Taking on The Tube:
An Examination of Parents' Media Behavior, Attitudes
and Discussion Regarding Television
by
Erica Weintraub Austin
Associate Professor
Edward R. Murrow School of Communication
Washington State University
Pullman, WA 99164-2520
(509) 335-8840
EMAIL: [log in to unmask]
FAX: (509) 335-1555
Yuki Fujioka, Paul Bolls, & Jason Engelbertson,
Graduate Students
Washington State University
with assistance from
Tracy Lee Clarke, Chris Knaus, Kristine Kay Johnson,
and Ana Meneguelli
Graduate Students, Washington State University
Manuscript submitted to the
Theory & Methodology Division of the
Association for Education in Journalism & Mass Communication
for 1996 Presentation, Anaheim, CA
This research has been supported in part by the
Alcoholic Beverage Medical Research Foundation
Running Head:
Mediation Styles
Mediation Styles
Taking on The Tube:
An Examination of Parents' Media Behavior, Attitudes
and Discussion Regarding Television
ABSTRACT
A random telephone survey (N=255) of Washington state parents of
children between the ages of 2 and 17 assesses parents' patterns of interaction
with their children regarding television, along with parental viewing habits and
perceptions of television content. Mediation is shown to be conceptually
distinct from coviewing and to have different motivations and implications when
it is positive or negative in valence. A typology of parental mediation
patterns emerges from the data, including nonmediators (parents who talk about
television with their children infrequently), optimists (those whose discussion
primarily reinforces television content), cynics (those whose discussion
primarily counters television content) and selectives (those who use both
positive and negative discussion strategies).
It is concluded that positive mediation may occur primarily by
happenstance rather than by intent, with negative mediation associating more
with critical viewing and protective motivations.
Taking on The Tube:
An Examination of Parents' Media Behavior, Attitudes
and Discussion Regarding Television
From Plato to the 1996 United States Congress, society has taken it
upon itself to control and censor the information available to children.
Whether via stories, regulations or V-chips, adults have always tried to protect
children from messages that might damage or threaten the impressionable
(Meyrowitz, 1985; Tatar, 1991). Nevertheless, the ubiquitous nature of
television has made it more difficult to limit children's access to ideas with
which their caregivers might disagree (Meyrowitz, 1985; Austin, 1993). Because
the flow of information is less easily controlled, other routes to parental
influence take on increased importance. Scholars' understanding of such
influences on children's interpretations and uses of media content, however,
remains limited. The purpose of this study, as a result, is to examine
parent-child interaction regarding television from the perspective of the parent
in an effort to better understand conceptual differences among their
communication activities as well as the motivations related to those activities.
Both the general public and parents frequently express concern about
television content. Mittal (1994), for example, reported that consumers
evaluated much advertising as misleading, boring, irritating, offensive, silly
and trivial. The public also blamed advertising for undesirable effects such as
spreading materialism, promoting sex and taking undue advantage of children.
According to a 1989 Gallup survey of parents, parents watching television with
their children experience some discomfort about the content of television. In
particular, parents of children under 18 objected to sex-related issues,
violence and bad language. Ridley-Johnson, Surdy and O'Laughlin (1991) also
reported parental fear over the effects of television violence on their
middle-school children, worrying that television contributed to the acceptance
of violence and made children feel upset and threatened.
Considering the apparent import of the issue to society as reflected
in political speechmaking and congressional action, comparatively little
attention has been paid to parent-child interaction regarding the mass media.
According to findings such as those from Austin, Roberts and Nass (1990),
parents can successfully reinforce or refute television content. According to
one recent review of the literature on family communication and the media
(Austin, 1993), concerned parents can exert direct or indirect control over
potential media influences via four primary routes. These include 1) rulemaking
(censorship); 2) modeling behaviors that confirm or disconfirm media messages
(coviewing and everyday behavior); 3) general communication norms that influence
children's information seeking patterns; and 4) active discussion of media
content, often called "mediation."
According to the existing literature, however, parents do little to
control or influence the messages their children receive from the mass media.
Parents tend to make few rules and discuss television content infrequently
(e.g., Bower, 1973; Comstock, 1975; Corder-Bolz, 1980; Lyle & Hoffman, 1972;
Austin, Roberts & Nass, 1990). Parents also frequently set a poor example for
children to imitate or cultivate communication patterns within the family that
do not foster the development of effective decisionmaking skills (e.g., Flay &
Sobel, 1983; Kandel & Logan, 1984; Alexander & Fry, 1990; Austin &
Nach-Ferguson, 1995).
It would be premature, however, to blame poor parenting for unwelcome
media effects on children. Many of the studies that have examined parent-child
interaction and the media are difficult to compare due to inconsistent
definitions and measurement techniques across studies, leading to apparently
contradictory and potentially misleading findings. As Austin (1992) recently
argued, for example, scholars frequently combine conceptually different
activities such as coviewing, the shared experience of media exposure by parent
and child; rulemaking, the encouragement or prohibition of certain media
experiences by the parent for the child; and mediation, the active discussion of
television content.
As a primary source of information for children, parents can help
children interpret television more critically through discussion (Austin, 1993;
Corder-Bolz, 1980; Salomon, 1981; Desmond, et al., 1985). According to Austin
(1993), relying on Messaris's (1982) conceptualization of parent's
television-related discussion, parental mediation involves three tasks:
categorization, showing whether and how television reflects the real world;
validation, involving endorsement or condemnation of portrayals; and
supplementation, showing how information from television can be useful by
supplying additional information or by engaging in related activities. All
three of these tasks can refute or reinforce television content, by suggesting
that television messages are/are not realistic; are right/wrong and are/are not
applicable to real-life situations. Desmond et al. (1985) also have noted that
parental communication about television tends to include both nonvalenced and
valenced (positive and negative) elements.
Whatever their potential for counter-influence, it appears that
parents frequently watch with their children without actually discussing the
content or by tacitly or explicitly endorsing it rather than refuting it.
Indeed, they may watch together because they like the content, not because they
plan to provide commentary. Scholars have suggested that coviewing may not be
sufficient to mediate children's perceptions of television (Austin, 1993; Bower,
1973; Desmond, et al., 1985). In addition, it appears that scholars need to
distinguish between positive and negative reinforcement of content, which can
have different motivations and different results.
For example, endorsement of the content ("positive mediation") may
have different effects than countering of the content ("negative mediation").
As Austin (1993) noted, parental alarm may produce a defensive reaction, with
endorsement producing a more approving response. A child may receive
recommendations more happily than coercion. A defensive strategy may meet with
resistance, then, giving negative mediation both the potential to backfire as
well as to cultivate skepticism and parentally endorsed mores. Yet, while
positive mediation may draw a child's attention to good things on television, it
also may reinforce easy acceptance of other messages as well. Both positive and
negative strategies, then have the potential both for success and for failure.
By examining parent-child interaction regarding television from the
perspective of the parent, this study represents a step toward examining the
implications of the potentially disparate effects of positive and negative
mediation, as well as their relationships to other aspects of parent-child
television-related interaction. The current literature suggests that parents'
primary response to challenging or offensive content is to turn the channel
rather than talk about it (Gallup, 1989). Nevertheless, some research suggests
that parents are more likely to discuss television when it conflicts with their
value system (Atkin, Heeter & Baldwin, 1989). The first finding could suggest
that interactions focused on television would tend toward the positive, whereas
the second finding could suggest that most of the mediation taking place would
be negative in valence. More likely, both conclusions represent
oversimplifications of parental motivations and behaviors.
Scholars have suggested that parents' perceptions of television's
influence will affect their patterns of parental guidance (e.g., van der Voort,
Nikken & van Lil, 1992; Bybee, Robinson & Turow, 1982). For example, Bybee et
al. found that parents who believed television had a negative impact on children
tended to make more rules and provide more explanation. Parents who held
positive views of television also tended to engage in discussion. In a
replication of Bybee and Turow's study, van der Voort et al. confirmed much of
Bybee and Turow's findings with a broader sample of parents. They also found
parental concern about television positively related to "unfocused guidance,"
essentially coviewing.
Thus, the literature suggests conceptual differences between negative
and positive communication strategies regarding television. Austin's finding
that more frequent mediation (largely operationalized as negative) predicted
greater skepticism among children, could suggest a specific role for negative
reinforcement. Meanwhile, Austin and Nach-Ferguson's finding that parental
mediation increased children's brand-specific knowledge about alcohol, which in
turn made children 7-12 more likely to report having tried an alcoholic drink,
led them to conclude that they were tapping positive (perhaps unintentionally
so) reinforcement. Since Austin's 1993 study also found that mediation
increased political involvement and public affairs media use along with
skepticism, suggesting the endorsement of television as a useful tool, this
suggests that mediation may well include both positive and negative
reinforcement components. Parents could be expected to use these mediation
styles in conjunction with their issues of concern, rather than using one style
exclusively. This suggests, as a result, that a) positive and negative
mediation are conceptually different and that b) parents may use one, both or
neither strategy in their interactions with their children. It is hypothesized,
then, that:
H1: In a factor analysis, positive and negative mediation behaviors
will separate into two
separate factors.
H2: Nonvalenced mediation will correlate positively with both
negative and positive
mediation.
H3: A typology of frequent/infrequent positive and negative mediation
will show significant
differences on a variety of constructs tapping parental media use
and opinions.
We can make some other predictions as well to tap the conceptual
differences underlying parental beliefs and behaviors regarding the television
medium. For example, Dorr et al. (1989) found that the use of television as a
socialization tool by parents, along with parental viewing encouragement,
predicted increased coviewing. Negative comments, meanwhile, did not predict
coviewing. Although the literature is not unanimous on this point (e.g., van
der Voort, et a., 1992) coviewing appears to occur because of positive views
toward television content rather than because of negative views. For example,
Dorr, Kovaric and Doubleday (1989) have found coviewing more likely when parents
value television content and think children should learn from television. This
leads to the expectation that:
H4: Positive attitudes toward television content and genres will
positively predict coviewing.
It appears that negative views of television have somewhat different
effects than positive views have. Bybee et al. (1982), for example, found that
while negative views predicted the use of both content-related discussion and
rulemaking, positive views predicted only discussion. Van der Voort, et al.
(1992) similarly found that negative views predicted discussion and also
"unfocused guidance," or coviewing combined with nonvalenced discussion. Van
der Voort et al. argued that parents' strong concerns about television
effects--whether good or bad--motivated parents to coview with children and
discuss the content to either protect them from negative effects or strengthen
positive effects. Atkin, et al. (1991), however, concluded from their results
parents are more likely to discuss television content if it presents values with
which they disagree. This would dovetail with Atkin et al.'s (1989) earlier
finding that discussion increases when parental values seem threatened. It
appears, then, that coviewing is motivated by positive views of
television--perhaps much by a desire to share an enjoyable experience, whereas
it is less certain that coviewing also is motivated by negative views of
television. If coviewing is motivated largely by positive views, then coviewing
also should be related to positive mediation and viewing frequency, reflecting
reinforcement of an enjoyable experience. Greater viewing time, say some
scholars, tends to imply less rulemaking and less critical viewing (Atkin, et
al., 1991; Desmond, et al., 1985; Medrich, 1979). The research on Family
Communication Patterns (e.g., Chaffee, McLeod & Atkin, 1971; Chaffee, McLeod &
Wackman, 1973) also has found that those who watch television the most tend to
be the least skeptical, suggesting that positive reinforcement would be more
likely than negative reinforcement among the frequent viewers.
H5: Coviewing will positively predict positive mediation.
H6: More frequent television exposure will positively predict
positive mediation.
The result, according to Austin and Nach-Ferguson (1995), is that
parents may unintentionally reinforce television content (such as alcohol
advertising) by communicating their own enjoyment and tacit approval of that
content to their children. Active discussion, then, appears to be motivated by
both positive and negative views. This, then leads to the following hypotheses:
H7: Negative attitudes toward television will predict negative
mediation.
H8: Positive attitudes toward television will predict positive
mediation.
If negative attitudes lead to negative mediation, then it follows
that parents with such attitudes will less likely approve of the use of
television as a babysitter. Research supports this expectation, given that
scholars such as Lin and Atkin (1988) have found that rulemaking and discussion
were strongly related. On the other hand, parents who approve of television
content should be more likely to approve of the use of television as a
babysitter. Coviewing and positive discussion will not necessarily negate the
use of television as a babysitter as well. Indeed, Medrich (1979) found that
parent-child television viewing patterns were more closely related in television
households in which overall viewing was less.
H9: Negative mediation will negatively predict parental agreement
with the idea that
television can be a good babysitter.
H10: Positive mediation will positively predict parental agreement
with the idea that
television can be a good babysitter.
It further follows that parents who communicate positively about
television think television has good lessons to reinforce. Thus, they are
likely to agree that television is a useful learning tool. Bower (1973), for
example, found that parents overwhelmingly saw education as the main benefit of
television. Similarly, Anderson and Collins (1988) have suggested that
mediation accomplishes little more than increasing children's ability to learn
from the medium, with Corder-Bolz (1980) suggesting that discussion
significantly enhances children's ability to learn from television. The
relationship of negative mediation and views of television as a useful tool,
however, is less clear. Negative mediation could lead parents who otherwise
enjoy television to use television as a tool to discuss what children should not
do (don't act like Bart Simpson, for example) but also lead other parents to
avoid television altogether. Thus, it is predicted that:
H11: Positive mediation will positively predict parental agreement
with the idea that
television represents a good learning tool.
Parents' own communication skills with regard to television also
should relate to their communication patterns. For example, skeptical parents
should encourage skepticism in their children more so than less skeptical
parents. Austin (1993) assumed this was the reason why increased mediation led
to increased skepticism among children. Austin and Nach-Ferguson, meanwhile,
surmised that the reverse relationship was behind their finding that mediation
was positively related to the number of beer brands a child could recite. In
other words, parents who enjoyed beer advertising and exhibited less skepticism
in effect encouraged their children to "buy" the brands advertised. Ward and
Robertson (1972) similarly suggested that television advertising complements
communication about product consumption within the family, with high levels of
communication about consumption positively related to favorable attitudes toward
advertising.
H12: Skepticism toward television programming and advertising will
positively predict
negative mediation.
H13: Skepticism toward television programming and content will
negatively predict positive
mediation.
METHOD
A statewide phone survey (N=255) was conducted in Washington state
using a purchased list of phone numbers generated by a random digit dialing
system. Additional phone numbers were created using the "plus one" method. The
phone survey was conducted by undergraduate communication students who completed
a half hour training session prior to making calls. The survey took
approximately eight minutes to complete and a response rate of 55% was obtained.
Respondents
Respondents were parents who had at least one child living at home
between the ages of 2 - 17. If a parent had more than one child in that age
bracket they were instructed to think of the child with the most recent birthday
when answering the questions. More female parents (58%, n = 144) than male
parents (42%, n = 107) responded. Respondents were predominantly Caucasian
(90%, n = 225), with a small representation of Native American (n = 9), Hispanic
(n = 7), Asian (n = 4) and African American (n = 2) ethnicities. Respondents'
ages ranged from 19 - 69 (M=40). The mean education level of the respondents
was 3.33, or some college. The mean income level was between $35,000-50,000 and
ranged from under $15,000 to over $50,000. Among the parents, 76 (n = 188) were
married, with 9 percent (n = 23) single, 6 percent (n = 15) divorced and
single, 8 percent (n = 19) divorced and separated, and 2 percent (n = 4)
widowed. The gender of the child used as a reference for each parent respondent
was fairly evenly split, with 52% girls (n = 132) and 48% boys (n = 123). The
child's age also was approximately balanced with each age 2 - 17 making up four
to nine percent of the sample (M = 9.45).
Instrument
The instrument was a survey designed to measure parental concern
about television genres and issues, positive mediation, negative mediation,
viewing habits, attitudes toward advertising, attitudes toward alcohol use and
demographics. A list of measures is included in the appendix. Parental
mediation measures were based on the mediation scale used by Austin and
colleagues (Austin, 1993; Austin & Nach-Ferguson, 1995) for use with children.
The measures were modified for relevance to parent respondents, and additional
measures intended to tap positive mediation and to make distinctions among
mediation for advertising versus mediation for programming were included.
Descriptive statistics may be found in Table 1.
-------------------------------
Table 1 About Here
-------------------------------
Factor analysis was performed on each index to confirm its
unidimensionality, and Cronbach's alphas were computed to verify reliability.
The scale measuring negative mediation consisted of five items placed on a
five-point likert scale (alpha = .76). Parental concern toward television
content was measured using a list of three genres, (news, advertising, and
entertainment), for which respondents indicated whether they thought the genres
generally teach really good things or really bad things to children, as well as
a list of three issues, (political, sexual, and issues concerning alcohol), for
which respondents indicated whether television teaches children really good or
really bad things about each issue. Cronbach's alpha for the concern toward
genres scale was .66. Television viewing habits were measured by asking parents
how often they had viewed prime time, sports, educational, and news programs in
the past week and were analyzed as individual items. Coviewing was assessed by
asking parents how often they watched television with their child in the past
week. Attitudes toward advertising were measured by asking parents how
entertaining, honest, realistic, and stereotypical they perceived ads to be.
Demographic measures included education level, marital status, ethnicity,
income, age, and gender.
Hypothesis testing
Hypothesis 1 was tested via factor analysis, using principal
components extraction. Hypothesis 2 was tested via Pearson's correlation, and
Hypothesis 3 via one-way analysis of variance, using the Student-Newman-Kuels
test for post-hoc differences among groups. The typology of four groups was
developed via the median split procedure for each of the two positive and
negative mediation variables, with the respondents then coded as high frequency
on one or both mediation valence, or low frequency on one or both. All other
hypotheses were tested via hierarchical regression, with control variables
(income and education) entered in the first block using the stepwise procedure.
Independent variables were entered in the second block using the stepwise
procedure. This combination of hierarchical and stepwise procedures allowed for
a theory driven, stringent test of each independent variable while preserving
power relative to the modest sample size.
RESULTS
Parents reported relatively high levels of television-related
interaction overall. For coviewing the mean was 3.5 on a five-point scale
(somewhat often). The means for nonvalenced, negative and positive mediation
were 3.8 (often), 4.2 (often to very often) and 3.3 (sometimes) respectively.
Hypothesis 1 predicted that factor analysis of the items measuring
mediation would produce two distinct factors representing positive and negative
parental mediation behavior. Principal components factor analysis produced two
distinct factors in which each item of the mediation scale had a factor loading
of at least .60 on its primary factor and less than .40 on the secondary factor.
The first factor loaded all six items that reflected negative mediation
(communication countering television content) and explained 35 percent of the
total variance. The second factor loaded the remaining two items which
reflected positive mediation and explained 15 percent of the total variance.
Positive mediation items included telling a child that they agree with something
on television or in ads. Parental mediation scale items and factor loadings are
displayed in Table 2. The parental mediation scale measured positive and
negative mediation; thus, Hypothesis 1 was supported.
-------------------------------
Table 2 About Here
-------------------------------
Hypothesis 2 predicted that positive and negative mediation each
would correlate with nonvalenced mediation. The correlation for positive
mediation and nonvalenced mediation was .26 (p<.001), and the correlation for
negative and nonvalenced mediation was .21 (p<.001), consistent with the
prediction.
Hypothesis 3 predicted that significant differences would be found on
items measuring parental media use and opinions. This hypothesis also predicted
that the differences would be based on a typology constructed of
frequent/infrequent, positive and negative parental mediation, as follows:
1. nonmediators: low on both positive and negative mediation
2. optimists: high on positive but low on negative mediation
3. cynics: high on negative but low on positive mediation
4. selectives: high on both positive and negative mediation
In the ANOVA performed to examine differences among typology groups
on media use and opinion items, significant differences were found on six of
the eight items measuring parental opinions toward television. For parental
media use, significant differences were found on two of five items.
On overall concern for what television teaches children,
(F[3,237]=7.15, p<.001), cynics expressed significantly more concern than
optimists and selectives. Additionally, optimists expressed significantly less
concern than nonmediators. Overall cynics expressed the most concern followed
by nonmediators, selectives, and optimists.
For talking about television (F[3,246]=7.32, p<.001), nonmediators
reported talking about television with their children significantly less than
optimists, cynics, and selectives. Selectives reported talking about television
the most followed by optimists, cynics, and nonmediators.
On the item stating that television is a good babysitter,
(F[3.246]=3.68, p<.01), optimists expressed significantly less disagreement
than cynics and selectives. Optimists expressed the least disagreement, with
the statement television is a good babysitter, followed by selectives,
nonmediators, and cynics.
On the item stating that television is a good learning tool,
(F[3,239]=5.25, p<.01), cynics expressed significantly more disagreement than
nonmediators, optimists, and selectives. Cynics expressed the most
disagreement, with the statement television is a good learning tool, followed by
nonmediators, selectives, and optimists.
For believing that television advertisements are honest,
(F[3,245]=8.25, p<.001), optimists perceived advertisements to be honest
significantly more than nonmediators, cynics, and selectives did. Optimists
perceived the most honesty in television advertisements followed by
nonmediators, selectives, and cynics. The same pattern emerged for perceiving
television advertisements to be realistic, (F[3,245]=7.58, p<.001).
Concerning parental media use, cynics reported viewing significantly
less prime time television than optimists and selectives (F[3,246]=4.34, p<.01).
Selectives viewed the most prime time television followed by optimists,
nonmediators, and cynics. Cynics also spent significantly less time watching
television with their child than selectives did, (F[3,246]=2.98, p<.05).
Selectives reported spending the most amount of time viewing television with
their child followed by optimists, nonmediators, and cynics.
In sum, significant differences between the typology groups were
found on over half the items measuring parental media use and opinions towards
television, as shown in Table 3.
-------------------------------
Table 3 About Here
-------------------------------
Contrary to the prediction of Hypothesis 4, as shown in Table 4,
positive attitudes toward television content did not positively predict viewing
television with the child.
-------------------------------
Table 4 About Here
-------------------------------
Hypothesis 5 stated that watching television with a child
(coviewing), would positively predict positive mediation. Coviewing was found
to be a positive predictor of positive mediation,(b=.173, p<.05), as expected.
Hypothesis 6 predicted that frequent TV viewing would lead to
positive mediation. The study, controlling for education, found the
significant result (b=.159; p<.05) produced by viewing primetime TV programs.
The result indicated that the more the parents watch primetime TV programs, the
more likely the positive mediation to be used. Exposure to other types of
programming (such as sports, advertising, news, educational programming) made no
difference. Hypothesis 6 is accepted with this result.
The hypothesis that negative attitudes toward television would
predict negative parental mediation, was not supported.
Hypothesis 8 predicted that positive mediation behaviors would
be predicted by parent's positive views of TV. Parental attitudes toward TV
(b=.261; p<.001) significantly led to positive mediation, as expected.
Hypothesis 9 stated that negative mediation will negatively predict
agreement with the statement television is a good babysitter. Results showed
that negative mediation was a significant negative predictor of the view
television is a good babysitter (b=-.178, p<.001). It appears that more
frequent negative mediation leads to less of a view of television being a good
babysitter, in support of the hypothesis.
Hypothesis 10 predicted that positive mediation would lead to more of
a view of television being a good babysitter. Stepwise regression results
failed to support this hypothesis.
Hypothesis 11 predicted positive mediation would lead to a view of
television as a good learning tool. Positive mediation positively predicted a
view of television as a good learning tool, (b=.267, p<.001), in support of the
hypothesis.
Hypothesis 12 predicted a positive relationship between parent's
skepticism toward television ads and parental negative mediation. Both
skepticism measures (b=-.208; p=<.01 for 'ads are realistic'; b=-.154; p=<.05
for 'ads are honest') significantly predicted negative mediation, as expected.
Hypothesis 13 predicted a relationship between positive mediation and
skepticism toward advertising. This hypothesis was supported (b=.17, p<.01), as
shown in Table 4.
DISCUSSION
This study investigated whether parent-child interaction surrounding
television takes a number of conceptually distinct forms, each of which has its
own implications in its relationship to other aspects of the television
interaction context. The study investigated nonvalenced, positive and negative
mediation styles, coviewing, and parental perceptions of televisions usefulness
as a babysitter and as a learning tool.
Overall, this study has demonstrated the existence of at least three
types of mediation: positive, negative and nonvalenced. It also has
demonstrated that parents may embrace both positive and negative mediation
strategies, as well as one or neither. It suggests that positive mediation may
occur more due to happenstance, with negative mediation associating more with
critical viewing and protective motivations.
One strength of this study was its generalizability, due to the
random statewide sample employed. Some limitations, however, also should be
noted. In focusing only on parental reports, this study measured perceptions
rather than actual behaviors. It also tapped only one parent's perspective
about one child, when every family member could have a different interpretation
of communication patterns within the family. In addition, measures for positive
mediation were limited to two items, whereas the negative mediation index
included six items. Further development of measures to tap parent-child
communication would be useful. This sample also probably was limited in its
cultural diversity, given the small proportion of minority respondents.
This study has demonstrated a number of conceptual distinctions among
aspects of parent-child interaction regarding television. The results indicate
that an assessment of mediation per se does not establish whether discussion is
negative or positive. In particular, the results demonstrated the existence of
four distinct styles of parental mediation patterns, each of which displayed
certain unique motivations and associated behaviors. The data showed striking
differences among the four mediation typology groups identified as 1)
nonmediators (low level of engagement in both positive or negative mediation);
2) optimists (high level of engagement in positive mediation but low in negative
mediation); 3) cynics (high level in negative mediation but low level in
positive mediation) and 4) selectives (high level of engagement in both positive
and negative mediation).
Generally, optimists possessed more positive opinions on
television and TV ads, followed by selectives, then nonmediators, and then
cynics. Intuitive results included: 1) nonmediators are less likely to talk
about TV with children; 2) optimists are less skeptical about ads ("ads are
realistic" and "ads are honest"); 3) cynics are least likely to consider TV a
good learning tool.
Based on the eight significant instances of typological
differences, characteristics include: 1) Cynics - their attitudes toward TV/ads
are the most negative; they are most skeptical about TV ads and least likely to
consider TV a useful learning tool or a good babysitter; they watch prime time
TV the least; they coview with children the least; cynics, however, talk about
TV with children as much as optimists and selectives do, but more negatively; 2)
Optimists - their attitudes toward TV/ads are the most positive; they are least
skeptical about TV ads and more likely to agree that TV can be utilized as a
good babysitter and as a good learning tool; they watch prime time TV and coview
with children as much as nonmediators and selectives do; and optimists talk
about TV with children just as much as selectives and cynics do, but more
positively; 3) Selectives - they have some commonalities with optimists in terms
of media use, nonvalenced mediation, coviewing and overall positive attitudes
toward TV/ads; selectives are, however, more skeptical about TV ads than
optimists; although selectives tend to agree that parents can use TV as a good
learning tool, they are more likely to disagree with using TV as a good
babysitter compared to optimists; 4) Nonmediators - nonmediators are placed
somewhere between cynics and selectives; they are more negative and skeptical
than optimists in overall views on TV and ads; nonmediators watch prime time TV
and coview with children the same as optimists and selectives do; nonmediators
also tend to agree that TV can be a good learning tool; but their level of
discussion with children is significantly lower than the other three groups.
In sum, cynics possess the most negative views on TV/ads, which may
make their media use and opinions differ from the other three groups. This is
particularly true when comparing cynics with optimists. Both talk about TV
with children, but with a different emphasis. Differences between optimists
and selectives appear to be derived from their level of skepticism. Differences
between nonmediators and optimists may be explained in optimists' more positive
attitudes and lower skepticism. Interestingly, the only significant difference
between nonmediators and selectives is the frequency of TV discussion.
An important result from this study was its support for the
contention that coviewing and mediation are conceptually distinct. That
nonmediators, for example, had the same levels of coviewing as the other groups
suggested that coviewing can exist without any concurrent discussion. Coviewing
alone appears unlikely to develop critical viewing skills in children.
This study also identified parents' positive views on TV as well as
active media consumption (watching prime time TV and coviewing with children) as
a significant predictor for positive mediation behaviors. Parental positive
reinforcement, then, led to their agreement with the idea that TV can be used as
a good learning tool. It appears that liking of television is not a motivating
force--instead, liking inspires more viewing, which results in more coviewing.
Then once parent and child are together in front of the television, they may
discuss it via happenstance more than by intent. Their discussion is likely to
be more positive than negative. Parents' positive attitudes toward television
also may motivate them to direct their children to television as a useful tool.
This suggests that parents use television as tool to reinforce lessons, rather
than as a source of examples of what not to do.
Although positive attitudes associated with coviewing and use of
television as a tool, negative attitudes did not predict the use of negative
mediation. The results bordered on significance in the predicted direction
(b=-.136, p=.06), however, suggesting that the lack of a result may be
measurement based. Negative mediation may result from dissatisfaction with a
specific portrayal of sexual issues, rather than with their view of television
content as a whole. Their concern, then, may vary widely across topics and
situations.
In support of this interpretation, the results did show that negative
mediators tended to be more critical viewers overall, and that they tended to
watch less television. This, then supports the Gallup (1989) finding that
parents tend to turn off the television if content offends them, but it also
supports Dorr et al.'s (1989) assertion that parents discuss television more
when their values appear threatened. These results suggested that while parents
may turn off what they don't like, they may nevertheless continue discussing it.
Negative mediators do appear less willing to use television as a
babysitter. Positive mediators, however, are not any more likely to do so.
This could be because positive mediators also tended to coview more and watch
more prime-time television. Thus, their children simply may be less likely to
view television alone. Negative mediators, on the other hand, use television
less, coview less and also appear to want their children around it less.
Future research should flesh out the distinctions among types of
mediation styles more fully, including the implications of these strategies for
children's perceptions and behaviors. This study focused exclusively on
parental perceptions, which limits its explanatory ability. Although based on
theory, this study is primarily descriptive and is limited in its predictive
value for explaining media uses and effects for children. One barrier to this
study's explanatory power, for example, is that children may view the same
family environment differently than their parents have reported to us. Some
studies have suggested, based on children's reports, that parents do not mediate
television to a great extent. These results, on the other hand, suggested that
parents think they do mediate television quite a bit (m=3.8 on a five-point
scale). It could be that children key more on the negative aspects of mediation
rather than the positive, whereas parents also consider reinforcement of
television content as "mediation." Children also may not consider coviewing as
tacit reinforcement, even though in this study coviewing tended to associate
with positive reinforcement of television content. This merits further
investigation, since coviewing could indeed represent tacit reinforcement given
the results of this study. These findings could suggest that advising parents
to watch television with their children is insufficient, and at times even
counterproductive, advice if the intent is to cultivate critical viewing skills
in children.
This study suggests a number of avenues for future research, given
the distinctions that emerged among various aspects of parent-child interaction
regarding television. It appears from these results that some parents do indeed
take on the tube, while others, more complacent, reinforce many of the lessons
their children see portrayed. The motivations and implications of these
parenting patterns merit more study. It seems unlikely that the mere
manufacture of a V-chip will displace the importance of actual communication
between parent and child about the realities and values of television content.
Table 1
Means, standard deviations, and reliabilities of variables
N Means S.D. Range Alpha
-------------------------- -----------------------------------
Age (child) 255 9.45 4.71 2-17
Index: Attitudes (pos/neg) 245 15.56 4.18 6-30 .66
TV_politics 250 2.71 1.07 1-5
TV_sexual issues 249 1.93 1.11 1-5
TV_alcohol use 251 2.28 1.22 1-5
TV_news 253 3.28 1.15 1-5
TV_advertisements 255 2.29 1.23 1-5
TV_entertainment 255 3.07 1.13 1-5
Index: Positive mediation 250 6.58 2.07 2-10 r=.33***
Agreement with TV 252 3.70 1.23 1-5
Agreement with TV ads 251 2.87 1.31 1-5
Index: Negative mediation 243 25.35 4.40 6-30 .74
TV is not OK 252 4.64 .80 1-5
TV ads are not OK 252 4.16 1.15 1-5
Things on TV are not real 253 4.53 1.01 1-5
TV ads aren't true 251 4.22 1.08 1-5
Tell more about TV 251 3.95 1.23 1-5
Explain ad's intention 251 3.76 1.36 1-5
Talk about TV 254 3.83 1.19 1-5
TV ads are honest 253 2.11 .94 1-5
TV ads are entertaining 254 3.46 1.02 1-5
TV ads are realistic 253 2.04 .78 1-5
TV ads teach stereotypes 251 3.98 .84 1-5
TV is a good babysitter 254 2.02 1.13 1-5
TV is a good learning tool 247 3.44 1.01 1-5
Watch prime time TV 254 2.99 1.22 1-5
Watch sports programs 253 1.98 1.03 1-5
Watch educational shows 252 3.06 1.25 1-5
Watch news programs 254 3.87 1.21 1-5
Coviewing 254 3.50 1.30 1-5
Index: Desirability 240 11.84 4.08 4-20 .83
People in ads_successful 251 2.70 1.28 1-5
People in ads_healthy 249 2.72 1.28 1-5
People in ads_fun 247 3.02 1.27 1-5
People in ads_attractive 249 3.50 1.19 1-5
Index: Norms (Real Life) 241 9.54 3.28 4-20 .86
Beer drinkers_successful 249 2.37 1.01 1-5
Beer drinkers_healthy 246 2.26 .91 1-5
Beer drinkers_fun 247 2.58 1.06 1-5
Beer drinkers_attractive 247 2.38 .92 1-5
Table 1 (Cont.)
Means, standard deviations, and reliabilities of variables
N Means S.D. Range Alpha
-------------------------- -----------------------------------
No. of children 251 1.96 1.02
Income 208 1.96 1.02 0-3*
Education 250 3.33 1.09 1-5**
Age (parent) 246 39.56 7.88 19-69
--------------------------------------------------------------
Note: High score indicates an answer in agreement with the statement
or a more frequent behavior
* Income : 0 = <$15,000; 1= over $15,000 but <$35,000; 2 = over
$35,000 but <$50,000; 3 = over $50,000.
** Education: 1=less than high school; 2=high school graduate;
3=some college; 4 =college graduate; 5 = graduate work;
*** P=<.001
Table 2
Factor loadings : Parental mediation
(Principal component analysis)
----------------------------------------
Factor Factor Communality
1 2 Estimates
===============================================================
Tell Ad's Intention 0.71723 -0.16484 .54159
Ad is not True 0.65419 -0.27289 .50244
TV Ad is not OK 0.64785 -0.18622 .45439
TV is not OK 0.64100 -0.00008 .41088
Tell More about TV 0.62186 0.03801 .38815
TV is not Real 0.61917 -0.22920 .43590
Agree with TV 0.36670 0.74539 .69007
Agree with TV Ad 0.39596 0.66264 .59588
--------------------------------------------------------------
Eignevalue 2.83431 1.18499
Variance
Explained by 35.4 % 14.8%
Each Factor
==============================================================
Table 3
Results of stepwise regression tests of hypotheses~
-------------------------------------------Dependent Vars R2
Prob
Independ. Vars Change Beta DF F (Model)
=================================================================
Coviewing
Education .030 -.173* (1,193) 5.975 .02
H4 Attitudes .000 .005 (2,192) 2.974 .05
Positive Mediation
Income .020 -.142* (1,195) 4.040 .05
H8 Attitudes .066 .261*** (2,194) 9.185 .000
Income .021 -.145* (1,202) 4.353 .04
H5 Coviewing .030i .173* (2,201) 5.379 .005
H13 Income .022 -.147* (1,201) 4.463 .04
Ads are realistic .028 .171** (2,200) 5.310 .006
Income .020 -.141* (1,200) 4.084 .04
H6 Watch Primetime .025 .159* (2,199) 4.678 .01
Negative Mediation
H7 Attitudes .018 -.136 (1,191) 3.578 .06
H12 Ads are realistic .043 -.208** (1,197) 8.967 .003
Ads are honest .022 -.162* (2,196) 7.067 .001
TV as a Babysitter
Income .020 -.143* (1,198) 4.143 .04
H9 Neg Mediation .032 -.178** (2,197) 5.435 .005
Income .024 -.154* (1,202) 4.915 .03
H10 Pos Mediation .011 .109 (2,201) 3.694 .03
TV as a Learning Tool
H11 Pos Mediation .071 .267*** (1,196) 14.990 .0001
=================================================================
( Note: * P=<.05; ** P=<.01; *** P=<.001)
~ Controlling variables (education and income) are entered first in
the equation when they are significant. Betas reported from the block of entry.
TABLE 4
Parental concern, coveiwing, attitudes toward TV/AD, and TV viewing by
mediation styles ~
Mediation styles
----------------------------------------
---------------
Non- F
DF
Mediators Optimists Cynics Selectives
(Model) =============================================================
================
Desirability of ad 11.35 12.06 12.56 11.58 1.00
(3,232)
Social norms 9.79 10.09 8.87 9.40 1.41
(3,233)
Attitudes about TV 15.09b 16.98ab 13.64ac 16.28c
7.15*** (3,237)
Freq of talk about TV 3.25abc 4.02a 3.96b 4.07c
7.32*** (3,246)
Ads are honest 2.04a 2.63abc 1.82b 2.00c
8.25*** (3,245)
Ads are entertaining 3.52 3.54 3.47 3.29 0.83
(3,246)
Ads are realistic 1.96a 2.41abc 1.74b 2.04c
7.58*** (3,245)
Ads do stereotype 4.04 3.87 4.16 3.90 1.54
(3,243)
TV as a babysitter 1.96 2.43ab 1.74a 2.01b 3.68**
(3,246) TV as a learning tool 3.45a 3.70b 3.00abc 3.56c 5.25**
(3,239)
Watch Primetime 2.96 3.09a 2.54ab 3.29b 4.34**
(3,246)
Watch Sports 2.00 2.09 1.86 1.99 0.48
(3,245)
Watch Education 2.85 3.20 2.95 3.20 1.30
(3,244)
Watch News 3.99 3.83 3.72 3.93 0.58
(3,246)
Coviewing 3.40 3.56 3.16a 3.82a 2.98*
(3,246)
=============================================================================
(n=67) (n=54) (n=57) (n=72)
( Note : * P=<.05; ** P=<.01; *** P=<.001 for the entire Model;
~ Sharing the same letter indicates significant differences between
the
groups.
Appendix
Measures used in the analysis
Coviewing:
five-point scale: never to often
How often do you and your child watch TV together?
Mediation Frequency:
five-point scale: never to often
A. Nonvalenced:
How often do you talk about TV with your child?
B. Negative Mediation:
I'd like to know how often tell your child more about something you've
seen on TV?
How often do you tell your child that something you've seen somebody
do on TV is not OK?
How often do you tell your child that something on TV is not real?
How often do you tell your child that an ad on TV says something that
isn't really true?
How often do you tell your child that something you've seen in a TV ad
is not OK?
How often do you explain to your child what ads on TV are trying to
do?
C. Positive Mediation:
How often do you tell your child that you agree with something you've
seen in a TV ad?
How often do you tell your child that you agree with something you've
seen on TV?
Attitudes Toward Television Content:
five point scale: teaches really bad things to teaches really good
things
A. Topics:
Politics
Sexual issues
Alcohol use
B. Genres:
TV News
Advertisements
Entertainment programs
Desirability of Alcohol Advertising:
five-point scale: strongly disagree to strongly agree
People in beer ads are successful.
People in beer ads are healthy.
People in beer ads are fun.
People in beer ads are attractive.
Perceived Social Norms for Alcohol:
five-point scale: strongly disagree to strongly agree
People who drink beer in real life are successful.
People who drink beer in real life are healthy.
People who drink beer in real life are fun.
People who drink beer in real life are attractive.
Skepticism Toward Television Advertising:
five-point scale: strongly disagree to strongly agree
TV ads are honest.
TV ads are realistic.
Attitudes Toward Advertising:
five-point scale: strongly disagree to strongly agree
TV ads are entertaining.
TV ads teach stereotypes.
Attitudes Toward TV Source Use:
five-point scale: strongly disagree to strongly agree
TV is a good babysitter.
TV is a good learning tool.
Viewing Habits:
five-point scale: not at all to every day
About how many days a week do you watch the following types of shows?
Prime Time TV?
Sports programs?
Educational shows?
News programs?
References
References
Alexander, A., & Fry, V. H. (1990). Interpreting viewing: Creating
an acceptable context. In W. A. E. S. Thomas, (Ed.), Communication and Culture:
Language, Performance, Technology and Media, pp. 236-243. Norwood, NJ: Ablex.
Anderson, D., & Collins, P. (1988). The impact on children's
education: Television's influence on cognitive development. Working paper No.
2, Washington, DC.: U.S. Department of Education, Office of Educational Research
and Improvement.
Atkin, D. J., Greenberg, B. S., & Baldwin, T. F. (1991). The home
ecology of children's television viewing. Journal of Communication, 41, 3,
40-52.
Atkin, D., Heeter, C., & Baldwin, T. (1989). How presence of cable
affects parental mediation of TV viewing. Journalism Quarterly, 66, 557-563,
578.
Austin, E. W. (1992). Parental mediation of television content: A
test of a measurement index. Paper presented to the Mass Communication Division
of the International Communication Association.
Austin, E. W. (1993). Exploring the effects of active parental
mediation of television content. Journal of Broadcasting & Electronic Media,
Spring, 147-158.
Austin, E. W., & Nach-Ferguson, B. (1995). Sources and influences of
young school-age children's general and brand-specific knowledge about alcohol.
Health Communication, 7, 1-20.
Austin, E. W., Roberts, D. F., & Nass, C. I. (1990). Influences of
family communication on children's television-interpretation processes.
Communication Research, 17, 4, 545-564.
Bower, R. T. (1973). Television and the Public. New York: Holt,
Rinehart and Winston.
Bybee, C., Robinson, D., & Turow, J. (1982). Determinants of
parental guidance of children's television viewing for a special subgroup: Mass
media scholars. Journal of Broadcasting and Electronic Media, 26, 697-710.
Chaffee, S., McLeod, J., & Wackman, D. 1973). Family communication
patterns and adolescent political participation. In J. Dennis, (Ed.),
Socialization to Politics: Selected Readings, pp. 349-364. New York: Wiley.
Comstock, G. (1975). The evidence so far. Journal of Communication,
25, 4, 25-34.
Corder-Bolz, C. R. (1980). Mediaiton: The role of significant
others. Journal of Communicaiton, 30, 3, 106-118.
Desmond, R. J., Singer, J. L., Singer, D. G., Calam, R., & Colimore,
K. (1985). Family mediation patterns and television viewing: Young children's
use and grasp of the medium. Human Communication Research, 11, 461-480.
Dorr, A., Kovaric, P., & Doubleday, C. (1989). Parent-child
coviewing of television . Journal of Broadcasting & Electronic Media, 33.
33-51.
Gallup, G. J. (1989). Parents disturbed by TV shows. San Francisco
Chronicle, 84.
Kandel, D., & Logan, J. A. (1984). Patterns of drug use from
adolescence to young adulthood. I. Periods of risk for initiation, continued
use, and discontinuation. American Journal of Public Health, 74, 7, 660-666.
Lyle, J., & Hoffman, H. R. (1972). Children's use of television and
other media. , 4. In E. A. Rubinstein, & G. A. & Murphey, J. P. Comstock,
(Ed.), Television and social behavior: Vol. 4. Television in day-to-day life:
Patterns of use, pp. 129-256. Washington, DC: U. S. Government Printing Office.
Medrich, E. A. (1979). Constant television: A background to daily
life. Journal of Communication, 29, 3, 171-176.
Messaris, P. (1982). Parents, children, and television. In G.
Gumpert, & R. Cathcart, (Ed.), Inter/Media. New York: Academic Press.
Meyrowitz, J. (1985). No sense of place: The impact of electronic
media on social behavior. New York: Oxford University Press.
Mittal, B. (1994). Public assessment of TV advertising: Faint
praise and harsh criticism. Journal of Advertising Research, 35-.
Ridley-Johnson, R., Surdy, T., & O'Laughlin, E. (1991). Paretn
survey on television violence viewing: Fear, aggression, and sex differences.
Journal of Applied Developmental Psychology, 12, 63-71.
Salomon, G. (1981). Introducing AIME: The assessment of children's
mental involvement with television. In H. Kelly, & H. Gardner, (Ed.), Viewing
children through television, pp. 89-102. San Francisco: Jossey Bass.
Tatar, M. M. (1991). Off with their heads! Fairy tales and the
culture of childhood. Princeton, NJ: Princeton University Press.
van der Voort, T. H. A., & Nikken, P. van Lil, J. E. (1992).
Determinants of parental guidance of children's television viewing: A Dutch
replication study. Journal of Broadcasting & Electronic Media, 36, 61-74.
Ward, S., & Robertson, T. S. (1972). Adolescent attitudes toward
television advertising: Preliminary findings. In E. A. Rubinstein, (Ed.),
Television and Social Behavior, Volume IV, pp. 526-542. Washington, DC: U. S.
Government Printing Office.
|
