, data were collected from the year 1950 onwards and the sought-after categories were best actor – motion picture drama, best actor – motion picture musical or comedy, best actress – motion picture drama, best actress – motion picture musical or comedy, best supporting actor, best supporting actress and best director. The third chosen awards are The People’s Choice Awards, which are consumer selected, voted on by the general public. The data are collected from the year 1975 onwards from the official website. People’s Choice Awards categories that are desired for this research are – favourite movie actor, favourite movie actress, favourite dramatic movie actor, favourite dramatic movie actress, favourite action movie star, favourite comedic movie actor and favourite comedic movie actress. After combining all the winners and nominees of these three awards, I gained a list of 1119 actors and actresses, and 220 directors.

In order to prepare the data for analysis, simple excel table, including variables the number of top stars, the opening week box office revenues, the total box office revenues, the metascore, the count of expert reviewers, the user score, the count of consumer reviewers, the director of the movie and whether the movie is a sequel, have to be set. This table can be transformed into statistical program.

(23)

23

4.3 Variables

4.3.1 Independent variable

In order to capture the star power of the movie, for every movie was calculated the number of award winning and nominated actors and actresses that were cast in this movie. The star has to obtain at least one nomination or win of one of the awards, in order to be included in the measure. This measure – the number of stars featured in a movie, will serve as independent variable.

4.3.2 Dependent variables

The study will analyse impact of independent variable on two different dependent variables. The first one is the opening week box office revenues, which are predicted to be positively influenced by the star power. The second independent variable is the total box office revenues. Here, I expect that the revenues will start to decrease, once the number of stars will reach some point.

4.3.3 Moderating variables

The focus is on two moderating variables. First is the director power. If the movie was directed by an award winning or nominated director, it was marked with number 1, otherwise 0. Second potential moderator is sequel. Here, similarly like with the director power, if the movie is a sequel, it was marked with number 1, otherwise 0. Both these variables are expected to positively moderate the relationship between star power and total box office revenues. 4.3.4 Mediating variables

I am going to investigate four potential mediators. First, it’s the count of expert reviews. Second is the score from expert reviews, called Metascore. As mentioned before, with the number of stars in the movie, the quality of it is expected to decrease. Experts focus on quality in their reviews, thus this mediator can weaken the relationship between the predictor and the

(24)

24

outcome. Third mediator is the count of consumer reviews and fourth, score from consumers. These are expected to positively mediate the relationship.

4.4 Analysis

Statistical software SPSS 20 will be used in order to analyse the set hypothesis. First I’ll conduct the descriptive statistics and correlation analysis for all the variables. Then, I’ll start with the hypothesis testing. When reporting the results, there are three points of interest that I need to focus at. The significance (p-value) describes whether results are reliable, and thus determine, whether hypotheses are supported. The beta (b) indicates the coefficient level, which interprets the change in dependent variable, when independent variables change. It also signifies the direction of the change. The last one, R2, describes how well the model fits.

The first hypothesis tests the relation between independent and dependent variable, here the number of top award winning stars in the movie and opening week box office revenues of the movie, respectively. Thus, I chose the linear regression analysis. It is expected, that the movie’s opening week box office revenues will increase, with increasing number of top stars featured in the movie. In the second hypothesis analyses, I test the impact of star power on my second dependent variable – the total box office revenues. The expected result is inverted U-shaped curve, which would indicate that the revenues increase with the number of top stars in the movie just to some point, where they start to decrease again. The linear regression analysis will be used again. The third hypothesis test the moderating effect of sequels, directors. This will done by the plug-in ‘Process’ in SPSS (Hayes, 2015). As I want to test a simple moderation, I’ll choose Model 1. The same steps will be used for both potential moderators. The fourth hypothesis will be again tested by the plug-in Process in SPSS. For a simple mediation, I’ll choose Model 4. The same procedure will be used for all four potential mediators, separately.

(25)

25

5 Results

5.1 Descriptive statistics

In Table 1 we can see results from descriptive statistics. The point of interest here is that the correlation values should be below .7 to exclude a probability of presence of problematic constructs (Pallant, 2011). As we can see in Table 1, after testing the bivariate correlations, only one of the correlations is above .7. It is the relation between the opening week box office revenues and the total revenues, r (1653) = .771, p < .01. None of the hypothesis is analysing the relation between these two variables, thus there is no need to exclude any of them.

When we look at means, we see that only 13% of movies with star power are directed by award winning and nominated director and 10% of them are sequels. The average metascore is 52 point on a 100 point scale, with average count of 28 expert reviewers. The standard deviation in the count of expert reviewers is 10, which indicates that approximately equal amount of expert evaluate every movie with star power, regardless on the number of stars in this movie. On the other hand, the standard deviation among the user reviewers is 150, which indicates that consumers are more influenced by the number of stars in their reviews. An average user score for movies with star power is 6.1, with average amount of 72 reviewers.

From the correlations provided in the Table 1, we can see that there is a significant relationship between the number of stars in a movie and the opening week box office, and also between stars and the total box office. Star power has a positive impact on all the other variables, on some stronger, on some weaker. For further hypothesis testing, it is important to look at the relation between the tested variables. Star power correlates significantly with all the possible mediators, and these mediators correlate significantly with the tested outcome, the total box office revenues.

(26)

Table 1: Descriptive statistics and correlations

Variables Mean S.D. 1 2 3 4 5 6 7 8 9

1. The number of stars in a movie 2.71 1.847 1

2. The opening week box office revenues 20804840.90 34234931.351 .344** 1

3. The total box office revenues 43758299.32 67535474.516 .267** .771** 1

4. Director power .13 .333 .254** .112** .090** 1

5. Sequel .10 .305 .107** .450** .347** -.011 1

6. Count of expert reviewers 27.61. 9.918 .387** .392** .311** .285** .138** 1

7. Metascore 52.19 17.659 .147** .132** .121** .241** -.007 .427** 1

8. Count of consumer reviewers 72.04 149.700 .253** .649** .489** .234** .261** .388** .268** 1

9. User score 6.093958 2.1201457 .124** .143** .140** .136** .046 .392** .347** .202** 1

(27)

5.2 Hypothesis testing

5.2.1 H1: The impact star power on the opening week box office revenues

To test the first hypothesis I ran the linear regression analysis. Table 2 presents the obtained outcomes. The results indicate that the independent variable. The number of stars in a movie, has explanatory effect, as its significance is below .05, which is a benchmark for the efficiency of the analysis. The R value, R = .344, represents the simple correlation between independent and dependent variable, which in this hypothesis are the number of stars in a movie and the opening week box office revenues. For these variables, there is a moderate positive correlation. The value of R2 = .119, tells us that the number of stars in a movie can account for 11.9% of the variation in the opening week box office revenues. The other 88.1% of the variation can be explained by other factors, then star power, that have not been tested in this research. From ANOVA, the analysis of variance, it is important to look at the F-ratio and associated significance value. For this data, F (1, 1653) = 222.58, p < .001, what tells us that this regression model predicts the opening week box office significantly well. Hypothesis 1 argues that the number of stars in a movie is positively related to the opening week box office revenues as the coefficient of the number of stars is significant (b = 6386020.836, p = .000). Hence, hypothesis 1 was supported.

Table 2: Regression analysis H1

Dependent variable: H1

The opening week box office revenues Beta Sig. Independent variable

The number of stars in a movie 6386020.836 .000**

Constant 3483482.878 .013*

R2 .119

Adjusted R2 .118

(28)

28

5.2.2 H2: The impact of star power on the total box office revenues

For second hypothesis, the linear regression analysis was used again. Here, the independent variable stayed the same, but the dependent variable is the total box office revenues, instead of the opening week box office revenues. Same as in hypothesis 1, the R value, R = 0.267, represents the simple correlation between independent and dependent variable, which in this hypothesis are the number of stars in a movie and the total box office revenues. This correlation is moderately positive, but the degree of correlation is lower than it was for the first hypothesis. The value of R2 = 0.071, tells us that the number of stars in a movie can account for only 7.1% of the variation in the opening week box office revenues. There might be many other factors that can explain this variation. Other 92.9% of the variation in the total box office revenues is unexplained. For the data from the output for the analysis of variance in the second hypothesis, the values are F (1, 1653) = 222.58, p < .001. As the significance is p > .05, it tells us that this regression model again predicts the opening week box office significantly well. The results show that the number of stars in a movie is positively related to the total box office revenues as the coefficient of the number of stars is significant (b = 9767882.869, p = .000). The results are presented in Table 3.

The hypothesis predicted that the revenues will start to decrease with increasing number of stars featured in a movie. From the results we see that the number of stars is moderately positively correlated to the total box office revenues, thus we can reject hypothesis 2.

(29)

29 Table 3: Regression analysis: H2

Dependent variable: H2

The total box office revenues Beta Sig. Independent variable

The number of stars in a movie 9767882.869 .000**

Constant 17264023.670 .000**

R2 .071

Adjusted R2 .071

**p < 0.01

5.2.3 H3: The moderating effects of the director power and sequels

The third hypothesis focus on the moderating effects of director power and sequels. As mentioned in the methodology section, I have run the regression analysis in the Process plug-in, and I have chosen Model 1, which is used for simple moderation. After obtaining the results, to see, if moderation occurs in the relationship between dependent and independent variable, we have to look at the significance of interaction effect. This interaction is the evidence of moderation, and for hypothesis 3a it represents interaction between the predictor, here, the number of stars and moderator, in this case, sequels. We can see that this effect is significant, b = 21819981.4, p = .0000, thus the hypothesis 3a can be confirmed. Hypothesis 3b tested moderating effect of director power on the relationship between the number of stars and total box office revenues of a movie. However, reading the interaction outline, obtained from the analysis, we observe that the interaction effect between the number of stars and the director power is not < 0.05, but b = 1606957.93, p = .7249. Therefore, since the interaction effect is not significant, we can conclude that there is no evidence of moderation, thus the hypothesis 3b is not supported. Table 4 gives an overview over the results from the moderation analysis conducted for hypothesis 3a and 3b.

(30)

30 Table 4: Moderating effect of sequels and director power

Variable Coefficient Standard error t p LLCI ULCI

Constant 42446865.9 1428097.17 29.7227 .0000** 39645793.4 45247938.4 Sequel 60733076.3 6553055.86 9.2679 .0000** 47879900.0 73586252.6 Stars 7556970.38 830854.177 9.0954 .0000** 5927331.41 9186609.36 Int_1 21819981.4 4441596.38 4.9126 .0000** 13108225.7 30531737.0 Constant 43507401.4 1746916.47 24.9053 .0000** 40080996.1 46933806.8 Director 3209575.46 7633373.63 .4205 .6742 -11762538 18181689.1 Stars 9499659.24 1247662.19 7.6140 .0000** 7052492.23 11946826.3 Int_1 1606957.93 4565988.57 .3519 .7249 -7348780.8 10562696.7 **p < .01

5.2.4 H4: The mediating effects of online movie reviews

Before analysing the mediation effect, three conditions have to be met (Baron & Kenny, 1986). First, the independent variable, the number of stars, must significantly predict the dependent variable, the total box office revenues, which will be presented as model 1. Second, the independent variable, the number of stars, must significantly predict the mediator variables, the count of expert reviewers, metascore, the count of user reviewers and user score, in model 2. Third, the mediator variables, the count of expert reviewers, metascore, the count of user reviewers and user score, must significantly predict the dependent variable, total box office revenues, in model 3. And last, the independent variable must predict the outcome variable less strongly, when the mediator is added as an independent variable. If significant beta changes to insignificant beta, and the mediator has a significant beta, then the second independent variable

(31)

31

is a mediator. The original independent variable is no longer significant. All the models are presented in Table 5.

Table 5: Conditions for mediation of online movie reviews

Variables Beta Sig. R2

Hypothesis 4a

Model 1 .267 .000** .071

Model 2 .387 .000** .150

Model 3 .311 .000** .097

2nd IV added .112

The number of stars .173 .000**

The count of expert reviewers .244 .000**

Hypothesis 4b

Model 1 .267 .000** .071

Model 2 .147 .000** .022

Model 3 .121 .000** .015

2nd IV added .078

The number of stars .255 .000**

Metascore .083 .000** Hypothesis 4c Model 1 .267 .000** .071 Model 2 .253 .000** .064 Model 3 .489 .000** .240 2nd IV added .262

The number of stars .153 .000**

The count of consumer reviewers .451 .000**

Hypothesis 4d

Model 1 .267 .000** .071

Model 2 .124 .000** .015

Model 3 .139 .000** .019

2nd IV added .083

The number of stars .254 .000**

User score .108 .000**

(32)

32

From the results presented in Table 5, we can see that in all four cases, the first three conditions have been met, but in none of the cases the last condition have been satisfied. When the potential mediator was added to the linear regression model as second independent variable, the significance of first independent variable stayed the same and beta was only slightly lower, from β = .267 to β = .255 and β = .254, p = .000, in hypothesis 4b and 4d respectively, where the second variable was the metascore and the user score. We can see a bit bigger change of beta in hypothesis 4a and 4c, where β = .267 was changed to β = .173 and β = .153, p = .000, when the added second independent variable was the count of expert reviewers and the count of user reviewers, respectively.

However, as none of the potential mediators changer the significant beta between independent and dependent variable to insignificant beta, we can conclude that the relationship between the number of stars in a movie and the total box office revenues is not mediated by the online movie review variables, specifically the count of expert reviews, the metascore, the count of user reviews and the user score. Hypothesis 4a, 4b, 4c and 4d were not supported.

(33)

33

6 Discussion

This research provides a new way of looking at an old issue, the star power. The empirical findings bring some interesting insight to this topic by including multiple stars in the measure of star power. While previous research was concentrated on the impact of a single star in a movie on its box office revenues, in this thesis, I aimed to focus mainly at the effect of the number of stars in a movie on its overall performance. I believe that star power in a movie can’t be represented by only one actor. According to Nelson and Glotfelty (2012), a presence of multiple stars could have increased the importance of only one star in the movie on the total box office revenues, even though more powerful actors and actresses were cast in the movie. This thesis brings support to the paper by Nelson and Glotfelty (2012) and shows that the total box office revenues do increase with rising number of award winning or nominated stars in the movie production

From the empirical results we see that, as expected, the opening week revenues are influenced by the number of stars, however they explain only approximately 12% of the variance of the revenues. This tells us that it is important to analyse also other factors that do influence the revenues, like movie reviews (expert and consumer), genre of the movie, production budget etc. (Desai, Basuroy, 2005). This thesis was trying to answer a research question: ’What is the impact of multiple stars on movies’ performance, in terms of box office revenues and online movie ratings? The main hypothesis predicted that the total revenues will start to decrease with increasing number of stars featured in a movie. This wasn’t proven to be true. Even though the relationship between the number of stars and the total box office revenues is weaker than in case of the opening week revenues, we could still see a positive linear relationship, which means that the revenues are, although more slightly, but still increasing with additional stars. The moderating effect of director power wasn’t found to be significant, which can be explained by lack of interest of the consumers in the director. Supporting Kim’s (2013)

(34)

34

claim, star power is more important than director power in consumers’ decision making whether or not to watch a movie. Sequels, on the other hand, were found to have a positive moderating effect on the relationship between the number of stars and the total box office revenues. Regarding the online movie reviews, neither of the tested variables, i.e. the count of expert reviewers, metascore, the count of consumer reviewers and user score, met the conditions to be considered a mediator of the relationship between the number of stars and the total box office revenues. However, from the conditional results where the linear regression analysis was performed, we saw that the star power has a bigger impact on the count of reviewers, mainly the user reviewers, and doesn’t influence the actual rating the movies achieve. Thus, the multiple stars in a movie attract bigger audience, but the scores that evaluate the quality of the movie are not influenced by the number of stars in a movie. From this finding we can conclude that multiple stars in a movie don’t necessarily lead to a high quality movies.

6.1 Academic relevance

The present research broadens up the literature by examining more in to depth the star power. Specifically, by its strong focus on the influence of the multiple stars in one movie production. By turning the attention on the number of stars cast in the movie, the study brings new method of measurement of star power of a movie. As mentioned above, the count of the award winning or nominated stars have not received great attention by scholars. Hence the thesis focuses on this research gap and offers interesting findings that may have been predicted, but have not been proven before. The results are consistent with the expectations and suggest the positive role of the number of stars as a predictor of opening week box office revenues. Furthermore, they show that the number of stars can be also seen as a positive predictor of the total box office revenues. The thesis also brings support to the study analysing director power (Kim, 2013), which is that director power is smaller predictor of the revenues than the star power.

(35)

35

6.2 Managerial implications

Regarding the managerial implications, an important value the managers should look at is the percentage of variance of star power, responsible for the box office revenues. Although it was proven that star power, specifically the number of stars has an impact on both opening week and total revenues, it explains only 12% of variance in the box office revenues. Furthermore, there are other findings important to managers. First, the study showed that with the number of stars increase the revenues, which tells the managers that featuring more star actor and actresses is beneficial for overall movie revenues. Second, the power of the director doesn’t moderate the relationship between star power and the revenues. For the managers it means that when they cast a lot of stars in a movie, it doesn’t matter whether it is directed by an award winning or nominated director, or by a regular one. The revenues won’t be influenced. Third, as sequels were found to positively moderate the relationship between the number of stars and the revenues, it is a good sign for managers in making a decision whether or not to release a sequel, when there are lot of stars involved. Last, from the online ratings point of view, managers will be happy to see that movies with lots of stars attract more consumers, thus generate bigger WoM.

6.3 Limitations and future research

The main limitation of the approach to measure star power in this research, is that I didn’t consider the specific power of each star, meaning, how many wins and nominations they have achieved and from which award. For any award or nomination they have achieved, they were considered a top star. However, the stars that actually won an award, logically, gained more power over the ones who were nominated, despite the fact that consumers appreciate already the nominated stars and are driven to see also movies featuring these nominated actors and actresses (Deuchert et al., 2005). Correspondingly, the study focused on three types of

(36)

36

awards – the peer selected Oscars, the expert selected Golden Globes and the consumer selected People’s Choice Awards. Even though it is advisable to consider all three types, each of these awards is taken differently inside the academic community and among people in general. Academy Awards have the most publicity and are the most known among the consumers. Thus, by winning an Oscar, the actor and actress gains acknowledgement among the peers and awareness among the consumers. This way it has an impact on the box office revenues. The advantage of the Golden Globes is that they are expert selected, so thanks to the expertise, the consumers see them as a signal of quality. Last there are the People’s Choice Awards that are the least known among the mentioned ones, but their selecting system, which is consumer based, ensures that this study will also consider the movies which may not be of a high quality, but they are popular among the consumers, thus generate high revenues. Considering the characteristics of the awards, in the calculations of stars power, the awards should have been weighted according to their importance. Future research should investigate this issue.

Future research should also focus on the genre of movie, in which multiple stars are featured. A previous study by Desai and Basuroy (2005) showed that star power has no impact on total movie revenues in case of familiar movie genre (comedies, dramas), only on the not so popular ones. It would be interesting to investigate the issue of multiple star in this hypothesis.

(37)

37

7 Conclusion

The purpose of this thesis was to gain insights to the topic power of multiple stars in one movie production. In the literature, this aspect of star power is relatively unexplored. The studies that research the issue of star power find mixed results and show that casting stars can be both beneficial and costly for the movie production. Although the recent studies agree on a fact that star power has a positive impact on the box office revenues, casting top stars can be costly for the production due to high compensations given to them. That’s why managers want to be certain that if they engage a top star in the movie, it will be successful. Especially, if they plan to cast multiple stars. However, the academic field did not yet address the issue of multiple stars. When many powerful stars meet at one set, they bring lot of ideas, experience to the production,but they can bring also conflicts and drama, which could lead to movies with lower quality. Based on this fact, this thesis investigated on the following research question: ’What is the impact of multiple stars on movies’ performance, in terms of box office revenues and online movie ratings?

This thesis, thus examined the impact of the number of stars in one movie on the box office revenues. The focus was mainly on the total box office revenues, as, even though the opening week revenues were expected to be higher with higher number of stars featured in the movie, the total box office revenues were expected to start to decrease in some point. The former hypothesis was proven to be true and can be explained by the fact that the popularity of the stars attract bigger audience the first week of airing a movie, however whether or not this movie is a success will be reflected on the total box office revenues. Therefore it was interesting to take one step further and look at the impact on the total revenues. Conversely, the latter hypothesis wasn’t supported, as there was a positive relationship also between the number of star and the total box office revenues. This thesis then analysed the moderating effect of the director power and sequels, and the mediating effect of the online movie ratings. A significant

(38)

38

moderation was found by the sequels, but the director power does not moderate the relationship between the number of stars in a movie and the total box office revenues. Regarding the online movie reviews, none of the variables, the count of expert reviewers, the metascore, the count of user reviewers and user score, seem to mediate the relationship.

The findings of the present research have both academic and managerial implications. First, examining the impact of multiple stars by using a new method of measurement of star power, taking into consideration wins and nominations for three types of movie awards – the peer selected Academy Awards, the expert selected Golden Globes and the consumer selected People’s Choice Awards. Second, it was proven that multiple stars generate high revenues and attract bigger audience, based on the count of reviewers evaluating the movies. To sum up, this thesis brings support to often stated finding from the previous research that star power, including multiple stars, has a positive impact on the movie performance.

(39)

39

8 References

Akdeniz M. B. & Talay M. B., Cultural variations in the use of marketing signals: a multilevel analysis of the motion picture industry, Journal of the Academy of Marketing Science, September 2013, Volume 41, Issue 5, pp 601-624

Baron, R.M. & Kenny, D.A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.

Basuroy, S., & Chatterjee, S. (2008). Fast and frequent: Investigating box office revenues of motion picture sequels. Journal of Business Research, 61(7), 798-803.

doi:http://dx.doi.org/10.1016/j.jbusres.2007.07.030

Chang, B., & Ki, E. (2005). Devising a practical model for predicting theatrical movie success: Focusing on the experience good property. Journal of Media Economics, 18(4), 247-269. doi:10.1207/s15327736me1804_2

De Jong I. K. E. & Burgers Ch., Do consumer critics write differently from professional critics? A genre analysis of online film reviews, Discourse, Context & Media, Volume 2, Issue 2, June 2013, Pages 75-83

Desai K. K. & Basuroy S., Interactive influence of genre familiarity, star power, and critics' reviews in the cultural goods industry: The case of motion pictures, Psychology & Marketing, Volume 22, Issue 3, pages 203–223, March 2005

De Silva, I. (1998). Consumer selection of motion pictures. In B. R. Litman (Ed.), The motion picture mega industry (pp. 144–171). Needham Heights, MA: Allyn Bacon.

(40)

40

Deuchert E., Adjamah K. & Pauly F., For Oscar Glory Or Oscar Money? Journal of Cultural Economics August 2005, Volume 29, Issue 3, pp 159-176

Duan W., Gu B. & Whinston A. B., The dynamics of online word-of-mouth and product sales— An empirical investigation of the movie industry, Journal of Retailing, Volume 84, Issue 2, June 2008, Pages 233-242

Elberse, A. (2007) The power of stars: do star actors drive the success of movies?, Journal of Marketing, 71, 102–20.

Elliott, C. and Simmons, R. (2008) Determinants of UK box office success: the impact of quality signals, Review of Industrial Organization, 33, 93–111

Flanagin, A. J., & Metzger, M. J. (2013). Trusting expert- versus user-generated ratings online: The role of information volume, valence, and consumer characteristics. Computers in Human Behavior, 29(4), 1626-1634.

doi:http://dx.doi.org/10.1016/j.chb.2013.02.001

Hayes, A. F. (2015). An index and test of linear moderated mediation. Multivariate Behavioral Research, 50, 1-22.

Hennig-Thurau, T., Houston, M. B., & Sridhar, S. (2006). Can good marketing carry a bad product? Evidence from the motion picture industry. Marketing Letters, 17, 205–219.

Hou Yong, Wang Tie-nan, & Li Xiang-yang. (2013). How successful movies affect performance of sequels: Signal theory and brand extension theory in motion picture industry. Management Science and Engineering (ICMSE), 2013 International Conference On, 798-806. doi:10.1109/ICMSE.2013.6586370

(41)

41

Jung, S., & Kim, M. H. (2010). Does the star power matter? Applied Economics Letters, 17(11), 1037-1041. doi:10.1080/00036840902817490

Kim, M. H. (2013). Determinants of revenues in the motion picture industry. Applied Economics Letters, 20(11), 1071-1075. doi:10.1080/13504851.2013.783678

Litman, B.,& Ahn, H. (1998). Predicting financial success of motion pictures: The early ’90s experience. In B. R. Litman (Ed.), The motion picture mega industry (pp. 172–197). Needham Heights, MA: Allyn Bacon.

Moon, S., Bergey, P. K., & Iacobucci, D. (2010). Dynamic Effects Among Movie Ratings, Movie Revenues, and Viewer Satisfaction. Journal Of Marketing, 74(1), 108-121. doi:10.1509/jmkg.74.1.108

Neelamegham R. & Jain D., Consumer Choice Process for Experience Goods: An

Econometric Model and Analysis, Journal of Marketing Research, Vol. 36, No. 3 (Aug., 1999), pp. 373-386

Nelson R. A., Donihue M. R., Waldman D. M. & Wheaton C., What's an Oscar worth? Economic Inquiry, Volume 39, Issue 1, pages 1–6, January 2001

Pallant, J. 2011. SPSS Survival Manual: A Step by Step Guide to Data Analysis using the SPSS Program. Allen & Unwin.

Ravid, S. A. (1999). Information, blockbusters, and stars: A study of the film industry. Journal of Business, 72, 463–492

(42)

42

Shon J. H., Kim Y. G. & Yim S. J., (2014) Classifying Movies Based on Audience Perceptions: MTI Framework and Box Office Performance, Journal of Media Economics, 27:2,79-106

Sochay, S. (1994). Predicting the performance of motion pictures. Journal of Media Economics, 7, 1–20.

Treme J. & Lee A. Craig (2013) Celebrity star power: Do age and gender effects influence box office performance?, Applied Economics Letters, 20:5, 440-445

Willemsen, L., Neijens, P.C. & Bronner, F., 2010. Perceived expertise vs .perceived trustworthiness: the suppressed effect of source type on review attitude. In: Okazaki, S. (Ed.), Advances in Advertising Research, vol.2. Gabler Verlag, Wiesbaden, pp.424–436.

Referenties

GERELATEERDE DOCUMENTEN

Plant Research International (PRI) beschikt nu echter over een apparaat dat meerdere spelden tegelijk kan vinden en dat aan een minieme hoeveelheid monster genoeg heeft om de dader

H2: Viewing satire will lead to a higher intention of political engagement in terms of (a) political participation and (b) interpersonal talk than watching a documentary or hard news

Moreover, we show that the limiting distribution of the joint workload process (as well as that of the virtual waiting time and the queue length processes) corresponds to the

It was clear that VTT needed partners to attain its aims to scientifically prove the health benefits of rye and bring the results to the attention of the larger population,

International journal of quality and reliability management, 14(9). Out of the crisis. Cambridge: Cambridge University Press. DOE see SOUTH AFRICA. Department of

The positive coefficient on DLOSSRDQ means that firm with negative earnings have a stronger relationship between credit ratings and risk disclosure quality compared to firms

The results confirmed the expected relation between the market value (measured using the market price to book ratio) and the credit rating, as well as relations between the CR

We construct the Multiple Linear Regression models for five dependent variables with metric data. In order to provide a comprehensive test of the hypotheses, four-step testing