What are the viewers’ reviews and emotions in Filmaffinity? A netnographic analysis

¿Cuáles son las valoraciones y emociones de los espectadores de cine en Filmaffinity? Un análisis netnográfico

José Serafín Clemente Ricolfe, Roberto Cervelló Royo, Víctor Rubio Aurrecoechea

Universitat Politècnica de València

Recibido: 26 de julio de 2021–Aceptado: 21 de septiembre de 2022–Publicado: 04 de julio de 2023

Forma de citar este artículo en APA:

Clemente Ricolfe, J. S., Cervelló Royo, R., & Rubio Aurrecoechea, V. (2023). What are the viewers’ reviews and emotions in Filmaffinity? A netnographic analysis. Revista Colombiana de Ciencias Sociales, 14(2), 485-508. https://doi.org/10.21501/22161201.4038

Abstract

The increasing importance of online user feedback is also influencing the film industry business. This work analyses the perceptions of film audience in Spain major using netnography. A total of 2187 opinions collected in a specialized film forum have been analyzed through factorial analysis of simple correspondences. The two dimensions obtained, together with the other results, show the importance of intrinsic variables. The first dimension allows to identify the personal action against the so-called environment. The second dimension contrasts positive perceptions against negative ones, allowing to detect market opportunities associated with a good soundtrack and a good plot. This second axis will also facilitate the detection of the worst rated films.

Keywords

Cinema; Ratings; Audience; Empirical research; Netnography.

Resumen

La creciente importancia de las valoraciones de los usuarios en la red está influyendo en el negocio del sector cinematográfico. El presente trabajo analiza las valoraciones de los espectadores de cine en España recurriendo para este fin a la netnografía. Un total de 2187 opiniones recogidas en un foro especializado en cine han sido analizadas mediante un análisis factorial de correspondencias simples. Las dos dimensiones obtenidas, junto al resto de resultados, muestran la importancia de las variables intrínsecas. La primera dimensión identifica la actuación personal frente al denominado entorno. La segunda dimensión opone valoraciones positivas frente a negativas y permite detectar una oportunidad de mercado asociada a buena banda sonora y buen argumento. Este segundo eje también facilita la detección de las películas peor valoradas.

Palabras clave

Cine; Valoraciones; Espectadores; Investigación empírica; Netnografía.

Introduction

Audiovisual productions carry a high risk. This affirmation is based on “audiovisual activities carry a high business risk since each product is unique and unrepeatable by making it from a combination, also unique, of productive elements” (Martí & Muñoz, 2001, p. 125). This makes investors in the audiovisual field, for example, have a lack of trust related to the uncertainty of investing in this sector and what might happen to their money, given that it is unlikely that they will be able to repeat the combination of material elements that gave them excellent results in the past. In addition, other uncertainties that may arise in the production process make the perceived returns insufficient to compensate for the risks taken, such as having amortization periods of a minimum of three to five years for such investments. In the case of Spain, not even the tax treatment applicable to investments in cinematographic productions in the light of the new regulation introduced by the Cinema Law 55/2007, of December 28, has made it more appealing. In other words, the deductions and bonuses for investments in cinematographic productions, their use by economic interest groupings and venture capital entities, and other tax-related aspects connected to these investments have not achieved the desired effect on potential investors. Thus, it can be stated that the cinematographic sector is one of the riskiest from an investor’s perspective. Therefore, the success or failure of film productions has been widely studied with mathematical models, pioneered by Litman (1983). Moreover, the number of films seems to decrease in the future due to the rising production and marketing costs of films, the appearance of substitute products, and the increasing concentration among theater operators (Hababou et al., 2016). In this context, the study of the viewers’ reviews becomes particularly relevant.

Multiple variables have been used to analyze the film spectator’s behavior, such as production costs, gender, release date, being nominated and winning an Oscar, Golden Globe, Golden Palm in the Cannes Film Festival, etc. Litman and Kohl (1989) pointed out that variables such as those above represent determining factors in the success of films or transmit information that reduces uncertainty. This type of work has relied upon databases, experiments, or surveys (Dellarocas et al., 2007). However, the Internet has modified the process of analysis and consumption in the current society by giving relevance to the comments and opinions of others. Within this context, cinema is one of the main sectors affected. As noted by Jacobs et al. (2015), the revision of online comments from spectators improves the understanding of how individuals come to conclusions about the quality of the artistic work of films.

Additionally, when researching the opinions of the film spectator, econometric and experimental models are often used (Cepeda, 2005). Nevertheless, the source of information used in this article will be netnography as consumers are increasingly, and to a greater extent, accessing communities, websites, and online platforms to obtain information more objective than that provided by the companies (Kozinets 2002, p. 61). Thus, netnography is a very useful tool to study film spectators since social media, forums, film databases, online cinema journals, etc., are on the rise, and their effect on films must be examined in detail (Hababou et al., 2016). It is also a source of information that allows obtaining naturally the personal and intangible experience represented by cinema (Stimpert et al., 2008). Furthermore, it is a way of studying social media while maintaining the complexities of cultural and experiential phenomena such as cinema (Kozinets, 2019).

Netnography has multiple applications, including the cultural field. Orea-Giner and Vacas-Guerrero (2020) studied the museum visitors’ experience. Vo Thanh and Kirova (2018) analyzed the enotouristsexperience. And Perkins (2010) observed the activities of online fans in the case of music concerts. In these works, netnography was used to study the consumers of creative products, such as cultural ones, and facilitated the identification of attributes and procurement of the dimensions that grouped them. In addition, they offered improvement opportunities in terms of efficiency or product design.

The new digital formats used by the cultural industry have also been analyzed, such as video streaming (Wang, 2020). Users of such platforms can comment on the videos available on the platforms and communicate among them through chats provided by the application. It also offers advantages over the traditional use of surveys to study cultural markets because netnography requires less time and money.

Among the questions concerning the users of these platforms, the following can be raised: Which factors motivate users to like, comment or share a video? Why do they find these videos attractive to watch? Why do they enjoy that format so much? What is the user experience like, and how do users interact with these environments?

It can be said that there is a gap in the literature that can provide answers to these questions since previous studies have not been able to provide a clear solution. Therefore, netnography is about solving these and other questions to close that gap in the current literature. It provides a set of instructions on how to specifically conduct qualitative research using a combination of different research practices (Kozinets, 2002). To do so, it analyzes the interactions and conversations generated on online sites to identify and understand the needs and decisions of relevant consumer groups.

However, netnography has hardly been used in the understanding of the evaluations and emotions of film spectators, so this is a novel approach.

Thus, considering all of the above, this article aims to analyze through netnography the emotions and reviews associated with different films in Spain using comments published on the Internet.

Conceptual framework

The production, distribution and exhibition of a film represent a complex, non-linear process involving a wide network of organizations and individuals. Therefore, its analysis requires a model that considers the different variables and forces involved in the making of a film. Thus, the model proposed by Hennig-Thurau et al. (2006) has been considered. Figure 1 shows the two types of variables considered in this model. On the one hand, there are intrinsic qualities, i.e., features or characteristics of the film (plot, characters, director, soundtrack, etc.) and, on the other hand, external factors (expert reviews, previous online comments, etc.).

Figure 1. General scheme for analyzing ratings of viewers of motion picture films.

Source: own elaboration based on Hennig-Thurau et al. (2006)

Among the intrinsic characteristics, some features of the film can be perceived before watching it, such as the plot, director, actors, or characters. Hadida (2009) stresses the importance of the plot because the audience can express positive reactions or not depending on the developed story. Nonetheless, although there is no guarantee in the film industry, there is a close positive relationship between the number of viewers and certain plots (Terry et al., 2011). For example, non-dramatic films (comedy, action, or animated) have more viewers than dramatic films (Weissmann, 2008).

The director is usually the most important person during the shooting of a film (Hennig-Thurau et al., 2001). Directors are in charge of coordinating the actors and the technical staff to make the film according to their plan regarding the development of the script and the story. Viewers can also evaluate films based on the participating actors. Thus, the visual appeal of a film actor has positive effects on the rating of a film (Wirtz et al., 2016). Moreover, if the actor is star power, the viewer has a heuristic device to decide to watch the film without having additional information (Desai & Basuroy, 2005). However, current empirical studies do not provide a clear conclusion about the impact of movie stars on the success of a film (Hofmann et al., 2017). For example, Hababou et al. (2016) did not observe that a movie star had a clear impact on the results of a film. Nonetheless, some authors note the interaction that may exist between movie stars and film plots, being much more decisive in less familiar genres (Desai & Basuroy, 2005).

Characterization is another variable relevant in cinematographic productions. According to Galán (2007), characterization models are focused on three axes: physical description (age, appearance, etc.), psychological (personality, temperament, etc.), and sociological (family relations, relationships, etc.). A character does not exist on their own, that is, isolated, they appear within a context and with certain influences depending on the historical period. For example, director David Lynch, in his film The Elephant Man, “integrates people with deformities and disabilities to the stereotypical world of Hollywood beauty” (Ponce, 2018, 289). His proposals question subjects hard to exploit, such as surrealism, human behavior, and mystery. Lynch’s characters cause his films to reach surprising places previously unknown to the viewer. In fact, his productions, far from trying to fill movie theaters, seek an effect on the audience, try to immerse them in his world of confusion and with his unique identity, which is known as film d’auteur or cult cinema.

Other intrinsic qualities of films cannot be perceived by viewers until they see the film. For example, the actors’ performance, the visual or sound style, or the emotions they might feel when watching the film. The first function of the cinema is to entertain, capture attention, and make the viewer focus completely on the story or narration appearing on screen. The interpretation provided by the cast is a variable that becomes stronger as it inspires more emotions in the viewer. These interpretations must be subjective since the cinema is art (Martínez-Salanova Sánchez, 1998). Furthermore, the audience is also attracted to the negative emotions (sadness or fear, for example) triggered by drama or horror films (Bartsch et al., 2010).

The visual style of the film and its soundtrack are vital elements in the cinema as a piece of art. For example, Dock Pope’s style, a British photography director, does not seek a striking visual impact based on the use of photographic elements that produce a theatrical, unreal, or spectacular result. On the contrary, it suggests the search for a photography at the service of the cinematographic narration (Cortés-Selva, 2014). The soundtrack is another tool that allows the transmission of different sensations and can influence the audience. For example, the album of the soundtrack of the film The Fellowship of the Ring, composed by Howard Shore, was certified gold (50,000 copies) before the film was released in movie theaters, eventually selling more than one million copies worldwide (Miguel de Bustos et al., 2004).

On the other hand, external factors include expert reviews, previous word-of-mouth comments, and online comments. Reviews by professional experts influence the perception of film viewers. Indeed, these critiques by experts modify the viewers’ opinions and create a positive relationship between the review of a film and its commercial success (Weissmann, 2008; Hababou et al., 2016). Once more, it is worth noting the interactions among the variables considered in the analysis of a film. Consumers try to find out from critics’ comments whether their high expectations for films featuring movie stars will be met or not (Desai & Basuroy, 2005).

Moreover, the comments from other moviegoers, both traditional word-of-mouth and online, also affect the reception of films. Word-of-mouth can have a positive or negative influence on the behavior of the consumer (Hennig-Thurau et al., 2001). Furthermore, consumers tend to trust the information provided by friends when deciding to watch a new film (Desai & Basuroy, 2005). However, along with traditional word-of-mouth, electronic word-of-mouth is added, which are two of the most influential factors on the consumer when it comes to choosing a film (Liu, 2006). The standardization of user ratings implemented by various websites (IMDb, Rotten Tomatoes, or Filmaffinity) through a system of stars or numeric scales also seems to have a great impact on viewer behavior, being this a good indicator to anticipate box office results (Dellarocas et al., 2007). Additionally, it is worth noting that online reviews of films are available in large quantities within a few hours of their release.

Nonetheless, to understand the rating of a viewer, it is necessary to implement a double approach: cognitive and emotional (Canini et al., 2009). The cognitive approach analyzes audiovisual elements, characters, etc., while the emotional tries to define the emotions transmitted in the film. Actually, one of the strengths attributed to cinematographic productions is their ability to generate emotions better than other media by combining images, texts, music, and narration, which engages the viewer cognitively and emotionally (García Amilburu & Ruiz Corbella, 2005). In the film Life Is Beautiful, happiness appears shortly after sadness. This film is a clear example of how cinema manages to guide the viewer throughout the film by transmitting the emotion required at each moment. Throughout most of the film, Roberto Benigni (director, actor, and scriptwriter of Life Is Beautiful) successfully conveys joy and hope to the viewer through the main character of the film. Then, that joy starts fading away and slowly transforms into sadness and, eventually, dejection (Gutiérrez et al., 2006). Sound also has a significant impact on the viewer’s emotions. For example, if an actor speaks energetically, excitement is conveyed (Soleymani et al., 2008). Films’ soundtracks also add an emotional dimension to them. For example, the soundtrack composed by H. Shore with the main themes sung by the Irish musician Enya and used in The Lord of the Rings trilogy transmitted a feeling of joy, happiness, or fear (Chełkowska-Zacharewicz & Paliga, 2019). Even the use of specific lighting for the same scene could cause a pleasant sensation if yellow tones were used, and a rejection feeling if grayish lighting was used (Canini et al., 2009).

Methodology

Approach and research questions

To explain the experience of a client from the consumer’s point of view, researchers can use two approaches: deductive or inductive. In the deductive approach, one starts with a logical relationship between the concepts and then moves to empirical evidence. In the inductive approach, the researcher starts with detailed observations and then makes generalizations. This article used the inductive approach because we intend to build the dimensions that explain the viewer’s ratings based on the classification of the information and considering the model proposed by Hennig-Thurau et al. (2006). Additionally, this inductive approach allows studying the ratings with an open point of view without predefining a certain number of frameworks (Igartua & Humanes, 2004).

Based on this premise, this article proposes two research questions:

1. Do viewers’ ratings revolve around the two variables posed in the model: intrinsic and external?

2. Are there associations of variables valued by viewers? If so, can underlying dimensions configured as the union of different variables be detected?

Netnography

Netnography begins with selecting the website and online community focused on viewers’ opinions. In this case, Filmaffinity was chosen due to three reasons. First, for its number of visits, it ranks 45th in the Spanish ranking (Alexa, 2019), having more than 700,000 users registered and more than 675,000 written reviews (Filmaffinity, 2019). Second, it is a cinema social network that does not belong to any communication media or group; that is, its independent nature gives it more credibility. Lastly, but not less important, it is an online community used in previous research on cinema (Gavilán et al., 2018).

Next, the second step in the development of netnography entails collecting data (Kozinets, 2002). The films selected for the study were the ten highest-grossing films in Spain in 2013 (Ministry of Education, Culture, and Sport, 2013), as shown in Table 1.

Table 1.

Ten highest-grossing films in Spain in 2013

Film

Genre

Viewers

Gross (euros)

The Croods (Sanders and De Micco, 2012)

Animated

2,105,441

13,777,179.91

Despicable Me 2 (Renaud and Coffin, 2013)

Animated

2,169,804

13,078,170.37

The Hobbit: The Desolation of Smaug (Jackson, 2013)

Adventure

2,342,046

13,077,298.41

World War Z (Forster, 2013)

Action

1,805,281

11,883,891.85

Frozen (Buck and Lee, 2013)

Animated

2,465,122

11,408,681.72

Monsters University (Scanlon, 2013)

Animated

1,863,404

11,241,002.08

Django Unchained (Tarantino, 2012)

Western

1,533,648

10,517,035.45

The Hunger Games: Catching Fire (Lawrence, 2012)

Science Fiction

1,650,573

10,385,738.34

Now You See Me (Leterrier, 2012)

Thriller

1,622,618

10,356,253.28

Iron Man 3 (Black, 2012)

Action

1,418,877

9,840,603.67

 

The sample consisted of 2187 valid reviews out of 2199 collected since 12 comments had to be discarded for not having information relevant to the study, such as

This film presents the great dilemma of Biology and Humanity. The need for survival of preserving the code, the laws in the face of the need for mutation, for change. In a world always devouring, always destroying itself. Looking for a tomorrow that can only be in Space, in the Stars. (Opinion 1)

Two criteria were considered regarding the quality of the data used in the research. First, to endorse the data collected, it was verified that there were no contradictions between the opinion expressed and the score given. For example, the review was rejected if the comment indicated that the viewer did not like the film or was angry, but the rating was high. Second, after a careful reading, the authors of the work individually observed the comments and described explicitly how they reached the interpretations. In the analysis of the comments, we tried to keep the original texts published in Filmaffinity to maintain their originality and spontaneity, which is one of the advantages of netnography.

The third stage of netnography is the analysis and interpretation of the collected messages. First, themes relevant to the research hypotheses were identified, and then emerging themes were compared with preconceptions derived from the literature. The variables used to analyze the viewers’ comments were based on the model proposed by Hennig-Thurau et al. (2006), which was previously explained. Thus, the variables included in the work are: Plot, Visual Appearance, Characters, Interpretation, Director, Movie Star, Soundtrack, Professional Review, Online and Word-of-Mouth Comments. Subsequently, they were divided into two opposite categories (for example, Good or Bad Interpretation) to include the positive or negative comments of each review in the corresponding category. However, the variable on the participation of a movie star and the emotions (joy, surprise, boredom, anger, fear, sadness, and disgust) were considered dummy variables, that is, whether or not they were present in the comment analyzed. In summary, all variables were categorical, a type of scale used in film research (Canini et al., 2009). That is, for each comment analyzed, the presence or absence of the variables derived from the literature was compared.

Statistical analysis

To study the dimensions and variables associated with each film, a factorial analysis of simple correspondence was carried out through the CAR program (Lorenzo-Seva et al., 2009). This data analysis technique will be used due to the presence of categorical variables and the possibility of elaborating a perceptual map, given that it is based on the association between objects and variables such as those used in this work. Specifically, the objects will be the ten films mentioned, and the variables will be those included in the comments of the viewers, which have been previously mentioned. In addition to studying the relationship between the variables, it is possible to understand how these relationships are structured.

RESULTS

Global ratings of the viewers

Analyzing the variables mentioned by the viewers (Table 2), Plot appears in the first place, supported by 81.5% of the comments. Although the positive and negative reviews of the films’ plots have similar values, it is the only variable where the negative reviews (42.5%) exceed the positive ones (39%). The following comments are examples of negative reviews: “the plot is too light and at the end one can’t help but wonder: That’s it?” (Opinion 2), “a fragmented mistake that presents in the form of a disjointed, uneven, and pastiche-like plot halfway between the prosaic and the light-hearted” (Opinion 3), or “a film whose plot, a children’s story, cannot bear the weight of 160 minutes of footage” (Opinion 4).

There are also positive reviews of the plot, “a cult film that will go down in history as a prodigious western, an amazing drama film and a brave historical document” (Opinion 3), “a plot that keeps you engaged from beginning to end” (Opinion 5), or “it proves a film of this genre can also have a good script” (Opinion 6). In short, the fact that most of the comments mention the plot and its confrontation implies the relevance of this variable for viewers.

Table 2.

Percentage of viewers that mentioned the variables analyzed

Variables

Total

Positive

Negative

Plot

81.5

39.0

42.5

Visual Appearance

60.9

50.2

10.7

Characters

53.8

31.7

22.1

Joy

50.6

Interpretation

49.8

42.2

7.6

Director

46.5

33.5

13.0

Movie Star

40.1

Soundtrack

29.7

21.9

7.8

Surprise

15.1

Boredom

11.4

Professional Review

3.8

2.7

1.1

Online Comment

3.4

1.7

1.7

Anger

3.2

Word-of-mouth

3.1

1.6

1.5

Fear

1.1

Sadness

0.5

Disgust

0.4

 

 

The visual appearance is in the second place, where almost half of the comments are positive. For example, “visually mindblowing” “impressive visual recreations” (Opinion 7), “it’s a visual show never seen before in this genre” (Opinion 8) or “flawless visual spectacle, a showcase of visual effects” (Opinion 9). The variable of Characters appears in the third place mentioned by more than half of the comments analyzed. The following comments demonstrate this: “How could you imagine that Tony Stark would ever feel fear? That’s how he appears throughout most of the movie, afraid, insecure and more human than ever,” (Opinion 10) or “the situations presented, the characters depicted and the resolution are so awesome, so amazing, so cool, that it doesn’t make any difference” (Opinion 11). In the case of Iron Man 3, the same character/actor, Tony Stark/Robert John Downey Jr., is consolidated as a franchise that has been exploited in continuous films confirming what Miguel de Bustos et al. (2004) noted.

Next, Joy appears as the most mentioned emotionin almost half of the comments analyzed. Some examples are: “Yesterday I had the pleasure of ENJOYING (in capital letters) this film, which seems small, almost harmless, but it’s the most entertaining and convoluted thing you can see,” “just to brighten up the afternoon of any day of the week” (Opinion 12) or “their work in this adaptation has been a great joy” (Opinion 13).

The interpretation is another variable mentioned in approximately half of the comments analyzed. There were mainly positive comments such as, in the case of Django Unchained, “Waltz, DiCaprio, and Foxx’s interpretations are worthy of the golden Oscar statuette, they are simply masterful” ” (Opinion 14). Although there were also negative comments: “it won’t be remembered as one of his best interpretations” (Opinion 15).

Reviews about the director were present in 46.5% of the users’ comments. Here are some examples: “Django Unchained maintains everything that Quentin Tarantino is, all his essence in each scene, but this time he achieves a complete adding it a bit of freshness” (Opinion 16) or “Peter Jackson outdoes himself as God of the camera” (Opinion 17).

Four out of ten comments highlighted the appearance of a movie star. The next comments are examples of this: “I thought this was going to be the classic zombie movie, but of course, starring Brad Pitt…” or “the greatness of an actor such as Robert Downey Jr” (Opinion 18).

The soundtrack of the films was mentioned in almost 30% of the comments analyzed. For example, “beautiful and catchy songs” or “the soundtrack is just sublime, it couldn’t be more appropriate” (Opinion 19).

The surprise feeling was mentioned in over 15% of the comments as shown below: “if there’s something interesting about this film, it’s that it surprises for not being what one expected” (Opinion 20).

The boredom feeling was mentioned in 11.4% of the comments analyzed: “all the segment about the city of the lake is boring, it makes it dense and long” (Opinion 21).

To a lesser extent, four variables were mentioned in between 3% and 4% of the comments, such as Professional Review, “most critics have praised it to the skies” (Opinion 22). Some online comments say “some criticize it for spoiling almost all the action scenes in the trailer (they are right about that) and others claim that it is an example of how to make a zombie movie without overusing gore” (Opinion 23). The anger feeling is reflected in comments such as: “a movie that made my indignation and anger increase as the footage went by” and the word-of-mouth: “the only thing I hear (and heard before watching it) is that it’s ‘very original’ and ‘it breaks Disney’s typical archetypes” (Opinion 24).

Finally, due to its minimal relevance, the fear, sadness, and disgust feelings were discarded for the rest of the research. This result may be attributed to the association of these emotions with movie genres not present in the analyzed sample, such as horror, drama, etc.

Association between variables and films

Among the analyzed comments, there are references to two or more variables simultaneously. For example, in “this time Tarantino surprises” there is a reference to the director of the film Django Unchained and the feeling of surprise. Or in a comment about Tarantino’s western, that reads, “you have a great time with his original story, incredible actors and sense of humor” (Opinion 25), that is, there is a mention of joy, plot and actor’s interpretation. Therefore, the factorial analysis of simple correspondences is applied to the data in Table 3 to detect this type of interrelationship. The study of axes 1 and 2 has been considered because their contribution to inertia is 63.2% (see Table 4). In fact, the inertia of these axes to analyze existing dependencies in the social sciences exceeds the minimum value of 50-60% (Hair et al., 2006). Moreover, the parallel analysis (Lorenzo-Seva, 2011) also recommended having two axes for this data set.

Table 3.

Level of association (number of comments) between the films and the analyzed variables (the numbers and letters appearing in this table will be used in the perceptual map to identify the films and variables analyzed)

A. The Croods

B. Despicable Me 2

C. The Hobbit: The Desolation of Smaug

D. World War Z

E. Frozen

F. Monsters University

G. Django Unchained

H. Now You See Me

I. Iron Man 3

J. The Hunger Games: Catching Fire

1. Joy

43

49

125

183

71

59

306

84

120

66

2. Surprise

7

5

13

56

41

9

91

44

49

16

3. Anger

5

2

9

6

9

1

5

16

11

6

4. Boredom

7

8

56

26

18

7

65

13

28

22

5. Good Plot

36

25

93

84

71

51

290

31

83

89

6. Bad Plot

22

26

155

199

68

32

155

112

111

50

7. Good Characters

30

53

120

51

74

52

161

14

89

50

8. Bad Characters

9

8

106

122

36

7

56

40

83

17

9. Good Interpretation

5

6

65

99

3

10

457

87

106

84

10. Bad Interpretation

2

10

12

48

0

0

40

13

25

16

11. Good Director

11

9

94

55

14

10

398

20

62

60

12. Bad Director

0

1

89

35

2

2

102

27

21

6

13. Good Visual Appearance

45

25

162

172

106

53

279

54

121

80

14. Bad Visual Appearance

2

2

50

62

1

0

65

24

15

13

15. Good Word-of-Mouth

0

2

1

3

12

1

4

3

5

3

16. Bad Word-of-Mouth

1

0

5

6

1

6

3

3

6

1

17. Good Professional Review

1

0

3

1

10

2

9

2

17

15

18. Bad Professional Review

0

0

4

2

2

4

5

6

1

1

19. Good Soundtrack

11

11

47

42

74

11

226

12

27

19

20. Bad Soundtrack

2

1

25

10

52

1

58

7

11

3

21. Movie Star

0

0

5

225

0

0

401

54

118

73

22. Good Online Comment

1

0

5

7

4

0

10

0

7

4

23. Bad Online Comment

0

0

8

11

0

0

11

2

4

1

 

 

Table 4.

Inertia explained by axes (in %)

%

Accumulated %

1

38.7

38.7

2

24.5

63.2

3

12.5

75.7

4

10.9

86.7

5

5.2

91.8

6

3.9

95.8

7

2.3

98.1

8

1.3

99.3

9

0.7

100

 

 

When interpreting the comments associated with each axis, we looked for those variables and films that show significant correlations. Thus, considering the results of Table 5 and the perceptual map (Figure 2), it can be noted that the first axis opposes variables such as Good Interpretation, Movie Star, and Good Director at the positive end of the axis, against Good Visual Appearance and Good Characters in the negative end. That is, this first dimension identifies variables focused on human performance (Director, Interpretation and Movie Star) and in opposition, Visual Appearance and Characters are placed as concepts of the film’s environment. In this case, on the one hand, the concept of the human performance is related to the film Django Unchained as the next comments exemplify: “if we put together a great director, great actors (no doubt that Christoph Waltz, Leonardo di Caprio and Samuel L. Jackson are 3 beasts on the screen) and a great story, we get what we were looking for” (Opinion 26) or

among the wonderful interpretations of all actors, each one immersed in their role (few people have praised the stoic Django brought to life by Jamie Fox, playing his role of hero of the West by the book), the director doesn’t move away from his role of revenge instigator; first he shows us with sadistic brutality the abuse suffered by the black slaves to see then Django take revenge, break his chains, with no less sadism than those who hurt him so much. This is the violence game that Tarantino offers us in this film. (Opinion 27)

On the other hand, the environment is associated with the four animated films present in the research: The Croods, Frozen, Monsters University, and Despicable Me 2. Some comments from the viewers of these films are the following: “I liked a lot the new Disney’s proposal, basically because of the magnificent design of stage and characters (that lovely snowman or the reindeer), its technique and animation” (Opinion 28) or “it’s very good visually and the main characters connect with the viewer” (Opinion 29).

Table 5.

Correlations of the variables and the films for the two axes analyzed

Variables/Films

Correlations with the axes

1

2

1. Joy

-0.552

-0.133

2. Surprise

-0.138

-0.183

3. Anger

-0.520

-0.278

4. Boredom

-0.541

0.075

5. Good Plot

-0.238

0.769

6. Bad Plot

-0.560

-0.792

7. Good Characters

-0.751

0.423

8. Bad Characters

-0.477

-0.767

9. Good Interpretation

0.925

0.111

10. Bad Interpretation

0.235

-0.673

11. Good Director

0.750

0.513

12. Bad Director

0.117

-0.288

13. Good Visual Appearance

-0.847

0.124

14. Bad Visual Appearance

0.219

-0.781

15. Good Word-of-Mouth

-0.584

0.300

16. Bad Word-of-Mouth

-0.445

-0.262

17. Good Professional Review

-0.222

0.261

18. Bad Professional Review

-0.326

-0.131

19. Good Soundtrack

0.041

0.810

20. Bad Soundtrack

-0.402

0.506

21. Movie Star

0.872

-0.262

22. Good Online Comment

-0.081

-0.005

23. Bad Online Comment

0.271

-0.720

A. The Croods

-0.729

0.275

B. Despicable Me 2

-0.651

0.190

C. The Hobbit: The Desolation of Smaug

-0.568

-0.204

D. World War Z

0.130

-0.873

E. Frozen

-0.720

0.439

F. Monsters University

-0.665

0.278

G. Django Unchained

0.852

0.496

H. Now You See Me

-0.016

-0.697

I. Iron Man 3

-0.140

-0.456

J. The Hunger Games: Catching Fire

0.298

0.282

 

 

Figure 2. Perceptual map of the variables and films (the numbers and letters are those used in Table 4 to identify variables and films correspondingly)

 

The second axis emerges from the opposition of the variables Good Soundtrack and Good Plot, which are located in the positive area, against Bad Plot, Bad Visual Appearance, Bad Characters, Bad Online Comments, and Bad Interpretation, located in the negative axis as seen in Figure 2. The second dimension opposes positive comments against negative comments, and that is why it could be labeled as bipolar since it has two opposite ends. Moreover, the negative comments are close to the films World War Z and Now You See Me, which obtained an average rating of 5.5 and 5.2, respectively (see Table 6). This result may indicate the variables that help the viewer assess the artistic quality of a film by differentiating between good and bad ratings. In addition, it is relevant because it seems to highlight the importance of the intrinsic variables in the final rating obtained by the film since most aspects are focused on features of the film, except for the online comments.

Table 6.

Rating of the 10 films analyzed, being 0 very negative and 10 very positive

Film

Average Score

The Croods

6.8

Despicable Me 2

6.7

The Hobbit: The Desolation of Smaug

6.6

World War Z

5.5

Frozen

6.6

Monsters University

6.8

Django Unchained

7.5

The Hunger Games: Catching Fire

6.9

Now You See Me

5.2

Iron Man 3

6.2

Discussion

Plot is the variable with the highest number of reviews, demonstrating the importance of answering the question “What?” from a film, that is, the content of the film (Aguilar, 2018). For example, The Platform, one of the most viewed films on Netflix in Spain, has made people continuously talk about it and comment on aspects related to the plot (Marín, 2020). Visual Appearance is the second attribute with the highest number of reviews, which along with the previous one, as noted by Aguilar (2018), suggests the plane of content cannot be separated from the plane of form, of the expression of that content, of the “How?” Additionally, the most mentioned emotion was Joy. It was also the positive feeling most mentioned in Aurier and Guintcheva’s (2015) research on emotions in films. Given that the rest of the emotions had a lower number of mentions, it is confirmed that Joy is a differentiating factor (Fernández Megías et al., 2011).

The interrelations obtained between variables and films confirm that animation and special effects are closely related, as suggested by Armenteros (2011). In fact, the technologies used to generate computer graphics in animated films can eventually extend to other entertainment products such as videogames consoles. Certainly, studios such as Disney and Pixar, creators of successful animated films, stand out in computer animation. Moreover, the connection between animated films and characters has also been analyzed. For example, Walt Disney Animation Studios has represented a specific image of women through its animated feature films. However, this representation has evolved throughout cinema history (López Iglesias & de Miguel Zamora, 2013). The Snow White character represents the submissive woman and her stepmother, the femme fatale. Nonetheless, the female animated characters of Disney mutate, and Dory, from the film Finding Nemo, becomes the leader, transforming into an autonomous character that takes control of her life (López Iglesias & de Miguel Zamora, 2013). In addition, this result is in line with previous works, such as that of Igartua (2008), where the identification with the characters through cinematographic fiction is emphasized.

Moreover, the first axis opposes animated cinema to the performance of people in Django Unchained. Tarantino cannot stop talent from jumping out of the screen toward the viewers who are captivated by his ability to tell stories (Fresneda, 2014). In fact, as noted by García Lozano (2013), Django Unchained confirms Tarantino’s talent as a director that can continue exploring a personal universe with a unique and distinctive style. Furthermore, the relationship between people, such as the director or actors of a film, reaffirms its use in web recommendation or suggestion systems for users (Carrer-Neto et al., 2012). The second axis opposes positive to negative comments, mainly intrinsic qualities, which appear to differentiate bad films due to their artistic quality. That is, following Lancaster’s model, the viewers seek to obtain the best set of attributes that maximize their well-being. This result was already obtained in works of art, where a painting was good or bad in terms of intrinsic variables such as composition, drawing, or color (Ginsburgh & Weyers, 1999). Moreover, the association between the variables Characters and Plot is confirmed; their interrelation has been noted already in the bibliography. Tal-Or and Cohen (2010) state that when the audience is deeply immersed in a story, they worry about the characters.

It is worth noting that none of the films analyzed was associated with Good Soundtrack and Good Plot. As Moreno (2015) notes, melodies in films are composed to strengthen the plots. Additionally, this result could also be interpreted as an opportunity to offer a musical with an appealing plot, such as La La Land, released after the analyzed films and with a positive reception. Therefore, as Kozinets et al. (2014) suggest, netnography is a non-intrusive cultural research that allows collecting data without making its presence visible to the members of the culture of the study. This would allow the cinematographic industry to make decisions based on viewers’ comments more naturally than if they were asked directly and detect perspectives unrelated to films.

Conclusions

Regarding the first question of this research, the first three variables most mentioned in the total viewers’ comments in Spain focus on intrinsic qualities: Plot (81.5%), Visual Appearance (60.9%), and Characters (53.8%). As Cepeda (2005) notes, the selection of a film depends just on some of its attributes, not all of them. Thus, this result emphasizes the relevance of the content and the form for the viewers and how they help build the final opinion of the film. Moreover, the cinematographic industry should focus its efforts on the raison d’être of its product. That is, cinema is one of the storytelling arts where the plot and how the story is told are relevant (Matamaros, 2009).

Regarding the second research question, interrelations between several variables were detected. The first of the relationships found opposes variables related to the performance of people to visual aspects and characters. Additionally, each type of variable is associated with a different film profile. For example, it confirms that the characters in animated films have a very relevant role (López Iglesias & de Miguel Zamora, 2013). Similarly, the second axis shows the consistency in the comments of viewers, differentiating positive variables from negative variables; the latter variables are associated with films with low global ratings. This result confirms the findings of Cheng and Huang (2019) since viewers use a limited set of elements to rate films. Knowing which elements of the reviews are relevant is important for the film industry. The industry should analyze these insights to understand the user and design the products better.

Finally, it is important to note the limitations of this work. First, it only includes international films. Therefore, a future line of research could be a study on national films, in this case, Spanish films. However, other countries could be studied too, since there may be cultural differences. Another limitation arises from the fact that the generalization of the results is limited to analysis at a specific moment in time. Longitudinal studies could offer an improvement in the analyzed reviews allowing us to observe their evolution through time. To conclude, and for future research lines, it would be desirable to include sociodemographic variables and cultural habits (for example, frequency of visits to the movie theater) with the aim to know better the audience’s opinions.

Researchers

In the research, José-Serafín Clemente-Ricolfe participated as principal investigator and his work focused on the design of objectives, theoretical framework, writing and data analysis. Roberto Cervelló-Royo, co-investigator, participated in the drafting and final revision of the manuscript. Victor Rubio-Aurrecoechea, as assistant, developed the field work and data collection.

Conflict of interest

The authors state that they do not have a conflict of interest with the institution or any commercial association.

References

Aguilar, S. (2018). Niveles narrativos cinematográficos. Revista Iberoamericana de Comunicación, (37), 29-54. https://ric.ibero.mx/index.php/ric/article/view/19/14

Alexa. (2019). Top Sites in Spain, consultado el 2 de septiembre de 2019, https://www.alexa.com/topsites/countries/ES/

Armenteros, M. (2011). Efectos visuales y animación. E-Archivos Universidad Carlos III de Madrid.

Aurier, P., & Guintcheva, G. (2015). The Dynamics of Emotions in Movie Consumption: A Spectator-Centred Approach. International Journal of Arts Management, 17(2), 5-18.

Bartsch, A., Appel, M. & Storch, D. (2010). Predicting Emotions and Meta-Emotions at the Movies: The Role of the Need for Affect in Audiences’ Experience of Horror and Drama. Communication Research, 37(2), 167–190. https://doi.org/10.1177/0093650209356441

Canini, L., Benini, S., Migliorati, P., & Leonardi, R. (2009). Emotional identity of movies. 2009 16th IEEE International Conference on Image Processing (ICIP). https://doi.org/10.1109/ICIP.2009.5413556

Carrer-Neto, W., Hernández-Alcaraz, M. L., Valencia-García, R., & García-Sánchez, F. (2012). Social knowledge-based recommender system. Application to the movies domain. Expert Systems with Applications, 39(12), 10990–11000. https://doi.org/10.1016/j.eswa.2012.03.025

Cepeda, A. (2005). El comportamiento del consumidor de cine: una revisión preliminar de la literatura. Revista Colombiana de Psicología, (14), 89-99.

Chełkowska-Zacharewicz, M., & Paliga, M. (2019). Music emotions and associations in film music listening: The example of leitmotifs from The Lord of the Rings movies. Roczniki Psychologiczne, 22(2), 151-175. https://doi.org/10.18290/rpsych.2019.22.2-4

Cheng, L.-C., & Huang, C.-L. (2019). Exploring contextual factors from consumer reviews affecting movie sales: an opinion mining approach. Electronic Commerce Research, 20(4), 807–832. https://doi.org/10.1007/s10660-019-09332-z

Cortés-Selva, L. (2014). El sello del estilo visual de Dick Pope y Mike Leigh: análisis de su obra cinematográfica. Historia y Comunicación Social, 18, 491-501. http://dx.doi.org/10.5209/rev_HICS.2013.v18.43983

Dellarocas, C., Zhang, X. M., & Awad, N. F. (2007). Exploring the value of online product reviews in forecasting sales: The case of motion pictures. Journal of Interactive Marketing, 21(4), 23–45. https://doi.org/10.1002/dir.20087

Desai, K. K., & Basuroy, S. (2005), Interactive influence of genre familiarity, star power, and critics’ reviews in the cultural goods industry: The case of motion pictures. Psychology & Marketing, 22(3), 203-223. https://doi.org/10.1002/mar.20055

Fernández Megías, C., Pascual Mateos, J. C., Soler Ribaudi, J., & Fernández-Abascal, E. G. (2011). Validación española de una batería de películas para inducir emociones. Psicothema, 23(4), 778-785. https://reunido.uniovi.es/index.php/PST/article/view/9159/9023

Filmaffinity (2019). ¿Qué es Filmaffinity y cómo funciona? Consultado el 21 de octubre de 2019, https://www.filmaffinity.com/es/site-guide.php

Fresneda, I. (2014). El Django de Quentin Tarantino. Pérgola, 2.

Galán Fajardo, E. (2007). Fundamentos básicos en la construcción del personaje para medios audiovisuales. Revista CES Felipe II, 7, 1-11.

García Amilburu, M., & Ruiz Corbella, M. (2005). Cine y (des)educación afectiva, en VV.AA.: cultivar los sentimientos. Propuestas desde la Filosofía de la Educación. Dykinson.

Igartua, J. J. (2008). Identificación con los personajes y persuasión incidental a través de la ficción cinematográfica. Escritos de Psicología, 2(1), 42-53.

García Lozano, F. J. (2013). Luces y sombras de la historia americana. Razón y fe, 267(1374), 359-362.

Gavilán, D., Martínez-Navarro, G. & Fernández-Lores, S. (2018). Influencia social en las comunidades de cine: filmaffinity como caso de estudio. Estudios Sobre El Mensaje Periodístico, 24(1), 551-565. http://dx.doi.org/10.5209/ESMP.59966

Ginsburgh, V., & Weyers, S. (1999). On the perceived quality of movies. Journal of Cultural Economics, 23(4), 269-283.

Gutiérrez, M. C., Pereira, M. C. & Valero, L. F. (2006). El cine como instrumento de alfabetización emocional. Teoría de la educación, 18, 229-260.

Hababou, M., Amrouche, N., & Jedidi, K. (2016). Measuring Economic Efficiency in the Motion Picture Industry: a Data Envelopment Analysis Approach. Customer Needs and Solutions, 3(3-4), 144–158. https://doi.org/10.1007/s40547-016-0069-0

Hadida, A. L. (2009), Motion picture performance: A review and research agenda. International Journal of Management Reviews, 11, 297-335. https://doi.org/10.1111/j.1468-2370.2008.00240.x

Hair, J., Black, W., Babin, B., Anderson, R., & Tatham, R. (2006). Multivariate Data Analysis, Prentice-Hall International, Upper Saddle River, NJ.

Hennig-Thurau, T., Houston, M. B. & Walsh, G. (2006). The differing roles of success drivers across sequential channels: An application to the motion picture industry. Journal of the Academy of Marketing Science, 34(4), 559-575. https://doi.org/10.1177/0092070306286935

Hennig-Thurau, T., Walsh, G. & Wruck, O. (2001). An investigation into the factors determining the success of service innovations: The case of motion pictures. Academy of Marketing Science Review, 6(1), 1-23.

Hofmann, J., Clement, M., Völckner, F., & Hennig-Thurau, T. (2017). Empirical generalizations on the impact of stars on the economic success of movies. International Journal of Research in Marketing, 34(2), 442-461. https://doi.org/10.1016/j.ijresmar.2016.08.006

Igartua, J. J., & Humanes, M. L. (2004). Imágenes de Latinoamérica en la prensa española. Una aproximación empírica desde la teoría del encuadre. Comunicación y Sociedad, 1, 47-75

Jacobs, R. S., Heuvelman, A., Ben Allouch, S. & Peters, O. (2015). Everyone’s a critic: The power of expert and consumer reviews to shape readers’ post-viewing motion picture evaluations. Poetics, 52, 91–103. https://doi.org/10.1016/j.poetic.2015.07.002

Kozinets, R. V., Dolbec, P.-Y., & Earley, A. (2014). Netnographic Analysis: Understanding Culture Through Social Media Data. The SAGE Handbook of Qualitative Data Analysis, 262–276. http://dx.doi.org/10.4135/9781446282243.n18

Kozinets, R.V. (2002). The field behind the screen: using netnography for marketing research in online communities. Journal of Marketing Research, 39(1), 61–72. https://doi.org/10.1509/jmkr.39.1.61.18935

Kozinets, R.V. (2019). Netnography: The essential guide to qualitative social media research. London:SAGE Publications Limited.

Litman, B. R. (1983). Predicting Success of Theatrical Movies: An Empirical Study. The Journal of Popular Culture, 16(4), 159-175. https://doi.org/10.1111/j.0022-3840.1983.1604_159.x

Litman, B. R. & Kohl, L. S. (1989) Predicting financial success of motion pictures: The ‘80s experience. Journal of Media Economics, 2(2), 35-50. https://doi.org/10.1080/08997768909358184

Liu, Y. (2006). Word of Mouth for Movies: Its Dynamics and Impact on Box Office Revenue. Journal of Marketing, 70(3), 74–89. https://doi.org/10.1509/jmkg.70.3.074

López Iglesias, M., & de Miguel Zamora, M (2013). La fémina Disney: análisis y evolución del personaje femenino en cuatro películas de la factoría Disney. Sociedad y Economía, (24), 121-142.

Lorenzo-Seva, U. (2011). Horn’s Parallel Analysis for Selecting the Number of Dimensions in Correspondence Analysis. Methodology, 7(3), 96–102. https://doi.org/10.1027/1614-2241/a000027

Lorenzo-Seva, U., Van de Velden, M., & Kiers, H. A. (2009). CAR: A MATLAB package to compute correspondence analysis with rotations. Journal of Statistical Software, 31(8), 1-14. https://doi.org/10.18637/jss.v031.i08

Marín, A. (01 de abril de 2020). Por qué la gente no deja de hablar de ‘El hoyo’, el filme español de Netflix que triunfa en el mundo. El País. https://elpais.com/elpais/2020/03/31/icon/1585667899_698507.html

Martí, F. P., & Muñoz, C. (2001). Economía del cine y del sector audiovisual en
España”. ICE: Revista de Economía, (792), 124-138.

Martínez-Salanova Sánchez, E. (1998). Aprender pasándolo de película. Comunicar, 6(11), 27–36. https://doi.org/10.3916/C11-1998-05

Matamaros, D. (2009). Las tres audiencias. En Distribución y marketing cinematográfico. Manual de primeros auxilios. UB Edicions.

Miguel de Bustos, D. J. C., Eizaguirre, R., López, E., & Oter, J. (2004). Factores de éxito de las series y sagas. Ámbitos. Revista Internacional de Comunicación, (11-12), 465–481. https://institucional.us.es/revistas/Ambitos/11-12/http-_dx.doi.org_10.12795_Ambitos.2004.i11-12.26.pdf

Ministerio de Educación, Cultura y Deporte. (2013). Anuario de Cine. Resumen ejecutivo 2013, consultado el 22 de mayo de 2017, http://www.mecd.gob.es/cultura/areas/cine/mc/cdc/anos-anteriores/ano-2013/c/cine-peliculas-recaudacion.html

Moreno Cardenal, L. (2015). El devenir del cine musical de Hollywood. Arbor, 191(774), a256. http://dx.doi.org/10.3989/arbor.2015.774n4011

Orea-Giner, A., & Vacas Guerrero, T. (2020). Textual analysis as a method of identifying museum attributes perceived by tourists: An exploratory analysis of Thyssen-Bornemisza National Museum in Spain. ESIC MARKET Economic and Business Journal, 51(3), 527–544. https://doi.org/10.7200/esicm.167.0513.2

Perkins, A. (2010). Identification in Popular Music: A Netnographic Exploration of Online Fan Communities. Proceedings of the Australian and New Zealand Marketing Academy Conference. http://anzmac2010.org/proceedings/pdf/anzmac10Final00030.pdf

Ponce, J. E. (2018). Neofiguración e iconoclasia en el cine de David Lynch. H-ART. Revista de Historia, Teoría y Crítica de Arte, 3, 273–310. http://dx.doi.org/10.25025/hart03.2018.11

Soleymani, M., Chanel, G., Kierkels, J. J. M. & Pun, T. (2008). Affective Characterization of Movie Scenes Based on Multimedia Content Analysis and User’s Physiological Emotional Responses. 2008 Tenth IEEE International Symposium on Multimedia. https://doi.org/10.1109/ISM.2008.14

Stimpert, J. L., Laux, J. A., Marino, C. & Gleason, G. (2008). Factors Influencing Motion Picture Success: Empirical Review And Update. Journal of Business & Economics Research (JBER), 6(11). https://doi.org/10.19030/jber.v6i11.2488

Tal-Or, N. & Cohen, J. (2010). Understanding audience involvement: Conceptualizing and manipulating identification and transportation. Poetics, 38(4), 402–418. https://doi.org/10.1016/j.poetic.2010.05.004

Terry, N., Butler, M., & De’Armond, D. A. (2011). The determinants of domestic box office performance in the motion picture industry. Southwestern Economic Review, 32, 137-148. http://dx.doi.org/10.4102/sajbm.v44i3.162

Vo Thanh, T., & Kirova, V. (2018). Wine tourism experience: A netnography study. Journal of Business Research, 83, 30–37. https://doi.org/10.1016/j.jbusres.2017.10.008

Wang, Y. S. (2020). Netnography in Live Video Streaming. In Netnography Unlimited (pp. 278-292). Routledge.

Weissmann, V. (2008). ¿Es la crítica un predictor del éxito de una película? Determinantes del nivel de audiencia de cine en Argentina. Palermo Business Review, (2), 33-44.

Wirtz, B. W., Mermann, M., & Daiser, P. (2016). Competencies and success of motion picture actors: A resource-based and competence-based empirical analysis. Journal of Media Practice, 17(1), 98-116. https://doi.org/10.1080/14682753.2016.1159452

Authors notes

José Serafín Clemente Ricolfe

Ph. D. Economics and Business Administration, Universitat Politècnica de València. Professor del Departamento de Economía y Ciencias Sociales, España. Contacto: jocleri1@upv.es ORCID: https://orcid.org/0000-0002-3962-3876, https://scholar.google.com/scholar_settings?hl=es#d=gs_hdr_drw&t=1686214942914

Roberto Cervelló Royo

Ph.D. Economics and Business Administration, Universitat Politècnica de València. Professor del Departamento de Economía y Ciencias Sociales, España. Contacto: rocerro@esp.upv.es. ORCID: https://orcid.org/0000-0002-8304-4177

Víctor Rubio Aurrecoechea

Degree in Business Administration and Management, Universitat Politècnica de València, España. Contact: vicruau@gmail.com. ORCID: https://orcid.org/0000-0002-5631-1516