Content-Type: text/html

Information Seeking & Behavior
 
 
 
 
 
The Ability of the AIDS Quilt to Motivate Information Seeking, Personal
Discussion and
Behavior as a Health Communication Intervention
 
 
 
by
 
Christopher Stephen Knaus
M.A., 1997
 
 
Bruce E. Pinkleton
Assistant Professor
 
and
 
Erica Weintraub Austin
Associate Professor
 
 
Edward R. Murrow School of Communication
Washington State University
Pullman, WA  99164-2520
 
Phone:  (509) 335-2795
Fax:  (509) 335-1555
E-mail:  <a href="/cgi-bin/wa?LOGON=A3%3Dind9812D%26L%3DAEJMC%26E%3D7BIT%26P%3D1980224%26B%3D--%26T%3DTEXT%252FPLAIN%3B%2520charset%3DUS-ASCII" >[log in to unmask]</a>
 
 
Manuscript submitted for presentation to the Communication Theory and
Methodology Division of the Association for Education in Journalism and Mass
Communication
August 1998, Baltimore, MD
 
 
Running Head:  Information Seeking and Behavior
 
Abstract
        Several seldom-used approaches have demonstrated significant effects with
regard to HIV and AIDS education and prevention. The NAMES Project Foundation's
AIDS Memorial Quilt is designed to encourage compassion and increased emotional
appeal, which is intended to lead to increased desire to seek information and
develop skills concerning the transmission and prevention of the disease.  A
field experiment (n=560) was used to examine the ability of the AIDS Quilt to
motivate information seeking, personal discussion and behavioral outcomes among
those who viewed it.  Results indicate that  the quilt intervention did explain
significant differences in information-seeking motivations and
information-seeking behavior.  Information-seeking motivation positively
predicted actual information seeking behavior, which in turn predicted increased
discussion and decreased risky behavior.  Information-seeking motivation in
itself did not predict discussion or behavior.  The results suggest that
campaigns designed primarily to increase information seeking motivation can
result in desired behavioral outcomes.
 
 
The Ability of the AIDS Quilt to Motivate Information Seeking, Personal
Discussion and
 Behavior as a Health Communication Intervention
        A large proportion of the governmental resources given to agencies fighting the
AIDS epidemic are spent on preventative measures (U.S. Department of Health and
Human Services, 1987).  Because the disease can affect people from any community
with disastrous results, multiple education efforts have been used with varying
degrees of success.  These efforts range from broad educational approaches that
attempt to influence the general public to narrowly targeted approaches that
tailor messages to individuals in targeted community groups (O'Donnell, San
Doval, Vornfett, & DeJong, 1994;  Stevenson & Davis, 1994).  The large number of
people with AIDS (Centers for Disease Control, 1996) combined with the fact that
the virus virtually is preventable points out the importance of developing and
testing a variety preventative approaches.
        Several seldom-used approaches have demonstrated significant effects with
regard to HIV and AIDS education and prevention.  These range from using theater
and drama, to using small group facilitation. The NAMES Project Foundation's
AIDS Memorial Quilt, for example, attempts to incorporate the strengths of
theater and art with identity building and emotional appeal in a non-threatening
way.  Rather than focus on the presentation of information or skill building,
the quilt is designed to encourage compassion and increased emotional appeal,
which is intended to lead to increased desire to seek information and develop
skills concerning the transmission and prevention of the disease.  In this
sense, the quilt is used as a tool to motivate individuals who view the quilt to
learn about AIDS and take the needed steps to alter their own behavior as well
as to help others do the same. Since message exposure and attention are key
aspects of decision making, it is important to determine whether the quilt is
successful at boosting information-seeking motivations and key behavioral
outcomes.  Therefore, the purpose of this study is to assess relationships among
viewing the AIDS Quilt and the motivation to seek additional information,
discuss AIDS with others and engage less in risky behavior associated with the
transmission of the disease.
Seeing the AIDS Quilt and Information Seeking
        Despite the importance of message exposure and audience attention in
successful health campaign outcomes (McGuire, 1989), many mass mediated
campaigns fail to positively impact people's health-related behaviors (Parrott,
1995).  Campaigns based on moral arguments, fear appeals and knowledge-only
approaches often are counterproductive or limited in their effectiveness.
Health campaign communication strategies that show greater potential are those
emphasizing social influence and social and cognitive skill development (Austin,
1995).  The key to this campaign strategy is the recognition that effective
campaign efforts must be designed to target individuals and the environments in
which they live. The AIDS quilt appears to exemplify the latter approach.
        As a single display, the AIDS quilt is a locally-targeted health campaign
designed to raise awareness so that additional information is sought and HIV
preventative behaviors are learned by campaign receivers.  The potential for
success of the quilt appears to hinge on its ability to boost audience
involvement and reduce the perceived social distance between the general public
and those who suffer with AIDS.  One of the most common failings of
communication campaigns is to  assume audiences are interested when they are not
or when they may be actively resistant to threatening messages (Mendelsohn,
1973; Solomon, 1989).  It is common for a majority of the public to lack
interest in communication campaigns, leading to apathy and ignorance on topics
that are potentially of great importance (Hyman & Sheatsley, 1947).  Langer
(1978) suggests that audience attention to campaign messages exists along a
continuum.  At one end of the continuum, audience attention is limited and
responses to campaign messages are likely to be a passive lack of interest.  At
the other end of the continuum, listeners actively attend to messages resulting
in thoughtful message consideration and knowledge gain.  As a result, boosting
involvement may be key, particularly for a campaign whose success requires a
thoughtful response from receivers.
        Accordingly, the Elaboration Likelihood Model developed by Petty & Cacioppo
(1981, 1986) suggests that audiences engage either in thoughtful scrutiny of an
argument or less thoughtful consideration of the cues in a persuasive
environment based on individual differences in information processing
strategies.  The most important predictor of the amount of  cognitive effort an
individual will exert in argument elaboration is individual involvement with the
subject of a message (Petty & Cacioppo, 1986; Stiff, 1986).  When involvement is
high, audience members are motivated to engage in argument elaboration resulting
in greater message attention and scrutiny.  When involvement is low, audience
members are unwilling to commit cognitive resources to argument elaboration
instead choosing to engage in a heuristic processing strategy (Petty & Cacioppo,
1986; Stiff, 1986).
         Although there have been differences in the conceptualization of the
involvement construct, in general, involvement is associated with the personal
relevance of a topic or message to the individual on either a long- or
short-term basis (see generally Johnson & Eagly, 1989; Pfau & Parrott, 1993;
Salmon, 1986; Zaichkowsky, 1985, 1986).  Research indicates that
involvement-relevant messages attract more attention and cognitive effort than
other messages (Celsi & Olson, 1988; Greenwald & Leavitt, 1984), and that
involvement associates with active, purposeful and deliberate information
seeking (Atkin, 1973; Donohew & Tipton, 1973; Gantz, Fitzmaurice & Fink, 1991).
Additional research by Roser (1990) indicates that attention to messages
increases learning and the perceived importance of message content, and that
higher perceptions of message relevance directly influence attitudes and
behaviors.
        In the current study, viewing the AIDS quilt should associate positively with
information seeking since information-seeking behaviors are associated with
higher levels of involvement, attention and learning.  Research indicates that
individual's communication-related activity increases when situational
involvement is stimulated by a temporary social situation (Kanihan & Chaffee,
1996) and active information seeking results when involvement is high (Grunig,
1979).  As a result, if individuals are motivated to seek additional information
about AIDS, they also should expand their use of a variety of relevant
information sources.
        H1:  Those who view the AIDS quilt will have greater motivation to seek
information     regarding AIDS than those who do not view the AIDS quilt.
        H2:  The motivation to seek more information regarding AIDS will positively
associate       with using information sources to learn more about  AIDS.
        H3:  Those who view the AIDS quilt will engage in greater information source
use     regarding AIDS than those who do not view the AIDS quilt.
        Information Source Use and Behavioral Outcomes
        According to decision making approaches, awareness and involvement may lead to
changes in behavior by encouraging individual movement among stages of change
(Holtgrave, Tinsley, & Kay, 1995).  For example, Maibach and Cotton (1995)
suggest that those in the precontemplative stage of decision making often fail
to recognize the targeted problem and lack motivation.  Campaign messages
designed to encourage self examination instead of behavior change tend to
increase involvement, moving message recipients through the decision-making
process in incremental stages.  Once people have moved into a contemplative
stage they seek information as a way to consider costs and benefits and develop
skills to overcome perceived barriers to change.  If this information search is
successful, its result will be the beginning of behavior change preparation such
as interpersonal discussion. Ultimately, a successful information search can
lead to behavior change attempts (Maibach & Cotton, 1995).
        A purposeful information search (Chaffee & McLeod, 1973; Grunig, 1979; McCombs,
1972) is likely to include interpersonal conversations and various, specific
mass media (Atkin, 1973; Pinkleton & Austin, 1997) resulting in greater media
use and situation-specific learning (Kanihan & Chaffee, 1996).  Related research
indicates that individuals with diverse information needs are motivated to
extend their information search to include sources that are more costly in terms
of effort, but also are more rewarding in terms of information provision (Chew,
1994).  Ultimately, the ELM predicts that attitudes formed as a result of
message scrutiny and cognitive effort are likely to be enduring, predictive of
an individual's behavior and resistant to change.  If this is the case, then
involvement should increase the use of information sources to learn about AIDS
ultimately leading to an increase in interpersonal discussion concerning the
disease.  In the same way, as involvement motivates information source use and
message elaboration, resulting attitudes concerning AIDS should be predictive of
individual behavior.
        H4:  Greater use of information sources to learn about AIDS will positively
associate with
        engaging in discussion regarding AIDS.
        H5:  Greater use of information sources to learn about AIDS will negatively
associate with
        engaging in sex without using a condom.
Method
Research design
        Since college undergraduates are at particular risk for HIV transmission
(Hollar & Snizek, 1996; Sheer & Cline, 1995;  Turner, Garrison, Korpita, Waller,
Addy, Hill, & Mohn, 1994), the sample consisted of undergraduates spanning
several academic departments in a rural location.  In this study, four randomly
selected clusters were assigned to experimental and control groups according to
the Solomon four-group design (Campbell & Stanley, 1963; Ragon, Kittleson, & St.
Pierre, 1995).   One group received a control intervention followed by a
posttest.  This controlled intervention consisted of attending a public speech
or training session relevant to the class.  Another control group completed a
pretest prior to the controlled intervention.  Two other groups observed the
same procedures as the first two, except that they received the experimental
intervention, which was  a visit to the AIDS quilt.  The final group did not
complete a pretest, but did attend the AIDS quilt display, followed by the
completion of the posttest.  Students received extra credit in their classes for
attending the assigned intervention.  The section teacher made no mention of the
other interventions with regard to extra credit.  Because this was a field
experiment with limited control,  students in control groups who attended the
quilt because they had wanted to, which was assessed in the survey, were
subsequently placed in an experimental group, depending on whether or not they
had completed a pretest.  Those who were in an experimental section, but chose
to not attend the quilt were classified as part of the control group, since they
had not attended the intervention.  This allowed for the maintenance and
integrity of the Solomon four-group design, while remaining flexible enough to
account for the realities of the natural field experiment situation.
Additionally, the use of the Solomon four-group study design allows for much
more control than using a standard control/experimental group design since it
delineates at least between pretest bias and intervention bias (seeing the quilt
on a voluntary basis, or as a way to earn extra credit).  Finally, the Solomon
four-group design addresses external validity concerns while accounting for some
of the interaction between the interventions and the instrument (Ragon, et al.,
1995).
        The current analysis focuses only on the posttest measures, although the
analysis of the full design has been reported elsewhere (Knaus & Austin, 1998).
Although the most powerful analysis of intervention effects would involve
assessing change between the pretest and the posttest for those who did or did
not view the quilt, this type of analysis is not feasible with the current data.
Due to the sensitive nature of the questions in the instrument, the
institutional review board responsible for approving the study prohibited the
tracking of individual respondents.  As a result, the study design makes it
possible to control for pretest contamination effects, but analysis of effects
must rely on posttest group differences.  Individual-level effects can be
determined only by analyzing relationships among posttest variables in the
context of known group-level differences.
        As a result, hypotheses regarding quilt effects were assessed via t-test, to
determine whether groups who viewed the quilt differed significantly from groups
who did not view the quilt.  Hypotheses regarding relationships among
decision-making steps such as information-seeking motivations and
information-seeking behavior were tested via regression analysis.  Once it has
been established whether variables such as information-seeking motivations have
been affected by the quilt-viewing intervention, it can be assumed that
significant differences reflected in these variables will be represented in
these variables' relationships with other variables.  Thus, if viewing the quilt
increases information-seeking motivation, and information-seeking motivation
positively predicts discussion of HIV/AIDS with a sexual partner, it can be
concluded that viewing the quilt has indirectly increased discussion by way of
its effects on information-seeking motivations.
        Regression analysis included the use of age as a control variable, since sexual
activity and information seeking about HIV/AIDS could be accounted for partly by
age.  Age was entered using the forward-stepwise procedure in the first block of
a hierarchical multiple regression procedure.  Independent variables such as
information-seeking motivation and information source use then were entered
using the forward-stepwise procedure in the second block of the equation.
Sample
        Due to some loss of data resulting from administrative difficulties (some
surveys were lost in the campus mail), there were unequal numbers of individuals
in the Solomon-four groups. A total of 62 respondents participated in the
control group and received only the posttest.  Another 187 control-group
participants received both pretest and posttest.  Of those receiving the
treatment (the viewing of the quilt), 229 received the pretest and the posttest,
with 82 received only the posttest.  Thus, among the 560 respondents, a total of
406 completed pretests.  A total of 311 received the treatment, with a total of
249 individuals not receiving the treatment.  Of the clusters that completed
pretests, the majority were first-year students (64.6%).  Slightly more of the
respondents were female (54%) than male.  The mean age was 19.41, with a
standard deviation of 1.80.  A significant majority of the respondents were 18
or 19 years old (68.7%).  The percentage of Caucasian students in the
university, as of fall 1996 was 88%, and for this study, 74.7% of the
respondents indicated they were Caucasian.
        Among the respondents completing the posttests, again the majority of the
respondents indicated they were first-year students, but to a lesser extent
(58.4%).  Again, a majority (53.8%) of the respondents were female.  The mean
age was slightly higher (M = 19.69), following the trend of fewer first-year
students than in the pretests.  The standard deviation, however, remained
constant when compared to the pretests, at 1.82.  The majority of the posttest
sample was representative of the overall university, in that 75.2% of the
respondents were Caucasian.
        Pencil and paper surveys were administered to students in their classes by
class professors and teaching assistants.  The pretests were given out
approximately four weeks before the interventions, with posttests given out
approximately three to four weeks after the intervention.
 
Instrument and operational definitions of concepts
Safe-behavior and discussion indices
        Safe-behavior as a concept was limited for the purposes of this study to
addressing sexual contact and transmission.  This was primarily due to a
priority for in-depth analysis of one method of transmission rather than a brief
look at several modes of transmission.  Further, an assumption was being made
that focusing on drug-use behavior and/or pregnancy rates of students in a rural
university would not yield significant numbers, whereas many individuals within
this target population engage in risky sexual behavior (Keeling & Engstrom,
1994; Sheer & Cline, 1995).  The safe behavior index was adapted from
Basen-Engquist (1994), as part of her reliable safer-sex behavior variable (with
an alpha of .72), and used a five response format, varying from often,
sometimes, rarely, never, and offering a don't know or not applicable response.
This index was broken into two parts, with the first assessing how frequently
respondents perceived themselves to engage in discussion about specific
behavior.  Items included how often respondent and a partner "discussed
HIV/AIDS," "sexually transmitted diseases," "safe sex," "sex," and "past
partners."
        The second part of the index asked respondents to report sexual activity,
focusing on sexual intercourse and oral sex without birth control.  Items
included how often respondents "engaged in sexual intercourse without any form
of birth control," "engaged in sexual intercourse without a condom," "engaged in
oral sex without a condom," "engaged in sexual activity with someone you didn't
know very well," and "engaged in behavior from which you could contract
HIV/AIDS."  These indices were prefaced by a question asking if the respondents
had engaged in any sexual contact within the past month.  The two indices were
broken down into discussion and reported behavior separately.
        Respondents with no sexual contact in the past month were advised to skip the
questions regarding sexual behavior and frequency of discussion.  As a result,
the number of respondents for these items is somewhat lower than for other
questions on the survey.  Factor analysis and the computation of Cronbach's
alpha established that the measures did not form a reliable index; thus,
individual variables were retained for further analysis.
Information source use indices
        Sources of information were conceptually defined in two ways; which sources
were used more often, and which were seen as more effective.  Only the
information source use measures are reported in the current analysis.  The
sources of information index was developed in order to assess which source of
information respondents used most frequently in order to get information about
HIV and AIDS.  It is a more expanded version of the scale comprised by
Dusenbury, Diaz, Epstein, Botvin, & Caton (1994), who did not report reliability
measures.  This was coded using an often, sometimes, rarely, never one through
four scale, and included items such as mass media, health centers, school,
peers, family, 1-800 numbers, religious sources, and the Internet.  Descriptive
statistics for all variables used in these analyses may be found in Table 1.
---------------------------------
Table 1 About Here
---------------------------------
Index Building
        The construction of each individual index began with the selection of the
original items designated from the survey instrument as conceptually related
(most from previously tested scales and indices).  Each pretest and posttest
index was then tested using factor analysis to determine whether items actually
held together as a single factor.  Items were kept parallel in both pretest and
posttest indices so as to best analyze across group type.  Thus, if an item was
dropped at pretest, for example, it was also dropped at posttest.  After factor
analysis was used to analyze the indices, computation of Cronbach's alpha was
used to determine reliability.
        Factor analysis indicated no relation among the items within the behavior
index, so the individual items were analyzed separately instead of as in index.
The discussion index was not altered at all, with all five items maintaining
relatively high alphas.  The alpha for discussion with partners in the pretests
was acceptable (.71), while the posttests again held together more closely
(.79).   In the pretest index, the discussion of sexually transmitted diseases
and discussion of HIV loaded above .89 in the single factor analysis, while the
discussion of sex, discussion of safe sex, and the discussion of past partners
held together above .52.  In the posttest index, the discussion of HIV,
discussion of sexually transmitted diseases, and discussion of past partners
loaded above .53, while the discussion of sex and discussion of safe sex had
loaded above .86.  Relevant discussion with partners, measured on an index
ranging from 5-25, also decreased slightly, from the pretests (M = 16.37, N =
139) to the posttests (M = 15.97, N = 186), suggesting that the pretest sample
discussed relevant issues more often than the posttest sample.
Results
        As predicted by Hypothesis 1 and shown in Table 2, those who viewed the quilt
reported greater motivation to seek information about HIV/AIDS (M=7.48)  than
did those who did not
---------------------------------
Table 2 About Here
---------------------------------
view the quilt (M=7.48, t(320)=-3.77, p < .001).  T-tests also demonstrated a
link between seeing the Quilt and information source use (t (307) = -3.66, p <
.001), as predicted by Hypothesis , suggesting that frequency of information
source used increased as a result of viewing the Quilt.
        As predicted by Hypothesis 2, and shown in Table 3, motivation to seek
information positively predicted reported information source use (b=.23,
p<.001).  Consistent with
---------------------------------
Table 3 About Here
---------------------------------
Hypothesis 4, increased information source use positively predicted discussion
of HIV/AIDS with sexual partners (b=.23, p<.01), and information source use also
negatively predicted engaging in sexual contact without the use of a condom, as
proposed by Hypothesis 5.
Information-seeking motivation, included in the regression analysis using the
forward-stepwise procedure, did not significantly predict discussion or risky
behavior.
        In sum, the quilt intervention did explain significant differences in
information-seeking motivations and information-seeking behavior.
Information-seeking motivation positively predicted actual information seeking
behavior, which in turn predicted increased discussion and decreased risky
behavior.  Information-seeking motivation in itself did not predict discussion
or behavior.
Discussion
        Although health campaign planners have been moving away from purely
information-based campaigns, individual information seeking nevertheless remains
important to program success.  This study has explored the potential of
information-seeking motivations for moving message recipients through stages of
decision making toward the ultimate goal of behavior change.  The NAMES Project
Foundation's AIDS Memorial Quilt was analyzed as an intervention designed to
increase information seeking by way of increased involvement.  The results
indicate that viewing the quilt does increase the desire to obtain information,
leading to increased use of information sources, and in turn to positive
behavior changes.
        The analysis of posttest results from a Solomon Four-Group Design field
experiment showed that individuals who viewed the quilt reported higher levels
of information-seeking motivations.  They both wanted to learn more about
HIV/AIDS and reported that they were trying to learn more.  The results further
indicated that this motivation predicted actual increases in information source
use.  This suggests that the quilt holds promise as an intervention, due to its
ability to increase receiver involvement.  Its goal of raising awareness through
the personal experience of viewing quilt panels designed to honor the lives of
individual AIDS victims appears to have more far-reaching effects.  The findings
add further support to literature suggesting that campaigns designed to
encourage self examination instead of directly targeting behavior change are
successful because they acknowledge that audience members must move through an
incremental decision-making process before implementing a change in behavior
(e.g., Maibach & Cotton, 1995).
        Individuals differ greatly in their motivation to seek and process information
via different channels, and their responses to health communication campaign
messages often depend on a variety of factors that differ significantly among
individuals.  Arbitrary or alternative message wording, for example, can
significantly alter the decisions and behaviors they produce in targeted
audiences (Holtgrave, et al., 1995).  While individuals generally engage in
decision making behavior because they seek a positive outcome (Yates, 1990),
long-term benefits such as avoiding AIDS can be overlooked in order to maximize
short-term benefits such as immediate physical gratification.  At other times,
individuals may be overwhelmed by the challenges presented by a decision,
finding it more difficult to broach the topic of HIV avoidance with a partner
than to simply abdicate their decision-making responsibility (Holtgrave, et al.,
1995).  As a result, communication campaign practitioners need to understand
message recipients' decision making processes and motivations, and take a
receiver-oriented approach when designing health campaign messages (Austin,
Pinkleton & Fujioka, 1998).
        A receiver-oriented approach acknowledges that individual motivations are an
important part of information source use (Reagan, 1995) and knowledge gain
(Pinkleton, Reagan, Aaronson, & Chen, 1997).   Indeed, the results of this study
indicated that information-seeking motivations did increase source use, with
positive behavioral results.  Greater use of information sources associated
positively with engaging in discussion with sexual partners regarding HIV/AIDS,
and source use also negatively associated with engaging in sex without a condom.
These findings support decision-making models of behavior change, which suggest
that campaigns must help individuals move from a less involved state toward the
development of skills and motivations that enable behavior change (Maibach &
Cotton, 1995).  By spurring information seeking, the AIDS Quilt encouraged
viewers to pursue their own individual information needs, thereby enabling them
to develop the skills and confidence to bring up the topic of HIV with their
partners and protect themselves more effectively from the disease.
        The study cannot account for individual change over time as a result of the
quilt intervention, due to the limitations of the data collection methodology.
Because tracking of individuals was not possible, the analysis of effects relied
on posttest group differences.  Nevertheless, the Solomon four-group design
enables the researcher to have more confidence in differences revealed via
posttest data analysis, due to its ability to control for potential pretest
bias.  Individual-level effects, meanwhile, can be determined by analyzing
relationships among posttest variables in the context of known group-level
differences.  Thus, once it was established that information-seeking motivations
were affected by the quilt-viewing intervention, it could be assumed that
significant differences reflected in motivations were represented in their
relationships with other variables.  Thus, because viewing the quilt increased
information-seeking motivation, and information-seeking motivation positively
predicted discussion of HIV/AIDS with a sexual partner, it could be concluded
that viewing the quilt has indirectly increased discussion by way of its effects
on information-seeking motivations.
        It is important, nevertheless, to consider the limitations of this study.
Because it was a field experiment, for example, it did not have as tight
internal control as a laboratory experiment would enjoy.  In addition, its focus
on one subpopulation (college students at a rural northwestern university)
limits its generalizability.  College students, however, are an important target
for HIV/AIDS interventions (Hollar & Snizek, 1996; Sheer & Cline, 1995; Turner,
et al., 1994), and rural areas are an especially problematic environment for
campaigns due to the lack of AIDS- related resources and the tendency for many
rural communities to view HIV and AIDS as a big-city problem (Berry, McKinney, &
McClain, 1996; Governor's Advisory Council, 1994; Half- million, 1996). Finally,
the variance explained by information seeking, while significant, was not large,
indicating that other factors play an important role in decision making.
        The major contribution of this study lies in its ability to provide an
empirical evaluation of a major, ongoing, nontraditional campaign strategy.
Besides demonstrating its effectiveness, the study puts its success in the
context of a theoretical foundation that can be applied to other health
communication campaigns.  The results demonstrate that campaign strategies
maximizing a receiver orientation and focusing on increasing involvement rather
than focusing purely on behavior change can achieve broader effects than simply
raising awareness.
References
        Atkin, C. K.  (1973).  Instrumental utilities and information seeking.  In P.
Clarke (Ed.), New models for mass communication research (pp. 205-242).  Beverly
Hills, CA:  Sage.
        Austin, E. W. (1995).  Reaching young audiences:  Developmental considerations
in designing health messages. In E. W. Maibach & R. L. Parrott (Eds.), Designing
messages:  Approaches from communication theory and public health practice (pp.
114-144).  Thousand Oaks, CA:  Sage.
        Austin, E. W., Pinkleton, B. E., & Fujioka, Y.  (1998, July).  Assessing public
service announcement effectiveness:  Effects of message quality, production
quality and persuasiveness.  Paper presented to the International Communication
Association convention, Jerusalem, Israel.
        Basen-Engquist, K. (1994).  Evaluation of a theory-based HIV prevention
intervention for college students.  AIDS Education and Prevention, 6, 412-424.
        Berry, D. E. McKinney, M. M., & McClain, M. (1996).  Rural HIV-service
networks:  Patterns of care and policy issues.  AIDS & Public Policy Journal,
Spring, 36-46.
        Campbell, D. T., & Stanley, J. C. (1963).  Experimental and quasi-experimental
designs for research.  Chicago:  Rand-McNally.
        Celsi, R. L., & Olson, J. C.  (1988).  The role of involvement in attention and
comprehension processes.  Journal of Consumer Research, 15, 210-224.
        Centers for Disease Control (1996). HIV/AIDS Surveillance Report.  Centers for
Disease Control and Prevention, 8 (2), 1-39.
        Chaffee, S. H., & McLeod, J. M. (1973).  Individual vs. social predictors of
information-seeking.  Journalism Quarterly, 50, 237-245.
        Chew, F.  (1994).  The relationship of information needs to issue relevance and
media use.  Journalism Quarterly, 71, 676-688.
        Donohew, L., & Tipton, L.  (1973).  A conceptual model of information seeking,
avoiding and processing.  In P. Clarke (Ed.), New models for mass communication
research (pp. 243-268).  Beverly Hills, CA:  Sage.
        Dusenbury, L., Diaz, T., Epstein, J. A., Botvin, G. J., & Caton, M. (1994).
Attitudes towards AIDS and AIDS education among multi-ethnic parents of
school-aged children in New York City.  AIDS Education and Prevention, 6,
237-248.
        Gantz, W., Fitzmaurice, M., and Fink, E. (1991).  Assessing the active
component of information-seeking.  Journalism Quarterly, 68, 630-637.
        Governor's Advisory Council on HIV/AIDS (1994).  1994 Annual Report of the
Governor's Advisory Council on HIV/AIDS.  Washington State Department of Health.
        Greenwald, A. G., & Leavitt, C.  (1984).  Audience involvement in advertising:
Four levels.  Journal of Consumer Research, 11, 581-592.
        Grunig, J. E.  (1979).  Time budgets, level of involvement and use of the mass
media.  Journalism Quarterly, 56, 248-261.
        Half-million U.S. AIDS cases show wide variation. (1996, January). AIDS Alert,
8-10.
        Hollar, D. S. & Snizek, W. E. (1996).  The influences of knowledge of HIV/AIDS
and self-esteem on the sexual practices of college students.  Social Behavior
and Personality, 24, 75- 86.
        Holtgrave, D. R., Tinsley, B. J., & Kay, L. S. (1995).  Encouraging risk
reduction:  A decision-making approach to message design. In E. W. Maibach & R.
L. Parrott (Eds.), Designing Messages:  Approaches from Communication Theory and
Public Health Practice (pp. 24-40).  Thousand Oaks, CA:  Sage.
        Hyman, H., & Sheatsley, P.  (1947).  Some reasons why information campaigns
fail.  Public Opinion Quarterly, 11, 412-423.
        Johnson, B. T., & Eagly, A. H.  (1989).  Effects of involvement on persuasion:
A meta-analysis.  Psychological Bulletin, 106, 290-314.
        Kanihan, S. F., & Chaffee, S. H.  (1996, August).  Situational influence of
political involvement on information seeking:  A field experiment.  Paper
presented to Association for Education in Journalism and Mass Communication
convention, Anaheim, CA.
           Keeling, R. P. & Engstrom, E. L. (1994).  Employing community based
intervention programs in public health education:  Responding to HIV in campus
communities.  In J. P. Van Vugt (Ed.), AIDS Prevention and Services:  Community
Based Research (pp. 129-152).  Westport, CT:  Bergin & Garvey.
        Knaus. C. S., & Austin, E. W.  (1998, July).  The AIDS Memorial Quilt as
preventative education:  A developmental analysis of the Quilt as a preventative
tool.  Paper presented to the International Communication Association
convention, Jerusalem, Israel.
        Langer, E. J. (1978).  Rethinking the role of thought in social interaction.
In J. H. Harvey, W. J. Ickes, & R. F. Kidd (Eds.), New directions in attribution
research (Vol. 2, pp. 35-58).  Hillsdale, NJ: Erlbaum.
        Maibach, E. W., & Cotton, D.  (1995).  Moving people to behavior change:  A
staged social cognitive approach to message design.  In E. W. Maibach & R. L.
Parrott (Eds.), Designing messages:  Approaches from communication theory and
public health practice (pp. 41-64).  Thousand Oaks, CA:  Sage.
        McCombs, M. E.  (1972).  Mass communication in political campaigns:
Information, gratification and persuasion.  In F. G. Kline & P. J. Tichenor
(Eds.), Current perspectives
in mass communication research (pp. 169-194).  Beverly Hills, CA:  Sage.
        McGuire, W. J.  (1989).  Theoretical foundations of campaigns.  In R. E. Rice &
C. K. Atkin (Eds.), Public communication campaigns (pp. 43-65).  Newbury Park,
CA: Sage.
        Mendelsohn, H.  (1973).  Some reasons why information campaigns can succeed.
Public Opinion Quarterly, 37, 50-61.
        O'Donnell, L., San Doval, A., Vornfett, R., & DeJong, W. (1994).  Reducing AIDS
and other STD's among inner city Hispanics:  The use of qualitative research in
the development of video-based patient education.  AIDS Education and
Prevention, 6, 140-153.
        Parrott, R. L.  (1995).  Motivation to attend to health messages: Presentation
of content and linguistic considerations.  In E. Maibach & R. L. Parrott (Eds.),
Designing health messages (pp. 7-23).  Thousand Oaks, CA: Sage.
        Petty, R. E. & Cacioppo, J. T.  (1981).  Attitudes and persuasion:  Classic and
contemporary approaches.  Dubuque, IA:  William C. Brown.
        Petty, R. E. & Cacioppo, J. T.  (1986).  Communication and persuasion:  Central
and peripheral routes to attitude change.  New York:  Springer-Verlag.
        Pfau, M. , & Parrott, R.  (1993).  Persuasive communication campaigns.  Needham
Heights, MA: Allyn & Bacon.
        Pinkleton, B. E., & Austin, E. W.  (1997, August).      Voter involvement,
disaffection and media use:  Testing a "spiral of disaffection."  Paper
presented to Association for Education in Journalism and Mass Communication
convention, Chicago, IL.
        Pinkleton, B. E., Reagan, J., Aaronson, D., & Chen, C.  (1997).  The role of
individual motivations in information source use and knowledge concerning
divergent topics.  Communication Research Reports, 14, 291-301.
        Ragon, B. M., Kittleson, M. J., & St. Pierre, R. W. (1995).  The effects of a
single affective HIV/AIDS educational program on college students' knowledge and
attitudes.  AIDS Education and Prevention, 7, 221-231.
        Reagan, J.  (1995).  The "repertoire" of information sources.  Journal of
Broadcasting and Electronic Media, 39, 297-312
        Roser, C.  (1990).  Involvement, attention, and perceptions of message
relevance in the response to persuasive appeals.  Communication Research, 17,
571-600.
        Salmon, C. T.  (1986).  Perspectives on involvement in consumer and
communication research.  In B. Dervin & M. J. Voigt (Eds.), Progress in
communication sciences, vol. VII (pp. 243-268).  Norwood, NJ: Ablex.
        Sheer, V. C., & Cline, R. J. (1995).  Individual differences in sensation
seeking and sexual behavior:  Implications for communication intervention for
HIV/AIDS prevention among college students.  Health Communication, 7, 205-223.
        Solomon, D. S.  (1989).  A social marketing perspective on communication
campaigns.  In
R. E. Rice & C. K. Atkin (Eds.), Public communication campaigns (pp. 87-104).
Newbury Park, CA: Sage.
        Stevenson, H. C., & Davis, G. (1994).  Impact of culturally sensitive AIDS
video education on the AIDS risk knowledge of African-American adolescents.
AIDS Education and Prevention, 6 (1), 40-52.
        Stiff, J. B.  (1986).  Cognitive processing of persuasive message cues: A
meta-analytic review of the effects of supporting information on attitudes.
Communication Monographs, 53, 75-89.
        Turner, J. C., Garrison, C. Z., Korpita, E., Waller, J., Addy, C., Hill, W. R.,
& Mohn, L. A. (1994).  Promoting responsible sexual behavior through a college
freshman seminar.  AIDS Education and Prevention, 6, 266-277.
        U.S. Department of Health and Human Services:  Public Health Service (1987,
March).  AIDS:  Information/education plan to prevent and control AIDS in the
United States.
        Yates, J. F.  (1990).  Judgment and decision making.  Englewood Cliffs, NJ:
Prentice- Hall.
        Zaichkowsky, J. L.  (1985).  Measuring the involvement construct.  Journal of
Consumer Research, 12, 341-352.
        Zaichkowsky, J. L.  (1986).  Conceptualizing involvement.
Journal of Advertising, 15, 4-14, 34.
 
Table 1
 
Descriptive statistics
 
                                                                                                Intercorr.
                                                                                                   Or
Variable                        N               M               SD              Range           Alpha
 
Source use frequency            311             16.38           4.89            8-32              .81
   (All for HIV educ.)
   Media                        320              2.44            .98            1-4
   Health centers               318              2.09           1.01            1-4
   School                       318              2.90           1.07            1-4
   Peers                        316              2.52           1.01            1-4
   Your family                  318              1.97            .94            1-4
   Hot lines                    318              1.50            .84            1-4
   Church                       319              1.42            .73            1-4
   Internet                     316              1.59            .87            1-4
 
Motivation                      326              7.10           2.01            2-10            .63***
   Want to learn more           326              3.97           1.08            1-5
    Am looking for more info    327              3.13           1.15            1-5
 
Discussion with partner 171             13.18           3.32            5-20            .81
   Discuss HIV/AIDS             179              2.09            .87            1-4
   Discuss sex. trans. disease  178              2.15            .85            1-4
   Discuss safe sex             176              3.12            .97            1-4
   Discuss sex                  177              3.36            .84            1-4
   Discuss past partners        173              2.50            .87            1-4
 
Engaged in sex  w/o condom      181              2.30           1.23            1-4
 
***p<.001. Table 2
 
T-tests assssing quilt-viewing intervention effects
 
Variable                            N           M               t               df
 
Source use
        Did not see quilt       135             15.27           -3.60***        307
        Did see quilt           174             17.25
 
Motivation
        Did not see quilt       140              6.64           -3.77***        320
        Did see quilt           182              7.48
 
 
***p<.001.
 Table 3
 
Regression tests of hypothesized model
 
Dependent Variable
   Independent Variable                 r2  Chge&               b&              df&             F@
 
Source Use
   Age                                  .02**           -.15**          1,306            6.89**
   Motivation                           .05***   .23*** 2,305           12.21***
 
Discussion
   Source Use                           .05**            .23**          1,160            8.86**
 
Sex without condom
   Age                                  .05*             .22*           1,171            8.72**
   Source Use                           .03*            -.16*           2,170            7.07**
 
 
*p<.05; **p<.01; ***p<.001.
Standardized betas, significance levels for R-square and F reported for the
block of entry.