The Big Picture for Large-Screen

Television Viewing
For Both Programming and Advertising, Audiences Are
More Attentive, More Absorbed, and Less Critical

Michael D. McNiven Large-screen televisions have gained prominence in the marketplace. Focus groups and a
Cumberland Advisors
national survey were used to investigate viewing of large-screen televisions as they relate to
michaelmcniven@
g.mail.com attitudes toward advertising and the way advertising and programming are viewed. Results
indicate that larger screens positively impact how advertising and television programming
Dean Krugman
University of Georgia are consumed. Large-screen television viewers were less skeptical of advertising than
dkrugman@uga.edu small-screen viewers; more positive toward advertising; and paid more attention to

Spencer F. Tinkham
both commercials and programming. Also, large-screen viewers were more absorbed
University of Georgia in television programming—a phenomenon that mediates the impact of screen size on
stinkham@uga.edu
attention, evaluation, and skepticism toward television advertising.

INTRODUCTION literature, especially to the advertising community

The current study used focus groups and a national interested in the impact of large-screens in present-
online survey to investigate attitudes toward ing information.
advertising and the differences in attention to With the impetus of technological, regulatory,
broadcast and commercial content between large- and market forces, worldwide total sales of large-
screen and traditional-screen television viewers. screen televisions grew 108 percent between 2005
Large-screen televisions have proven to be a and 2007. The tremendous growth of large-screen
viable and high-quality television technological television adoption will continue as flat screens
enhancement that continues to gain prominence in are becoming part of the staples of society and
the television marketplace. Larger screens coupled essential products for the global consumer (Lucas,
with digital reception and wide screens represent a 2010). Large-screen worldwide sales were 33 mil-
significant change from the past in how television lion units in 2007 and were estimated to be 61 mil-
fare is delivered to viewers. Both programming and lion in 2010 (Abowd, 2007). A recent survey of U.S.
advertising are delivered in a visual format that is adults shows that more than half of HDTV sales
superior to traditional television, thus enhancing are in the large-screen category of 40 inches or
the prospects for storing and remembering infor- larger (Crain, 2010). Moreover, U.S. installations
mation (Reeves, Lang, Kim, and Tatar, 1999). for high-end home theaters requiring large screens
Large-screen television adoption at some level have been projected to grow from 190,000 per
indicates a changing relationship—or commit- year in 2008 to 277,000 in 2013 (Weston, 2008). In
ment—toward television. The cost, technical com- the United States, the percentage of large screens
plexity, and advanced technologies attached to increased as older analog televisions were replaced
large screens can be associated with the commit- in conjunction with the FCC’s 2009 change to digi-
ment viewers have to the viewing process. Study- tal transmissions.
ing the impact of large screens as a new, distinct The rapid adoption rate of large-screen tele-
television innovation can make a meaningful and visions indicates a prolonged, important change
relevant contribution to the media-consumption in television’s home infrastructure through the

DOI: 10.2501/JAR-52-4-421-432 December 2012  JOURNAL OF ADVERTISING RESEARCH  421

The Big Picture for Large-Screen Television Viewing

coming decades. A conservative estimate distance showed that, for television In part, these works focused on atten-
of large-screen television product life- images displayed in high resolution, tion to the set and activities that take place
span is 15 to 20 years (“LCD TVs versus the subjects preferred larger displays during the viewing process. For example,
Plasma TVs”, 2007). Even though techno- (Neuman, 1990). Abernethy (1991) analyzed several studies
logical innovations will continue to alter • Another study found that large screens and concluded that viewers avoid approx-
the media landscape, the current forms of (90-inch picture) created greater arousal imately 32 percent of the advertisements.
large-screen units will persist for the fore- and memory in viewers than a 22-inch Krugman et al. (1995) found that viewers
seeable immediate future. screen (Detenber and Reeves, 1996). were visually connected to programming
• A study exploring the effects of screen size 62 percent of the time but only 33 percent
BACKGROUND and message content on attention and of the time during commercials. In relation
Large-screen television owners have the arousal indicated that larger screens ini- to previous studies, they found that atten-
opportunity to enhance their viewing tiated more peripheral vision, which, in tion to programming was relatively stable
experiences based on the quality and reso- turn, resulted in increased input beyond over time. Attention to commercials, how-
lution associated with larger screens. It is the initial visual focus, visual searching ever, had diminished in comparison to
important to understand whether (and is greater to take in all of the information, previous findings. In addition, the authors
how) the positive associations with large- and the larger mental images produced also found a correlation between stronger
screen television extend to program and by larger screens positively influence program viewing and commercial view-
advertising viewing. Prior work examin- recall (Reeves et al., 1999). ing presumably via a “lead-in” effect.
ing attention to television, the absorbing At this juncture, there has been little
nature of television, and attitudes toward From the outset, television long has been work examining advertising within the
advertising help provide a background for noted for its unique “absorbing” quality context of large-screen viewing. Though
understanding the differences in viewing innate to the medium (Schramm, 1961). work has been done to investigate reac-
between large- and small-screen viewers. Large screens accentuate that absorbing tions to large-screen screens themselves
quality, becoming a new variable of “pres- and to examine attention to screen for pro-
Experiments and Lab Studies ence” in the home. gramming and advertising via traditional
With a growing segment of television television viewing (less than 40-inch screen
audiences viewing programming through Presence means feeling and acting like you size), the authors believe that no published
digitally enhanced, wide-screen, and large- are having a non-mediated experience—the work specifically has examined how adver-
screen panels, there exists an immediate media user therefore responds (directly) to tising is regarded by large-screen viewers.
need to better understand the relationship objects, events, and people in that environ- The literature on attitudes and skepti-
between screen size and attention. ment rather than (indirectly) to what seem cism toward advertising is applicable to
Experiments or other forms of testing to be only symbolic representations or rec- large-screen viewing, especially when
in laboratory situations have been used reations of objects, events, and people that compared to small-screen viewing. Con-
to study the impact of large screens. For (if they exist at all) are in fact somewhere sumer and public views have been an
instance: else (Lombard et al., 1997). important part of understanding how peo-
ple relate to and understand advertising.
• One study that examined screen size Visual Attention Studies for Programming Specifically:
as a variable in viewing found that and Advertising
large-screen television viewing pro- Several studies or overviews have used • Pollay and Mittal (1993) modeled and
vides more favorable and more intense visual attention to examine attention to the developed a scale to understand con-
viewing (Lombard, Ditton, Grabe, screen for programming and/or advertis- sumer beliefs about advertising.
and Rich, 1997). Big screens produced ing (Abernethey, 1991; Allen, 1965; Ander- • Obermiller and Spangenberg (1998)
greater memory recall, greater physi- son, Lorch, Field, Collins, and Nathan, argued that skepticism toward adver-
ologic response, and more intense direct 1986; Bechtel, Achelpohl, and Akins, tising impacts whether people believe
response to images on the screen. 1972; Collet and Lamb, 1986; Krugman, such product claims. The authors devel-
• Research that manipulated the variables Cameron, and White, 1995; Krugman and oped and validated a nine-item scale to
of screen size, resolution, and viewing Shamp, 1992; Steiner, 1966). examine skepticism toward advertising.

422  JOURNAL OF ADVERTISING RESEARCH  December 2012

 The Big Picture for Large-Screen Television Viewing

• More recently, Diehl, Mueller, and Ter- during the programming and respondents, were held as part of the first
lutter (2008) used a combination of the commercials than do small- phase of the project. The second phase of
two scales to examine pharmaceuti- screen viewers? the study used a national online survey
cal advertising. An underlying part of yielding N = 1,328 respondents.
these scales assumed that attitudes and An extension of the concept of atten- In a multi-method design, both quali-
skepticism toward advertising provide tion seeks to find out whether large-screen tative and quantitative techniques are to
a background lens by which people television viewing offers a more absorb- be used to study specific research ques-
view the industry and its products ing experience (Lombard et al., 1997). In tions (Wimmer and Dominick, 2006).
(advertisements). short, the large-screen television poten- Each research technique contributes to the
tially offers a more engaging medium depth and understanding of the study’s
RESEARCH QUESTIONS for connection to televised experience. results and, as such, each technique
Given the cost, technical complexity, and This transcendent effect—one that moves employed—either qualitative or quantita-
associated technologies attached to large beyond the notions of attention or absorp- tive—is central to the methodology of the
screens, the authors of the current study tion—could be described as the concept of research. Qualitative techniques provide
initially asked whether there were basic “presence.” general impressions, context, and depth
demographic profile differences between that principally address the questions of
large- and small-screen television viewers. RQ3: Do people feel more absorbed “why” and “how” large- and small-screen
in the television viewing on televisions are used. Information from the
RQ1: What is the demographic pro- a large-screen television than focus groups set up key study areas for
file of large- and small-screen when viewing small-screen the survey. The survey provides much
viewers? television? more generalizable data in understanding
the differences between large- and small-
The experiments and lab studies provided Finally, we know very little about the screen viewers. As such, the quantitative
a solid base to understand how people way large-screen television viewers feel technique describes larger populations
approach large-screen televisions but did about advertising. Understanding atti- and helps answer the questions more
not make advertising a focal point of their tudes toward advertising (Pollay and broadly (Wimmer and Dominick, 2006).
study (Detenber and Reeves, 1996; Reeves Mittal­, 1993) and skepticism of advertising
et al., 1999; Nueman, 1990; Lombard et al. (Obermiller and Spangenberg, 1998) will Focus Groups
1997). Visual-attention studies provided a provide a background for understanding Five focus groups were conducted to gen-
background for understanding attention how viewers of large-screen television erate a general understanding of large-
to advertising and programming prior to accept advertising as part of the enhanced screen viewing. Participants in the focus
introduction of large-screen televisions viewing process. groups represented diverse ages of the
(Allen, 1965; Anderson et al., 1986; Bechtel adult population. The average age of the
et al., 1972; Collet and Lamb, 1986; Krug- RQ4a: Do large-screen viewers have a participants across all five focus groups
man and Shamp, 1992; Steiner, 1966). more positive attitude toward was 43 years. However, the median age of
At this juncture, it is important to enrich advertising than small-screen 46 years and the mode of 56 years suggest
the area by providing a specific compari- viewers? a slight skew toward the older popula-
son between large- and small-screen tele- tion in the sample. The ages ranged fairly
vision viewers using a method that moves RQ4b: Are large-screen viewers less evenly between 18 years and 76 years, giv-
beyond experimental conditions or direct skeptical of advertising than ing the sample a robust cross-generational
observation and offers a more generaliz- their small-screen counterparts? representation. Twenty-five of the 36 par-
able sample. Additionally, it is important ticipants were male.
to make advertising a focal point of the METHODOLOGY The overall makeup of the respond-
study. The current study employed a multi- ents was relatively homogeneous with
method approach using focus group respect to marital status, race, education,
RQ2: Do large-screen viewers pay interviews and a national online sur- and income. Twenty-nine of the 36 partici-
more attention to television vey. Five focus groups, with a total of 36 pants were married; 2 of 36 were divorced,

December 2012  JOURNAL OF ADVERTISING RESEARCH  423

The Big Picture for Large-Screen Television Viewing

and 5 of 36 were single (never married). if applicable, how they would describe entry into the panel. Panelists are limited
Twenty-four of 36 participants had chil- being “engaged” in the programming. The to one survey invitation per week. Pan-
dren living in their home. Thirty-four of latter two variables were operationalized elists who do not complete any surveys in
the 36 participants were white (Cauca- as specific questions for the survey. a 3-month time period are removed.
sian), with only one African-American and Attention to programming, attention
one multi-racial participant represented. National Online Survey Panel to advertising, and absorption in televi-
The participants were largely from the Using preliminary conclusions from focus sion scales were developed from focus
more educated segments of society. Sixteen groups, a survey instrument was con- group data and pretested. Established
of the 32 respondents reported attending or structed, tested, and administered. scales for advertising skepticism—the
completing graduate school. Twenty-seven The research design used an online sur- first nine items (Obermiller and Spangen-
of the 32 respondents reported attending vey panel, with respondents self-reporting berg, 1998)—and attitudes toward adver-
or graduating from undergraduate college. their attention to broadcast programming tising—the last three statements (Pollay
Five of the 32 respondents did not have an and commercials. The online survey was and Mittal, 1993)—were included as sur-
undergraduate or graduate degree. a comparative instrument to test the dif- vey questions. Respondents were asked
Last, the participants were largely from ferences between large-screen owners and to mark their agreement or disagreement
the more prosperous segments of society. traditional-screen owners. The research with the 12 statements about advertising.
Sixteen of the 30 respondents who pro- goal was to profile large- and small-screen This 12-item scale was validated by Diehl,
vided income data reported household viewers and to test what changes (if any) Mueller, and Terlutter (2008).
income of $100,000 or more each year. Nine large-screen televisions brought to their The questionnaire employed five-point
of 30 respondents reported household attention to and absorption with program- Likert scale responses using an umbrella
income of between $50,000 and $99,999. ming and commercials. format adapted from Dillman (2007). Two
Five of 30 people responding to the ques- Generalized attitudes toward advertis- pretests were conducted. Paper-based sur-
tion earned less than $50,000 per year. ing and skepticism toward advertising vey testing was conducted using 20 adults.
The respondents for the five focus were also sought for each viewing group. The survey was programmed into the
groups represented a group of television MarketTools, Inc.’s Zoomerang based in online platform and checked for errors in
viewers that consumed “cutting-edge” San Francisco was selected as the panel translation between a paper-based instru-
television. Focus-group data were essen- provider for the research. The national ment and an online instrument. The online
tial in understanding the basic viewing panel is well accepted, and MarketTools survey was then sent to individuals via an
process that takes place when consum- is noted in the Honomichl Top 50 issue of e-mail invitation asking them to complete
ing traditional television and large-screen Marketing News (“2007 Honomichl Top 50,” the survey. Forty-two respondents com-
television. Respondents were asked to 2007). A key advantage of the panel is the pleted the online survey pretest. Feedback
describe their overall use of television and ability to identify plasma or LCD owner- gained from the online test led to import-
the way large-screen viewing takes place. ship. Identification of LCD and plasma ant clarifications and refinements, which
Respondents were of critical importance owners enabled for pre-identification of were subsequently incorporated into the
in defining what constitutes a large-screen sample prospects who may fall into the final online instrument.
television and for understanding the tech- large-screen owner category. In addition to the preceding items, sur-
nologies associated with such viewing. Three million respondents in the United vey respondents answered demographic
The consensus of the focus groups put States are actively managed. The panel questions. Information was used to pro-
the large-screen entry-point at (or around) undergoes a continuous and rigorous pro- vide a basic profile comparing demo-
40 inches. A large-screen television was cess of online verification to ensure accu- graphic variables and respondents’ screen
minimally defined as 37 inches or larger. rate responses. By contractual agreement, size. The questions addressed age, gender,
In practical terms, however, respondents no panel members were recruited on “sur- race, marital status, number of people
agreed that a measurement of 40 inches or veys for cash” Web sites. A small rewards living in the home, the presence of chil-
larger became the definition, as few sets program was offered as an incentive for dren in the home, education, and gross
are sold in increments between 37 and 39 survey completion, but the rewards gained income.
inches. Respondents were asked to define were nominal and were meant to discour- One of the key profile designations for
how they paid attention to the set and, age professional panelists from gaining all panel members includes the types of

424  JOURNAL OF ADVERTISING RESEARCH  December 2012

 The Big Picture for Large-Screen Television Viewing

television owned. A follow-up survey sent affect normal viewing patterns during the owners of televisions 40 inches or larger,
to update attributes inquired whether the 3 days of survey implementation. 475 owners of televisions 30 to 36 inches,
panelists owned a large-screen LCD or a Each of the three phases of survey invi- and 378 owners of televisions 29 inches
large-screen plasma television. From this tations was e-mailed to panelists in the or smaller. In general, the sample under-
inquiry into the overall panel, a list of early morning. The invitation to complete represented the least affluent segment
approximately 4,000 owners of LCD large the survey arrived in the morning when (less than $24,999) and over-represented
screens and approximately 1,500 owners the panelist would readily recall the situ- both the average Zoomerang panel and
of plasma large screens was generated. ation of their prior evening’s activities. A the U.S. Census in the middle- to upper-
The combined group of 5,500 panelists 24-hour survey close was placed on each middle-class categories ($50,000–$99,999).
was then randomly listed in a large-screen of the three survey tranches to provide This was an expected finding due to the
television sampling pool. Thus, a compar- even response across the 3-day period. high prestige and price of large-screen
ative sampling frame was generated with The overall response rate for the sur- televisions’ drawing a more affluent con-
large-screen and traditional-television vey was 12.8 percent which is within the sumer base.
owners represented. middle range (5 percent to 20 percent)
The sampling frame was further divided of the estimated overall panel response RQ1: What is the Demographic Profile of
to control for gender. The study employed, rates (Wilner, 2008). As will be noted later, Large- and Small-Screen Viewers?
therefore, four sampling pools: the decision to close the responses after Survey data were used to compare the
24 hours likely suppressed response rates. demographic variables based on the
• females owning large-screen televisions, The first tranche of the survey yielded respondents’ screen size. The variables
• males owning large-screen televisions, a higher-than-expected response by both of age, number of people in the home,
• females owning traditional televisions, male and female traditional televisions education level, and gross income were
and owners (Table 1). Based on those response compared parametrically using one-way
• males owning traditional televisions. rates, the number of invitations sent to analysis of variation (ANOVA). The varia-
each sampling pool was adjusted for the bles of race, marital status, and presence of
A sequential probability sampling and last two surveys to more accurately reach children in the home were analyzed using
disproportionate stratified sampling were the completion parameters of the study. Pearson’s chi square and cross-tabulations.
employed in the study. The online sur- There was no significant age difference
vey was implemented in three separate FINDINGS between owners of large-screen televisions
tranches over a 3-day, weekday period. The total sample N for the online survey and other television owners (F = 0.864, df = 1,
Due to the questions referring to watch- of 1,328 usable responses included 475 p = 0.353, eta = 0.001). There also was no
ing television yesterday, the sample was
implemented in split releases over Tues-
day, Wednesday, and Thursday. This
Table 1
ensured that the respondents would be Online Survey Invitations and Response
referencing their television viewing on Female/ Female/ Male/ Male/
Monday through Wednesday. Large- Standard Large- Standard
Research by Nielsen Media has shown screen TV screen TV Totals
that the most primetime viewing occurs Tuesday Invitations 921 2,127 429 2,241 5,718
on Sunday night, followed by Tuesday 6/10/2008 Completes 105 252 51 274 682
night, Monday night, and then Wednesday
Wednesday Invitations 625 624 660 421 2330
night (Average U.S. Home Now Receives a 6/11/2008 Completes 68 85 83 77 313
Record 118.6 TV Channels, 2008). The sur-
Thursday Invitations 625 624 661 421 2,331
vey implementation occurred during the
6/12/2008 Completes 68 105 86 76 335
early part of the week when primetime
Totals Invitations 2,171 3,375 1,750 3,083 10,379
viewing was consistent across all catego-
Completes 241 442 220 427 1,330
ries of viewers. There were no news events
Response Rate 11.1 13.1 12.6 13.9 12.8
or other abnormal occurrences that would

December 2012  JOURNAL OF ADVERTISING RESEARCH  425

The Big Picture for Large-Screen Television Viewing

evidence in the data that there was any traditional television households reported and they reported spending more money
difference in race between large-screen incomes in that range. per month on various television services.
and smaller screen owners. (Note that the Finally, the survey results showed a
overall sample contained 89 percent white statistically significant but negligible dif- RQ2: Do Large-Screen Viewers Pay
respondents, which is more than the white ference in education levels between large- More Attention to Television during the
population, according to the U.S. Census.) screen versus traditional screen owner Programming and Commercials Than Do
Large-screen television owners were (F = 5.037, df = 1, p = 0.025, eta = 0.004). The Small-Screen Viewers?
more likely to be married than were tradi- small mean difference and the low effect Survey respondents were asked to self-
tional screen owners (chi square = 21.273, score indicate that there are only marginal report their attention levels to the screen
df = 1, p = 0.000, phi = –0.127). Sixty-seven differences in educational levels between using a five-point scale ranging from “Pay
percent of large-screen owners were mar- the two groups. No Attention” to “Pay Full Attention” A
ried, whereas only 54 percent of standard- In summary, large-screen owners did one-way ANOVA was used to compare
sized television respondents (36" or less) not vary according to age, and there was means and estimate effects sizes.
were married. Large-screen owners also no conclusive information to suggest any The results showed that large-screen
were more likely to have children in the meaningful difference in education level viewers paid more attention to the screen
home than were traditional screen own- or race. for traditional broadcast and cable/
ers (chi square = 8.300, df = 1, p = 0.004, Large-screen television households satellite programming (F = 25.536, df = 1,
phi = 0.079). Thirty-seven percent of large- were more likely to be married, have p = 0.000, eta = 0.019). The effect size score
screen television owners had children 17 teenagers and older children in the home, indicated a modest relationship. The actual
years old or younger. Only 29 percent of live with more people, and make more means of the responses (4.01 for large-
standard television owners had children money each year than traditional televi- screen respondents, 3.74 for traditional
17 years old or younger at home. Large- sion households. Large-screen television screen respondents), however, exhibited
screen owners had more teenagers (11–17 households also tended to own more tele- a difference of 0.25, which indicates rea-
years old) in the home (chi square = 7.912, visions than did small-screen households, sonable separation in the data (Table 2).
df = 1, p = 0.005, phi = –0.077) and children
older than 18 (chi square = 5.510, df = 1, Table 2
p = 0.019, phi = –0.064). Consistent with
those findings, large-screen owners were
Attention to the Largest Home Television
more likely to live with more people than Large-screen (40" or more) Television Owners (percent)
Pay No Pay Full
were traditional screen owners (F = 12.733,
Attention Attention
df = 1, p = 0.000, eta= 0.010).
1 2 3 4 5 Mean
Large-screen television households also
Traditional broadcast  2.1  1.9 19.2 46.1 30.5 4.01
earned significantly more money per year
cable programs
than the traditional television households
(F = 111.184, df = 1, p = 0.000, eta = 0.077). Commercials 23.2 36.0 25.7 10.5  4.0 2.36
The results indicated a moderate effect “I get absorbed in the Strongly Disagree Strongly Agree
size for the statistical relationship. The programming”  0.8  7.2 28.4 30.1 33.5 3.88
differences in gross income were found Standard screen (36" or less) Television Owners (percent)
on the lower and upper extremes of the Pay No Pay Full
responses. In the lower incomes, 49 per- Attention Attention
cent of traditional screen owners reported 1 2 3 4 5 Mean
a gross income of less than $50,000, Traditional broadcast  3.8  4.6 26.0 44.8 20.8 3.74
whereas only 24 percent of large-screen cable programs
owners reported an income less than Commercials 29.0 40.9 21.0  6.4  2.2 2.12
$50,000. In the higher incomes, 29 percent
“I get absorbed in the Strongly Disagree Strongly Agree
of large-screen owners reported incomes
programming”  7.5 15.4 31.1 30.9 15.1 3.31
above $100,000, whereas only 13 percent of

426  JOURNAL OF ADVERTISING RESEARCH  December 2012

 The Big Picture for Large-Screen Television Viewing

The separation was consistent with with previous work (Krugman and John- of being absorbed into the programming
previous television technology stud- son, 1991; See Table 2). (F = 86.192, df = 1, p = 0.000, eta = 0.061).
ies that used scale items to distinguish The consistency of the two findings The means of the responses (3.88 for large-
among viewing groups (Krugman and for both program viewing and advertis- screen respondents, 3.31 for traditional
Johnson, 1991). The self-report meas- ing viewing, therefore, indicated that the screen respondents) exhibit a difference of
ure of attention, therefore, indicated that large-screen television viewers are more 0.57 (Figure 2).
large-screen televisions positively impact attentive. For this relationship, the effect size score
attention levels to broadcast programming indicated a stronger relationship between
(Figure 1). RQ3: Do People Feel More Absorbed in the variables than was observed for the
In a similar finding, the results also the Television Viewing on a Large-Screen attention scores. Large-screen respond-
showed that large-screen viewers paid Television Than When Viewing Small- ents were found to report greater absorp-
more attention to commercials than the Screen Television? tion levels than the comparative group of
traditional standard-screen television Respondents also were also asked about smaller screen viewers (See Figure 1).
viewers (F = 17.304, df = 1, p = 0.000, the absorbing quality of their largest tel-
eta = 0.013). Similar to the foregoing, the evision and its ability to draw them into RQ4a: Do Large-Screen Viewers Have a
actual means of the responses (2.36 for the programming. Specifically, the ques- More Positive Attitude toward Advertising
large-screen respondents, 2.12 for tradi- tion asked the respondents to report the Than Small-Screen Viewers?
tional screen respondents) exhibited a dif- degree to which they got “absorbed” in
ference of 0.24, which indicated reasonable the programming. Large-screen viewers RQ4b: Are Large-Screen Viewers Less
separation in the data and was consistent also reported significantly greater levels Skeptical of Advertising Than Their Small-
Screen Counterparts?
A principal component factor analysis was
Absorption in Programs
employed for the two groups of advertis-
Attention to Programs
ing questions and their responses. Each
Attention to Advertising group of items loaded onto a separate
4.5
component, producing a two-component
factor structure. The nine advertising-
Estimated Marginal Means (5-point Scale)

4.0 skepticism statements loaded with an

Eigenvalue of 7.136, and the three attitude-
toward-advertising statements loaded
3.5
with an Eigenvalue of 2.678. Eigenvalues
of 1 or larger are considered significant.

3.0 Cronbach’s alpha was also calculated for

both scales and showed similar results
(Cronbach’s alpha for advertising skep-
2.5 ticism = 0.967; Cronbach’s alpha for
attitude-toward-advertising = 0.939).
With the components established, the
2.0
advertising-statement responses were
averaged into one score for each variable
1.5 (advertising skepticism, attitude toward
Standard Screen Large Screen advertising), and one-way ANOVA
was used to compare mean differences
between large-screen and traditional tele-
Figure 1  Attention to Advertising and Attention to/Absorption
vision owners. The scores for advertising
in Programming Among Owners of Standard- and Large-Screen skepticism were reverse-coded to reflect a
Televisions higher score signifying higher skepticism.

December 2012  JOURNAL OF ADVERTISING RESEARCH  427

The Big Picture for Large-Screen Television Viewing

Traditional small-screen television own- owners (F = 15.558, df = 1, p = 0.000, income). Moreover, large-screen owner-
ers statistically were found to have greater eta = 0.012)—in other words, large-screen ship is positively associated with two
skepticism toward advertising than large- owners embraced advertising more than reported television-viewership responses
screen viewers (F = 9.976, df = 1, p = 0.002, traditional-screen television owners (See (level of attention to programming and
eta = 0.007). People with smaller tele- Figure 2). ANOVA confirmed the ini- level of absorption in programming).
visions, therefore, were more skeptical of tial finding that large-screen viewers (40 Because these multiple significant bivar-
advertising than people with large-screen inches or more) embraced advertising iate relationships were observed, it seems
televisions. more than medium- and smaller-screen quite likely that one or more of them may
The authors separated the respondents owners (36" or less; F = 7.961, df = 2, function as covariates or mediators of the
into three groups (large, medium, and p = 0.000, eta = 0.012). relationships between large-screen owner-
small screens) and, using ANOVA, showed ship and assessments of television adver-
that significance was found between large- COVARIATES AND MEDIATORS OF tising (attention paid to commercials,
screen (40 inches and larger) and smaller ADVERTISING RESPONSES attitude toward television advertising,
screen (29 inches or smaller; F = 6.306, After examining the research questions, and advertising skepticism).
df = 2, p = 0.002, eta = 0.009; mean differ- the authors report heretofore that large- Specifically, the relationships between
ence of 0.2219 between large- and small- screen ownership is significantly and screen size and advertising responses
screen viewing significant at 0.001 level positively related to a number of demo- might be explained by antecedent demo-
using Tukey HSD, See Figure 2). graphic characteristics of the sample graphic characteristics related to both or by
Traditional-screen television owners (marital status-married, presence of teens behaviors correlated with those character-
also were found to have lower attitudes and/or presence of older children in the istics (e.g., number of televisions owned,
toward advertising than large-screen home, number of people in the home, and dollars spent on television services). Addi-
tionally, the relationships may operate
through a mechanism in which general
television viewership characteristics (e.g.,
4.5 Skepticism toward Advertisinga
attention to programming, absorption in
Attitude toward Advertisingb
programming) mediate respondents’ more
Estimated Marginal Means (5-point Scale)

4.0 specific evaluations of advertising seen on

large versus small screens.
To more fully explain the relationship
3.5
between large-screen ownership and the
assessments of advertising, the authors
3.0 conducted a series of exploratory analyses
in which the significant demographic vari-
ables and their correlates were controlled
2.5 as potential covariates, while the significant
television viewership responses were exam-
ined as potential mediators of each of the
2.0
advertising evaluations (see Kenny, 2009; or
Zhao, Lynch, and Chen, 2010, for thorough
1.5 discussions of mediation analysis).
Small Screen Medium Screen Large Screen
Notes: Covariance Analysis
a
Skepticism of small vs. large is significant at p = 0.001
b
Attitude of small vs. large is significant at p = 0.001; Attitude of medium vs. large is significant at p = 0.004 The first exploratory analysis consid-
ered potential antecedent variables that
might serve as covariates explaining the
Figure 2 Skepticism and Attitude toward Advertising among observed relationships between screen size
Owners of Small-, Medium-, and Large-Screen Televisions and evaluations of broadcast advertising.

428  JOURNAL OF ADVERTISING RESEARCH  December 2012

 The Big Picture for Large-Screen Television Viewing

Included in this analysis are the following Observed original relationships between large-
demographic or socioeconomic variables
found to be related to screen size: screen ownership and greater attention to
• number of televisions in the household, commercials, less skepticism toward advertising,
• dollars per month spent on television
services, or more positive attitude toward advertising.
• marital status,
• number of people in the household,
• income, the authors conducted a four-step media- significantly predicts each of the adver-
• presence of young teenagers, and tion analysis for each of the three advertis- tising evaluations), the original observed
• presence of older (adult) children in the ing evaluations (Kenny, 2009). relationships between screen-size owner-
household. For reported attention to television ship and attention to commercials, adver-
programming, a pattern similar to that tising skepticism, and advertising attitude
Of these, at the bivariate level, marital sta- reported for the covariance analyses was were found to no longer be significant
tus, income, and presence of older children observed. That is, although screen size was when the programming absorption meas-
were not observed to be significantly pre- a significant predictor of greater attention ure served as a statistical control.
dictive of any of the three broadcast adver- to television programming and attention to This finding supports the argument that
tising evaluations, so they were eliminated television programming was observed the more favorable broadcast advertising
from the covariance analysis. to be a significant predictor of each of the evaluations among large-screen owners
When the remaining potential covari- three broadcast advertising evaluations can be explained through the mechanism
ates were regressed (using stepwise (when controlling for screen size), the of their greater levels of absorption in tele-
inclusion) along with a dummy variable relationship between screen size and the vision programming.
representing screen size, screen size still three broadcast advertising evaluations Simply stated, reported programming
was observed to be a statistically signifi- remained significant in the expected direc- absorption completely mediates the orig-
cant predictor of each of the broadcast tion consistent with the earlier bivariate inal observed relationships between large-
advertising evaluations: analyses even when controlling for the pro- screen ownership and greater attention
gramming attention measure, specifically: to commercials, less skepticism toward
• attention to commercials (p < 0.001); advertising, and more positive advertising
• skepticism toward advertising (p < 0.01); • attention to commercials (p < 0.01); attitude.
and • skepticism toward advertising (p < 0.05);
• attitude toward advertising (p < 0.01). and DISCUSSION
• attitude toward advertising (p < 0.01). The onset of television technologies such
Thus, none of the potential covariates of as the digital video recorder has been seen
screen size “explains away” the original In other words, programming attention as a threat to advertising viewing. Because
findings that large-screen ownership is does not mediate the observed original consumers are in more control of the view-
predictive of greater attention to television relationships between large-screen owner- ing process, there is a continuing concern
commercials, less skepticism toward ship and greater attention to commercials, that commercials will be fast-forwarded or
advertising, and a more positive attitude less skepticism toward advertising, or deleted. The current study, which exam-
toward advertising. more positive attitude toward advertising. ined large-screen viewing as another
In contrast, reported absorption in tel- aspect of viewing technology, brings con-
Mediation Analysis evision programming was observed to siderably better news in terms of advertis-
Based on the argument that reported play a significant mediating role. Not only ing attention and receptivity.
attention to television programming and does reported absorption in television pro- Viewers conceptualize “large-screen”
absorption in television programming gramming meet the statistical criteria as a as “40-include-plus” screen size. Large-
provide a context in which television potential mediator (i.e., it is significantly screen viewers are more television-centric
advertising evaluations might be couched, predicted by screen-size ownership and than their small-screen counterparts.

December 2012  JOURNAL OF ADVERTISING RESEARCH  429

The Big Picture for Large-Screen Television Viewing

Large-screen viewers are more attentive The data clearly indicate that viewers of large-screen
to advertising and programming, more
absorbed in advertising and programming, television pay more attention to television commercials,
and overall less critical of advertising.
The findings of the current study showed have more positive attitudes toward advertising, and are
improved audience attention to both pro-
gramming and commercials for large- less skeptical of advertising than small-screen viewers.
screen viewers. The findings also showed
an even stronger absorbing quality of the
large-screens that draws the viewers into
the content. This suggests that the supe-
rior television experience was matched by Similar arguments have been made for $1,600; 32-inch LCD, $450; and 19-inch
enhanced audience engagement. other advertising media in support of LCD, $199 (Taub, 2008a, 2008b). In 2011,
Large-screen television owners also claims of greater advertising effectiveness. the approximate prices for leading brands
were shown to respond to the enhanced Of particular relevance related to the cur- were 50-inch plasma, $999; 40-inch LCD,
experience through greater attention lev- rent findings are the out-of-home media, $584; 32-inch LCD, $378; and 19-inch LCD,
els and absorption. The findings portray a which have long argued that “large size” $199 (Best Buy, April 25, 2011). Though
class of television viewing that is more vis- offers a substantial persuasive advan- large-screen prices have fallen, however,
ually active and, therefore, more engaging tage. For example, one 1980 study theo- they remain considerably higher than most
for both program content and advertising. rized that because outdoor advertising is small screens.
The data clearly indicate that viewers “larger than life,” it conveys a message of The data also reveal that large-screen
of large-screen television pay more atten- “conviction” from the advertiser, which, owners have both higher incomes and are
tion to television commercials, have more in turn, contributes positively to persua- more interested in television. This seg-
positive attitudes toward advertising, sion goals (Bernstein, 1980). Another piece ment—with both the means and predispo-
and are less skeptical of advertising than of research interpreted the “larger-than- sition to buy the best available technology­/
small-screen viewers. This is an important life” effect as acting through the “higher services­
—self-selects and gains greater
finding because it indicates that the large- impact” of large outdoor advertisements satisfaction from more enhanced tele-
screen environment is more welcoming to (Moriarty, Mitchell, and Wells, 2009). A vision products; therefore, the segment is
advertisers. third referred to the “giant canvas” of the more television centric.
Furthermore, although the effect sizes of outdoor medium (Maskulka, 1999). The authors believe the segment is an
the observed relationships between screen The screen-size findings of the current enduring part of the television-viewing
size and receptivity to advertising are mod- study suggest that advertising effective- population as evidenced by earlier studies
erate, they are quite robust. The authors’ ness is enhanced through its ability to that found similar patterns in both income
covariance analyses did not diminish the generate greater “absorption” in the pro- and interest (e.g., Ducey, Krugman, and Eck-
relationships between large-screen own- gramming that, in turn, enhances advertis- rich, 1983; Krugman and Eckrich, 1982). In
ership and more positive assessments of ing attention and attitudes. Recently, Peter this regard, television centricity exists irre-
advertising. Greater level of absorption in Gardiner, chief media officer at New York spective of the attitude toward advertising.
large-screen television programming was advertising agency Deutsch Inc., made
observed to be a complete mediator of the similar claims that “niche” programs that LIMITATIONS OF THE STUDY
advertising evaluations. generate high viewer loyalty have much The current study used a national online
Though the “medium” may not actu- greater commercial retention, persuasion, panel company with a reputation for
ally be the “message,” its ability to gen- and recall (Dumenco, 2010). quality-assured samples. The survey sam-
erate greater absorption in the message The overall cost of large-screen tel- ple under-represented the least affluent
positively influences how advertising in evisions has been reduced significantly segment of society (less than $24,999) and
that medium is perceived. And, in this in recent years. For example, in 2007, over-represented the panel and the U.S.
instance, the “presence” provided by the leading brands sold for approximately Census in the middle- to upper-middle-
medium is good news for advertisers. 50-inch plasma, $2,600; 40-inch LCD, class categories ($50,000–$99,999). This is

430  JOURNAL OF ADVERTISING RESEARCH  December 2012

 The Big Picture for Large-Screen Television Viewing

an expected finding due to the prestigious FUTURE RESEARCH and audience factors. He has conducted both survey

nature and price of large-screen televisions, Future researchers would do well to repli- and experimental studies sponsored by the Cox

drawing a more affluent consumer base. cate the study to further verify the findings Institute and the American Academy of Advertising. His

As the socioeconomic status of large-screen that the size of television screen impacts research has appeared in Journal of Marketing, Journal

television households became a central attention levels to both programming and of Advertising, Journal of Advertising Research,

point in the findings of the study, a word commercials. In addition, other methodol- Journal of Broadcasting & Electronic Media, Political

of caution about a small upward bias in ogies—including eyes-on-screen or physi- Communication, and Journal of Consumer Psychology.

the sample on income should be noted. ological measures—could be gainfully

The use of a sequential sampling employed to further analyze the changes
method offered greater flexibility to “over- in television viewing that occur due to the References
sample” on the second and third days in adoption of larger and more impressive
those subgroups with lower response rates televisions in homes. 2007 Honomichl Top 50. Marketing News (2007).
on the first day. This approach could limit At the same time, it remains centrally
the generalizabilty of descriptive popula- important that advertisers understand the Abernethy, A. M. “Television Exposure:
tion characteristics. fundamental viewing process rather than Programs vs. Advertising.” Current Issues &
A priority of the study, however, was only the “reactions” to each new technol- Research in Advertising 13 (1991): 1–2.
to compare responses across screen-size ogy and service. 
subgroups and other relationships within Abowd, R. “Flat panel display market research-
the data. This priority argued for sampling Michael D. McNiven is senior vice president/portfolio World.” Unpublished Report, Pacific Media Asso-
procedures that achieved acceptable sta- manager at Cumberland Advisors, Vineland, New Jersey. ciates (2007).
tistical power within subgroups. A judg- He began his career at the Financial Times in New York

ment was made in the study to stagger the working in market research, business development, and Allen, C. “Photographing the TV Audience.”
online survey and select response dead- advertising sales. McNiven, who received his PhD from Journal of Advertising Research 5 (1965): 2–8.
lines requiring rapid responses across a the University of Georgia, was an assistant professor at

3-day period starting on Tuesday and end- Rowan University prior to joining Cumberland Advisors Anderson, D., E. P. Lorch, D. E. Field, P. A.

ing on Thursday. The three-tranche survey and also has taught and assisted at the University of Collins, and J. G. Nathan. “Television View-

over 3 days was an effort to eliminate the Georgia, Utah Valley University, and Brigham Young ing at Home: Age Trends, Visual Attention, and

impact of any news events or television University. Time with TV.” Child Development 57 (1986):

events on any one day that could skew the 1024–1033.

viewing measures of the study. As such, Dean M. Krugman is professor emeritus in the depart­

each tranche of the survey was closed after ment of advertising and public relations at the University Average U.S. Home Now Receives a Record 118.6 TV

24 hours, not allowing for responses after of Georgia. Dean’s research includes audiences and Channels. New York: The Nielsen Company, 2008.

that time. advertising, new media, and health communication. His

Closing the surveys after 24 hours, how- work has been published in such places as Journal of Bechtel, R. B., C. Achelpohl, and R. Akins.

ever, also suppressed response rates. The Advertising, Journal of Advertising Research, Journal “Correlates Between Observed Behavior and

survey response rate of 12.8 percent likely of Public Policy & Marketing, and the American Journal Questionnaire Responses on Television View-

would have been larger if multiple requests of Public Health. Dean’s work on media and audiences ing.” In Television and Social Behavior, Reports and

for response were sent and time for com- regularly has appeared in the Journal of Advertising Papers, Volume IV: Television in Day-to-Day Life:

pletion were spread over several days. Research over a 30-year period. He is the former Pat. Terms of Use, A Technical Report to the Surgeon

Although some online panel response rates president (2009) of the American Academy of Advertising General’s Scientific Advisory Committee on Televi-

were reportedly high in the early days of and received the “Outstanding Contribution to Research sion and Social Behavior, A. C. George and P. M.

online panels, online panel response rates Award” award from that organization. John, eds. Washington, DC: U.S. Government

have declined, significantly trending lower Printing Office, 1972.

than response rates in traditional survey Spencer Tinkham is professor in the department of

forms, prompting research on how to maxi- advertising and public relations at University of Georgia. Bernstein, D. Larger than Life. Educational film
mize online response rates (Neslin, Novak, Tinkham’s research, in part, has focused on persuasive produced by Osmond Commercial Productions, an
Baker, and Hoffman, 2006). communication, particularly with respect to message educational service of the Gannett Foundation, 1980.

December 2012  JOURNAL OF ADVERTISING RESEARCH  431

The Big Picture for Large-Screen Television Viewing

Best Buy.com. Retrieved October 25, 2011, Krugman, D. M., and K. F. Johnson. “Dif- Ostendorp, P., S. Foster, and C. Calwell.
from http://bestbuy.com/site/TV-Video/ ferences in the Consumption of Traditional “Televisions: Active Mode Energy Use and
Televisions/abcat0101000.c?id=abcat0101000 Broadcast and VCR Movie Rentals.” Journal of Opportunities for Energy Savings.” Natural
Broadcasting & Electronic Media 35, 2 (1991). Resource Defense Council Issue Paper: March
Collet, P., and R. Lamb. “Watching people 2005.
Watch TV. Report to the Independent Broadcasting Krugman, D. M., and S. Shamp. “Observations
Authority, London, 1986. of Audience Behavior during Television Com- Pollay, R. W., and B. Mittal. “Here’s the
mercials.” Proceedings of The 1992 Conference of The Beef: Factors, Determinants, and Segments in
Crain Communications. CEA: HDTV Pen- American Academy of Advertising (1992): 156–160. Consumer Criticism of Advertising.” Journal of
etration at 28% Study Suggests Average HDTV
Marketing 57, 3 (1993): 99.
Price, Location and Uses, James Hibberd, LCD TVs versus Plasma TVs. Media Post,
TV WEEK (2010). 2007: [URL: http://www.flattvpeople.com/ Reeves, B., A. Lang, E. Y. Kim, and D. Tatar.
tutorials/lcd-vs-plasma.asp] Retrieved Septem- “The Effects of Screen Size and Message Con-
Detenber, B. H., and B. Reeves. “A Bio-
ber 10, 2007. tent on Attention and Arousal.” Media Psych-
Informational Theory of Emotion: Motion and
ology 1, 1 (1999): 49.
Image Size Effects on Viewers.” Journal of Com-
Lombard, M., T. B. Ditton, M. E. Grabe, and
munication 46, 3 (1996): 66.
R. D. Reich. “The Role of Screen Size in Viewer
Schramm, W. L. Television in the lives of our chil-
Responses to Television Fare.” Communication
Diehl, S., B. Mueller, and R. Terlutter. “Con- dren. Stanford, CA: Stanford University Press,
Reports 10, 1 (1997): 95–106.
sumer Responses Towards Non-Prescription 1961.

and Prescription Drug Advertising in the US

Lucus, K. “Make Money in Non-Negotiable
and Germany.” International Journal of Advertis- Steiner, G. A. “The People Look at Commer-
Technology.” Daily Markets, March 30, 2010:
ing 27, 1 (2008): 99–131. cials: A Study of Audience Behavior.” Journal
of Business of the University of Chicago 39 (1966):
Maskulka, J. “Outdoor Advertising: The
Dillman, D. A. Mail and Internet Surveys: The 272–304.
Brand Communication Medium.” Special
Tailored Design Method, 2nd ed. Hoboken, NJ:
Report. Washington, DC: Outdoor Advertising
Wiley, 2007. Taub, E. A. “Flat Panel TV Prices Plummet.” New
Association of America, 1999.
York Times, October 25, 2008a. Retrieved April
Ducey, R. V., D. M. Krugman, and D. Eck- April 22, 2011, from http://bits.blogs.nytimes.
Moriarty, S., N. Mitchell, and W. Wells.
rich . “Predicting Market Segments in the Cable com/2008/10/25/tv-prices-falling-faster/
Advertising: Principles and Practices. Upper Sad-
Industry.” Journal of Broadcasting 27, 2 (1983):
dle River, NJ: Pearson Prentice Hall, 2009.
155–161.
Taub, E. A. “TV Prices Falling Faster.”
New York Times, Bits, December 2, 2008b.
Dumenco, S. “The Tyranny of Smart TV.” Neslin, S., T. Novak, K. Baker, and D. Hoff-
Retrieved April 22, 2011, from http://Nytimes.
man . An Optimal Contact Model for Optimizing
Details 28, 10 (2010): 94–98.
com/2008/12/02/tecnologytechspecial2/
Online Panel Response Rates. Riverside, CA:
02flat.html/adxnnl=1&page…
Kenny, D. A. Mediation, 2009. Retrieved April Sloan Center for Internet Retailing, University

21, 2010, from http://davidkenny.net/cm/ of California-Riverside, 2006.

Weston, M. “Looking to HDTV Content Insider
mediate.htm
Neuman, W. R. Beyond HDTV: Exploring Subjec- #102.” CEN, Consumer Electronics Net. April 21,

tive Responses to Very High Definition Television. A 2008.

Krugman, D. M., G. T. Cameron, and C. M.
White. “Visual Attention to Programming and Research Report for GTE Labs and the TVOT Con-
Commercials: The Use of in-Home Observa- sortium. Cambridge: MIT, 1990. Wilner, M. Zoomerang Account Manager. San

tions.” Journal of Advertising 24, 1 (1995): 1. Francisco, CA: MarketTools, Inc.

Obermiller, C., and E. R. Spangenberg. “Devel-

Krugman, D. M., and D. Eckrich. “Differ- opment of a Scale to Measure Consumer Skepti- Wimmer, R. D., and J. R. Dominick. Mass Media
ences in Cable and Pay-Cable Audiences.” Jour- cism Toward Advertising.” Journal of Consumer Research: An Introduction, 8th ed. Belmont, CA:
nal of Advertising Research 22, 4 (1982): 23–29. Psychology 7, 2 (1998): 159. Wadsworth, 2006.

432  JOURNAL OF ADVERTISING RESEARCH  December 2012

Copyright of Journal of Advertising Research is the property of Warc LTD and its content may not be copied or
emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.