This article was downloaded by: [University of Waikato]

On: 29 May 2012, At: 15:10

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

The Communication Review

Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/gcrv20

Understanding Variation in Audience

Engagement and Response: An
Application of the Composite Model to
Receptions of Avatar (2009)
a b c
Carolyn Michelle , Charles H. Davis & Florin Vladica
a
School of Social Sciences, Faculty of Arts and Social Sciences, The
University of Waikato, Hamilton, New Zealand
b
School of Radio and Television Arts, Faculty of Communication and
Design, and Entrepreneurship and Strategy Department, Ted Rogers
School of Management, Ryerson University, Toronto, Canada
c
School of Radio and Television Arts, Faculty of Communication and
Design, Ryerson University, Toronto, Canada

Available online: 24 May 2012

To cite this article: Carolyn Michelle, Charles H. Davis & Florin Vladica (2012): Understanding
Variation in Audience Engagement and Response: An Application of the Composite Model to
Receptions of Avatar (2009), The Communication Review, 15:2, 106-143

To link to this article: http://dx.doi.org/10.1080/10714421.2012.674467

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.
 The Communication Review, 15:106–143, 2012
Copyright © Taylor & Francis Group, LLC
ISSN: 1071-4421 print/1547-7487 online
DOI: 10.1080/10714421.2012.674467

Understanding Variation in Audience

Engagement and Response: An Application
of the Composite Model to Receptions
of Avatar (2009)

CAROLYN MICHELLE
School of Social Sciences, Faculty of Arts and Social Sciences, The University of Waikato,
Downloaded by [University of Waikato] at 15:10 29 May 2012

Hamilton, New Zealand

CHARLES H. DAVIS
School of Radio and Television Arts, Faculty of Communication and Design,
and Entrepreneurship and Strategy Department, Ted Rogers School of Management,
Ryerson University, Toronto, Canada

FLORIN VLADICA
School of Radio and Television Arts, Faculty of Communication and Design,
Ryerson University, Toronto, Canada

Although much research documents variations in viewers’

responses to screen media, the basis for divergent receptions
remains relatively poorly understood and inadequately conceptu-
alized. One possible theoretical schema is offered in the composite
multidimensional model, which charts 4 distinct modes of recep-
tion that shape the specific form and content of audience responses
in different contexts. In this study, the core distinctions charted
in the composite model were tested in a Q methodology study
of cross-cultural receptions of Avatar (2009). 120 respondents
from 27 countries modeled their subjective responses to this pol-
ysemic text by rank-ordering 32 items and then commenting
on their selections. Through factor analysis, 4 discrete responses
toward Avatar were identified among participants, accounting

The authors express their sincere gratitude to Steven R. Brown for his thoughtful
reflections on an earlier version of this article.
Address correspondence to Carolyn Michelle, School of Social Sciences, Faculty of Arts
and Social Sciences, The University of Waikato, Private Bag 3105, Hamilton 3240, New
Zealand. E-mail: [email protected]

106
 Variation in Audience Response 107

for 74% of all respondents. Each factor clearly reflects key ele-
ments of the transparent, referential, mediated, and discursive
modes identified in the composite model, indicating that the model
is reasonably accurate in identifying broad distinctions in the
underlying approaches to meaning-making that can be adopted
by different viewers. Suggestive associations between viewers’ sub-
jective orientations and demographic characteristics, social group
memberships, and discursive affiliations were also documented.

Reception scholars from various disciplines have long sought to account for
and understand the bases of divergent audience decodings of media texts.
Following the tradition of British cultural studies, a plethora of qualitative
Downloaded by [University of Waikato] at 15:10 29 May 2012

audience ethnographies have documented the active, creative, and critical

capacities of audiences (e.g., Dahlgren, 1988; Johnson, del Rio, & Kemmitt,
2010; Liebes & Katz, 1989; Morley, 1980), whereas uses-and-gratifications
scholars have elucidated the motivations and satisfactions associated with
media usage through extensive qualitative and quantitative research (see
Blumler, 1979; Levy & Windahl, 1984; Ruggiero, 2000). Those concerned
with media effects have focused on understanding the psychological pro-
cesses governing viewers’ engagement with popular media and on factors
that cultivate the potential for narrative persuasion, delineating concepts
such as involvement, identification (Cohen, 2001), transportation (Green &
Brock, 2002), and narrative engagement (Busselle & Bilandzic, 2008).
Given that most reception scholars refute the notion that variations in
audience response are entirely personal or idiosyncratic, researchers have
attempted to conceptualize and categorize broad differences in viewers’
engagements with media texts (e.g., Busselle & Bilandzic, 2008; Hall, 1980;
Liebes & Katz, 1990; Schrøder, 2000; Staiger, 2000; Suckfüll & Scharkow,
2009). Yet, when such models are examined and compared, most fall short
of offering a comprehensive, multidimensional account of the range of pos-
sible audience responses to the growing diversity of media genres, because
each neglects dimensions of reception that other models from different dis-
ciplinary perspectives postulate as centrally important. At present, there is
no commonly accepted conceptual schema that adequately charts the com-
plexity and diversity of audience engagement and response while enabling
identification and analysis of the underlying correspondences between
seemingly idiosyncratic responses to ostensibly unique texts.
The absence of an agreed analytical framework for identifying, inter-
preting, and analyzing divergent receptions is the product of, and a key
factor in perpetuating, the increasing degree of unproductive fragmen-
tation in the wider cross-disciplinary field of audience reception studies
(Barker, 2006; Michelle, 2007), which today includes scholars working
in media and communication studies, cultural studies, rhetorical studies,
 108 C. Michelle et al.

sociology, media psychology, anthropology, cultural geography, and else-

where. Although this fragmentation largely relates to increasing differentia-
tion in our research objects and objectives, it is evident also in the historic
qualitative/quantitative divide and, perhaps more pertinent, in the ongoing
ethnographic/experimental divide (e.g., compare the different approaches
taken to understanding audience receptions of television and film adopted
by Glaser, Garsoffky, & Schwan [2012], informed by media and cognitive
psychology, and the richly ethnographic work of Friedman [2006], situated
in cultural anthropology). Fragmentation in the field is further manifested
in the common failure to acknowledge correspondences between findings
that pertain to different audience-text encounters, particularly when derived
from divergent disciplinary traditions, perhaps because many scholars are
somewhat disinclined to engage with studies that are grounded in a rad-
ically different disciplinary and epistemological perspective. To date, little
Downloaded by [University of Waikato] at 15:10 29 May 2012

work has been done in terms of synthesizing key findings across the wider
field in order to develop a more cohesive and holistic understanding of the
specific forms of variation in audience engagement and response that have
now been consistently documented.
Fragmentation in the field is further compounded by some general
tendencies that can also be observed, such as the trend for ethnographic
research to be conducted in the absence of any explicit analytical concep-
tualization of the different approaches to sense-making that can potentially
be adopted by readers and viewers, often because researchers do not wish
to impose predetermined categories on their data (e.g., see Johnson et al.,
2010). In such cases, scholars frequently present interesting findings in great
detail but also in isolation, as though they bore little or no relation to the
wider body of evidence compiled to date. Some researchers attempt to
construct conceptual schemas to help make sense of their voluminous and
unwieldy data set but usually do so with little reference to previous efforts
to do the same (e.g., see Andacht, 2004). The value of such efforts is thus
greatly undermined. Although fascinating accounts of particular instances
of the audience-text encounter are frequently generated, this somewhat ad
hoc and unsystematic case study approach tends toward descriptive anec-
dotalism (Barker, 2006; Morley, 2006), and seems a rather ineffective way
to derive a deeper understanding of audiences and processes of audience
engagement; nor does it necessarily facilitate the development of a more
adequate theoretical understanding of the nature of reception per se, as
Barker (2006) and others have noted.
Seeking to facilitate a reconciliation among scholars interested in the
individual, psychological, social, textual, and contextual influences and pro-
cesses that generate variation in audience interpretation and response in
a way that might productively contribute to informing a general theory of
reception, Michelle (2007) offered a composite multidimensional model of
audience reception that synthesizes, refines, and conceptually maps some
 Variation in Audience Response 109

of the existing categories and analytical distinctions drawn from a diverse

and cross-cultural array of reception studies focused on various genres of
television and (less frequently) film and literature. In this study, we used an
innovative research procedure, Q methodology, to test the theoretical and
conceptual categories charted by the composite model through an analysis
of divergent receptions of director James Cameron’s 2009 award-winning
feature film, Avatar. In this respect, our purpose in conducting this research
is somewhat unusual and worthy of further comment.
Our primary purpose here is not to elucidate in great detail all of the
complexity and nuances of audience engagements with this particular text;
indeed, some readers may be troubled by the absence of rich detail that nor-
mally accompanies studies of film and television reception published in this
journal. Rather, we adopt an approach that reflects the theoretical position
that audience receptions of Avatar are but one manifestation of a wider set
Downloaded by [University of Waikato] at 15:10 29 May 2012

of reception processes that come into play in every audience-text encounter;

these reception processes are our primary object of interest. This approach
reflects our observation that actual audiences, composed of real individuals,
engage with an increasingly diverse range of screen media during the course
of their everyday lives: television, films (occasionally watched at the cinema
but more often in their own homes on DVD), various forms of visual adver-
tising, video games, music videos, online streaming content, video-capable
smart phones, and more. It is no longer feasible, we contend, to imag-
ine that those individuals reserve entirely unique processes of interpretation
and engagement for each of these separate mediums. Reception, although
shaped in response to different media texts and technologies, is not deter-
mined by them, since it is fundamentally a process that pertains to socially
located individuals. Those individuals draw from their own general pool of
experiences and understandings, psychological tendencies, discursive alle-
giances, and cultural competencies as they serially engage with different
visual media over the course of each day. Although there are certainly some
genre- and medium-specific competencies that may pertain to particular for-
mats and textual encounters, those competencies are attached to real people
and draw from each person’s existing pool of interpretive resources, which
contains a wealth of other sense-making materials that will be applied across
a wide range of textual encounters.
Hence, our primary focus here is on contributing to our understand-
ing of audience reception in general, rather than in the particular, which
we seek to do by establishing the degree to which one proposed analyt-
ical model of reception, the composite model (Michelle, 2007), is able to
anticipate and account for the actual forms of variation in audience engage-
ment and response in this particular case. For this reason, attention to
the more idiosyncratic elements of individual receptions of Avatar is set
aside in favor of exploring the analytical power of the composite model,
specifically in terms of its ability to accurately identify the core distinctions
 110 C. Michelle et al.

that characterise and differentiate the underlying subjective frameworks or

modes of reception that are most commonly adopted by different view-
ers among Avatar’s global English-speaking audience. The value of such
an approach lies, we believe, in its potential to contribute to building a
theoretical understanding of the reception process, thereby enriching and
hopefully progressing audience research in its various cross-disciplinary
manifestations.

The Composite Model

As previously outlined in this journal, the composite model charts four
broad modes of audience engagement and response: transparent, referen-
tial, mediated, and discursive. These categories are informed by, synthesize,
Downloaded by [University of Waikato] at 15:10 29 May 2012

and refine existing analytical distinctions identified in relevant scholarship

and are supported by extensive reviews of empirical research on audience
receptions of screen media texts across various cultural contexts (for a fuller
description and discussion of the composite model and its application, see
Michelle, 2007, 2009).
Focusing here on the model’s application to fictional texts such as
Avatar, a transparent mode reflects a close subjective relation between
viewer and text whereby viewers temporarily suspend disbelief and criti-
cal distance to grant fictional worlds the status of real life, entering fully into
the story to derive the specific forms of pleasure and enjoyment intended
by the text’s makers. This kind of subjective response to media narrative
has been well documented in the existing scholarship and is alluded to
in the concepts of inferential reading (Worth & Gross, 1974), transparency
reading (Richardson & Corner, 1986), and strong involvement (Schrøder,
1986). Media effects researchers have gone further in elucidating this mode
of engagement in relation to entertainment media, citing as a key aspect
the experience of being fully transported into a fictional world by the nar-
rative, as described by Green and colleagues (Green & Brock, 2000, 2002;
Green, Brock, & Kaufman, 2004). Viewers in this mode experience a high
degree of cognitive and emotional engagement and full immersion in the
text, are swept away by the story, and may experience strong feelings of
identification with the central characters or textual themes (Cohen, 2001;
Liebes & Katz, 1990; Wilson, 1995). Such identification, Wilson (1995) sug-
gested, facilitates acceptance of a text’s “related prescriptions for action . . .
and persuasion of what is the case” (p. 12). Green et al. (2004, p. 313) simi-
larly noted that narrative transportation can be transformative in the sense of
changing social beliefs and perceptions. Michelle (2007) suggested that any
such effects are rendered possible because, when reading in a transparent
mode, viewers rely on information supplied within the text itself as primary
resources for its decoding, rather than drawing on extratextual information.
Because textual meanings are therefore read straight, a dominant/preferred
 Variation in Audience Response 111

position in relation to the text’s ideological content (Hall, 1980) can be

assumed of those reading solely in this mode.
In a referential mode, the text is primarily understood in relation to
viewers’ experiential knowledges and perceptions of the text’s relevance
(or lack thereof) to an extratextual reality. The specific quality of this sub-
jective mode can be discerned by consolidating insights offered by other
scholars, in particular Schrøder’s (1986) notion of indicative involvement,
the concept of trivial/random personal association (Dahlgren, 1988), and
referential reading (Liebes & Katz, 1989, 1990). In a referential mode, view-
ers perceive the text as standing alongside the real world, and often make
comparisons and analogies between that depicted reality and the world as
they see it. In so doing, they typically draw from aspects of their own cul-
tural milieu and existing body of experiences, observations, and knowledges
to assess the accuracy of textual depictions of people and events and the
Downloaded by [University of Waikato] at 15:10 29 May 2012

version of reality presented. Friedman (2006) richly documented this pro-

cess in relation to receptions of the film Twin Bracelets in China, among
audiences who “recognized that they were watching a version of them-
selves, even as they contested how they were being represented” (p. 604).
Similarly, Staiger (2000) noted the significant role that verisimilitude and
real-world expectations can play in shaping some viewers’ responses in cer-
tain contexts. The fact that referential knowledge may be used to affirm,
question, or reject textual realism means that a viewer’s position in rela-
tion to the text’s preferred ideological meanings cannot be predicted on
the basis of their adoption of this viewing mode. This may be because, as
Busselle and Bilandzic (2008) suggested, negative evaluations of textual real-
ism disrupt viewers’ engagement and hence undermine a text’s persuasive
power.
What distinguishes a mediated mode of reading is its explicit focus
on the constructed nature of the text as an aesthetic object and media
production—as an elaboration of established media codes and conventions
by particular authors and creators. This subjective orientation is described
by the categories of attributional reading (Worth & Gross, 1974), analytic
decoding (Neuman, 1982), mediation reading (Corner & Richardson, 1986),
media awareness/demystification discourse (Dahlgren, 1988), syntactic crit-
icism (Liebes & Katz, 1989, 1990), and discrimination (Schrøder, 2000); see
also Staiger’s (2000) description of aesthetic order and variation activities.
Synthesizing the key elements of these categories, mediated readings reflect
a more distant or separate relation between text and viewer (although the
reverse may be true of hardcore fans), in which viewers may praise or dis-
parage the quality of production, particular aesthetic or generic features of
the text, or the perceived intentions of its authors or producers. In adopting a
mediated mode of reception, viewers characteristically draw on (often quite
considerable) knowledge of aspects of media production, aesthetic ideals,
generic conventions and textual formulae, intertextual references, and the
 112 C. Michelle et al.

functions and motivations of the film and television industries. This knowl-
edge may interrupt the process of identification and potentially also militate
against viewers’ serious engagement with the text’s narrative or message
content. Because such receptions require specific knowledges, discursive
competencies, and media literacies, Michelle (2007) suggested that some
viewers will have greater access and allegiance to this mode of viewing than
others.
Last, receptions framed in a discursive mode primarily and overtly
address the text’s propositional or message content—that is, its ideo-
logical connotations. Other scholars have also noted responses of this
kind. Neuman (1982) described this approach as an interpretive decoding,
whereas Richardson and Corner (1986) noted some viewers’ identification
of a text’s manipulative intent. Liebes and Katz (1989, 1990) described this
response in terms of semantic criticism, whereas Schrøder (2000) identi-
Downloaded by [University of Waikato] at 15:10 29 May 2012

fied related processes in his dimensions of comprehension and position.

Although it is clear that all receptions have a discursive element, responses
primarily framed in this mode give particular credence to the text’s perceived
attempt to communicate a particular message about the wider social world,
and represent the viewer’s response to that message. In assessing the con-
notative meaning of the text through the lenses of their unique stock of prior
beliefs, assumptions, and discursive allegiances, the composite model sug-
gests viewers may adopt one of the three positions theorized by Stuart Hall
(1980)—preferred/dominant, negotiated, or oppositional. However, opposi-
tionality is understood in relation to the preferred meanings of the texts in
question, which cannot be assumed to affirm hegemonic interests at each
and every moment (Schrøder, 2000).
Although the composite model acknowledges that viewers potentially
have access to all four modes and may oscillate, or commute (Schrøder,
1986) between them during any given media encounter (see also Staiger,
2000, p. 21), and although modes of response may also change over
time (e.g., as prefigurative influences shape expectations, repeated tex-
tual exposure increases textual familiarity, and alternative perspectives are
encountered via ongoing social interactions), these four modes remain rel-
atively distinct and, at times, contradictory registers of meaning that use
unique sets of cultural and discursive competencies at different moments.1
Much of the variability of audience response, this model suggests, is

1
Almost all of the studies cited in this discussion of existing models of reception document clear
distinctions in the discursive resources and approaches to sense-making that are utilized by different
groups of viewers, and in some cases the same viewer at different moments (e.g., see the exemplary
case of a commuting viewer, Michael, detailed in Michelle, 2007, pp. 213–215). That these are distinct
and, at times, contradictory modes of reception has been noted by Schrøder (1986) and also Andacht
(2004), who identifies the commuting process in relation to young people’s receptions of Big Brother in
Latin America, where
 Variation in Audience Response 113

attributable to the propensity for different segments of the audience to adopt

(and, in some cases, move between) distinct viewing modes which, work-
ing in tandem with the parameters imposed by textual encoding, define and
delimit the kinds of readings that are likely to be produced by differently
positioned audience members in particular contexts.
In seeking to test the utility of this analytic framework, we sought
a methodology that could independently verify whether these four dis-
tinct modes are indeed evident among the viewing audience and whether
they take the specific forms described earlier. Although little known among
communication scholars, Q methodology is commonly used in various aca-
demic fields specifically to discern and describe people’s shared subjective
viewpoints and understandings, as this is considered its greatest strength.
Elsewhere, we have argued that Q methodology should take its place
among the growing number of new methods that have emerged in media
Downloaded by [University of Waikato] at 15:10 29 May 2012

and communication studies in recent years as researchers seek innovative

ways to explore audience engagements in an increasingly complex con-
vergent digital media environment spanning a growing array of formats,
in which audiences are increasingly not only consumers but also produc-
ers of media content (Davis & Michelle, 2011). Although any number of
new and established research methods could have been used in a study of
Avatar’s reception, we considered Q methodology to be the most suitable
means of empirically verifying the distinct interpretive modes theorized in
the composite model because it offered the relatively unique capacity for
participants to independently chart a holistic representation of their mul-
tidimensional response to viewing the film, rather than registering their
reaction to discrete issues or topics consecutively as typically occurs with
most other methods. Q methodology also acknowledges the inherent social-
ity of reception in that even the most seemingly idiosyncratic responses are
not formulated in a vacuum, but rather emerge within a wider discursive
context that necessarily informs individual understandings of media texts,

the most noticeable tendency . . . is a fair split, even an oscillation between a firm belief
in the genuineness of the format’s index appeal, and the strong suspicion that its most
memorable moments are carefully staged by the participants of Big Brother in complicity
with the producers. (pp. 124–125)

Hill (2005, p. 177) similarly notes that (some) viewers of reality TV may commute between at least
two different modes in the same response: “audiences are able to switch from appreciation of these
ordinary people and their experiences, to awareness of the staged nature of their experiences created
for television.”’ Clearly, a reading presuming that a media text faithfully reproduces authentic events or
behavior is predicated on a different set of assumptions about the text and its production to one that
proposes those same events or behaviors have been deliberately engineered or fabricated. That in these
examples viewers are described as being split, oscillating, or switching reflects the inherent difficulty of
reconciling incompatible positions, such that moving between these modes requires rejecting one set of
assumptions and beliefs and adopting another, if only momentarily. See also Barker and Mathijs (2005)
and Staiger (2000) for examples of viewer responses that contain inherently contradictory assumptions
about the status of the text as viewers shift between different levels or modes of reading.
 114 C. Michelle et al.

their content, and production. Further advantages of Q methodology are

subsequently discussed.
Through our investigation we aimed to further elucidate the underly-
ing principles that generate interpretive divergence among screen media
audiences, using cross-cultural receptions of director James Cameron’s inter-
nationally successful 2009 feature film, Avatar, as a case study to test
whether the composite model’s major conceptual and analytical distinctions
are sufficiently inclusive and robust to provide a more widely applicable
schema of audience engagement and response. A further aim was to explore
the relation between shared subjectivity and the complex intersection of
respondents’ culture, ethnicity, gender, age, socioeconomic class, education,
religion, and political beliefs, thereby addressing a key area of research
and debate in the field of reception studies. We comment only briefly
on these findings here, which indicate potentially fruitful areas for future
Downloaded by [University of Waikato] at 15:10 29 May 2012

research.

Avatar: Divergent Responses to a Highly Polyvalent Text

Most readers will be aware of the extraordinary level of discussion
and debate surrounding the 2009 award-winning science fiction fantasy
film, Avatar. Those unfamiliar with the film can find a brief synop-
sis at http://movies.msn.com/movies/movie-synopsis/avatar.2. Written and
directed by James Cameron, Avatar melded narrative with advanced
computer-generated imagery in a 3D format to create what was—for many
viewers around the world—a highly realistic and compelling screen expe-
rience. Various commentators have identified Avatar’s overt themes of
environmental degradation and human(oid) displacement in the service of
economic development, (neo)colonial imperialism and capitalist exploita-
tion, violent military repression, transformation and bodily transcendence,
and new-age spirituality. In terms of Avatar’s intended or preferred meaning,
Cameron has explicitly acknowledged his desire to raise public conscious-
ness about environmental issues through the medium of entertainment, and
conceived the film as an allegory for humanity’s exploitation and destruction
of the natural environment in the context of our collective failure to avert
the impending global ecological disaster (Cameron, 2010).
Yet, in preparing for this study via an extensive “cultural trawl” (Stenner
& Marshall, 1995, p. 626) of online professional and lay film reviews, online
fan discussions, Facebook message board comments, and international news
coverage and media commentary, it became clear that audience responses
to Avatar varied greatly and that differently located viewers were making
widely diverging readings of the film. It is clear that Avatar is a highly
polysemic text, or to be more accurate, a polyvalent one (Condit, 1989),
inviting multiple interpretations through its combination of several different
themes and story elements. This makes it a particularly appropriate choice
 Variation in Audience Response 115

of text for a study aimed at testing a model of interpretative divergence.

Further, much like Dallas in the 1980s, Avatar’s global success2 appears to
reflect this openness and “universality, or primordiality, of some . . . themes
and formulae,” which renders Avatar “psychologically accessible” (Liebes &
Katz, 1990, p. 5). Olson (2004) asserted that the international appeal of U.S.
popular culture is linked to its polysemy combined with a high degree of
“narrative transparency” or a tendency to “manifest narrative structures that
easily blend into other cultures. Those cultures are able to project their own
narratives, values, myths, and meanings into the American iconic media”
(p. 114). Such texts are no longer experienced as entirely foreign.
Ensuring a high degree of polyvalence and narrative transparency
was clearly central to Avatar’s conceptualization and production. Cameron
explicitly acknowledges using familiar archetypes to aid the universal acces-
sibility of the film’s environmental messages (Cameron, 2010) but also
Downloaded by [University of Waikato] at 15:10 29 May 2012

alludes to various other encoded messages and themes, as will become

evident later in this article. It is also significant to note Cameron’s choice
of genre: Science fiction is described by Lee (2006) as “close to being the
quintessential universal movie genre” (p. 274).
As seemingly intended, then, Avatar has provided considerable plea-
sure to a heterogeneous global audience. Comments on fan websites such
as avatar-forums.com typically reveal a high degree of emotional engage-
ment, identification with the trials and tribulations of story characters, and
full immersion in the fictional world, and describe viewing Avatar as a
compelling, at times overwhelming, and in some cases transformative expe-
rience. For some, the film evidently caused some discomfort, whether
physical in the form of headaches and nausea related to the film’s 3D pre-
sentation, or emotional in that many serious fans claim to have experienced
some form of post-Avatar depression.3
Somewhat differently, other viewers primarily regarded Avatar as echo-
ing the environmental destruction occurring within their own communities
in the interests of economic development: The film’s tall blue humanoids, the
Na’vi, have become icons of resistance with global resonance, their plight
being compared with that of local populations resisting the efforts of mining
and oil companies in Canada (EdmontonJournal.com, 2010), India (Hopkins,
2010; Thottam, 2010), and South Africa (Clarke, 2010). Activists in Indonesia
and the Amazon basin have similarly appropriated the symbolic power of
the Na’vi as part of protests against deforestation. Cameron (2010) affirmed
that the film consciously references the economic and military imperialism
of the colonial and present periods, a message that has been interpreted

2
As of June 13, 2011, Avatar has grossed US$2,782,275,172 worldwide, making it the highest grossing
film of all time (unadjusted for inflation).
3
For example, see the four-part discussion on “Ways to Cope With the Depression of the Dream of
Pandora Being Intangible” in the “General” forum at www.avatar-forums.com.
 116 C. Michelle et al.

as relevant to various contexts, including the ongoing conflict in Palestine

(Associated Press, February 12, 2010).
Alongside responses reflecting an appreciation of Avatar’s allegorical
relevance to current environmental and human rights concerns are a range
of alternative decodings. Cameron acknowledges that Avatar contains delib-
erate textual references to, and implicit critique of, the U.S. wars in Iraq and
Vietnam—most notably in dialogue drawing on the “shock and awe” ter-
minology coined by the Bush administration—leading some conservative
critics to slate the film as antimilitary, anti-American, and unpatriotic (Nolte,
2009; Podhoretz, 2009). Avatar’s spiritual themes have also been subject to
criticism, most notably by the Vatican, which decried the film’s pantheism
and worship of the natural world (Associated Press, January 13, 2010). Still
others have lambasted the recycling of the paternalistic White Messiah fable
in which a White man becomes the savior of a tribe of noble savages whose
Downloaded by [University of Waikato] at 15:10 29 May 2012

appearance, gestures and rituals embody stereotypical visions of Earth’s

tribal peoples (Brooks, 2010; Gates, 2010; Washington, 2010; Zizek, 2010).
These are just some of the more commonly offered responses to Avatar
circulating within the wider discursive field. It is clear that some of these
interpretations pertain to the denotative level of meaning and evaluate the
text in terms of its allegorical relevance to historical, environmental, and
political issues here on Earth, while others address the film’s connotations in
terms of its perceived ideological or message content and possible societal
impacts. A distinctly different set of readings, however, primarily relate to
Avatar as a constructed media entertainment product. Avatar has received
much praise for its technological advances, effective use of performance cap-
ture, computer-generated imagery, 3D and other special effects (e.g., Ebert,
2009). Indeed, Avatar is regarded by some as a game-changer for the film
industry given the technological advances it makes in 3D filming processes
and exhibition (Acland, 2010). But while many fans saw Avatar as a highly
enjoyable and technically accomplished Hollywood blockbuster, it has also
been critiqued for its derivative storyline, weak dialogue, clichéd charac-
terizations, and mawkishness (e.g., Turan, 2009). Such sentiments comprise
a significant component of the wider discursive field—or in Q terms, the
concourse—surrounding this film.
In our reception study, we investigated how casual viewers as well as
Avatar fans and haters responded to the film, and in the process aimed
to test the composite model’s accuracy and validity as a holistic, gener-
alized schema of the form and content of potential responses to screen
media. This study thus reflects the theoretical position that, considered as
a whole, the body of existing reception research documents consistently
observed patterns in terms of the manner in which different audience mem-
bers engage with and make sense of texts of multiple genres. Although the
specific content of audience response is to some degree textually contin-
gent (in that textual encoding determines the subject matter and attempts
 Variation in Audience Response 117

to define and delimit interpretations), the mode of engagement ultimately

establishes the parameters for textual decoding at the level of reception.
Audience responses to a particular text thus need to be understood in rela-
tion to the underlying subjective orientations to the process of reception,
or modes of engagement, audience members may potentially adopt when
encountering any such text. Further, because receptions are always situated
and contextually specific, simultaneously individual and social in nature,
they are highly likely to be patterned in discernable ways based on viewers’
demographic characteristics, social group memberships and discursive affili-
ations. Clearly establishing the underlying subjective orientations that govern
sense-making is thus an important first step in efforts to more clearly under-
stand how, and under what conditions, shared identities and social locations
shape individual receptions, which remains an unresolved question for many
reception scholars.
Downloaded by [University of Waikato] at 15:10 29 May 2012

METHOD

Like Livingstone (2003) and others, we are mindful of the dangers and dif-
ficulties posed by cross-cultural research, but still wish to acknowledge its
importance to the wider project of reception studies as a means of testing
the broader applicability of analyses and conceptual understandings. Our
research design thus reflects the “meta-theoretical” solution advocated by
Swanson (1992, as cited in Livingstone, 2003), in that we have adopted “a
common theoretical framework that identifies abstract concepts or dimen-
sions” (p. 490) in relation to which the responses of differently located
viewers can be analyzed—this being the composite multidimensional model
of audience reception, outlined earlier (Michelle, 2007). The composite
model thus provides the theoretical and analytical framework underpinning
the research, with Q methodology being the research procedure used to
objectively and independently identify the range of subjective responses to
Avatar among the research sample.

Q Methodology
Although Q methodology is now relatively frequently used in psychol-
ogy, environmental studies, political science, policy studies, and health
research, it remains less familiar to reception scholars and is only occa-
sionally applied in studies of media audiences (see Davis & Michelle, 2011,
for a full review of existing Q research in this area). However, the inven-
tor of Q methodology, William Stephenson (1976, 1978, 1995–1996), was
clearly interested in its application to advertising, television and film recep-
tion. Esrock (2005) suggested that Stephenson’s approach serves to remind
mass communication scholars that “although media institutions disseminate
 118 C. Michelle et al.

texts, whether for information or persuasive purposes, ultimately individuals

are the consumers of those texts. And ultimately, individual perceptions
and interpretations reveal true meaning, no matter what may have been
intended” (p. 249). We thus perceived a natural fit between Q methodol-
ogy’s focus on the expressed subjective perspectives of human subjects and
an audience-centered approach to the study of media reception and use.
For those unfamiliar with this method, Q is a rigorous qualitative
research methodology that uses factor analysis to discern and describe peo-
ple’s shared subjective viewpoints and understandings. Guided by a suitable
theoretical framework, Q potentially offers a means of examining “the rela-
tions between and among the ‘reading positions’” (Barker, 2006, p. 130) of
differently situated viewers. These positions, as Barker noted, are developed
in response to the assessments of others and draw from the existing dis-
course or concourse that circulates around a given text, or set of texts. By
Downloaded by [University of Waikato] at 15:10 29 May 2012

asking participants to preferentially rank-order a set of statements carefully

chosen to represent (as far as practically possible) the universe of possible
responses, Q allows each individual to model his or her subjective viewpoint
in a holistic sense, expressing where it fits in relation to others viewing the
same text. Q is thus self-referential: Individuals rank each statement and
determine their placement within the typal array in the manner that best
reflects their unique subjective perspective.
Sorts are factor analyzed by person, a procedure that locates sets of
like-minded respondents, making it possible to systematically identify and
compare the variety of viewpoints shared by individuals within a wider
public and their unique components (for further details, see Brown, 1980,
1994; McKeown & Thomas, 1988; Watts & Stenner, 2005). As Watts and
Stenner (2005) explained, Q is “essentially a gestalt procedure” that can
“show us the particular combinations or configurations of themes which are
preferred” (p. 70) by different segments within a wider sample of individuals.
Furthermore, with Q, “subjective input produces objective structures” (Watts
& Stenner, p. 85), or factors that are statistically significant (nonrandom)
patterns in the responses of participants, indicating that they share a similar
perspective on a given topic.
Although we do not have space to discuss in-depth the advantages that
Q methodology offers over more traditional approaches that might have
been used in this research, we would note that Q combines many of the
strengths of qualitative and quantitative approaches, while mitigating their
key limitations (for further detail, see Davis & Michelle, 2011). Others have
noted the significant difficulties that qualitative research can pose for audi-
ence researchers (see Schrøder, 2000), and while software packages such as
NVivo now make it much easier for researchers to classify and organise large
qualitative data sets, and can even independently identify commonalities in
terms of recurring words and content, they do not offer a way of identifying
shared multidimensional subjective perspectives or viewpoints in the more
 Variation in Audience Response 119

holistic sense that Q does. In contrast, quantitative approaches to attitude

measurement such as semantic differential tests are not designed to permit
subjects to model their own viewpoints. Instead, subjects’ ratings on vari-
ous validated scales are used as independent variables to predict dependent
variables. Such tests lack a qualitative dimension, do not consider the dif-
ferent interpretations that individuals may make of the terms or statements
selected for inclusion, and offer no opportunity for participants to articu-
late their own views, responses, and priorities. Q methodology, on the other
hand, explicitly solicits participants’ comments in order to gain further insight
into their responses and to guide the interpretations that researchers make
of the different factors. This qualitative data can be analyzed in its own right,
and there is no reason why Q could not be used in conjunction with other
qualitative and quantitative research tools. Indeed, many researchers have
suggested Q methodology can be used to effectively bridge these different
Downloaded by [University of Waikato] at 15:10 29 May 2012

approaches (see Davis & Michelle, 2011, for further discussion).

Because Q methodology is explicitly aimed at revealing the underly-
ing structure and form of people’s subjective opinions and beliefs (Brown,
1986), and because the composite model assumes that the form and con-
tent of audience reception is shaped by underlying modes or interpretative
“orientations,” Q was considered highly appropriate for our research on
audience receptions of Avatar. If the composite model’s analytical frame-
work is a reasonably accurate and holistic representation of natural variations
in audience members’ subjective orientations toward making sense of screen
media texts, including Avatar, then the viewpoints that emerge should
broadly map onto the distinct modes of reception identified within the
model. However, this does not necessarily mean that only four viewpoints
will be identified through factor analysis, given that different positions are
possible within some modes: A mediated reading may be celebratory or crit-
ical, while a referential reading may find the text accurate or unrealistic in its
rendering. Either way, the viewpoints that emerge should be recognizable
and clearly explicable using concepts derived from the model, given that
they will reflect its different aspects. At the very least, factor analysis reveals
significant patterns in the rank ordering of statements, reflecting shared
subjective viewpoints, and also identifies which individuals are strongly
correlated with each factor.

RESEARCH PROCEDURE

A structured Q sample of 32 statements was devised drawing on an extensive

cultural trawl of dominant and marginal themes, opinions, intellectual and
emotional reactions articulated in the responses of causal viewers, media
commentators and film critics, which form part of the wider concourse that
quickly emerged around Avatar (described earlier). In conducting this cul-
tural trawl, we made no distinctions between the origin of the different
 120 C. Michelle et al.

viewpoints that were considered for inclusion, because its primary purpose
was to gain a broad and inclusive impression of the full range of things being
said about Avatar across the wider public domain, including within official
and unofficial sites of discussion, in order that the items selected for inclu-
sion in the Q-sample represent as far as possible the diversity of viewpoints
circulating around this film, its production, content, and possible impact. An
important methodological principle of Q is that the statements considered
for inclusion take the form of subjective opinions rather than facts; as such,
those opinions are considered valid irrespective of their origin.
Because the selected statements were chosen to reflect as much as
possible the kinds of diverse perspectives on the film expressed among
differently located subjects, we tried to preserve the original expression of
authors wherever possible. At first, around 260 raw statements articulat-
ing the most frequently recurring ideas, opinions, or reactions from various
Downloaded by [University of Waikato] at 15:10 29 May 2012

different groups were cut and pasted into an Excel file. This step in the
process aimed to be as inclusive as possible and continued until a signifi-
cant degree of repetition and redundancy among the statements began to
emerge. Because it is not usually practical to include a large number of
statements in a Q survey, we needed to make further selections from this
initial pool of potential statements. To guide the selection process, the state-
ments were reviewed and loosely categorized in terms of the four modes
of response identified in the composite model and their various subcate-
gories. A few additional statements were added by the researchers to reflect
important aspects of modality that, although implicitly articulated within the
wider concourse, were not clearly or adequately expressed in one existing
statement.
It is important to note that in Q terms, this kind of structured approach
is seen as helpful in ensuring a theoretically justifiable basis for the selec-
tion of statements, and in making an explicit link to theory may assist in
bridging the gap between quantitative and qualitative research traditions
(Kerlinger, 1972). Further, a systematically structured Q sample is not con-
sidered to either predefine or predetermine participants’ responses, because
of the huge number of possible combinations of the chosen items and the
fact that participants must review, sort, and rank their statements indepen-
dently. Although in our case all statements encapsulated a distinct response
to a particular aspect of Avatar and its reception, individuals could strongly
agree or disagree with each statement in turn, feel indifferent, or consider a
statement irrelevant or perhaps even laughable; how they chose to respond,
sort and rank order the statements was entirely in their hands. The state-
ments remained open to interpretation and could potentially shift in meaning
when read in relation to each other on the typal array. Each participant was
also free to rearrange their Q-sort before it was finalized and submitted.
This means that the statements could be collectively deployed to holisti-
cally convey a general impression of each individual’s subjective orientation
 Variation in Audience Response 121

on a number of dimensions simultaneously—emotional, visceral, aesthetic,

intellectual, political, and so on—while the ranking process clarified which
statements were more immediately salient for those individuals.
Taking care to preserve as much of the original diversity of responses as
possible, an initial cull systematically reduced the number of Q statements
to 48. In the course of a pilot study conducted with 24 respondents in the
city of Hamilton, New Zealand (population: 131,000), these statements were
progressively reevaluated and revised to ensure each expressed relatively
unambiguous ideas or themes and to eliminate redundancy among the state-
ments. The statements were gradually whittled down to the final Q sample of
32 statements (Appendix). Each mode theorized in the Consolidated Model
is loosely represented by eight statements reflecting its different aspects.
Although the final Q sample is not inclusive of all perspectives drawn from
the cultural trawl, it does represent our best effort to reflect the predominant
Downloaded by [University of Waikato] at 15:10 29 May 2012

ideas, opinions, reactions, and concerns expressed by fans as well as crit-

ics, while simultaneously operationalizing the categories and subcategories
of the composite model. Among the themes and issues not represented are
those relating to the film’s depiction of disability and gender, the physical
effects of 3D technology, and the emotional impact of the musical score and
dragon flight scenes.
Participants performed their Q-sorts online using the free software
application, FlashQ, and almost all also provided details of their gender,
age, nationality, ethnicity, education, occupation, and religious and political
affiliations. Immediately before carrying out their Q-sort, participants read
a set of instructions on screen which stated that the researchers wished to
know how they responded to and felt about Avatar when they first saw
it. As the condition of instruction, they were instructed as follows: “Please
recall the thoughts and feelings you had immediately after watching Avatar
while you rank the following statements.” After first reviewing and sorting
all the statements into three piles—agree, disagree, or neutral—they were
instructed to arrange the 32 statements on a continuum in a forced normal
distribution, with the number of statements under each score, as follows:

Most strongly disagree Most strongly agree

−4 −3 −2 −1 0 +1 +2 +3 +4
2 3 3 5 6 5 3 3 2

Respondents were also invited to provide a written explanation for why

they selected the four statements they placed at each end of the scale. These
qualitative comments were crucial in guiding our interpretation of the sub-
jective viewpoints that emerged, and examples of the kinds of statements
offered are incorporated into the written analysis of each factor to illustrate
and provide richer qualitative insight into participants’ subjective responses.
 122 C. Michelle et al.

Although we believe the condition of instruction encouraged partic-

ipants to consider their initial response to the film, there is no way of
guaranteeing that this was the basis on which they sorted the statements,
nor that their recollection of their initial response was untainted by subse-
quent thought, discussion and reflection on the film and its public reception.
There are inherent limitations that arise from treating participants’ responses,
in some cases solicited well after the initial moment of viewing, as evi-
dence of an underlying subjective orientation adopted or stimulated during
the viewing encounter which shaped viewers’ engagement and response
to the film, as the composite model assumes. Those immediate reactions
and responses are notoriously difficult to access in their ‘pure’ state, given
that the mechanisms used to access them and the context in which such
efforts occur may well alter the reception process itself. Hence, almost all
researchers rely to some extent on articulated rather than natural responses
Downloaded by [University of Waikato] at 15:10 29 May 2012

to texts viewed during or some time before the research encounter. All such
articulated responses are constructed accounts produced in a social context,
and thus may be influenced to varying degrees by the following:

1. The researcher effect, whereby participants may attempt to anticipate

and respond to what they believe the researcher is seeking, rather
than expressing their own views; in this research, researcher effect was
minimized because the researchers were physically absent.
2. Participants’ existing understandings relating to genre, medium, their posi-
tion on the themes and issues raised by the text, their education, social
class, gender, religion, and so on, any of which may provide particular
cultural competencies that inform their talk about the text.
3. The wider concourse surrounding the text in terms of how it is commonly
talked about within public and private domains, particularly if it is already
subject to some discussion and debate.
4. Impression management, especially in the context of focus group discus-
sions, where individuals may censor or manage their responses according
to the expressed views of others, or in an effort to present themselves in
a particular manner. Because the Avatar Q-sorts were completed inde-
pendently and anonymously online, impression management was not
considered to be a concern in this research.

Hence, while we recognize the limitations inherent in claiming that the

typal arrays constructed by individual respondents generally represent their
initial subjective response to the film, the same critique can be made of
virtually all audience reception research, none of which is able to access
cognitive content in a pure, unmediated way. But it would be a mistake
to assume on this basis that people’s subsequently articulated reflections
offer little or no useful insight into their original subjective states, because
such a view implies respondents know nothing of their own minds and are
 Variation in Audience Response 123

unable to reflect on past cognitive and emotional states with any degree of
accuracy.
Furthermore, we note that the consolidated model is informed by a
wide range of studies utilizing a variety of research methods—including
ethnographic studies that attempted to solicit immediate, relatively unmedi-
ated responses to viewed texts, such as that of Liebes and Katz (1990). It
is important to note that the model reflects consistent patterns in subjec-
tive responses across temporal dimensions. That is to say, the same modes
can be identified at the during and after stages of reception, and there is
some basis in the literature on the role of prefiguration in film reception
(see Barker & Mathijs, 2008) to argue that anticipatory reactions, which can
be established well before viewing, similarly reflect some of the core dis-
tinctions theorized in the composite model, and further, can potentially shift
and change during and subsequent to the viewing encounter.4
Downloaded by [University of Waikato] at 15:10 29 May 2012

The Avatar Q-sort went live online on May, 18, 2010, and was open
to English-speaking participants with Internet access anywhere in the
world, being hosted on a server at Ryerson University’s Digital Value Lab.
Our sample is thus self-selecting rather than representative, and excludes
respondents unable to communicate in English as we did not have sufficient
resources to enable translation of the research tool and resulting data. Since
Avatar fans were considered more highly motivated to participate, we
actively encouraged participation from a wider range of viewers, including
Avatar’s critics, posting invitations on the “Avatar Sucks!” Facebook group
message board and on general film discussion boards such as rottentoma-
toes.com. Special interest groups were also targeted through Facebook
groups such as Survival International, Military.com, and the Rainforest
Action Network.
Since respondents were primarily recruited through fan networks and
online communities, the sample consists of people who are active on
the Internet. Recent estimates suggest that this now accounts for 75–
90% of people in most Western developed nations (Internetworldstats.com,

4
Because prefiguration may shape but does not and cannot determine audience response to any
given textual encounter, it is not considered necessary to include it in a comprehensive model of audi-
ence engagement and response. For a similar reason, the focus of uses-and-gratifications research on
motivation, selectivity, and gratifications of media use is considered less immediately relevant to an
understanding of the reception process, and is not explicitly addressed in the composite model. That
said, these dimensions are clearly likely to offer considerable insight into why particular modes of recep-
tion rather than others are adopted by particular individuals. Presumably, a viewer who is motivated
by diversion or escape may be more likely to adopt a transparent mode, while one whose motivation
is surveillance or value reinforcement may be more likely to adopt a referential or discursive mode.
However, one’s original motivations for embarking on an encounter with any given media text do not
necessarily determine the outcome of that encounter. Many readers will have had the experience of
spontaneously viewing a randomly selected movie with the explicit motivation of meeting their needs
for social integration—to spend time with friends or family—only to experience completely unanticipated
levels of emotional or cognitive arousal that became far more significant and enduring than the social
interaction which provided the reason and context for viewing.
 124 C. Michelle et al.

2009). Nonetheless, the research is likely to overrepresent the views of

Internet-savvy “‘cosmopolitans’ [Hannerz, 1990]: people whose life orien-
tation revolves around global interconnectedness rather than their local
communities” (Kuipers & Kloet, 2009, p. 104), and these are likely to be
relatively economically empowered individuals, especially in developing
nations. This bias appears to be a significant limitation of online research
generally, and is evident in our sample: Half of our respondents were uni-
versity students, while relatively few others reported earning low incomes
or having non–white collar occupations (see Table 2). However, the lack of
representativeness of our sample is not necessarily problematic, because
we make no claims about the distribution of any discovered subjective
orientations in the general population.
Downloaded by [University of Waikato] at 15:10 29 May 2012

RESULTS

We analyzed 120 responses received between May and August 2010 from
individuals of 41 self-identified nationalities, races, or ethnicities residing in
27 countries. To conform with Q methodology conventions, we sought a
factor analysis solution that accounted for the greatest number of responses
defined uniquely (i.e., excluding “confounded” and statistically insignificant
sorts) by the smallest number of factors. A four-factor solution best satisfied
these requirements. Factor loadings equal to or greater than 0.46 are signif-
icant at the 1% level with 32 items (see McKeown & Thomas, 1988, p. 50).
We used a commercial dedicated Q methodology software package, PCQ,
to perform the analysis. PCQ uses the centroid method of factor extraction
and permits varimax or judgmental factor rotation (we used the former).
This four-factor solution accounts for 89 (74%) of the 120 respondents.
47 out of 120 respondents loaded significantly on Factor 1, 16 on Factor
2, and 13 each on Factors 3 and 4. 26 respondents constructed mixed or
confounded Q-sorts (i.e., they loaded on more than one factor) and five
were not significant. The presence of confounded sorts potentially reflects
the tendency (acknowledged in the composite model) for some viewers
to commute between modes, in which case they may load significantly on
more than one factor. Two of the factors had a small number of negative or
inverted loadings. The typal sort for each factor is shown in Table 1, while
the aggregate scores assigned to each statement within all factors are shown
in Appendix A. Four consensus statements emerged from the factor analysis
(#1, #3, #8, #28), with the strongest consensus emerging for statement 1: “My
enjoyment of Avatar was marred by its negative portrayal of the U.S. mili-
tary, which I felt was anti-American”, with which most respondents strongly
disagreed (−3).
Three of the four identified factors strongly correspond with one of the
modes identified in the composite model, in that five of the eight statements
 Variation in Audience Response 125

TABLE 1 Typal Arrays

Factor 1: Narrative transportation with emotional resonance (transparent mode)

−4 −3 −2 −1 0 1 2 3 4

11 1 2 17 10 4 16 6 5
12 9 3 18 13 8 25 30 26
14 7 19 22 15 29 32
20 23 27
21 24 31
28
Factor 2: Opposition to form/negotiation of message content (discursive mode)
−4 −3 −2 −1 0 1 2 3 4

29 1 3 10 13 7 6 4 2
32 18 20 19 16 8 11 5 14
26 27 21 22 15 12 9
28 24 17
Downloaded by [University of Waikato] at 15:10 29 May 2012

31 25 23
30
Factor 3: Appreciation of real-world relevance (referential mode)
−4 −3 −2 −1 0 1 2 3 4

11 1 2 13 10 7 6 4 17
12 9 3 15 16 8 18 5 23
14 32 27 25 20 19 24
29 26 21
30 28 22
31
Factor 4: Appreciation of technical excellence marred by a weak narrative
(mediated mode)
−4 −3 −2 −1 0 1 2 3 4

7 1 18 3 2 4 9 10 15
32 11 19 21 8 5 25 13 31
12 20 22 17 6 29 16
23 24 14
28 26 30
27

ranked at +2, +3 and +4 levels were originally classified by the researchers

as belonging to the same mode. In the case of Factor 2, there is an even
split between statements initially classified as discursive and mediated, with
a slight difference in the ratings in favour of the discursive. However, when
the particular statements preferred by these respondents are read in relation
to each other and interpreted holistically, the basis for our interpretation of
Factor 2 as a discursive response becomes more clearly evident.

FACTOR 1: NARRATIVE TRANSPORTATION WITH EMOTIONAL RESONANCE

(TRANSPARENT MODE)
The largest group of 47 subjects (53% of all significant Q-sorts) expressed
a high degree of narrative transportation, suspension of disbelief, emotional
 126 C. Michelle et al.

engagement, and strong agreement with the film’s core messages. These
characteristics suggest Factor 1 reflects the adoption of a transparent mode
of reception. This group of respondents felt transported to an amazing
new world—Pandora—in which they became pleasurably submerged and
engrossed (+3 on #30). They were mesmerized by the beauty and realism
created through the film’s visual effects (+2 on #25), likening it to a moving
work of art (+2 on #16), and their overall experience of the film was exhil-
arating and awe inspiring (+2 on #29). For example, one respondent said
the following:
I loved Pandora, the visuals of it, the floating mountains, the stone
arches, the bioluminescence, everything! Experiencing Pandora was an
entirely new experience for me. I was completely submerged in this
beautiful and breathtaking world . . . .
Downloaded by [University of Waikato] at 15:10 29 May 2012

—17-year-old Irish man

For these subjects, the beauty and visual hyperrealism of the film
undoubtedly facilitated full immersion in the viewing experience, to the
point where Pandora and the Na’vi were temporarily ascribed the status of
real life and related to as though real. Having become drawn into the film in
the intended manner, these viewers experienced intense emotional involve-
ment with the characters and scenario depicted, and most strongly agreed
with statement 26 (+4), which expresses a sense of caring deeply for the
world of Pandora and its inhabitants:
First I was awestruck by the visuals, but soon became very attached to
these characters and cared deeply for their well-being. I shed tears a few
times and I don’t normally for movies.
—38-year-old Canadian man

For some respondents, their emotional response was unexpectedly strong

and unprecedented:
This film reached me on a deep level, one that I cannot totally explain.
—17-year-old American woman
At first I was a little skeptical about Avatar, but after destruction of the
Tree of Voices and then Home tree there were nothing but tears—and
then ∗ RAGE.∗
—20-year-old Russian man
I could not stop thinking about this film after viewing it. It became very
intense . . . . I spent a great deal of time learning all I could about the
film and even joined a forum, something I have never done. It moved
me to see it 11 times in the theatre, no other film has done that.
—38-year-old Canadian man

Factor 1 respondents were so transported into the world of Pandora

that they felt sad at the close of the film, and expressed a desire to become
Na’vi (+3 on #32):
 Variation in Audience Response 127

My life felt like nothing compared to being a Na’vi on Pandora. I fell into
sort of a depression, like a lot (most) of the people on the Avatar-forums
—18-year-old Canadian man

Reflecting their adoption of the preferred viewing position, these view-

ers read “with” the text and expressed strong agreement with the film’s
core messages. They very strongly agreed with statement 5, “Avatar con-
veys a very important message: Here on Earth we are destroying the natural
world that our species relies on, and we need to change our ways,” (+4)
and also statement 6, “I appreciated Avatar’s key message that everything
is connected—all human beings to each other, and us to the earth” (+3).
Many of these respondents strongly aligned themselves with Avatar’s envi-
ronmental messages, offering evidence of a relationship between transparent
receptions and story-consistent attitudes and suggesting Avatar may indeed
Downloaded by [University of Waikato] at 15:10 29 May 2012

have had the transformative potential Cameron intended:

I admired the fact they would die to protect the forest, everything about
the Na’vi took my life by storm forcing me to try and change my ways.
—21-year-old Mexican man living in the United States

We might not have the “physical” connection to our planet like the Na’vi
do, but if we were to look after it like we should, Earth could be just
as beautiful. Why destroy it? Rather than looking for other planets in the
universe that can sustain life, why not look after the one we have now?
—-21-year-old Australian woman

Respondents expressing this viewpoint strongly disagreed with state-

ments discounting the film’s value on the grounds of predictability, senti-
mentality, clichéd and derivative storyline, and commercial orientation (−4
on #11; −3 on #9, and #14). They most strongly rejected the notion that
Avatar was overhyped (−4 on #12) and instead believed the film fully
deserved to be successful, given its overall excellence, with many consid-
ering Avatar a masterpiece. More important than technical excellence or
aesthetic value for these viewers was the film’s emotional power, ability
to transport them (in many cases repeatedly) to a new world, and vital
environmental message, all consistent with the theorized transparent mode.

FACTOR 2: OPPOSITION TO FORM / NEGOTIATION OF MESSAGE CONTENT

(DISCURSIVE MODE)
Sixteen respondents (20% of the 89 significant sorts) expressed the view that
while Avatar contained serious messages, it was basically boring commer-
cial hype that recycled material and themes already abundantly exploited
by Hollywood. They most strongly agreed with statement 14: “I found
Avatar to be trite, predictable, two-dimensional, and overly sentimental”
 128 C. Michelle et al.

(+4) and also with statement 9: “It’s a shame the storyline was so clichéd
and unoriginal; Avatar reminded me of other films I’ve seen” (+3). They
also expressed agreement at +2 level with statements 11 and 12, suggesting
Avatar has been greatly overrated and is not worthy of viewers’ time and
attention, having primarily been made for commercial purposes. Although
these findings might suggest a critical mediated mode is being adopted, this
viewpoint reveals more immediate concern with the message content of this
film (statements 1, 2, 3, 4, 5, and 6 all register either significant agreement
or disagreement—see Appendix A), and respondents are clearly critical of
certain aspects of those messages. Overall, this viewpoint appears to reflect
a negotiated or even oppositional position in relation to Avatar’s message
content and form, and thus appears consistent with the theorized discursive
mode of reception.
Downloaded by [University of Waikato] at 15:10 29 May 2012

This paradoxical position can be seen in the high level of agreement

(+4) with statement 2: “Avatar expresses the White Messiah myth where
some White guy becomes the ‘most awesome’ member of a non-White
culture, and was quite patronizing.” In the case of Factor 2 respon-
dents, the reiteration of the White Messiah myth is read discursively as
having overtly racist elements, and not merely as an overused narrative
trope:

It’s true—the indigenous people themselves only resist because of the

hero, who is white. No coincidence all the Na’vi were played by black
actors.
—48-year-old Briton living in Singapore
That was the point of the movie. Athletic spectacular White man
outshines the indigenous.
—18-year-old American woman

Yet, there is also a significant level of agreement (+3) with statement

5 addressing the need for greater conservation of Earth’s natural resources.
This viewpoint also rated positively statement 4 (+3) regarding the need
to protect indigenous cultures, and statement 6 (+2) regarding the con-
nectedness of all life. Positive ratings of these statements suggest that these
respondents are primarily engaged with the film’s message content, accept-
ing certain aspects of it while questioning others. On the whole, these
respondents express partial, qualified agreement with textual messages and
many are quite critical of the film’s representations and implied moral
order:

Obviously, humans (Americans, capitalists, army, government) were the

bad guys in the movie, and the nature-worshiping aborigines who lived
in harmony with other creatures were the good guys.
—28-year-old Israeli woman living in the United States
 Variation in Audience Response 129

Part of this more circumspect response to Avatar seems to reflect a

perception of the film as overly simplistic and inadequate to the task of
communicating serious messages effectively, or in a sufficiently sophisti-
cated and thus interesting manner. As one respondent remarked, “It’s a
crude mix of Native American history and eco-conscious story presented
like an average action movie” (42-year-old German man). This viewpoint is
also unique in failing to respond to Avatar’s technical execution and visual
beauty. For these viewers, the film was not at all exhilarating or inspir-
ing (−4 on #29) and they were not moved by the plight of the Na’vi,
nor did they want to be like them (−3 on #26; −4 on #32). Further,
they strongly rejected the notion that Avatar has any personal relevance
to them (−3 on #18, −2 on #20). In sum, Factor 2 expresses critical
distance and emotional detachment from the film, and is most strongly
characterized by discursive engagement with its ideological or message
Downloaded by [University of Waikato] at 15:10 29 May 2012

content.

FACTOR 3: APPRECIATION OF REAL - WORLD RELEVANCE ( REFERENTIAL MODE )

Thirteen (15%) of the 89 significantly loaded Q-sorts expressed positive or

negative variants of Factor 3, which in its positive variation shares with
Factor 1 the sense that Avatar conveys an important message, but focuses
on the film’s messages relating to imperialism and environmental destruction
here on Earth, not on a Pandoran utopia. Avatar was thus interpreted from
a perspective framed by recognition and assessments of textual realism, and
as an allegory for real-world events, consistent with the theorized referential
mode of reception. This group of subjects agreed most strongly (+4) with
statements 17 and 23: “Avatar’s scenario of economic development at the
expense of people and the environment is very similar to what is happening
in my own community today,” and “Avatar is a realistic depiction of how
Western imperialists have subjugated indigenous peoples around the world.”
Factor 3 thus emphasized the realism of Avatar’s portrayal of indigenous
people on Earth, and of real-world social conflict in the context of Capitalist
imperialism, with environmental destruction as a secondary consequence
(+3 on #5). These viewers also agreed with statements alluding to Avatar’s
historical and contemporary relevance (+3 on #24) and they expressed
strong agreement with Avatar’s message about “needing to protect indige-
nous cultures against imperialism, militarism, and economic exploitation”
(+3 on #4).
Uniquely for these viewers, Avatar’s scenario of economic development
at the expense of people and the environment was perceived as similar
to events happening within their own communities (+2 on #18 and #19).
Several Factor 3 respondents expressed personal experience of the conflicts
portrayed in Avatar:
 130 C. Michelle et al.

I’m from Mexico, and many of the current problems are due to Mexicans
who control in concert with transnational companies the life of Mexico
without taking care of people and environment.
—33-year-old Mexican man living in Singapore
In my community [Pondoland] people have sacrificed their lives, people
are still prepared to die for their land. Everything has been done to stop
the mining. Lawyers have been organised to represent us. Marches have
been done to protest against the decision. Media has been used at all
fronts. We have not given up.
—41-year-old South African man

As with those associated with Factor 1, Factor 3 respondents disagreed

with statements critiquing the film’s commercialism, triteness, unwarranted
hype and oversentimentality (−4 on #11 and #12; −3 on #9 and #14).
Downloaded by [University of Waikato] at 15:10 29 May 2012

However, these viewers did not respond to the film’s spectacular and mes-
merizing special effects (−1 on #13 and #15; 0 on #16, #25, and #31), nor did
they experience the same depths of narrative transportation and emotional
engagement that characterized Factor 1 respondents (−1 on #30), again sug-
gesting a more distanced, cognitive rather than emotional response. They
disagreed with statements suggesting that Avatar expresses mainly myth-
ical themes (−2 on #2 and #3) and they did not feel sad at the end of
the film or wish to become Na’vi (−2 on #32). Nonetheless, it is clear that
these viewers considered Avatar worthy of praise and attention because of
its contemporary real-world relevance. Three subjects correlated negatively
with this factor, supporting the composite model’s claim that in a referen-
tial mode, viewers can also reject the text as a highly unrealistic depiction,
drawing on their own contrary understanding of social reality.

FACTOR 4: APPRECIATION OF TECHNICAL EXCELLENCE MARRED BY A WEAK

NARRATIVE (MEDIATED MODE)

Subjects correlated with Factor 4 celebrated Avatar’s technical prowess and

the resulting visual spectacle, while lamenting weaknesses in the storyline
and characterization. These characteristics render this factor consistent with
a mediated mode of reception. Thirteen respondents (15% of significant
Q-sorts) expressed this view. They most strongly agreed with statements
referencing the film as “a major technical accomplishment” with special
effects that were “spectacular and groundbreaking” (+4 on #15) and also
believed “Avatar was implemented excellently and has opened up the 3D
film market, like it was designed to do” (+3 on #13). Such sentiments are
reflected here:

The film was a long time in the making with substantial investment in
researching and developing enabling technology especially 3D cameras.
 Variation in Audience Response 131

The whole package came together well and seemed very life-like, natural
and smooth, including the action sequences and trip through the jungle.
—45-year-old Australian man

This group of respondents also expressed strong agreement with state-

ments 10 (+3), 16 (+3), and 31 (+4), which collectively emphasize the film’s
visual beauty, aesthetic qualities, excitement and power, making Avatar
“worth the price of admission” (#31). They felt “mesmerized by the amaz-
ingly detailed, highly realistic world of Pandora and all the creatures living
in it” (+2 on #25) and found the experience of viewing exhilarating and
inspiring (+2 on #29). However, these respondents also found the film lack-
ing in certain aspects relating to its narrative execution, as one respondent
explained: “Avatar is first and foremost a technical masterpiece. The story
plays second fiddle to mesmerizing computer-generated characters, effects
Downloaded by [University of Waikato] at 15:10 29 May 2012

and scenery” (17-year-old Singaporean man; emphasis added). Thus, this

viewpoint regards Avatar as having been somewhat let down by Cameron’s
clichéd and unoriginal storyline (+2 on #9):

I’d have loved to have seen the world of Pandora act as a setting for a
better story—it deserves it. All that work, all that intellectual and artis-
tic effort, only to use some of the most clichéd science fiction tropes
ever—seriously, “unobtanium?” That one’s been overplayed even for
camp value.
—24-year-old American man
Avatar is a rehash of Pocahontas and Dances with Wolves, with an
added sci-fi element that doesn’t alter the old “White Messiah” story
significantly.
—17-year-old Singaporean man

It is important to note that criticisms of the White Messiah trope

expressed by these respondents were primarily expressed in the con-
text of disparaging the film’s narrative quality and uniqueness, and did
not appear to reflect concerns about the racist aspects of that depiction.
Indeed, for these viewers, serious engagement with the message content
of Avatar was subordinated to evaluations of Avatar’s form and aesthetic
quality. Thus, they expressed neutrality or very mild agreement regard-
ing the film’s core messages about protecting indigenous peoples against
imperialism and saving the environment (0 on #8; +1 on #4 and #5).
Further, these respondents strongly disagreed with claims that Avatar con-
stitutes a critique of the War on Terror (−4 on #7) or a negative portrayal
of the American military (−3 on #1), perhaps because from their per-
spective, the film carried few if any serious messages whatsoever. They
further disagreed with statements affirming the personal relevance of the
film (−2 on #18, #19, and #20). Although greatly appreciating the film
 132 C. Michelle et al.

as an impressive visual spectacle and finding it a powerful piece of

Hollywood filmmaking, these viewers were not sad when it was over and
did not wish to be Na’vi (−4 on #32). They expressed neutrality on the
issues of emotional engagement and identification with story characters
and the plight of the Na’vi (−1 on #28; 0 on #26, #27). This muted emo-
tional response appears related to the text’s deficient characterization and
storyline:

Truth be told, it was hard to muster any emotional involvement in

the film. There was never any doubt how the story would end, who
would live, who it might be permissible to kill, who would ultimately
be defeated . . . Too many clichés prevented me from actually being
interested in the ways characters might develop emotionally or change
their worldviews.
Downloaded by [University of Waikato] at 15:10 29 May 2012

—24-year-old American man

Linking Divergent Receptions and Social Location

On its own, Q methodology is not suitable for estimating the distribution
of holders of viewpoints throughout a population. We cannot estimate, for
example, how many individuals within the wider global audience appre-
hended Avatar in a transparent mode. Q methodology only permits us
to claim that in the population who participated in the study, four dis-
tinct subjective orientations were revealed that appear to closely resemble
the four modes of reception theorized in the composite model, and that if
our research procedure were repeated with a similar population of respon-
dents, these four modes would likely be uncovered again. We note also that
in an independent study of receptions of the animated short film, Ryan,
four factors were identified that similarly resemble the theorized modes,
despite these researchers having drawn on a radically different concep-
tual schema—Holbrook’s framework of consumer value—in selecting their
Q-sample statements (see Davis & Vladica, 2010).
We can, however, gain insight into the social characteristics of the hold-
ers of each of the identified viewpoints in the population we examined,
and conjecture that these characteristics are germane to the media recep-
tion experience communicated to us by each of the respondents. However,
because the numbers of respondents who loaded on Factors 2, 3, and 4 are
relatively small, it is not possible to make any general observations based on
the associations between modes and social characteristics among our partic-
ipants; further research is clearly needed in this area. One-way analysis of
variance tests identified only two statistically significant differences among
the four factors: The number of respondents viewing the film in 3D, and the
number who had purchased or intended to purchase the film on DVD or
Blu-Ray.
 Variation in Audience Response 133

Selected social characteristics of the 89 significant respondents are pre-

sented in Table 2. Those who expressed the transparent mode (Factor 1)
were on average slightly younger, less highly educated, predominantly men
(79%), and were the group most likely to have undertaken military service
(49%). Youthfulness combined with lower levels of education potentially
means these respondents had less extratextual knowledge available to

TABLE 2 Q-Sort Subject Characteristics

One way
analysis of
Subject Total Factor 1 Factor 2 Factor 3 Factor 4 variance test of
characteristics (N = 89) (n = 47) (n = 16) (n = 13) (n = 13) significanced

Average age 29 27 29.3 33.4 30.8 .63

Downloaded by [University of Waikato] at 15:10 29 May 2012

Female (%) 25.9 21.3 43.8 23.1 15.4 .50

Ethnic 11.2 8.5 18.7 23.1 0 .68
minorities (%)
Viewings (M) 6.7 11.3 1.3 4.3 1.6 .46
Viewing Avatar in 76.4 78.7 68.8 61.5 92.3 .005
3D (%)
Purchased or 51.7 83.0 0 30.8 23.1 .03
intend to
purchase DVD
or Blu-Ray (%)
Years of 14.5 13.5 15.2 16.4 15 .53
education (M)
Managerial, 40.4 36.1 43.7 53.8 38.4 .29
semiprofessional
or professional
occupations (%)a
Nonprofessional 2.2 4.2 0 0 0 .72
occupations (%)b
Students (%) 50.6 51.1 56.2 38.5 53.8 .12
Lower incomes 9.1 13.0 0.0 0.0 16.7 .59
(nonstudents:
n = 4) (%)c
Conservative/ 2.2 4.2 0 0 0 .72
republican
political
affiliations (%)
Politically 38.2 29.8 62.5 53.8 23.1 .34
active (%)
Religion is very 24.7 25.5 12.5 38.5 23.1 .44
important (%)
Military service (%) 42.7 48.9 37.5 30.8 38.5 .14
a
Occupations included: creative artist or musician, media producer, graphic designer; government official;
manager or business executive; professional such as school teacher, nurse, accountant, or public servant;
higher professional such as doctor, lawyer, professor, scientist, or engineer.
b
Occupations included: unskilled manual worker; skilled manual worker; tradesperson; small business
owner; clerical or administrative worker; sales worker; information and communications technology
worker; call centre worker.
c
Defined as lower income/unpaid; or lower middle income. The other income categories were middle
income; higher-middle income; and high income.
d
Calculated on the basis of actual occurrences, not on the basis of percentages.
 134 C. Michelle et al.

inform referential or discursive readings, while military experience and gen-

der may have rendered this group more likely to personally identify with
the film’s hero, Jake, a young U.S. marine. It is remarkable that Factor 1
respondents report having viewed the film an average of 11 times; repeated
pleasurable consumption of the same film narrative appears consistent with
the high level of narrative transportation reported by this group of viewers.
These respondents also reported lower levels of political activity than those
associated with Factors 2 and 3.
The discursive mode (Factor 2) was expressed by twice as many women
as the other viewpoints (44%). These respondents had the lowest level of
declared importance of religion (12.5%), the second lowest reported rate
of viewing the film in 3D (69%), and the highest reported level of political
activity (62.5%). A relatively high proportion (19%) of these respondents self-
identified as ethnic minorities. The latter two characteristics may be related
Downloaded by [University of Waikato] at 15:10 29 May 2012

to this group’s negotiated position regarding the meanings and messages of

the film. Most holders of this viewpoint saw the film only once.
Those associated with Factor 3, the referential mode, were on average
older and better educated than those expressing the other viewpoints, and
were the second most politically active group at 54%. These respondents had
the highest proportion of persons who self-identified as belonging to an eth-
nic minority (23%) and also the highest proportion of individuals for whom
religion was “very important” (38.5%), suggesting that political, religious and
ethnic affiliations all provide frames of reference for media response.
Last, the mediated mode (Factor 4), which emphasized the technical
excellence of the film, was the group most strongly dominated by men
(85%). This group had the highest rate of viewing in 3D (92%) and the lowest
level of political activism (23%) of the four viewpoints, which may help
explain the expressed neutrality of this group in relation to core message
content. No self-identified ethnic minorities expressed this viewpoint, again
suggesting that ethnicity was an important frame of reference for viewers
of Avatar, encouraging the adoption of discursive and referential modes of
reading.
The associations evident here are, however, particular to this sample.
Further research is needed to ascertain whether these kinds of links between
preferred modes of reception and aspects of identity and social location are
consistently evident in the same ways, or, as we suspect is more likely the
case, textually and contextually contingent.

DISCUSSION

Our results provide considerable empirical support for the composite model
(Michelle, 2007). Factor analysis revealed four distinct subjective orientations
characterized by differing degrees of emotional and cognitive involvement,
 Variation in Audience Response 135

as well as different foci in terms of the most salient issues and concerns for
viewers. These factors in turn exhibit notable similarities with the four modes
of reception theorized in the composite model. Whereas Factor 1, which we
interpret as the transparent mode of reception, is marked by suspension
of disbelief, feelings of being transported to the amazingly realistic world
of Pandora, emotional engagement with the Na’vi’s plight, and agreement
with preferred messages, Factor 2, the discursive mode, exhibits estrange-
ment and emotional detachment, with these viewers rejecting Avatar as
an overcommercialized Hollywood entertainment product while also engag-
ing critically with the film’s ideological or message content. While Factor
3, the referential mode, focuses cognitive attention on the film’s similar-
ity and relevance to past and present struggles occurring in the real world
against Western imperialism, militarism, and capitalist exploitation of natu-
ral resources, Factor 4, the mediated mode, relates primarily to Avatar as
Downloaded by [University of Waikato] at 15:10 29 May 2012

a constructed entertainment media product which is aesthetically pleasing

and technologically remarkable, but has significant shortcomings in terms of
script and storyline. Each mode reflects a very different subjective approach
to making sense of the same polysemic and polyvalent text, and focuses on
different levels of denotative and connotative meaning. The adoption of dif-
ferent modes of reception in turn produces considerable, but still clearly
patterned, forms of interpretive divergence among different segments of
our sample audience, with these patterns and the core distinctions between
subjective viewpoints being accurately predicted by the composite model.
Although this study represents our first attempt at operationalizing the
modes and their subcategories in this way, it nonetheless provides empir-
ical support for the composite model as an accurate, comprehensive, and
potentially widely applicable schema of audience engagement and response.
However, there are some limitations to this research and potential criti-
cisms that could be raised; some of which have been anticipated earlier.
Some may argue that the seemingly close fit between the four theorized
modes of reception and the identified factors is a methodological artifact
stemming from subjective biases influencing the selection of Q-sample state-
ments, which, having been chosen in part to reflect different aspects of
the model, may have predetermined the outcome of our research—perhaps
by limiting the ability of respondents to express their own unique views.
However, again we stress that the selection of statements was guided in the
first instance by Q’s methodological imperative to represent the diversity of
ideas and themes already existing within the wider concourse (Brown, 1980).
Although it was not possible to capture every issue or theme, we believe
the statements reflect forms of response to Avatar that were commonly
articulated, as well as more marginal viewpoints. Further, while respondents
may have been prevented from articulating a radically divergent perspective
using the statements on offer, the potential for variation was more than suf-
ficient for most individuals to complete a Q-sort that broadly reflected their
 136 C. Michelle et al.

views. Participants were invited to comment on the statements they ranked

most highly and lowly to further clarify their perspectives, and to add addi-
tional comments at the end of the questionnaire. These comments largely
confirmed the salience of the issues and themes selected for inclusion. In
the case of this research, only 5 out of 120 participants constructed Q-sorts
that did not load significantly on any of the factors; 26 constructed mixed or
confounded sorts that loaded on more than one factor, possibly as a result
of their commuting between modes.
It is also important to note that although our focus has been on iden-
tifying similarities in the subjective modes of response of participants, Q
methodology retains the capacity to analyze respondents’ sorts collectively
and individually. Hence, it is possible to identify similarities between indi-
viduals through factor analysis, while also determining which individuals
had truly unique responses. The Q-sorts and comments of those individuals
Downloaded by [University of Waikato] at 15:10 29 May 2012

could then be subjected to intensive analysis and perhaps explored further

through individual interviews. Although this has not been our focus in this
article, identifying and analyzing the truly divergent response is vitally impor-
tant in terms of extending and refining our theoretical conceptualization of
the reception process. To date, however, largely because of the absence of
an accepted analytical schema of predominant modes of reception, it has
been impossible to systematically differentiate between what is and is not a
truly atypical form of audience engagement and response.
Also, as noted earlier, regardless of the statements chosen, there remains
significant potential for the expression of diverse viewpoints using Q, as
each statement can be interpreted differently and meanings may shift and
change when statements are read in relation to each other. The number of
possible combinations of 32 statements within the rank ordered distribution
is, for all practical purposes, infinite5 . That 89 different individuals indepen-
dently ranked the 32 statements in ways that were significantly similar to the
rankings of a number of others is thus strong evidence of shared subjectiv-
ity, and reveals clear differences in the approach to sense-making among
different segments of our population of respondents.
An additional area of possible critique stems from presenting respon-
dents with a wide range of statements expressing ideas and viewpoints that
might not have otherwise occurred to them, perhaps leading respondents
to reshape their perspective on the film. However, such cross-fertilization
reflects a process that already occurs within the wider social context, and
in focus groups, where people’s initial views may change in response to
hearing the views of others. With Q, there is some safeguard against this
revision process in that, within the context of evaluating and charting their
response to a large number of Q statements, some or many of which may

5
The number of possible variants of a rank-ordered list of 32 statements is 2.63130837 × 10 to the
35th power. If the statements were randomly ordered, it would not be possible to identify factors.
 Variation in Audience Response 137

be highly salient in terms of reflecting respondent’s existing views, it seems

unlikely that previously unconsidered ideas would produce a strong enough
reaction to feature among the statements ranked at either extreme, and it is
these statements that are most significant.
This issue, however, raises a problem in terms of testing the composite
model, because it also implies that respondents’ Q-sorts may not necessarily
reflect an objectively existing psychological orientation toward reception as
such, but rather an acquired cognitive perspective derived from encounters
with the views and responses of others. In response, as mentioned earlier,
the likelihood of multiple individuals independently sorting 32 statements
in significantly similar ways without there being some kind of underlying
affective and cognitive orientation guiding the ranking process is remote. It
is also improbable to imagine respondents are sorting statements according
to their understanding of an already existing, clearly defined perspective on
Downloaded by [University of Waikato] at 15:10 29 May 2012

the film within the public domain, because this article is likely the first to
chart multidimensional viewpoints on Avatar in any detail.
By undertaking further studies that use the same conceptual and ana-
lytical schema provided by the composite model, reception researchers
may have a sounder basis for identifying variations as well as congruen-
cies in the form and content of audience engagement and response across
diverse cultural contexts, and in relation to a variety of texts of different
genres and mediums. Although some readers may balk at the use of an
analytical schema that they perceive as classifying in advance the range
of modes of response that audience members are most likely to adopt in
any given textual encounter, we believe that reception research has now
generated a significant body of descriptive evidence, but has yet to pay ade-
quate attention to how all of this evidence should be collectively evaluated,
synthesized and analyzed (see also Barker, 2006; Schrøder, 2000). Rather
than continuing to proliferate descriptive studies of specific audience-text
encounters—a potentially endless endeavour, given the overwhelming vol-
ume of content production in the current digital media environment, and not
ideally suited to the accumulation of knowledge about audience reception
per se—a different approach is now required, one that seeks to formulate
and refine an analytical model that can given some order and coherence
to seemingly disparate findings from across the field, without obscuring the
specific nuances of each individual case. Indeed, such nuances may be more
clearly perceived once the general and typical are identified and analyt-
ically delineated. Further, because Q methodology is specifically oriented
toward understanding human subjectivity from the unique perspectives of
research subjects in an objective, rigorous way, it offers considerable but still
largely unrealized potential for those scholars who are interested in explor-
ing audience-centered approaches to reception studies, but who also seek
to utilize some of the benefits of quantitative modes of analysis (for further
discussion, see Davis & Michelle, 2011). In that sense, Q may provide one
 138 C. Michelle et al.

useful means of reconciling the disparate approaches currently taken in the

broader cross-disciplinary field of reception studies.

REFERENCES

Acland, C. (2010). Avatar as technological tentpole. Flow TV , 11(6). Retrieved from

http://flowtv.org/2010/01/avatar-as-technological-tentpole-charles-r-acland-
concordia-university
Andacht, F. (2004). Fight, love and tears: An analysis of the reception of Big Brother
in Latin America. In E. Mathjis & J. Jones (Eds.), Big Brother International:
Formats, critics and publics (pp. 123–139). New York, NY: Wallflower Press.
Associated Press. (2010, January 13). Vatican says Avatar is no masterpiece.
3 News. Retrieved from http://www.3news.co.nz/Vatican-says-Avatar-is-no-
Downloaded by [University of Waikato] at 15:10 29 May 2012

masterpiece/tabid/418/articleID/137043/Default.aspx
Associated Press. (2010, February 12). Palestinian protestors pose as Na’vi from
‘Avatar.’ Haaretz.com. Retrieved from http://www.haaretz.com/news/palestin
ian-protesters-pose-as-na-vi-from-avatar-1.263250
Barker, M. (2006). I have seen the future and it is not here yet...; or, on being
ambitious for audience research. The Communication Review, 9, 123–141. doi:
10.1080/10714420600663310
Barker, M., & Mathijs, E. (2005). Understanding vernacular experiences of film in
an academic environment. Art, Design & Communication in Higher Education,
4(1), 49–71.
Barker, M., & Mathijs, E. (2008). Watching The Lord of the Rings: Tolkein’s world
audiences. New York, NY: Lang.
Blumler, J. (1979). The role of theory in uses and gratifications studies. Communi-
cation Research, 6, 9–36.
Brooks, D. (2010, January 7). The Messiah complex. The New York Times. Retrieved
from http://www.nytimes.com/2010/01/08/opinion/08brooks.html
Brown, S. R. (1980). Political subjectivity: Applications of Q methodology in political
science. New Haven, CT: Yale University Press.
Brown, S. R. (1986). Q technique and method: Principles and procedures. In W. D.
Berry & M. S. Lewis-Beck (Eds.), New tools for social scientists: Advances and
applications in research methods (pp. 57–76). Beverly Hills, CA: Sage.
Brown, S. R. (1994). Q methodology as the foundation for a science of subjectivity.
Operant Subjectivity, 18, 1–16.
Busselle, R., & Bilandzic, H. (2008). Fictionality and perceived realism in expe-
riencing stories: A model of narrative comprehension and engagement.
Communication Theory, 18, 255–280. doi: 10.1111/j.1468-2885.2008.00322.x
Cameron, J. (2010). Interview with Charlie Rose. Retrieved from http://www.
charlierose.com/view/interview/10866#frame_top
Clarke, J. (2010, January 18). Comment posted by John Clarke, Jan 18th, 2010
3:22 pm. Scene-Stealers. Retrieved from http://www.scene-stealers.com/print-
reviews/movie-review-avatar
Cohen, J. (2001). Defining identification: A theoretical look at the identification of
audiences with media characters. Mass Communication & Society, 4, 245–264.
 Variation in Audience Response 139

Condit, C. (1989). The rhetorical limits of polysemy. Critical Studies in Mass

Communication, 6, 103–122.
Corner, J., & Richardson, K. (1986). Documentary meanings and the discourse
of interpretation. In J. Corner (Ed.), Documentary and the mass media
(pp. 141–160). London, England: Edward Arnold.
Dahlgren, P. (1988). What’s the meaning of this? Viewers’ plural sense-making of TV
news. Media, Culture & Society, 10, 285–301.
Davis, C. H., & Michelle, C. (2011). Q methodology in audience research: Bridging
the qualitative/quantitative ‘divide’? Participations: Journal of Audience and
Reception Studies, 8, 559–593.
Davis, C. H., & Vladica, F. (2010). Consumer value and modes of media reception:
audience responses to the computer animated psychorealist documentary Ryan
and its own documentation in Alter Egos. Palabra Clave, 13(1), 13–30.
Ebert, R. (2009, December 11). Avatar. Chicago Sun-Times. Retrieved from http://
rogerebert.suntimes.com/apps/pbcs.dll/article?AID=/20091211/REVIEWS/91211
Downloaded by [University of Waikato] at 15:10 29 May 2012

9998
EdmontonJournal.com (2010, March 4). Canadian firms upset with oilsands-
slamming ad in Variety. Retrieved from http://www.edmontonjournal.com/
business/Canadian+firms+upset+with+oilsands+slamming+Variety/2642388/
story.html
Esrock, S. (2005). Review and criticism: Research pioneer tribute—William
Stephenson: Traveling an unorthodox path to mass communication dis-
covery. Journal of Broadcasting & Electronic Media, 49, 244–250. doi:
10.1207/s15506878jobem4902_8
Friedman, S. (2006). Watching Twin bracelets in China: The role of spectator-
ship and identification in an ethnographic analysis of film reception. Cultural
Anthropology, 21, 603–632. doi: 10.1525/can.2006.21.4.603
Gates, C. (2010, January 1). Avatar recycles indigenous ‘stereotypes.’ Retri-
eved from http://www.stuff.co.nz/entertainment/film/3201437/Avatar-recycles-
indigenous-stereotypes
Glaser, M., Garsoffky, B., & Schwan, S. (2012). What do we learn from docutain-
ment? Processing hybrid television documentaries. Learning and Instruction,
22(1), 37–46. doi: 10.1016/j.learninstruc.2011.05.006
Green, M., & Brock, T. (2000). The role of transportation in the persuasiveness of
public narratives. Journal of Personality and Social Psychology, 79, 701–721.
Green, M., & Brock, T. (2002). In the mind’s eye: Transportation-imagery model
of narrative persuasion. In M. Green, J. Strange, & T. Brock (Eds.), Narrative
impact: Social and cognitive foundations (pp. 315–341). Mahwah, NJ: Erlbaum.
Green, M., Brock, T., & Kaufman, G. (2004). Understanding media enjoyment:
The role of transportation into narrative worlds. Communication Theory, 14,
311–327. doi: 10.1111/j.1468-2885.2004.tb00317.x
Hall, S. (1980). Encoding/decoding. In S. Hall, D. Hobson, A. Lowe, & P. Willis
(Eds.), Culture, media, language (pp. 128–138). London, England: Hutchinson.
Hill, A. (2005). Reality TV: Audiences and popular factual television, London,
England: Routledge.
Hopkins, K. (2010, February 8). Indian tribe appeals for Avatar director’s help to stop
Vedanta. The Guardian. Retrieved from http://www.guardian.co.uk/business/
2010/feb/08/dongria-kondh-help-stop-vedanta
 140 C. Michelle et al.

Internetworldstats.com. (2009). Internet World Stats: Usage and population statistics.

Retrieved from http://www.internetworldstats.com/stats.htm.
Johnson, A., de Rio, E., & Kemmitt, A. (2010). Missing the joke: A reception anal-
ysis of satirical texts. Communication, Culture & Critique, 3, 396–415. doi:
10.1111/j.1753-9137.2010.01077.x
Kerlinger, F. (1972). Q methodology in behavioral research. In S. R. Brown & D.
J. Brenner (Eds.). Science, psychology, and communication: Essays honoring
William Stephenson (pp. 3–38). New York, NY: Teachers College Press.
Kuipers, G., & de Kloet, J. (2009). Banal cosmopolitanism and The Lord of the Rings:
The limited role of national differences in global media consumption. Poetics,
37, 99–118. doi:10.1016/j.poetic.2009.01.002
Lee, F. (2006). Cultural discount and cross-culture predictability: Examining the
box office performance of American movies in Hong Kong. Journal of Media
Economics, 19, 259–278. doi: 10.1207/s15327736me1904_3
Levy, M., & Windahl, S. (1984). Audience activity and gratifications: A conceptual
Downloaded by [University of Waikato] at 15:10 29 May 2012

clarification and exploration. Communication Research, 11, 51–78.

Liebes, T., & Katz, E. (1989). On the critical abilities of television viewers. In E.
Seiter, H. Borchers, G. Kreutzner, & E. Warth (Eds.), Remote control: Television,
audiences and cultural power (pp. 204–221). London, England: Routledge.
Liebes, T., & Katz, E. (1990). The export of meaning: Cross-cultural readings of
Dallas. New York, NY: Oxford University Press.
Livingstone, S. (2003). On the challenges of cross-national comparative media
research. European Journal of Communication, 18, 476–500. doi: 10.1177/
0267323103184003.
McKeown, B. F., & Thomas, D. B. (1988). Q methodology (Quantitative Applications
in the Social Sciences series, Vol. 66). Newbury Park, CA: Sage.
Michelle, C. (2007). Modes of reception: A consolidated analytical framework. The
Communication Review, 10, 181–222.
Michelle, C. (2009). On (re)contextualising audience receptions of reality TV.
Participations: Journal of Audience and Reception Studies, 6, 137–170.
Morley, D. (1980). The “nationwide” audience. London, England: British Film
Institute.
Morley, D. (2006). Unanswered questions in audience research. The Communication
Review, 9, 101–121. doi: 10.1080/10714420600663286
Neuman, W. (1982). Television and American culture: The mass medium and the
pluralist audience. Public Opinion Quarterly, 46, 471–487.
Notle, J. (2009, December 11). Review: Cameron’s ‘Avatar’ is a big, dull,
America-hating, PC revenge fantasy. Big Hollywood. Retrieved from http://
bighollywood.breitbart.com/jjmnolte/2009/12/11/review-camerons-avatar-is-a-
big-dull-america-hating-pc-revenge-fantasy
Olson, S. (2004). Hollywood planet: Global media and the competitive advantage
of narrative transparency. In R. C. Allen & A. Hill (Eds.). The television studies
reader (pp. 111–129). London, England: Routledge.
Podhoretz, J. (2009, December 28). Avatarocious. The Weekly Standard. Retrieved
from http://www.weeklystandard.com/Content/Public/Articles/000/000/017/350
fozta.asp
Richardson, K., & Corner, J. (1986). Reading reception: Mediation and transparency
in viewers’ reception of a TV programme. Media, Culture & Society, 8, 485–508.
 Variation in Audience Response 141

Ruggiero, T. (2000). Uses and gratifications theory in the 21st century. Mass
Communication and Society, 3(1), 3–37.
Schrøder, K. (1986). The pleasure of Dynasty: The weekly reconstruction of self-
confidence. In P. Drummond & R. Paterson (Eds.), Television and its audience:
International research perspectives (pp. 61–82). London, England: British Film
Institute.
Schrøder, K. (2000). Making sense of audience discourses. European Journal of
Cultural Studies, 3, 233–258. doi: 10.1177/136754940000300205
Staiger, J. (2000) Perverse spectators: The practices of film reception. New York, NY:
New York University Press.
Stenner, P., & Marshall, H. (1995). A Q methodological study of rebel-
liousness. European Journal of Social Psychology, 25, 621–636. DOI:
10.1002/ejsp.2420250603
Stephenson, W. (1995–1996). Quantum theory media research: 1. Reception analysis.
Operant Subjectivity, 19, 1–11.
Downloaded by [University of Waikato] at 15:10 29 May 2012

Stephenson, W. (1976). Q methodology: Conceptualization and measurement of

operant effects of television viewing. JCATS: Journal of the Centre for Advanced
Television Studies, 4, 17–18.
Stephenson, W. (1978). Applications of communication theory IV. Immediate
experience of movies. Operant Subjectivity, 1, 96–116.
Suckfüll, M., & Scharkow, M. (2009). Modes of reception for fictional films.
Communications, 34, 361–384. doi: 10.1515/COMM.2009.023
Thottam, J. (2010, February 13). Echoes of Avatar: Is a tribe in India
the real-life Na’vi? Time. Retrieved from http://www.time.com/time/world/
article/0,8599,1964063,00.html
Turan, K. (2009, December 17). Review: ‘Avatar’ restores a sense of won-
der to the moviegoing experience. Los Angeles Times. Retrieved from
http://herocomplex.latimes.com/2009/12/17/avatar-restores-a-sense-of-
wonder-to-the-moviegoing-experience
Washington, J. (2010, January 11). ‘Avatar’ critics see racist theme. Associated Press.
Retrieved from http://www.huffingtonpost.com/2010/01/10/avatar-critics-see-
racist-theme_n_418155.html
Watts, S., & Stenner, P. (2005). Doing Q methodology: Theory, method
and interpretation. Qualitative Research in Psychology, 2, 67–91. doi:
10.1191/1478088705qp022oa
Wilson, T. (1995). Watching television: Hermeneutics, reception and popular culture.
Cambridge, England: Polity Press.
Worth, S., & Gross, L. (1974). Symbolic strategies. Journal of Communication, 24(1),
27–39.
Zizek, S. (2010, March 4). Return of the natives. New Statesman. Retrieved from
http://www.newstatesman.com/film/2010/03/avatar-reality-love-couple-sex
 142 C. Michelle et al.

APPENDIX
Scored Statements

Factor 1 2 3 4

1. My enjoyment of Avatar was marred by its −3 −3 −3 −3

negative portrayal of the U.S. military, which I
felt was anti-American.
2. Avatar expresses the White Messiah myth −2 4 −2 0
where some White guy becomes the “most
awesome” member of a non-White culture, and
was quite patronising.
3. Avatar suggests we should worship nature −2 −2 −2 −1
rather than God, and in this sense is
anti-Christian.
4. I liked the message in Avatar about needing to 1 3 3 1
Downloaded by [University of Waikato] at 15:10 29 May 2012

protect indigenous cultures against imperialism,

militarism, and economic exploitation.
5. Avatar conveys a very important message: here 4 3 3 1
on Earth we are destroying the natural world
that our species relies on, and we need to
change our ways.
6. I appreciated Avatar’s key message that 3 2 2 1
everything is connected; all human beings to
each other, and us to the Earth.
7. Avatar critiques the way the US is conducting −2 1 1 −4
its War on Terror, and that really appealed to
me.
8. I was very interested in, or concerned about, 1 1 1 0
the messages conveyed by Avatar.
9. It’s a shame the storyline was so clichéd and −3 3 −3 2
unoriginal; Avatar reminded me of other films
I’ve seen.
10. Avatar is an exciting, visually arresting, and 0 −1 0 3
occasionally powerful piece of Hollywood
filmmaking.
11. Avatar is a big, dumb movie built to make −4 2 −4 −3
money, but hardly worthy of my time or
attention.
12. Avatar has been greatly overrated and doesn’t −4 2 −4 −3
live up to the hype. I was disappointed, and
didn’t think it was worth the money I paid.
13. Avatar was implemented excellently and has 0 0 −1 3
opened up the 3D film market, like it was
designed to do.
14. I found Avatar to be trite, predictable, −3 4 −3 1
two-dimensional, and overly sentimental.
15. Avatar is an amazing technical accomplishment. 1 1 −1 4
The computer generated imagery, 3D, and
other special effects were spectacular and
groundbreaking.
16. Avatar is a visual masterpiece. It was like 2 0 0 3
watching a moving work of art.
(Continued)
 Variation in Audience Response 143

APPENDIX (Continued)

Factor 1 2 3 4

17. Avatar’s scenario of economic development at −1 1 4 0

the expense of people and the environment is
very similar to what is happening in my own
community today.
18. I saw in Avatar a powerful reflection of my −1 −3 2 −2
own people’s resistance to those who wish to
destroy our homes and take away our lands.
19. Avatar effectively showed many similarities −1 −1 2 −2
with people and events in my own life.
20. I saw similarities between the Na’vi and my −1 −2 1 −2
own people in terms of our customs, history,
spirituality, or way of life.
21. Watching Avatar, I discovered symbols that −1 −1 1 −1
draw from my own religious or spiritual
Downloaded by [University of Waikato] at 15:10 29 May 2012

worldview.
22. The Na‘vi’s struggle against powerful forces is 0 0 1 −1
mine too, and I felt good for them when they
defeated their high-tech enemy.
23. Avatar is a realistic depiction of how Western 0 1 4 −1
imperialists have subjugated indigenous
peoples around the world.
24. I was particularly struck by Avatar’s relevance 0 0 3 0
to historical or contemporary events.
25. I was completely mesmerised by the amazingly 2 0 0 2
detailed, highly realistic world of Pandora and
all the creatures living in it.
26. I felt very emotionally involved in this film and 4 −3 0 0
truly cared about what happened to Pandora,
and to the Na’vi.
27. I could really relate to the characters, and 1 −2 −1 0
strongly identified with one of them.
28. I identified with the Na‘vi’s plight to the point 0 −1 0 −1
that it made me think, “what if this happened
to me?”
29. Experiencing Avatar was exhilarating and 2 −4 −1 2
inspiring, and left me awestruck.
30. I felt like I was taken to another world, and 3 0 −1 1
became submerged in an amazing new reality.
31. Just the breathtaking visual beauty of Pandora 1 −1 0 4
was worth the price of admission—the huge
trees, the wonderful floating mountains, the
soaring waterfalls.
32. I felt sad when it was over, and wished I could 3 −4 −2 −4
become Na’vi.