ArticlePDF Available

Abstract and Figures

Findings on European countries show non-uniform decreasing trends of film theatres’ audiences, the decline being more severe in Spain. This research presents a multifaceted perspective of Spanish filmgoers focused on motives for and barriers to film theatres attendance. Two comprehensive scales (motives, barriers) are proposed. First, motives and barriers are optimally scaled with principal components analysis (PCA); and, second, we identify segments of filmgoers with latent class modelling (LC). PCA recommended a five-factor solution for motives (education, film popularity, film quality, social interaction, and mood) and a seven-factor solution for barriers (film offerings, venue’s features, perception, preference and place, substitute activities, financial restrictions, recreation time disposability). LC analysis suggested three segments: mainstream filmgoers who watch films in multiplexes in shopping centers and in the center of the city; art-house filmgoers; and filmgoers who go to film theatres to watch films in original version. The socioeconomic and behavioral covariates complete the profile of the clusters, and the findings are consistent with the existing evidence on film audiences. Increasing cultural participation is the objective of many governments’ cultural policies and a more comprehensive understanding of film audiences can contribute to this.
No caption available
No caption available
No caption available
No caption available
No caption available
Content may be subject to copyright.
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
Marketing Department, Universitat de València, Spain
Corresponding author
Economics Department, Universitat de Girona, Spain
Applied Economics Department, Universitat de València, Spain
Findings on European countries show non-uniform decreasing trends of film theatres’
audiences, the decline being more severe in Spain. This research presents a multifaceted
perspective of Spanish filmgoers focused on motives for and barriers to film theatres
attendance. Two comprehensive scales (motives, barriers) are proposed. First, motives and
barriers are optimally scaled with principal components analysis (PCA); and, second, we
identify segments of filmgoers with latent class modelling (LC). PCA recommended a five-
factor solution for motives (education, film popularity, film quality, social interaction, and
mood) and a seven-factor solution for barriers (film offerings, venue’s features, perception,
preference and place, substitute activities, financial restrictions, recreation time disposability).
LC analysis suggested three segments: mainstream filmgoers who watch films in multiplexes
in shopping centers and in the center of the city; art-house filmgoers; and filmgoers who go to
film theatres to watch films in original version. The socioeconomic and behavioral covariates
complete the profile of the clusters, and the findings are consistent with the existing evidence
on film audiences. Increasing cultural participation is the objective of many governments’
cultural policies and a more comprehensive understanding of film audiences can contribute to
Keywords: filmgoers, motives, barriers, latent-class models, PCA
JEL classification: M310, Z11
1. Introduction
In a report for the EU, Paris (2014) has shown that the decrease in film theatres’ audiences
has reached its two lowest levels since the beginning of the century, the first in 2005 and the
second in 2009, in spite of being a favorite way of spending spare time, especially among
young people (De Bruyn and Cillessen, 2008). According to Paris (2014), this decreasing
trend was not uniformly distributed, as some countries, such as Spain, have experienced a
very severe decline in their cinema audiences (more than 46% in the period 2001-2013),
while others have managed this trend more effectively. In Spain, this evolution was observed
in the previous century too, with more than half of the film theaters being closed from the
‘60s to the early ‘90s (Cuadrado and Frasquet, 1999). This evidence invites the exploration of
a deeper understanding of the filmgoers’ behavior to help marketers and owners of film
theatres design new marketing strategies meant to increase filmgoers’ attendance.
From a consumer behavior perspective, motives are taken to be at the origin of the
decision process which manifests itself in actions intended to satisfy needs (Crompton and
McKay, 1997:425; see also Murray, 1964; Iso-Ahola, 1980). Jackson (1997) developed the
‘constraints’ paradigm, according to which the desire or need to participate is inherent in
human condition, but individual choices are limited by constraints. Thus, “in order to fully
understand leisure involvement we need to understand both facilitators and constraints, and
how they work together to produce participation and non-participation and their
accompanying experiences” (Raymore, 2002:38).
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
With respect to individuals’ engagement with cultural activities, in general, and films, in
particular, many studies have sought to explain the motives underlying filmgoers’ behavior
paying relatively less attention to the barriers. As a matter of fact, even motives have often
been determined from a comparative analysis with the audiences of other media, such as
television (Austin, 1986:117; Austin, 1983). Some authors have argued that the main reason
why fewer studies have paid attention to the motives of cultural audiences (i.e., filmgoers)
was that researchers (i.e., film reviewers) were interested in addressing a more general
audience (Dyer, 1981). This could explain why “in the motion picture industry, the consumer
is the great unknown” (Wierenga, 2006:674). In the case of films, audiences are particularly
important because it is “through the existence of an audience that film acquires social and
cultural importance” (Gripsrud and Lavik, 2008:455).
This research seeks to analyze the behavior of Spanish filmgoers, with a special focus on
motives and barriers of attendance. With this research, we would like to contribute to the
literature on cultural participation by providing a multifaceted characterization of filmgoers.
The empirical analysis builds on data collected via a self-administered questionnaire to a
sample of Spanish population, of 18 years of age and above, selected according to a non-
probabilistic method, based on quotas of gender and age (INE, 2011). The methodological
approach consists of two steps: first, motives and barriers are optimally scaled with principal
components analysis (PCA) and, at a second stage, we employ exploratory latent class (LC)
modelling, a method generally recommended for the study of hedonic products’ consumption
(Botet and Wedel, 1999), with the purpose of identifying specific patterns of film theatres’
attendance. Thus, based on preferences for film theatres and their environmental
characteristics, we identify three segments of filmgoers: one that prefers multiplexes in
shopping centers and in the center of the city, a group for which going to the film theatres
means leisure and entertainment; art-house filmgoers; and filmgoers that prefer multiplexes
showing films in original version. Moreover, the method allows for considering additional
restriction to segment the sample and we do so here by introducing an economic constraint
(see also Swanson, Davis and Zhao, 2008). The key findings are intended to add to the
literature on experiential goods consumption, such as films, for a better understanding of film
audiences’ behavior.
The article unfolds as follows: section 2 provides a brief review of the literature on
filmgoers’ consumption behavior, section 3 is dedicated to the methodology, section 4
presents the main results, and in section 5 we conclude.
2. A Consumer Behavior Perspective of Film Theatres’ Audiences
The motives and barriers approach
While a significant bulk of the research on film audiences has focused on the type of films
(i.e., ‘mainstream’ vs. ‘art house’ films), a particular stream of research has paid special
attention to the motives underlying individuals’ decisions to participate/not participate in
leisure and cultural activities in general, film theatres attendance being one of them. In this
line, Chuu, Chang and Zaichkowsky (2009) offer, for example, a comprehensive overview of
the motives driving audiences for art versus commercial films. Several features of film
consumption, consistent across various studies, were identified by these authors: 1) there is a
higher frequency of attendance for art film audiences compared to those of commercial films
(see also Faber, O’Guinn and Hardy, 1988); 2) socialization and/or entertainment are not the
main motives for art filmgoers but the films per se (see also Vahemetsa, 1970; Austin, 1984);
3) art film audiences seem to be quite ‘self-determined’ (Chuu, Chang and Zaichkowsky,
2009:216), as their attendance behavior is less dependent on others’ opinions or company (see
also Faber, O’Guinn and Hardy, 1988); 4), commercial film audiences are more likely to
watch films about which they know more, such as popular movies that are extensively
advertised, while the audiences of art films go to film theatre to watch a film just because it is
an art film (see also Chamberlin, 1960); 5) art film audiences appreciate the ‘cultural value’
transmitted by films (see also Vahemetsa, 1970,); 6), regarding venues’ features, art film
audiences were found to be less demanding than commercial film audiences with respect to
location or facilities (i.e., sound, seats, parking, etc.), given their main interest in the quality
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
of the film (Adler, 1959; Austin, 1984); last, art filmgoers who prefer more drama and
original version films, are also motivated to watch a director’s work, and put more weight on
reviewers’ critiques than on advertising (see also Adler, 1984; Austin, 1984; Faber, O’Guinn
and Hardy, 1988).
As for barriers, research on film theatres’ attendance indicates the preference for other
ways (digital TV, internet downloads, DVD discounts, etc.) of enjoying films (De Vany and
Walls, 2007; Silver and McDonnell, 2007); the preference for film substitutes (videogames)
or other viewing environment (mall entertainment) (Silver and McDonnell, 2007), or the
existence of domestic constraints, such as lack of time, family obligations, etc. (Collins, Hand
and Ryder, 2005). Hart, Kerrigan and Vom Lehn (2016), have analyzed film consumer’s
hedonic experience with an introspective (diary) research method and identified three
dimensions of film consumer behavior that support existing evidence: film characteristics
(artistic, commercial), viewing environment (home or film theatre) and situational
environment (time, mood, companions, etc.). Their findings also confirm the importance of
previous film experience on filmgoers “future sense making and film consumption
experiences” (pp. 388).
Types of film theatre audiences
This stream of research distinguishes among various segments of filmgoers. Vahemetsa
(1970) found, for example, four segments of art filmgoers: the ‘cultural prestige type’, for
whom film is a specific cultural expression; the ‘first cognitive type’ (films are important
information sources contributing to increasing life experiences); the ‘second cognitive type’
whose motive for art film attendance is to escape from reality; and the ‘aesthetic type’ who
perceives art films as ‘creative products'. Sedgwick and Pokorny (2012:329-330) worked with
a historic micro-dataset (early 1940s) on 22 film theatres in Philadelphia, in the US, and
identified three types of filmgoers: a non-selective segment, consuming films more as a
generic commodity, but who go to the film theatre for various motives –film-viewing habits,
the experience of viewing films in a movie theater rather that the film per se, or
accompanying friends who want to see a specific film; the selective film consumers, whose
preference for recommended films is directly related to their social status; and a small
segment of filmgoers who based their choice on personal recommendations.
In Spain, Cuadrado and Frasquet (1999) found three segments of filmgoers –social
viewers, apathetic viewers and filmgoers– and, the benefits of going to film theatre to watch
films, ranged from having a good time and feeling emotions to having fun. García-Álvarez,
Filimon and López-Sintas (2007) focused on filmgoers’ choices of films by country of origin
–US, Spain and other countries– and identified three typologies of film theatres audiences: a
majority of filmgoers with a clear preference for US films, especially families and younger
individuals; an audience for Spanish films, integrated mainly by middle-age and middle-class
filmgoers; and a social and intellectual elite that preferred European films. The dominant
preference for US films was explained by the fact that these were perceived by audiences as
synonymous of familiar and reliable entertainment, and in Spanish language; in contrast,
these qualities were not all met by Spanish and European film productions. In a research
closer to the approach presented here, Fernández-Blanco, Orea and Prieto-Rodríguez (2009)
worked with a 1998 Spanish dataset containing self-rated valuations of both US and Spanish
films. The authors applied LC models and identified two groups of filmgoers, differentiated
by variables such as income, ticket price, education, and age, among others. Overall, their
results advocate in favor of introducing socioeconomic indicators and self-rated preferences
in the analysis, for a more complete identification of the hidden consumer preferences.
Socioeconomic context and film audiences
From a temporal perspective, the research evidence seems to indicate that there has been a
change in the social context surrounding the experience of going to the film theatre to watch a
movie (Tudor, 2013), with a significant change in the age profile of the filmgoers over the last
twenty years (1993-2012): while young people (11-14 and 15-19 age segments, in particular,
and to a lesser extent, the 20-24 year-old segment) go to the film theatre less frequently, those
above 50 years of age have increased their consumption of films (Paris, 2014:12-13). Other
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
demographics, such as gender, point to a rather even trend of film theatre attendance for men
and women (Sedgwick and Pokorny, 2012; Tudor, 2013); and, the inverse correlation
between film theatre attendance and filmgoers’ social status, education level and income, has
also weakened over time (Tudor, 2013).
In UK, Chan and Goldthorpe (2007) analyzed visual arts and found evidence for gender
effects only in the cases of theater, cinema and dance, with women exhibiting a more
omnivorous pattern of consumption than men. Chuu, Chang and Zaichkowsky (2009) found
that art filmgoers, compared to commercial filmgoers, were, on average, more educated (high
cultural capital), more likely to be single, and men, although gender differences have proven
to be an inconsistent feature across the research on film audiences and elitist cultural
consumption. From a cross-country perspective, Governo and Teixeira (2014) explored the
determinants of the consumer demand for art house films vs. mainstream film theatres
offerings and found no significant relationship between social and cultural status indicators,
(the relative demand for art films was usually associated with higher income and education
levels of film consumers). While these findings are in line with the cultural omnivorousness
setting (Peterson and Kern, 1996), other authors (see Katz-Gerro, 1999; López-Sintas and
García-Álvarez, 2006, etc.) did find evidence supporting cultural stratification thesis
(Bourdieu [36]).
Finally but not lastly, according to Morley (1992:157-158), cited in Meers (2001:140),
“picture palaces” and “domestic context” are considered to be different film watching
experiences inviting to take into account the ‘context’ (place) of film consumption (see also
Hart, Kerrigan and Vom Lehn, 2016). Richins (1997) also argues that in the case of goods
embedding meanings (i.e., experiential goods like films), context is particularly important for
emotions. The place was found important also in the case of music consumption: Roose and
Vander Stichele (2010), differentiated among music listening and attending behavior in
Flanders, through activities or practices that embed a socially visible status marker (i.e., a
concert hall) and others that do not (i.e., home), among others.
Overall, existing research indicates that it is necessary to look into the relationship
between the film theatre audiences and the role of motives/barriers, and socioeconomic
variables to gain a more complete understanding of filmgoers’ decision-making behavior.
3. Methodology
Data and variables
The data were gathered with a personal survey based on a structured questionnaire using a
quota sampling method that yielded information from 516 individuals of both genders, aged
18 or older in 2013. The questionnaire was divided in three parts: filmgoers’ consumption
habits, motives/barriers and sociodemographic variables. A comprehensive number of self-
rated items registered the motives for (17 items) and the barriers to film theatre attendance (28
items) (see Table 1).
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
Socioeconomic and behavioral characteristics of cinemagoers
The survey comprised information on several representative indicators of individuals’
positions in the social hierarchy: socioeconomic status and education level, all informative
with respect to the respondents’ social, economic and cultural capital endowments. The
occupational status, at the time of the survey, was used as a proxy for the socioeconomic
status, as the survey did not elicit information on income level. The categories were as
follows: 1) employed, 2) unemployed, 3) retired, 4) student, 5) house work, and 6) others.
Educational attainment, recorded with four levels, was used to operationalize the cultural
capital: 1) primary school, 2) secondary school, 3) technical college, 4) university studies. A
special five-point Likert scale question was meant to measure whether the interviewees had
enough disposable income to enjoy leisure pursuits. For calculation purposes, the five levels
of responses were reduced to three: disagree, neutral, agree. The profile of the respondents
was completed with sociodemographic indicators (see Table 2).
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
The frequency of cinema attendance was organized in four levels: 1) once a year or less, 2)
once or twice every three months, 3) twice a month, 4) once a week or more. A five-level
Likert type question –strongly disagree, disagree, neutral, agree, and strongly agree– asked
the respondents to assess whether they had enough time to enjoy leisure activities (see Table
Motives for cinema attendance
A set of 17 items (see Table 1) were dedicated to understanding the motives for cinema
attendance. In order to determine the adequate number of components to retain in the
analysis, we applied a principal components analysis (PCA). Basic assumptions on the
suitability of the data for this type of analysis were checked with the Kaiser-Meyer-Olking
(KMO) statistic, which should be greater than 0.600 (KMO=0.744) and the Bartlett’s test,
which was significant (Chi-squared=1382.486; df=136; p-value=0.000). The PCA procedure
recommended a five-factor solution explaining 54.07% of the total variance (TVE). Table 3
shows the final five-factor solution (Varimax Procedure), consisting of 17 items selected
based on eigenvalues, TVE, loadings and interpretability. Except for the ‘mood’ component,
integrated by only two items –to relieve boredom and to look for relaxation– which returned
high communalities (>0.500), the other components were determined by at least three items
(see also Hager and Winkler, 2012, for similar reports of factors with only two items).
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
For each factor extracted, the items were summated, and in order to reduce the sparseness
of the data, the five-level Likert scale was collapsed into two levels, taking as reference the
mean of each subscale (1-disagree/below the average; 2-agree/above the average).
Barriers to cinema attendance
In a similar fashion, the barriers were registered on a five-point Likert scale by a set of 28
items (see Table 1). The PCA procedure converged to a final seven-factor solution (see Table
5 below), explaining 58.2% of the total variance and retaining 26 items out of 28, based on
eigenvalues, the items’ loadings, TVE and interpretability. The tests of the initial hypotheses
(KMO=0.859 and Chi-squared=3091.634; d.f.=190; p-value=0.000) confirmed that a PCA
analysis could return significant factor structures. After a first PCA iteration, indicating a
suitable seven-factor solution (TVE=56.52%), two items (nobody to go with and multiplexes
are far away from home) with very low loadings were eliminated, and the final solution is
presented hereafter. A similar procedure, as the one in the previous section, was applied to
rescale the items and to reduce the sparseness of the data.
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
Type of film theaters and audiences
Film theatre attendance was recorded for four types of venues: a) multiplexes in shopping
centers, b) multiplexes in city centers, c) film theaters, and d) multiplexes showing films in
original version. Each respondent was asked to rank the four types from 1 to 4, according to
the frequency of attendance. A paired samples t-test was conducted to evaluate whether the
respondents scored differently on these variables. The results indicated that there was not a
significant difference in the scores for multiplexes in shopping centers (M=1.79; SD=0.932)
and multiplexes in city centers (M=1.75; SD=0.728) at the 0.05 level of significance (t=0.564;
df=447; p=0.573). In this case, we would not reject the null hypothesis that the scores for
these two variables are the same, and a new variable –multiplexes in shopping and city
centers– was calculated using the mean scores of these two variables. Finally, in order to
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
control for the sparseness of the data, the frequency ranks were re-coded in two levels: most
preferred and less preferred film theatres, respectively (see Table 5).
Data analysis and parameters of the model
We are interested here in exploring the unobserved data heterogeneity associated to
filmgoers’ patterns of attendance emerging from the type of film theaters usually chosen to
watch a film. These patterns are next analyzed in association with filmgoers’ motives for and
barrier of attendance, and other variables such as, socioeconomic indicators, behavior
(frequency of film theatre attendance), generational patterns (age), film preferences, etc., with
the purpose of showing that they respond to socioeconomic and generational structures and
this applies also to filmgoers’ motives/barriers of attendance (for a previous version of this
research, without PCA analysis for motives and barriers see Cuadrado et al., 2013). To do
this, we employed the latent class (LC) method (Lazarsfeld and Henry, 1968) to test the
consumer behavior model according to which, individuals consume patterns of products
(Boter and Wedel,1999). Intuitively, the LC method splits the sample in T clusters or latent
classes and estimates, for each one, a set of parameters. In the LC model defined in the
equation below (see Vermunt, 2010), Y stands for the whole set of indicators Y1, Y2, and Y3,
corresponding to the three types of film theatres defined in Table 5, and Z, for the set of so-
called active covariates (e.g., financial constraint) as they will condition the final number of
The model estimates, for each cluster, its size and the probabilities of the indicators (Y),
conditioned to cluster membership. Once obtained the clusters and the parameters, each
individual (observation) is assigned to a cluster only (Magidson and Vermunt, 2001) based on
the calculated membership probabilities (see also Vermunt and Magidson, 2008).
Model selection
The parameters of the LC model are estimated by first fitting the null, or restricted, model
(T=1), with one latent class and the unrestricted model (T=2) with two latent classes. A
likelihood ratio test (L2) is performed, such that if the null model is rejected, the process
continues by incrementing, each time, the number of latent classes by one. This process aims
at finding the model that provides an adequate fit for the data and stops once we fail to reject
a null model. Table 6 gives the statistics used to assess the goodness of fit of the LC model
used in the analysis: the chi-squared likelihood-ratio statistic L2; the Bayesian information
criterion (BIC) and the Consistent Akaike’s Information Criterion (CAIC) are both based on
L2 (see also Raftery, 1986; Fraley and Raftery, 1998). The L2 statistics indicate, for each
model, the amount of unexplained association among the variables (hence, lower values are
preferred). The data in Table 6 show that the addition of a latent class (from the model with
one cluster to the model with two clusters) contributed to reduce the L2 statistic by 75.6%,
while the contribution of the additional latent classes (the models with three to four classes)
was much lower (22.5%; 1.2%). The p-value (>0.05), indicates that the model with three
clusters provides an adequate fit for the data. Additional statistics (Dayton,1998), such as BIC
(which takes parsimony into account) and CAIC, also suggest that the model with three
clusters would be a better fit for the data (the lower their values, the better).
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
4. Main Findings and Discussion
The parameters of the model
Table 7A presents the estimates of the parameters for the three-cluster model. The first
row presents the proportion of how many individuals have been classified in each
cluster, that is, the cluster’s relative size, and the following rows indicate the
probability of filmgoers’ behavior (preference for a type of film theatre), given their
classification in that cluster in percentages. If take, for example, a filmgoer classified
in cluster 1, he or she has a probability of 99.97% of going very frequently to watch films in
multiplexes in shopping and city centers, a probability of 99.96% of not going to multiplexes
showing films in original version (OV) and a 32.78% probability of facing financial
constraints on cultural and leisure activities. Thus, based on these conditional probabilities
and mean values, we can characterize the probabilistic behavior of the Spanish filmgoers
regarding the type of film theater preferred, constrained by the disposable income for cultural
The parameters estimated suggest one big cluster (70.53% of the sample) of filmgoers that
basically prefer multiplexes in shopping centers and in the center of the city (mainstream
filmgoers). A second cluster (15.87% of the sample) stands for art-house filmgoers; according
to Chuu, Chang and Zaichkowsky (2009:214), “[a]rt films are frequently selected to be shown
at major film festivals and are often the winners of distinctive film awards. Many of these
films are made in non-English speaking countries… In terms of the theatrical release of art
films, they are screened primarily in art-house and repertoire theatres only”. The third cluster
(13.6%) represents filmgoers that prefer to watch films in original version (OV filmgoers).
An alternative, and easier, interpretation of filmgoers’ profiles is based on whether the
individuals classified in cluster t, , are over- or underrepresented among individuals
with that behavior (similar to a row profile table, see bold values in Table 7B). Thus,
filmgoers in cluster 1 (70.53%) are overrepresented among those choosing the very frequent
level (Yes) of attendance for multiplexes in shopping and city centers (mainstream filmgoers),
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
and for whom film theatres mean leisure and entertainment, and the less frequent level (No)
of attendance for venues showing films in original version (OV filmgoers). Cluster 2 (15.87%)
is overrepresented among individuals who prefer film theaters, an elitist segment of filmgoers
for whom watching films at the movie theatre is an experience (art-house filmgoers); cluster 3
(13.6%) is overrepresented among filmgoers who have a clear preference for watching films
in multiplexes showing them in original version (OV filmgoers).
As already mentioned, the LC model estimated here also takes into consideration the
income available for cultural pursuits (see i.e., Fernández-Blanco and Baños-Pino, 1997;
Fernández-Blanco, Orea and Prieto-Rodríguez, 2009). The estimates show that while
filmgoers in cluster 1 (mainstream filmgoers), who prefer multiplexes in shopping and city
centers, are subject to financial constraints, those in the other two clusters (art-house and OV
filmgoers) ‘agree’ with having enough income for cultural pursuits, as they are
overrepresented in these indicators.
Motives for going to film theatres
According to the results presented in Table 8, for cluster 1 (mainstream filmgoers), films’
popularity (i.e., prizes, intensive advertising, favorite actors and successful box-office
records) and mood (to relieve boredom and to relax) are the main drivers of cinema
attendance. They do not seem to be influenced by films’ quality (good reviews, recommended
by others, or better image and quality of the exhibition) or by educational purposes or
socialization needs. All in all, on average, they seem to fit well in the profile of filmgoers who
prefer commercial and popular films for entertainment. Cluster 2 (art-house filmgoers)
exhibits a different motivational pattern: as expected, they are overrepresented in indicators
related to motives of personal education achievements and social interaction (share emotions,
experiences and socialize with filmgoers sharing common preferences for art films) (see
Swanson, Davis and Zhao, 2008; Hager and Winkler, 2012); neither films’ quality nor
popularity seem to be strong motives for choosing film theatres, as they seem to be a well-
informed and rather specialized film audience and go mainly for the movies alone. Cluster 3
(OV filmgoers) exhibits a pattern similar to cluster 2, except for the fact that these filmgoers
do take into account films’ quality (reviews, recommendations, better sound and image). It
appears that film theatres allow them to benefit more from the impact of technical innovations
on film releases. This result is consistent with the findings of Governo and Teixeira (2014)
predicting, on average, a positive relationship between the countries’ level of technological
development and the domestic demand for art films.
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
Barriers for attending film theatres
The results in Table 9 indicate that clusters 1 and 2 (mainstream and art-house filmgoers)
are, on average, the most affected by the barriers analyzed. Thus, the mainstream segment is
sensitive to factors such as: film offerings (too many films to choose, difficult to get tickets,
etc.), perception about the films released (very similar or bad films), their preference for other
cultural activities and places; for this audience, the distinction between home and film theatre
‘contexts’ is important (see Morley, 1992; Sedgwick and Pokorny, 2012); while financial
resources act as a constraint (see Fernández-Blanco, Orea and Prieto-Rodríguez, 2009), they
face no leisure time restrictions. Overall, these barriers seem to fit well with this profile of
filmgoers, looking mainly for entertainment, and that exhibits features similar to the
commercial films’ audiences (see also Chuu, Chang and Zaichkowsky, 2009). Cluster 2 (art-
house filmgoers) shares some of the barriers of cluster 1 (film offerings, perception,
preference and place), but in contrast, they are not affected by financial restrictions. They do
not face leisure time restrictions either, and as expected, are sensitive to film theatres’ features
(i.e., smelly food, mobile phone noise, nostalgia for former type of film theatres, etc.). In this
respect, Tudor (2013) argues, for example, that an important part of social life in the past
century was closely related to ‘going to the movies’ ritual, and the change in this social
context surrounding the experience of going to the film theatre eventually affected attendance.
Finally, cluster 3 (OV filmgoers) is affected by two barriers only, that is, leisure time
restrictions and the preference for other substitute activities (social networking, videogames).
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
Socioeconomic and behavioral profile of cinemagoers in clusters
The estimates for the socioeconomic context presented in Table 10 indicate that
individuals in cluster 1 (mainstream filmgoers) are, on average, more likely to be men below
50 years of age, some unemployed and with a low education level (primary and secondary
school), and single with or without children; they are not affected by leisure time restrictions,
as they are either unemployed or students (most of this group being foreigners residents in
Spain at the moment of the interview); they are also very frequent filmgoers. Cluster 2 (art-
house filmgoers.) is overrepresented among women, who are more likely to be unemployed or
retired, but with a higher cultural level (technical and university studies) than the mainstream
filmgoers, and some dedicated to housework, with a partner and children; in contrast to
cluster 1, they are above fifty years of age and mainly Spanish. They do not go very often to
the film theatre, also due to leisure time constraints. Cluster 3 (OV filmgoers) exhibits a
different socioeconomic profile: mainly foreign women, above 50 years of age, highly
educated (university studies), employed or with another occupational status, and with partner
and no children. They are frequent filmgoers and have enough time for leisure pursuits (see
also Fernández-Blanco et al., 2009, for the significant impact of socioeconomic indicators on
film theatres’ audiences). These findings are also in line with the cultural stratification
framework (Bourdieu, 1979) confirming previous research evidence on cultural consumption
(see López-Sintas and García-Álvarez, 2006).
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
5. Conclusions
This research explored the behavior of Spanish filmgoers in order to disentangle their
tastes and habits, with a special interest in the motives and barriers of going to film theatres.
The empirical results suggest three clusters of filmgoers with the following probabilistic
patterns of behavior: a segment of mainstream filmgoers, for whom going to film theatres to
watch a film means leisure and entertainment; a segment of art-house filmgoers, and a
segment that enjoys watching films in their original version (OV filmgoers). The
segmentation of the data set in three clusters was conditioned by filmgoers’ financial
constraints showing that they may affect filmgoers’ behavior.
The PCA findings identify five groups of motives –educational purposes, film popularity,
film quality, socialization and mood– and seven groups of barriers: film offerings, venue
features, perception, preference and place, substitute activities, financial barriers and
recreation time disposability. The socioeconomic and behavioral (frequency) indicators
complete the profile of the filmgoers. These findings add to the existing research evidence in
favor of using self-rated preferences and socioeconomic variables to determine (film)
consumer groups. This paper illustrates the different filmgoers segments, and in doing so,
illustrates the specific barriers and motivations for each of these groups. These insights can
assist cultural policy makers and film theatres’ managers in designing specific actions and
market strategies to better meet the needs and preferences of each of these distinct segments.
Finally but not lastly, these findings invite to expanding filmgoers’ analysis to other national
contexts, to allow for further cross-country comparisons.
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
Authors gratefully acknowledge the contribution of Finola Kerrigan and Andrea Rurale on
a previous version of this research.
Adler, K.P. (1959), “Art Films and Eggheads”, Studies in Public Communication, Vol. 2 (Summer),
Austin, B.A. (1983), The Film Audience: An International Bibliography of Research with Annotations
and an Essay, Scarecrow, Metuchen, NJ.
Austin, B.A. (1984), “Portrait of an Art Film Audience”, Journal of Communication, Vol. 34 No. 1, pp.
Austin, B.A. (1986), “Motivations for Movie Attendance”, Communication Quarterly, Vol. 34 No. 2,
pp. 115-126.
Boter, J. and Wedel, M. (1999), “ Segmentation of Hedonic Consumption: An Application of Latent
Class Analysis to Consumer Transaction Databases”, Journal of Market-Focused Management,
Vol. 3 No. (3/4), pp. 295–311.
Bourdieu, P. (1979). La Distinction. Critique Social du Jugement, Les éditions de minuit, Paris.
Chamberlin, P. (1960), “Allies, Not Enemies: Commercial Nontheatrical Experience on the West
Coast”, Film Quarterly, Vol. 14, No. 2, pp. 36-39.
Chan, T.W. and Goldthorpe, J.H. (2007), “Social Stratification and Cultural Consumption: The Visual
Arts in England”, Poetics, Vol. 35, pp. 168-190.
Chuu, S.L.H., Chang, J.C. and Zaichkowsky, J.L. (2009), “Exploring Art Film Audiences: A Marketing
Analysis”, Journal of Promotion Management, Vol. 15 No. 1-2, pp. 212–228.
Collins, A., Hand, C. and Ryder, A. (2005), “The Lure of the Multiplex? The Interplay of Time,
Distance, and Cinema Attendance”, Environment and Planning, Vol. 37 No. 3, pp. 483-501.
Cuadrado, M. and Frasquet, M. (1999), “Segmentation of Cinema Audiences: An Exploratory Study
Applied to Young Consumers”, Journal of Cultural Economics, Vol. 23 No. 4, pp. 257-267.
Cuadrado, M., Filimon, N., Kerrigan, F. and Rurale, A. (2013), “Exploring Cinema Attendance
Facilitators and Contraints. A Marketing Research Approach”, Paper presented at the 5th
Workshop on Cultural Economics and Management, 24-25 October 2013, Cadiz, Spain.
Daenekindt, S. and Roose, H. (2014), “Ways of Preferring: Distinction Through the ‘What and the
How’ of Cultural Consumption”, Journal of Consumer Culture, Vol. 17 No. 1, pp. 25-45.
Dayton, M.C. (1998), Latent Class Scaling Analysis, Sage University Paper Series on QASS, 07-126,
Thousand Oaks, Sage Publishers, CA.
De Bruyn, E.H. and Cillessen, A.H.N. (2008), “Leisure Activity Preferences and Perceived Popularity
in Early Adolescence”, Journal of Leisure Research, Vol. 40 No. 3, pp. 442-457.
De Vany, A.S. and Walls, W.D. (2007), “Estimating the Effects of Movie Piracy on Box-office
Revenue”, Review of Industrial Organisation, Vol. 30, pp. 291-301.
Dyer, R. (1981), “Entertainment and Utopia”, in Altman, R. (Ed.), Genre: The Musical, Routledge and
Kegan Paul, London, pp. 175-189.
Faber, R.J., O’Guinn, T.C. and Hardy, A.P. (1988), “Art Films in the Suburbs: A Comparison of
Popular and Art Film Audiences”, in Austin, B.A. (Ed.), Current Research in Film: Audiences,
Economics and Law, Vol. 4, Ablex, Norwood, NJ, pp. 45-53.
Fernández-Blanco, V. and Baños-Pino, J.F. (1997), “Cinema Demand in Spain: A Cointegration
Analysis”, Journal of Cultural Economics, Vol. 21 No. 1, pp. 57-75.
Fernández-Blanco, V., Orea, L. and Prieto-Rodriguez, J. (2009), “Analysing Consumers’
Heterogeneity and Self-reported Tastes: An Approach Consistent with the Consumer’s Decision
Making Process”, Journal of Economic Psychology, Vol. 30 No. 4, pp. 622–633.
Fraley, C. and Raftery, A. (1998), “How Many Clusters? Which Clustering Methods? Answers via
Model-based Cluster Analysis”, Computer Journal, Vol. 41, pp. 578-588.
García-Álvarez, M.E., Filimon, N. and López-Sintas, J. (2007), “Reliable Entertainment: Spanish
Consumers’ Preferences Regarding a Film’s Country of Origin”, International Journal of Non Profit
and Voluntary Sector Marketing, Vol. 12 No. 3, pp. 217-229.
Governo, F. and Castro Teixeira, A.A. (2014), “Marketing and Technology Sophistication as Key
Determinants of the Demand for ‘Art House’ Cinema Films: A Cross Country Analysis”, Tékhne,
Vol. 12 No. 1, pp. 15-25.
Gripsrud, J. and Lavik, E. (2008), “Film audiences”. In J. Donald and M. Renov (Eds.), The SAGE
Handbook of Film Studies (pp. 454-470). London: SAGE Publications Ltd doi:
Cuadrado-García M., Filimon N., Montoro-Pons F.J., Regional Science Inquiry, Vol. X, (2), 2018, pp. 45-60
Hager, M.A. and Winkler, M.K. (2012), “Motivational and Demographic Factors for Performing Arts
Attendance across Place and Form”, Nonprofit and Voluntary Sector Quarterly, Vol. 41 No. 3, pp.
Harry, H. (1986), “Motivation Research", Marketing Intelligence & Planning, Vol. 4 No. 5, pp. i-234.
Hart, A., Kerrigan, F. and Vom Lehn, D. (2016), “Experiencing Film: Subjective Personal
Introspection and Popular Film Consumption”, International Journal of Research in Marketing, Vol.
33 No. 2, pp. 375-391.
Hawkins, B.A., Peng, J Hsieh, C.M. and Eklund, S.J. (1999), “Leisure Constraints: A Replication and
Extension of Construct Development”, Leisure Sciences: An Interdisciplinary Journal, Vol. 21 No.
3, pp. 179-192.
Instituto Nacional de Estadistica/National Institute of Statistics (INE). (2011), Survey on Population
Demographics, available at (accessed 10 January 2011).
IsoAhola, E. (1980), Social Psychological Perspectives on Leisure and Recreation. W. C. Brown
Company Publishers.
Jackson, E.L. (1997), “In the Eye of the Beholder: A Comment on Samdahl and Jekubovich (1997), ‘A
Critique of Leisure Constraints: Comparative Analyses and Understandings’”, Journal of Leisure
Research, Vol. 29 No. 4, pp. 458-468.
Katz-Gerro, T. (1999), “Cultural and social stratification: Leisure activities, musical tastes, and social
location”, Social Perspectives, Vol. 42 No. 4, pp. 627-646.
Lazarsfeld, P.F. and Henry, N.W. (1968), Latent Structure Analysis, Houghton Mifflin Company,
López-Sintas, J. and García-Álvarez, M.E. (2006), “Patterns of Audio-Visual Consumption: The
Reflection of Objective Divisions in Class Structure”, European Sociological Review, Vol. 22 No.
4, pp. 397-411.
Magidson, J. and Vermunt, J.K. (2001), “Latent Class Factor and Cluster Models, Bi-plots, and Related
Graphical Displays”, Sociological Methodology, Vol. 31, pp. 223–264.
Meers, P. (2001), “Is there an Audience in the House? New Perspectives on (European) Film
Audiences”, Journal of Popular Film and Television, Vol. 29 No. 3, pp. 138-144.
Morley, D. (1992), Television, Audiences and Cultural Studies, Routledge, London. Murray, E.J.
(1964), Motivation and Emotion, Prentice-Hall, Englewood Cliffs, NJ.
Paris, T. (2014), New Approaches for Greater Diversity of Cinema in Europe? Analyses of
Experiments Launched in the Context of the Preparatory Action1 "Circulation of European Films in
the Digital Era" (European Parliament & European Commission)”, available at (accessed 12 July 2015).
Peterson, R.A. and Kern, R.M. (1996), “Changing Highbrow Taste: From Snob to Omnivore”,
American Sociological Review, Vol. 61, pp. 900-907.
Raftery, A.E. (1986), “Choosing Models for Cross-Classifications”, American Sociological Review,
Vol. 51, pp. 145-146.
Raymore, L.A. (2002), “Facilitators to Leisure”, Journal of Leisure Research, Vol. 34, pp. 37-51.
Richins, M.L. (1997), “Measuring Emotion in the Consumption Experience”, Journal of Consumer
Research, Vol. 24 No. 2, pp. 127-146.
Roose, H. and Vander Stichele, A. (2010), “Living Room vs. Concert Hall: Patterns of Music
Consumption in Flanders”, Social Forces, Vol. 89 No. 1, pp. 185-207.
Sedgwick, J. and Pokorny, M. (2012), “Film Consumer Decision-Making: The Philadelphia Story,
1935-36”, Journal of Consumer Culture, Vol. 12 No. 3, pp. 323-346.
Silver, J. and McDonnell, J. (2007), “Are Movie Theatres Doomed? Do Exhibitors See the Big Picture
as Theatres Lose their Competitive Advantage?”, Business Horizons, Vol. 50 No. 6, pp. 491-501.
Swanson, S.R., Davis, C.J. and Zhao, Y. (2008), “Art for Art's Sake? An Examination of Motives for
Arts Performance Attendance”, Nonprofit and Voluntary Sector Quarterly, Vol. 37, pp. 300-323.
Tudor, A. (2013), Image and Influence: Studies in the Sociology of Film, Routledge Library Editions:
Cinema, London.
Vahemetsa, A. (1970), “A Typology of the Recipients of Artistic Film”, Society and Leisure, Vol. 2,
pp. 87-101.
Vermunt, J.K. (2010), “Latent Class Modelling with Covariates: Two Improved Three-Step
Approaches”, Political Analysis, Vol. 18, pp. 450-469.
Vermunt Jeroen K., and Jay Magidson. (2008), LG-Syntax User’s Guide: Manual for Latent GOLD
4.5 Syntax Module. Belmont, MA: Statistical Innovations Inc.
Wierenga, B. (2006), “Motion Pictures: Consumers, Channels, and Intuition. Invited Commentary”,
Marketing Science, Vol. 25 No. 6, pp. 674-677.
... In this research we apply Latent Class Analysis (LCA) to identify unobserved profiles of Spanish book readers, also called latent classes or "hidden groups" (Lazarsfeld and Henry, 1968). LCA 2 is an exploratory approach which can also be applied to categorical latent variables (see i.e., Daenekindt and Roose, 2014;Cuadrado-García, Filimon and Montoro-Pons, 2018;López-Sintas et al., 2014;etc.). ...
Full-text available
El trabajo identifica tres perfiles de lectores de libros con metodología de clases latente: activos (leen casi a diario); frecuentes (solo leen 1-2 veces por semana); ocasionales (leen libros alguna vez al mes o trimestre), aportando evidencia sobre los hábitos de lectura de libros de los españoles. Indicadores socioeconómicos y culturales completan el perfil de los lectores. Una mejor comprensión de los hábitos de lectura de libros, de las preferencias literarias, así como de su contribución al bienestar de los lectores, podrían ser de interés para los responsables de políticas culturales y para las editoriales
Conference Paper
Full-text available
The paper is focused on the issue of tourism destination competitiveness. It tracks the application of two most-cited models of destination competitiveness in research practice defined by the various research papers published in journals involved in ProQuest Central and EBSCO databases from 2010. The aim of the paper is to identify the principal factors of destination competitiveness and create an aggregate model in a form of a causal loops diagram (CLD). Moreover, the paper pays attention to a role of destination stakeholders in the competitiveness models. Thus, it enriches the aggregate model with the stakeholder collaboration factor. It examines connections between the factor of stakeholder collaboration and other related competitiveness factors and identifies the leverage points affecting the dynamics of destination competitiveness. In such a way the paper provides a recipe for destination managers enabling them to define tourism policy so to be able to utilize cooperation with stakeholders as a source of competitive advantage, and thus to systematically increase destination competitiveness and improve the competitive position of the destination.
Technical Report
Full-text available
Digital technology, the new forms of distribution and consumption of works, piracy, increasing saturation of screens, the arrival of new economic players - these are some of the elements that are transforming the cinema landscape. The European Parliament has launched a programme of experiments to transform these threats into opportunities, with a view to improving the circulation of films in Europe. These experiments, conducted by several players in the European film industry, gave rise to several film releases using innovative sequences of distribution windows, and with a multi-territorial approach. While the number of spontaneous experiments has increased over recent years, particularly in the United States and the United Kingdom, and concerning simultaneous theatrical and VOD releases (Day-and-Date), VOD previews (ultra-VOD) or exclusive VOD releases (Direct-to-VOD), the experiments launched by the European Parliament also strove to include a multi-territorial dimension so as to encourage the simultaneous release of films in several European countries.
At first sight, the audience might seem an extremely neglected aspect of film research. Many film scholars stress the scarcity of audience studies (e.g., Austin, Barker, Stacey). Others, however, point to the fact that there has always been a certain interest in the film audience, albeit in divergent traditions. In quantitative terms, scholarly research and writing about some dimension of audiences clearly outnumber publications about any other aspect of the film medium, such as film production or the aesthetics of film (Gripsrud 202).
How and why audiences consume films is a much-researched yet inconclusive area of film marketing. Film is an experiential product and qualitative research methods are a suitable way of gaining insight into how people choose between different film offerings and how they assess their film viewing experience. Before we can understand others' choices and experiences, we first must understand ourselves. We therefore begin our investigation by taking a snapshot of our experiences facilitated by Subjective Personal Introspection (SPI) to gain insight into how the lead author makes sense of his film consumption. The key findings complement and advance current debates in film and experiential consumption. Indeed, the theoretical contribution is two fold; the development of a film consumption experience model based on three-interrelated classification dimensions (film characteristics, viewing environment, situational environment), which collectively impacts the lead author's consumption behavior, and our expansion of Schmitt's (1999) SEMs model.
The purpose of this article is to address some of the issues raised by Samdahl and Jekubovich (1997) in their critique of leisure constraints research. I suggest that the authors have based their criticism on a rather narrow and outdated conception of leisure constraints, that the concept of constraints is not an artificial construct, and that the new data and interpretations put forward by Samdahl and Jekubovich are better viewed as potentially important refinements to thinking about constraints to leisure rather than as grounds for abandoning constraints research.
Previous research seeking to explain why people attend live performing arts events has focused primarily on demographic explanations, such as the age or income of patrons. Recent research reinvigorates an old debate regarding the veracity of psychological motivations for attendance, such as an appreciation for beauty or the opportunity to socialize with others. We add to this line of inquiry by investigating how demographic and motivation explanations fare across three cities, and for three different art forms. We find that demographic variables fare better overall than psychological motivations across place and form. Despite evidence of strong consistency of the magnitude of variable coefficients across place and form, we conclude with warnings that inconsistencies of significant effects between place and form should cause us to question whether results in one or several sites can provide stable and useful generalizations for managers and arts marketers.