This paper was presented at the Association for Education in Journalism and
Mass Communication in San Antonio, Texas August 2005.
If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
rakyat [ at ] eparker.org. For an explanation of the subject line,
send email to
[log in to unmask] with just the four words, "get help info aejmc," in the
body (drop the "").
(Jan 2006)
Thank you.
Elliott Parker
====================================================================
Jung-Sook Lee Competition
In a word: A qualitative exploration of health information-searching behaviors
among healthy and diabetic women
The experiences of a woman looking for health information on the Internet can
vary based on the resources on which she relies and the search
terminology that she uses.
The search phrase "breast cancer" garners more government health
agency sites on
Yahoo! than on Google. Altering the search phrase by adding
"treatments" narrows the
first page of results to information from large, reputable cancer
organizations and the
National Institutes of Health, while adding "alternatives" tends to
pull up pages of
homeopathic remedies and often medically suspect "cures" from .coms.1
The Internet
offers thousands of health Web sites on variety of health topics
(Eysenbach, Sa &
Diepgen, 1999; Marwick, 2000) in consumer-oriented formats (Hall,
1998) from the
common to the extraordinary (Solovitch, 2001). The quality of
information on these Web
sites may be good, bad or indifferent (e.g., Berland et al., 2001;
Cline & Noland, 2002;
Evans, 2001; Eysenbach, Powell, Kuss, & Sa, 2002; Griffiths &
Christensen, 2000; Hong
& Cody, 2002; Impicciatore, Pandolfini, Casella, & Bonati, 1997;
Meric et al., 2002;
Sandvik, 1999). However, the questions remain about the specific
techniques health
consumers use to find this information.
Several large national surveys have investigated health information searching
behaviors on the Internet (Fox & Rainie, 2002; Kommers & Rainie,
2002; Taylor &
Leitman, 2002, 2003, 2004; Taylor, 2004). While these surveys have
explored why
people search for health information online, what information they
seek, where they
search, and how they evaluate online content, little work has been
done on how they
1 Based on a search of "breast cancer" on Yahoo! and Google and a
search of "breast cancer treatments"
and "breast cancer alternatives" on Google on March 30, 2005.
In a word -- 2of 20
actually conduct online health information searches. This question of
how people search
is important to understanding the information-seeking process. The
choice of search
engine or Web site and the choice of word or phrasing can
dramatically affect what
information health consumers find, how they evaluate that
information, whether the
information adequately fulfills their information needs and/or how
lengthy or complex
the search becomes. Part of a larger project investigating how
healthy and chronically ill
women find and evaluate health information on the Internet, this
study examined some of
the particular search behaviors used by women in their online
health-information seeking.
This paper will describe the previous research in the area of online
health-information
seeking behaviors, and detail the methodological procedures used to
answer the research
questions. The paper will summarize the main themes of the
qualitative investigation and
discuss the implications, limitations and future research suggested
by the findings.
Previous Research on Online Health-Information Searching
Previous research into health information seeking on the Internet has
found that
search engines are the preferred online resource. Almost 90% of
participants in a Pew
Internet & American Life poll started with a search engine rather
than a bookmarked
health site (Fox & Rainie, 2002). Only about 8% of health consumers
started their latest
searches on a specific medical Web site (Fox & Rainie, 2002) despite
the fact that a third
of health consumers have a health-related Web site bookmarked in
their Internet browser
(Fox & Rainie, 2002). People who visit the Internet for health
information more
frequently were more likely than occasional users to bookmark Web
sites in their
browsers (Fox & Rainie, 2002). The average health information search lasts
approximately 30 minutes (Fox & Fallows, 2003), and most consumers
(86%) investigate
2003).
recognized (Fox & Rainie, 2002).
In a word -- 3of 20
two to five Web sites during a search (Fox & Rainie, 2002). A Harris
Poll found fewer
people relying on search engines to look for health information the
first time (Taylor &
Leitman, 2003). Approximately 51% of Harris Poll respondents said
they used a search
engine or a portal, 23% used a health-related site, and 14% used a
general information
site that has a section devoted to health care (such as About.com)
(Taylor & Leitman,
For consumers who use a search engine, they tend to have different
strategies for
sorting through the list of search results. Forty-five percent of Pew
respondents simply
started at the top of a list and worked their way down the results
(Fox & Rainie, 2002).
Approximately 39% chose results from the list that they perceived as
being most relevant,
while 12% chose Web sites that were associated with sponsoring
organizations they
Large national surveys are not the only sources for information on
online health
searching behavior. Studies using think-aloud protocols, focus groups
and/or interviews
have examined search-related behavior. Hansen, Derry, Resnick and
Richardson (2003)
asked teen-agers to search the Internet to find answers to specific,
predetermined health
questions. The searches were deemed successful if the students found
a "correct and
useful answer" (Hansen et al., 2003, p. 2). Of the six searches
conducted by each of the
12 participants, approximately 69% were considered successful. Most
often, students
answered the questions by following a link found in the first nine
results on a search
query list. Common causes of incorrect answers were search term
misspellings and
search strategies that relied on summary information from a search
engine's results list
rather than direct access of a Web page (Hansen et al., 2003).
Eysenbach and Kohler
specific health-related Web sites.
specific Web site. These gaps suggested the study's research question.
do they choose search times? How are searches revised?)
Data Collection and Analysis
In a word -- 4of 20
(2002) used focus groups, usability tasks and interviews to examine
participant search
behavior (Eysenbach & Kohler, 2002). The researchers found that consumers only
examined the first few links of any search results list (Eysenbach &
Kohler, 2002).
The chronically ill conduct lengthier and more complex searches than
people who
have not been diagnosed, according to previous research (Freimuth,
Stein, & Kean,
1989). Millard and Fintak (2002) talked to more than 10,000 people in
a survey of the
online behavior of the chronically ill. Women with chronic illnesses
spent more time
looking for health information online than men with chronic illnesses
(Millard & Fintak,
2002). While health consumers in general start looking for health
information using a
search engine, the chronically ill are more likely than the general
populace to visit
Previous research suggested that search engines are preferred by
healthy people
but did not investigate search term selection. Prior studies also did
not identify prompts to
revise searching or selection of alternate search phrases. While
previous work suggests
the chronically ill are likely visit health-specific Web sites, it
does not describe the
circumstances when the chronically ill are more likely to use a
search engine versus a
RQ1: What specific procedures do diabetic and healthy women use to search for
health information online? (e.g., How do they choose online resources? How
This study used a triangulation of three qualitative methods to
investigate the
research questions: focus groups, in-depth interviews and think-aloud
protocols. The
think-aloud protocols, which involved the primary investigator
observing participants as
behavior (Morgan, 1997).
In a word -- 5of 20
they talked through and preformed three preset tasks on the Internet,
offered a way for the
researcher to directly observe participant search behavior. Each
think-aloud protocol
participant then was interviewed about their think-aloud searches in
particular and their
health information searching in general. The in-depth interviews
allowed the researcher
to explore search behaviors at great length (McCracken, 1988). This
study used focus
groups to help elaborate on the findings from the think-aloud
protocols and in-depth
interviews and to identify and explore issues that may not come up as
part of the in-depth
interviews. According to Morgan (1997), small group discussions allow
participants to
explore topics Internet searching by offering participants the
opportunity to react to and
expand on the answers of other people in the group. This interaction
may be especially
important to tease out nuances and explanations of otherwise
habitual, near-mechanical
Qualitative research does not strive for random samples, instead it
seeks the most
experienced and knowledgeable participants about the phenomena in
question (Lindlof,
1995). This study used a variation of maximum variation sampling,
called statistically
nonrepresentative stratified sampling. In maximum variation sampling,
"cases are usually
selected serially, with each adding a different, contrasting element
to the overall sample"
(Lindlof, 1995, p. 126). In statistically nonrepresentative
stratified sampling, essential
variables are determined and participants are selected to fill out
the subsets of the
essential variables (Lindlof, 1995). In this study, prior research
had identified health
status – healthy versus chronically ill – as an important determinant
of search behavior.
The sample was selected to draw from the population subsets of
healthy and diabetic
cities.
graduate to multiple graduate degrees.
In a word -- 6of 20
women. Participants were recruited through community groups in three
Southeastern
The researcher conducted four focus groups: two healthy focus groups (12
participants) and two diabetic focus groups (six participants).
Think-aloud protocols and
in-depth interviews were conducted with six health participants and
six diabetic
participants. A total of 30 women participated in this study.
Participants ranged in age
from the late 20s to early 60s. Most of the women in the study were
Caucasian; three
were African-American. Education level of the participants ranged
from high school
Focus groups, think-aloud protocols and in-depth interviews were
audio-taped and
transcribed. In addition, the think-aloud protocols and the focus
groups were videotaped.
In order to make sense of the vast amounts of data collected in the
study, the data analysis
method used was a variant of a style variously described as grounded
theory (Morse,
1994), constant comparative method (Lindlof, 1995) and analytic
induction (Taylor,
1994; Taylor, Hoy, & Haley, 1996). It is based on Glaser and Strauss'
initial work The
Discovery of Grounded Theory (1967).
The first step of the data analysis is conduct an intraparticipant
microanalysis, in
which a starting text is read line by line to catalog emic
descriptions (Lindlof, 1995, p. 96
(n3)), and also to begin examing underlying values, attitudes and
meanings. The data
from this first intraparticipant microanalysis on a focus group
transcript was pulled into
as many categories as possible (Lindlof, 1995). The primary
investigator made initial
notes and developed briefs descriptions (Lindlof, 1995). The next step was
interparticipant microanalysis, in which additional text are compared
to the categories
organizing the data (Morse, 1994).
and the findings.
Findings
this study would then appear as "I work for [name of employer]."
In a word -- 7of 20
created in the first analysis stage (Morse, 1994). Subsequent texts
expanded on categories
and meanings, and the researcher added categories where appropriate.
New categories
were created if subsequent texts suggested new or different
variations (Lindlof, 1995).
Coding in this manner helps researchers with labeling, separating,
compling and
The primary researcher wrote a memo to describe the categories once they
reached saturation – a point at which additional transcripts were not
adding anything new
to the data collection (Glasser & Strauss, 1967; Lindlof, 1995). A
highly trained second
coder read approximately 10% of the transcripts, developed her own
categories and
description of emergent themes, and then discussed the material with
the primary
investigator. The comments of the second coder were incorporated into
the data analysis
Names and identifying information have been changed to protect participant
identities. Names that contain a "d" indicate diabetic participants
(i.e. Amanda, Candace,
Deirdre, and Lydia). Names that contain an "l" indicate healthy
participants (i.e. Allison,
Dolly, Evelyn and Lilly). Other identifying information within quoted
material has been
removed or has been replaced with a bracketed phrase identifying what
was replaced. For
example, if the participant said "I work for the Acme Corporation"
then the words Acme
Corporation were replaced with the bracketed phrase "[name of
employer]." The quote in
RQ1: What specific procedures do diabetic and healthy women use to search for
health information online? (e.g. How do they choose online resources? How
do they choose search times? How are searches revised?)
included AOL, Alta Vista, Ask Jeeves and Dogpile.
protocols and when she might pick a search engine:
In a word -- 8of 20
Starting points. Study participants started or reported starting
health information
online in two basic ways: going to a search engine/search portal and
going to a specific
site. Among participants, search engines were more popular starting
places than Web
sites. The think-aloud protocol participants mostly relied on search
engines to complete
the tasks. Google and Yahoo! were the most popular search engines. As
Heidi said
"Google is the answer." Brenda said, "I pretty much stick with Google
. . . Mostly
because it is easy to remember and easy to type." Paula said, "I
normally use Yahoo!
because you can narrow that search and be more specific." The
participants viewed the
search engines/portals as having different strengths. Google was
identified as a search
engine of choice when the participant needed to cast a very wide net. When the
participants needed more specific results quickly they chose other
search engines. Sally's
attitude was typical: "I chose MSN because unlike Google, it doesn't
give you 5 million
[sites] . . . it's a way of weeding first. If I can't find what I'm
looking for, then I might
go to Google or Yahoo!" Other search engines mentioned or used by participants
Think-aloud protocol participants Beverly, Dolly, Edith, Amanda and Colleen
started health information search tasks by going to specific Web
sites. Beverly, Amanda
and Dolly started at WebMD, an online health information site that
Amanda described as
"the King." Dolly described her reasons for picking WebMD for the think-aloud
Participant Dolly: . . . With Web MD, I know it is specifically about
medical stuff
so I don't have to worry about burrowing animals. I know it is much more
specific and I have used it enough to think that it is pretty
reliable. With Google, I
know it is just a broad, catch-all kind of thing and I'm going to get
just a little bit
of everything. But lots of times. . . there are lots of things that I
don't know about
so I just kind of look and see what else is out there. Like they might mention
information for her daughter she started at her pediatrician's Web site.
In a word -- 9of 20
magazines, other things to look at, I might use it for that, but if I
want real. . . if
I'm under pressure for time, I would go to something like Web MD first. And
then if I have the time, I will just kind of do a little more browsing.
Flora also said she usually goes online to WebMD when she is need of
specific disease
information. Focus group participant Sally said that when she
searched for health
Several participants said they choose different search engines for
separate tasks.
Google was generally seen as a way to get a broad range of search
results. Other search
engines like Yahoo! and MSN were described as being better for more
specific results.
Participants who started by looking at a particular site, like WebMD,
reported using a
search engine next if the specific site did not provided the needed
information.
An additional starting point was other media content. Several
participants said
they typed in Web site addresses that they found in magazines and
newspapers. Lydia
heard a Web address in a radio advertisement for a weight-loss
product. Ilene used a
newspaper article to start a search for stress and meditation information.
Participant Ilene: . . . . But low and behold I picked up the (local
newspaper) one
morning and on the Lifestyles page was a story about the relationship between
meditation and stress reduction. So, um, I probably would not have
gone online to
look for that information but the Web site was provided for me, right
there. So I
went directly to the Web site the newspaper listed. And it was easy
to find stuff. I
clicked on several links and went to several different pages.
Choosing search terms. Participants used and reported using several types of
initial search terms. Some participants searched for symptoms such as
"itchy elbow";
some participants searched for disease or condition names such as
"melanoma," "chicken
pox" or "migraines"; other participants searched for health
information sources such as
the American Cancer Society or the Centers for Disease Control. The
choice of initial
search term depended in part on the participants' familiarity with
the topic for which they
In a word -- 10of 20
were searching. Many participants were like Brenda and tried to get
very specific search
terms: "When I know that it's something way too broad for a good
goal, I try to get it
down to something that it is going to know what I'm talking about."
Ellen said, "If I put
technical language in it will be a higher brow result." Although the
primary investigator
used the phrase "skin cancer" when assigning one of the think-aloud
tasks, several
participants searched for the more specific "melanoma" instead of
"skin cancer."
Continuing to search. If initial search terms did not bring up
relevant or credible
sources, then participants used or reported using several strategies.
Some participants
altered search terms: choosing a broader search term, choosing a more
specific search
term or adding a word to their current search term. Deirdre changed
one word in a
combination or changed positions of the words.
Participant Deirdre: . . . I, no matter what the topic is, I'll try
many different
ways, because you never know. Especially if there is somebody that's written a
really great article from India, they might not put "eating right."
They might put
"eating well" or "eating healthfully" or, or something like that. So,
I'm pretty
aware of linguistics.
Gail also reported trying similar variations in terminology when
looking for information
on her son's symptoms – "child's cough," "child cough" "child
nighttime cough." In
order to find alternate search terms, participants said they went to
Ask Jeeves, which
allows for natural language searches, or WebMD to get more specific
language to use in
Google or other search engines.
In addition to altering the search term itself, some participants
said they used
advanced search pages and added search limiters such as quotes,
pluses and minuses to
narrow in on desired results. Paula said she chose Yahoo! because
"you can go to
advanced search and tell it to look for these words specifically. It
will look for dogs and
In a word -- 11of 20
cataracts together, not just dogs." Kendra said whether she used the
advanced search page
depended on the type of information for which she was searching. "It
just depends. Like
if it is the name of the medication, I'll just go to the basic
search. But if it was more
complicated, I'll go to the advanced search." In practice, just three
of the think-aloud
protocol participants used any type of search limiters. Only one of
the think-aloud
protocol participants used the advanced search function on any of the
search engines to
limit results, increase relevancy or weed out unwanted site types. To
help with searches
and terminology, several participants said they kept a pad and pen
next to their computer
to take notes on search results.
Search result lists. Once participants typed a term or phrase into a
search engine,
they employed several techniques to sort through the search result
list. Most skipped over
sponsored links at the top and/or side of the results list. Dolly
clicked on a sponsored link
to get to a medication home page. Another participant clicked on a
sponsored link to the
MD Anderson Cancer Center – the perceived credibility of the cancer
center overcame
her general reluctance to look at sponsored links. Amanda, an
inexperienced Internet
user, clicked on several sponsored links during her think-aloud
protocol and clicked
almost immediately back to the search results page because the
sponsored links lead to
pages with no bearing on her search or to pages selling products.
After skipping over the sponsored links, participants scrolled
through the search
lists. Some participants said they always clicked on the first link
or the first three or 10
links. Olive said she always clicked the first link, figuring that it
would be the most
relevant to her search. For Paula, the number of sites that she
clicked depended on the
number of sites her search returned.
In a word -- 12of 20
Participant Paula: "If you finally get just 20 things that open up, I
will probably
open up all 20 of them. If I get 1,000 or 20 pages worth, then I am
just going to
go through and read those brief descriptions, and see if it is something I am
looking for. If not, I will go on to the next one.
Other participants scanned the list and clicked only on links to Web
sites they
thought would be relevant and/or credible. Participants said they
looked at the heading,
the description or the Uniform Resource Locator (URL), which contains
the Web address
for the link. Many participants looked at a combination of the
heading, description and
URL.
A majority of study participants said they would usually look only at
the first two
pages of search results before trying either another search phrase or
another source.
Participants said that after the first two pages the search results
were less relevant.
Participant Ellen: I usually just, like if I do a search I'll put in
whatever. I look
through the first page. If I don't find something on the first page
I'm changing my
search. Umh, only because I feel that the best ones might be in the
beginning and
then their just going to taper off to you know the more obscure sites.
Other participants reported mining the search results lists further
than the first few pages.
Allison said that she continued to look past the second page of
search results as long as
she was still finding relevant and credible sites. Paula said that
she had looked past the
first two pages on several searches and how far she went depended on
the seriousness of
the search and how badly she wanted to know the answer.
Participant Paula: Oh, I have been known before to go through 50 pages. When I
was looking for this one certain item for my husband. There were 55 pages out
there, and I read all 55 pages trying to find this item.
Lydia, an inexperienced Internet searcher, looked at several pages
based on an unusual
system.
Participant Lydia: The one in the middle, first and last. In this
group of 18 pages,
what will it offer me? Will it be the first page? I want to think
that but I want to
give the last page a chance, and I will go halfway down. And if
number 9 is pretty
In a word -- 13of 20
good, I will then go back to 8 or 7. And then if I do 8, I'll jump to
2 and then I'll
go back to 17 and check. I'll just make little comments on the
notepad beside me.
Factors affecting search parameters and procedures
In addition to the factors discussed earlier that affect the extent
of searches, other
factors can influence search parameters and procedures. Again, like
cost and benefits,
connection speed affected how participants searched for health
information online. Dialup
connection slows the process and may limit the number and type of
sites that are
visited as part of the search. Graphic-intense Web sites may not be
consulted for health
information if they do not load quickly. Connection speed impacted
the amount of time
some think-aloud protocol participants spent on the three tasks. The
time available to
search may alter the search process by limiting the number of sites
consulted or curtailing
the depths of the search. The amount of conflicting information or
medical uncertainty
surrounding a health issue may also change the search process. Health
questions around
which there is more medical uncertainty may stimulate greater search
scope and depth.
As Gail said, "I think it would make me want to search more."
Uncertainty may have the
opposite effect as well. Participants reported giving up on searches
that led to conflicting
health information or turning to other sources to settle
controversial health questions.
Discussion
Repeatedly in national surveys of online health seeking, respondents
say that they
use search engines to find content. Some of the national surveys even
discuss how users
select Web sites from the search results list. However, little work
has been done on how
individuals pick and modify their search terms. Key findings in this
area include
participants using three major categories of health search terms:
symptom descriptions,
In a word -- 14of 20
diseases/treatment names and source designations. Many participants
reported modifying
search terms after looking through one or two pages of results.
Common modifications
included adding a word, choosing a more specific term (and checking
with medical Web
sites to find more specific terminology), changing search engine,
going to a specific Web
site, and using more advanced search techniques like advanced search
pages, pluses,
minuses and quotes, and Boolean operators. This study suggests that
Internet users have
developed not only complex rules for identifying credible
information, but also complex
heuristics about how to search, where to search, and when to search again.
The typical Web searchers' habit of using search engines more often
than specific
medical sites to find health information (Fox & Rainie, 2002; Fox &
Fallows, 2003;
Taylor & Leitman, 2003) was also noted in this study. However, not
all search engines
and not all medical sites are created equally. Participants described
the reasons why they
chose one search engine over another and when they would choose a
health-specific Web
site over a search engine. Earlier studies had found that chronically
ill users were more
likely to use health-specific Web sites rather than search engines.
The diabetic women in
this study, however, seemed as reliant on search engines as their
healthy counterparts.
The study participants also processed search results in similar ways
to participants
in previous studies (Fox & Rainie, 2002a): starting at the top and
working down the list,
choosing results that seem relevant, picking sites from recognizable
sponsoring
organizations. However, unlike in other studies, the participants in
this study reported
choosing relevant sites more often than starting at the top and
clicking down the list. It
also was obvious from discussing search habits with participants that
recognizing the
source of the information was not a wholly separate way to choose
sources, but was, for
In a word -- 15of 20
many participants, part and parcel of determining which sites on a
search result list were
"relevant."
Although some research projects do not lend themselves to practical
applications,
the results of this study suggest several important lessons for
health care Web site
designers interested in reaching an interested population of health
information users.
When designing health care Web sites, this study suggests paying
attention to several
aspects of design, including page titles, page descriptors, URLs,
consumer-oriented
language, and advertising and commercial content.
Page titles. The dominance of search engines as the primary method of finding
health information online means that page titles, which are reported
in search results lists,
are important for users to determine relevance and credibility. Page
titles should include
the full name of the Web site or sponsoring organization, should be
written in consumeroriented
language and should be descriptive of the type of information found on that
page.
Page descriptors. Search engines rely in part on descriptors provided
by the Web
sites themselves in order to answer search queries. Based on the ways
that participants
conducted search term selection and modification, this study suggests
that the page
descriptor or metatags include words related to disease symptoms,
disease names and any
common nicknames (i.e. mad cow disease as well as bovine spongiform
encephalitis),
and treatment options. More specific words should be included along with broad
descriptors (i.e. melanoma diagnosis and suspicious moles). The exact
name of the
organization, department or company should be in the page descriptors
as well as more
generic descriptions (i.e. American Diabetes Association and diabetes
organization).
In a word -- 16of 20
URLs. Since site-type extensions are part of the way that
participants chose Web
sites and part of the way they evaluated the credibility and
relevance, organizations
should use .org extensions and not choose .com or .net extensions.
Specific descriptors in
the Web address also are useful for attracting search engine users.
For Web sites that are
.coms, a seal of approval from a consumer group or a governmental agency would
improve credibility assessments.
Limitations
The methods employed in this study were qualitative in nature and were used in
order to get thick, rich description. The purpose of this study was
to explore the
phenomenon of online health searches and the interrelations between
the elements of
searching. The purpose was not to prove causality or even suggest
statistical correlation.
Quantitative methods, such as experiments and surveys, would be more
appropriate to
describe precise associations. The purposive sampling used in this
study means that the
sample results cannot be generalized to other populations.
Participants in this study were self-selecting. It is possible that
women who do not
often use the Internet for health care information did not choose to
participate because
they felt they had something to add to the discussion. These more
rare users of the
Internet might have different search procedures or habits.
This study also chose diabetics as the chronically ill study
population of interest.
Other chronic illnesses may affect the search procedures in ways that
diabetes did not.
For example, chronic illnesses that cause physical disabilities may
affect the length and
complexity of the search procedure.
In a word -- 17of 20
Future Research
Many factors seemed to influence the search process itself. Future
studies could
look at what elements prompted affected source selection, search term
selection and
modification, and search complexity.
Similarly, added research could explain why many participants do go
further than
two pages while some report that they have looked at as many as 50
pages to find what
they need. Do the data miners have common characteristics distinct
from data skimmers?
Are there identifiable factors that influence someone to dig deeper
rather than return and
modify the search?
In a word -- 18of 20
References
Berland, G. K., Elliott, M. N., Morales, L. S., Algazy, J. L.,
Kravitz, R. L., Broder, M. S.,
Kanouse, D. E., Munoz, J. A., Puyol, J., Watkins, K. E., Yang, H. &
McGlynn, E.
A. (2001). Health Information on the Internet [Electronic version].
Journal of the
American Medical Association, 285, 2612-2621.
Cline, R. J. W., & Noland, V. J. (2002, November). The dark side of
"informationseeking"
on the Internet: Interactive health communication and risk promotion.
Presented, National Communication Association, New Orleans, LA.
Evans, W. (2001). Mapping mainstream and fringe medicine on the
Internet. Science
Communication, 22, 292-299.
Eysenbach, G., & Kohler, C. (2002). How do consumers search for and
appraise health
information on the world wide web? Qualitative study using focus groups,
usability, tests and in-depth interviews. British Medical Journal,
324, 573-577.
Eysenbach, G., Powell, J, Kuss, O., & Sa, E. R. (2002). Empirical
studies assessing the
quality of health information for consumers on the World Wide Web: A
systematic review. Journal of the American Medical Association (JAMA), 287,
2691-2700.
Fox, S., & Fallows, D. (2003, July). Internet health resources:
Health searches and email
have become more commonplace, but there is room for improvement in searches
and overall. Retrieved March 31, 2004, from The Pew Internet & American Life
Project Web site: http://www.pewinternet.org
Freimuth, V. S., Stein, J. A., & Kean, T. J. (1989). Searching for
health information: The
Cancer Information Service model. Philadelphia: University of Pennsylvania
Press.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded
theory: Strategies for
qualitative research. Chicago, IL: Aldine Publishing.
Griffiths, N.,& Christensen, H. (2000). Quality of Web-based
information on treatment of
depression: Cross sectional survey, British Medical Journal, 321, 1511-1515.
Retrieved August 28, 2001, from
http://www.bmj.com/cgi/content/full/321/7275/1511
Hansen, D. L., Derry, H. A., Resnick, P. J., & Richardson, C. R.
(2003). Adolescents
searching for health information on the Internet: An observational
study. Journal
of Medical Internet Research, 5(4), e25. Retrieved Feb. 8, 2004 from
the Journal
of Medical Internet Research Web site: http://jmir.org/2003/4/e25/
In a word -- 19of 20
Hong, T., & Cody, M.J. (2002). Presence of pro-tobacco messages on
the Web. Journal
of Health Communication, 7, 273-307.
Impicciatore, P., Pandolfini, C., Casella, N., & Bonati, M. (1997).
Reliability of health
information for the public on the World Wide Web: Systemic survey of advice on
managing fever in children at home. British Medical Journal, 314, 1875-1879.
Retrieved on August 28, 2001, from http://www.bmj.com/cgi/content/
full/314/7098/1875.
Kommers, N., & Rainie, L. (2002, May 8). Use of the Internet at major
life moments.
Retrieved June 28, 2002, from The Pew Internet & American Life Project Web
site: http://www.pewinternet.org.
Lindlof, T. (1995). Qualitative communication research methods.
Thousand Oaks, CA:
Sage Publications.
McCracken, G. (1988). The long interview. Thousand Oaks, CA: Sage
Publications.
Meric, F., Bernstam, E. V., Mirza, N. Q., Hunt, K. K., Ames, F. C.,
Ross, M. I., Kuerer,
H. M., Pollock, R. E., Musen, M. A., & Singletary, S. E. (2002).
Breast cancer on
the World Wide Web: Cross-sectional survey of quality of information and
popularity of Websites. British Medical Journal, 324, 577-581.
Millard, R. W. & Fintak, P. A. (2002). Use of the Internet by
patients with chronic
illness. Disease Management and Health Outcomes, 10(3), 187-194.
Morgan, D. L. (1997). Focus groups as qualitative research (2nd ed.).
Thousand Oaks,
CA: Sage Publications.
Morse, J. M.(1994). Emerging from the data: The cognitive processes
of analysis in
qualitative inquiry. In J. M. Morse (Ed.), Critical issues in
qualitative research
methods (pp. 23-41). Thousand Oaks, CA: Sage.
Morse, J. M.(1998). What's wrong with random selection. Qualitative
Health Research,
8(6), 733-735.
Morse, J. M. (2004). Editorial: Qualitative significance. Qualitative
Health Research,
14(2), 151-152.
Sandvik, H. (1999). Health information and interaction on the
Internet: A survey of
female urinary incontinence. British Medical Journal, 319, 29-32. Retrieved
August 28, 2001, from http://www.bmj.com/cgi/content/full/319/7201/29.
Taylor, H. (2004, January 21). Harris Poll no. 4. Online activity
grows as more people
use Internet for more purposes. Retrieved February 8, 2004, from the Harris
Interactive Web site: www.harrisinteractive.com/harris_poll/
index.asp?=PID=433.
In a word -- 20of 20
Taylor, H., & Leitman, R. (Eds.) (2002, May 8). Cyberchondriacs
continue to grow in
America. Health Care News, 2(9), 1-4. Retrieved December 14, 2002, from the
Harris Interactive Web site: http://harrisinteractice.com/news/newsletters/
healthnews/HI_HealthCareNews2002Vol2_Iss09.pdf.
Taylor, H., & Leitman, R. (Eds.) (2003, March 28). No significant
change in the numbers
of "cyberchondriacs" – those who go online for health care information. Health
Care News, 3(4), 1-4. Retrieved February 8, 2004, from the Harris Interactive
Web site: http://harrisinteractice.com/news/newsletters/healthnews/
HI_HealthCareNews2003Vol3_Iss04.pdf.
Taylor, H., & Leitman, R. (Eds.) (2004, April 12). No significant
change in the number
of "cyberchondriacs" – those who go online for health care information. Health
Care News, 4(7), 1-4. Retrieved June 18, 2004, from the Harris Interactive Web
site: http://harrisinteractive.com/news/newsletters/healthnews/
HI_HealthCareNews2004Vol4_Iss07.pdf.
Taylor, R. E. (1994). Qualitative research. In M. W. Singletary (Ed.), Mass
communication research (pp. 265-279). New York: Longman.
Taylor, R. E., Hoy, M. G., & Haley, E. (1996). How French advertising
professionals
develop creative strategy. The Journal of Advertising, 25(1), 1-14.
|
