Reflection Paper (Assignment)

 REFLEXTION PAPER
(DUE BY FEBRUARY 23, BY 2 PM)

Please read this article and answer these questions:

(1) What was the purpose of the study?

(2) How do the authors define self-injury?

(3) Is there a difference in prevalence rate of self-injurious behavior between boys and girls? If so what might that reason be?

(4) What do the authors suggest is the way in which the internet has increased the rate of self-injury among adolescence?

(5) Do you agree or disagree with the authors’ conclusions? Can you think of alternative explanations for the outcomes of their results?

 Reflection papers must be typed, double-spaced throughout, 1-inch margin on all sides, 12 font size, and Times New Roman font type. 

BOOK “INVITATION TO LIFE” Kathleen Stassen Berger (CHAPTER 3, pp 103-104)

References in APA format (2-3 references would be enough)

Three-page minimum (excluding title page and reference page) 

  

The Virtual Cutting Edge: The Internet and Adolescent Self-Injury

Janis L. Whitlock, Jane L. Powers, and John Eckenrode
Cornell University

The 2 studies reported here use observational data from message boards to investigate how adolescents
solicit and share information related to self-injurious behavior. Study 1 examines the prevalence and
nature of these message boards, their users, and most commonly discussed topics. Study 2 was intended
to explore the correlations between content areas raised for discussion. Both studies were intended to
shed light on the role of message boards in spreading information about self-injurious practices and
influencing help-seeking behavior. More than 400 self-injury message boards were identified. Most are
populated by females who describe themselves as between 12 and 20 years of age. Findings show that
online interactions clearly provide essential social support for otherwise isolated adolescents, but they
may also normalize and encourage self-injurious behavior and add potentially lethal behaviors to the
repertoire of established adolescent self-injurers and those exploring identity options.

Keywords: adolescence, self-injury, Internet, mental health

I think my greatest fear is to be forgotten. A teacher I had last year
doesn’t even remember my name—it makes me think that no one
remembers me. How do I know I exist? At least I know I exist when
I cut. (Self-injury message board post)

Because adolescents use the Internet for the purpose of connect-
ing with others at higher rates than any other age group (Lenhart,
Rainie, & Lewis, 2001), a better understanding of how Internet use
affects their social and emotional development is an important line
of scientific inquiry. Indeed, a small but growing body of research
is beginning to examine the implications of various electronic
forums for social interaction (e.g., chat rooms, news groups, mes-
sage boards) on adolescent behavior (e.g., Gross, 2004; Subrah-
manyam, Greenfield, & Tynes, 2004; Tynes, Reynolds, & Green-
field, 2004). The Internet may have particular relevance for
adolescents who feel marginalized, because it provides a low-risk
venue for finding others who share their perceived or real differ-
ences and exchanging information that is difficult to convey in
person or when using one’s real identity (McKenna & Green,
2002). Adolescents who intentionally injure themselves are one
such group. Although research is nascent, adolescent self-injury
appears to be increasingly common (Welsh, 2004; Yates, 2004)
and, as this article shows, is a practice around which many virtual
communities have formed. To date, almost nothing has been
written about the existence of self-injury Internet forums, the types

of exchanges that occur there, or the ways in which these ex-
changes may affect the development of adolescents and their
ability to cope with distress.

Adolescent Internet Use

Computer access and use among adolescents have grown expo-
nentially over the past decade (Becker, 2000). More than 80% of
American youth 12 to 17 years of age use the Internet, and nearly
half log on daily (Lenhart, Madden, & Hitlin, 2005). Once con-
nected, adolescents engage in a wide variety of activities, includ-
ing doing schoolwork, playing games, shopping, and downloading
music. Research shows, however, that adolescents use the Internet
primarily for social reasons (Gross, 2004; Roberts, Foehr, & Ride-
out, 2005). The Internet has become a virtual meeting place where
teens hang out with their peers to pass time. Many adolescents
reportedly prefer being online to other media, including the tele-
phone, TV, and radio (2002 Gallup Survey, cited in Heitner, 2002).
According to data from the Pew Internet and American Life
Project (Lenhart et al., 2005), the vast majority (89%) of teens use
e-mail and 75% use instant messaging (IM), which allows them to
have multiple simultaneous conversations with a defined group of
peers. More than 50% of teens possess more than one e-mail
address or screen name, which they can use to send private
messages to friends or to participate anonymously in online fo-
rums, such as chat rooms (Lenhart et al., 2001).

Some studies suggest that Internet use may facilitate social
interaction by making it easier for individuals to connect with
others they know as well as with strangers. It serves also as a
powerful resource for youth desiring information about socially
sensitive topics such as sexuality and interpersonal relations (Su-
zuki & Calzo, 2004). This form of communication may be espe-
cially advantageous for shy, socially anxious, or marginalized
youth, enabling them to practice their social skills without the risks
associated with “on the ground” interactions (Heitner, 2002; Mc-
Kenna, Green, & Gleason, 2002; Subrahmanyam et al., 2004).
Additionally, online communication may encourage more truthful

Janis L. Whitlock and John Eckenrode, Family Life Development Cen-
ter and Department of Human Development, Cornell University; Jane L.
Powers, Family Life Development Center, Cornell University.

This study was supported by funding from the Cornell College of
Human Ecology Seed and Innovation Grant. We thank Amanda Purington,
Alexis Matusiewicz, and the Cornell Self-Injury Study Team for their
substantial work on data collection and analysis. We also thank Elizabeth
R. Woods for her thoughtful reading of an earlier version of this article.

Correspondence concerning this article should be addressed to Janis L.
Whitlock, Family Life Development Center, Cornell University, Beebe
Hall, Ithaca, NY 14853-4401. E-mail: jlw43@cornell.edu

Developmental Psychology Copyright 2006 by the American Psychological Association
2006, Vol. 42, No. 3,

407

– 417 0012-1649/06/$12.00 DOI: 10.1037/0012-1649.42.3.407

407

exchanges; many people report a greater willingness to share
thoughts and feelings online than they would in face-to-face situ-
ations (Lenhart et al., 2001; McKenna & Bargh, 2000). Clearly, the
Internet is transforming the social world of adolescents by influ-
encing how they communicate, establish and maintain relation-
ships, and find social support. Nevertheless, the developmental
consequences of adolescent Internet use is an area about which
little is known (Wartella, Caplovitz, & Lee, 2004).

The Internet and Adolescent Social and Emotional
Development

Three central tasks are integral to healthy social development
during adolescence: (a) to establish caring, meaningful relation-
ships; (b) to find acceptance and belonging in social groups; and
(c) to establish interpersonal intimacy (Baumeister & Leary, 1995;
Reis & Shaver, 1988; Sullivan, 1953). Peers play a crucial role in
this process, because a positive relationship with peers is important
for psychological well-being and social adjustment (Bishop &
Inderbitzen, 1995; Hartup, 1996), whereas peer rejection is linked
to serious problems, including delinquency, drug abuse, and de-
pression (Hartup, 1996; Merten, 1996).

Early studies of the influence of online interactions on adoles-
cent development suggested that high levels of Internet use may
inhibit healthy social development by linking frequent use to social
isolation and depression, especially among teenagers (Kraut et al.,
1998; Nie & Erbring, 2000). However, these findings have been
disputed. A follow-up study conducted by Kraut et al. (2002)
found that the effects documented in their earlier study had largely
dissipated. They did find, however, that effects of Internet use on
depression differed for introverts and extroverts: Extroverts were
more likely to feel greater social connection as a result of Internet
use, whereas introverts became more depressed and withdrawn.
Heitner (2002) found that adolescents who use the Internet to
connect with others in real-time social exchanges tended to possess
higher peer status, more social skills, and greater social integration
than their more socially introverted and withdrawn peers, who
spent most of their Internet time in solitary activities. Additionally,
adolescents who used chat rooms exhibited lower peer status and
had fewer social skills than those who did not. Similarly, Gross,
Juvonen, and Gable (2002) found that teenagers with strong social
connections use e-mail and IM to reinforce preexisting bonds,
whereas those with less developed social networks use the anon-
ymous features of the Internet to find new friends and social
outlets, perhaps compensating for what they lack offline. This
suggests that chat rooms and similar venues in which adolescents
share experiences anonymously may provide a safe forum for less
socially adept adolescents to practice social interaction.

Research finds that online exchange decreases social isolation
among adolescents and helps them connect with people and ex-
plore their identity (Maczewski, 2002; Suzuki & Calzo, 2004).
This helps to explain how the Internet may serve as a virtual peer
support group where adolescents under stress can express feelings
and exchange information about modes of coping. Adolescence is
also a time of increased feelings of distress (e.g., depressed mood)
and increased access to modes of coping with stress that are
independent of parents (Arnett, 1999; Compas, 1987; Petersen,
Kennedy, & Sullivan, 1991). To the extent that Internet use can
reduce feelings of social isolation, help normalize feelings of

distress through a process of self-disclosure and social comparison,
and serve as a venue for giving and receiving social support, it may
also provide a positive coping resource for distressed youth. Al-
ternatively, Internet use may maintain or increase distress if the
information exchanged reinforces negative views of self or sug-
gests destructive or otherwise ineffective coping strategies.

This article examines the role of the Internet among adolescents
who use self-injurious behavior (SIB) as a method of coping with
distress. The opportunity to explore different identities and roles
through the Internet may be particularly important for individuals
with stigmatized identities, such as self-injurious youth, who feel
that important aspects of their selves need to be concealed in their
day-to-day lives. These youth may be especially motivated to
participate in electronic forums that allow them to express them-
selves in a safe and anonymous environment and to find support
from others who share their sense of marginalization and under-
stand their behavior. To date, however, not only is very little
known about self-injury in the adolescent population, but nothing
is known about how this group uses the Internet to connect with
others about their practice.

Adolescent Self-Injury as a Developmental Phenomenon

The increasing number of stories in the mainstream press and
popular media, as well as the growing number of anecdotal reports
by physicians, therapists, and school counselors, suggest that self-
injury may be “the next teen disorder” (Welsh, 2004). Although
operationally elusive, scholars differentiate self-injury from cul-
turally sanctioned forms of self-mutilation, such as piercing or
tattooing, by intention rather than form. Alternatively called de-
liberate self-harm, self-injury, self-mutilation, or cutting, self-
injurious behavior typically refers to a variety of behaviors in
which an individual purposefully inflicts harm to his or her body
for purposes not socially recognized or sanctioned and without the
obvious intention of committing suicide (Alderman, 1997;
Favazza, 1996). Self-injury, which is most often associated with
the term cutting, also includes intentional carving or cutting of the
skin and subdermal tissue, scratching, burning, ripping or pulling
skin or hair, swallowing toxic substances, bruising, and breaking
bones. Although not typically a suicidal gesture, self-injury is
statistically associated with suicide and can result in unanticipated
severe harm or fatality (Claes, Vandereycken, & Vertommen,
2003; Favazza, DeRosear, & Conterio, 1989).

There are currently no reliable estimates of the prevalence of
self-injury among the general U.S. adolescent population. The vast
majority of research on self-injury has been conducted in clinical
populations or using small, unrepresentative community samples.
These studies generally find that cutting and other forms of self-
injury are evident in approximately 20% of the clinical population
(Deiter, Nicholls, & Pearlman, 2000) and are linked to high levels
of pathology (Brodsky, Cloitre, & Dulit, 1995; Ross & Heath,
2003). The few studies that have been conducted in U.S. commu-
nity samples of young adults and adolescents are limited by small
convenience-based samples and vary in estimates of self-injury
prevalence from 4% to 38% (Briere & Gil, 1998; Favazza, 1992;
Gratz, Conrad, & Roemer, 2002; Kokaliari, 2005; Muehlenkamp
& Gutierrez, 2004). Large studies in Britain estimate that approx-
imately 10% of youth 11 to 25 years of age self-injure. A British
report on the national scope of the problem documents a dramatic

408 WHITLOCK, POWERS, AND ECKENRODE

increase in disclosures of self-injury to national children’s help
lines over the 5 years before the study, noting a 65% increase in the
last 2 years (Young People and Self Harm: A National Inquiry,
2004).

The reasons for this apparent increase are unclear. Although
general awareness of self-injury may lead to increased willingness
to disclose the behavior, it may also reflect a true increase in
incidence. Social contagion has been identified by some (Yates,
2004; Rosen & Walsh, 1989), because self-injury follows
epidemic-like patterns in institutional settings such as hospitals
and detention facilities (Matthews, 1968; Ross & McKay, 1979;
Taiminen, Kallio-Soukainen, Nokso-Koivisto, Kaljonen, & Hele-
nius, 1998). The possibility that self-injury is communicable may
reflect a pattern similar to what Brumberg (1992) argued with
regard to the spread of anorexia nervosa in the 1980s, when
heightened cultural visibility through the mass media rendered
anorexia nervosa an available emotional outlet for individuals with
receptive predispositions.

Although no single self-injurer profile has emerged, there is
general consensus that self-injury is most common among adoles-
cents. Self-injurious behavior may parallel other problem behav-
iors, which begin during early adolescence, peak during mid- to
late adolescence, and then decline in adulthood (Briere & Gil,
1998; Favazza, 1999; Favazza & Conterio, 1989). However, there
is some evidence that self-injurious behavior may follow two
distinct patterns similar to the life course-persistent and
adolescence-limited trajectories evident in antisocial behavior
(Moffit, 1993). One of these begins in early childhood and persists
through adulthood. The other follows a typical adolescence-limited
course that emerges in early adolescence and declines in late
adolescence or early adulthood (Dubo, Zanarini, Lewis, & Wil-
liams, 1997; Nixon, Cloutier, & Aggarwal, 2002). Although some
research finds females more likely to injure themselves than males
(Conterio & Lader, 1998; Favazza, 1999), other studies suggest
that the gender gap may be narrower (Briere & Gil, 1998; Deiter
et al., 2000; Dulit, Fyer, Leon, Brodsky, & Frances, 1994; Galley,
2003; Martin, Rozanes, Pearce, & Allison, 1995). The difficulty in
accurately assessing gender differences in self-injurious behavior
may arise from variation in how male and female self-injurers are
identified, how they injure themselves, and whether or not they
seek treatment (Alderman, 1997; Connors, 2000).

The relationship between self-injury and suicide is important but
not clearly understood. Persons who engage in self-injury are more
likely to consider or attempt suicide (Walsh & Rosen, 1988;
Gardner & Cowdry, 1985; Hawton, Fagg, Simkin, Bale, & Bond,
2000). Nevertheless, in the vast majority of cases self-injury is
used to alleviate distress temporarily rather than to signal the
intention to end one’s life (Favazza, 1996; Rosen & Walsh, 1989;
Tantam & Whittaker, 1993). Indeed, some scholars see it as a
highly functional alternative to suicide (Alderman, 1997; Strong,
1998). Among clinical populations, self-injury is comorbid with
borderline personality disorder, eating disorders, posttraumatic
stress disorder, depression, anxiety disorders, and a history of
abuse or trauma (Alderman, 1997; Connors, 2000; Conterio &
Lader, 1998; Holmes & Nadelson, 2000; Sansone & Levitt, 2002;
Yates, 2004). Indeed, some researchers have called for a new
Diagnostic and Statistical Manual of Mental Disorders impulse-
control disorder, deliberate self-harm syndrome, which would
include self-injurious behavior (Favazza et al., 1989). Most, how-

ever, see self-injury as a manifestation of mental or emotional
disorders or heterotypic manifestation of childhood trauma (Con-
ners, 2000; Conterio & Lader, 1998; Favaro & Santonastaso, 2002;
Strong, 1998; Yates, 2004).

Whether a trauma-linked developmental model will hold true
for nonclinical populations is unclear. However, the proximal
causes for those who engage in self-injurious behavior in both
clinical and community settings are strongly linked to difficulties
in regulating strong emotions and coping with stress (Favazza,
1996; Rosen & Walsh, 1989; Ross & Heath, 2002; Tantam &
Whittaker, 1993). The cumulative impact of stressful life events
that trigger self-injurious episodes, the subsequent shame associ-
ated with the self-injurious response, and the secrecy surrounding
the behavior create fertile ground for the development of margin-
alized perceptions of the self. The Internet provides self-injurers
with an anonymous venue for sharing actual and fabricated aspects
of their true identity among a community of similar others. Be-
cause young people with depressive symptomatology are much
more likely to talk with strangers online and to disclose personal
information than those without depressive symptomatology (Ybarra,
Alexander, & Mitchell, 2005), self-injurers may be particularly
receptive to using the Internet to locate virtual communities. Al-
though such virtual communities may provide a much-needed
source of support and cathartic sharing, the possibility that self-
injury is communicable suggests that the Internet may also serve to
spread or deepen the practice in adolescent populations.

Aim of the Present Studies

Two studies were designed to explore and document adolescent
use of online message boards to share, solicit, and receive infor-
mation and advice related to self-injurious behavior. Message
boards are an electronic venue in which individuals (i.e., “posters”)
register with the site under a chosen name and are allowed to post
thoughts, ask questions, and respond to other posters. Unlike
e-mail and IM, which permit private exchange between selected
individuals, message boards and chat rooms (in which exchanges
occur in real time) are entirely public: All postings are available to
all members and, in many cases, nonmembers as well, although
nonmembers cannot post material. When individuals log on to a
message board, the content varies but typically includes informa-
tion about the message board purpose and rules as well as direct
links to “threads,” posts that have been made by users sorted by
subject title. Anyone who logs on to a message board may read
posts, but only those registered may make a post. Blogs, which are
becoming increasingly common, are essentially public electronic
diaries. Message board posters frequently link their blog site to
their membership identity or the signature line in their posts.

The goal of the first study was to investigate the prevalence and
nature of self-injury message boards and their users with the
intention of better understanding the general subject categories
most commonly raised for discussion. The goal of the second study
was to explore the correlations between content areas raised for
discussion. Both studies were intended to shed light on the role of
message boards in spreading information about self-injurious prac-
tices and influencing help-seeking behavior. Studies of the Internet
are effectively studies of culture. Participants are free to construct
their age and identity as they wish, and studies show that teen
Internet users often construct themselves as somewhat older than

409SPECIAL SECTION: VIRTUAL CUTTING EDGE

they actually are (Gross, 2004). Both studies were approved by the
University Internal Review Board and have abided by the stipu-
lation that participant quotes would be paraphrased rather than
exact quotes. All quotes used here comply with this requirement.
Because the studies use observational methods, active consent was
not required.

Study 1: Prevalence and Nature of Self-Injury Message
Board Use

The first study was undertaken to describe: (a) the prevalence of
self-injury message boards, (b) the number of self-injury message
boards advertised jointly with message boards for another disorder
or behavior (such as eating or bipolar disorders), (c) the ratio of
active and inactive posters, (d) the age range of users, as revealed
in their self-descriptions, (e) date of appearance on the Web, (f) the
relationship between using message boards and blogs, and (g) the
nature of and variation in the content of postings on active message
boards.

Method

Identification and selection. For this study, there were two units of
analysis: the message board and individual posts in a selected set of
message boards. First, to identify the prevalence of self-injury message
boards, five Internet search engines were used: Yahoo, Google, MSN,
AOL, and Gurl.com. Terms searched included self-injury, self-harm, self-
mutilation, and cutting. More than 400 self-injury message boards were
identified using this method; of these, 10 were selected for in-depth content
analysis. These boards represented the first 10 listed in each search engine,
which are generally those Web sites with the most activity. The moderation
level of each of these boards varied as well. Moderation level refers to the
degree to which posters are actively monitored for potentially damaging
content (such as sharing techniques for self-injury). For example, when a
particularly graphic or triggering post is made to a high- or medium-level
moderation board, it is blocked altogether (high moderation) or labeled as
a potential trigger (medium moderation). We were most interested in
message boards with medium and low moderation, because content would
be less likely to be censored. Detailed membership and moderation level
information for the 10 sites selected was recorded (see Table 1 for a

description of the sites; names are withheld to protect confidentiality). A
total of 3,219 individual posts, examined over a 2-month period, were
coded for themes of interest and used in the content analysis that follows.

In addition to individual-level posts, we were also interested in gathering
descriptive information about message board characteristics. Although
most message boards do not contain information about date of inception,
cross-listing with other topics, and membership characteristics, one of the
Internet service providers did permit access to this information. The iden-
tity of this search engine is hidden to protect confidentiality. All 140
message boards identified through this engine were examined for dates of
inception, number of members, and cross-listing with other conditions.

Measures for message board analysis. Depending on the analysis, the
following information was obtained for each message board: date of
establishment, mean self-reported user age, number of active and inactive
members, percentage of posters with blogs, and colisting with message
boards for other behaviors. Date of establishment was determined using
information from the home page or, if this was absent, from the date of the
first post. The mean range of self-described age was determined by
selecting 50 members at random from the site introduction board, where
basic member profiles are stored. The number of active and inactive posters
was determined by comparing the number of members who have registered
but never posted with that of members who have posted. Moderation level
was determined by identifying whether there were any instructions to users
about allowable and unallowable posts and whether potentially triggering
posts (i.e., events that provoke self-injurious behavior) were labeled as
such. Cross-listing with other topics refers to the extent to which a message
board was advertised as a virtual meeting place for individuals interested in
or affected by the intersection of two or more issues (e.g., self-injury and
suicide). Cross-listing was measured by examining the formal message
board title and description of each board. When a cross-listed board was
identified, the cross-listed topic was noted and the message board was
considered as a cross-listed board.

Measures for content analysis of postings. The content analysis of
posts was conducted using a set of binary (present–absent) codes. These
codes were drawn from a synthesis of the research cited previously, from
interviews conducted with 15 self-injurers by Janis L. Whitlock, and from
pilot message board observations that spanned a 2-week period. A team of
three coders systematically reviewed posts and used the constant compar-
ative method (Glaser & Strauss, 1967) to inductively monitor themes. Once
the initial set of observations had been coded, thematically grouped clusters
were identified (e.g., motivation). The 11 broad categories identified en-

Table 1
Membership Breakdown of Selected Self-Injury Message Boards

Site

Age (year

s)

% Female No. members % w/Blogs

Members w/posts (%)

Mod levelaRange M Never 2–10 �100

A 12–44 18.7 88.1 844 30 41.0 20.0 9.5 Low
B 13–54 18.3 90.0 5,259 25 57.0 15.5 7.2 Med
C 14–47 19.4 88.0 5,082 15 14.0 44.1 17.3 High
D 12–37 17.6 74.0 NA 25 31.5 32.8 2.0 Low
E 14–22 17.5 91.5 70 15 45.7 41.5 0 Low
F 14–36 19.6 87.5 6,656 10 63.4 10.9 8.5 Med
G 16–47 20.5 90.0 153 20 65.0 9.8 9.8 Low
H 14–28 18.1 80.0 4,021 30 41.3 15.9 21.0 Med
I 15–46 23.9 80.6 1,427 15 30.4 27.8 9.5 Med
J 13–26 16.4 78.0 2,581 10 50.8 25.0 10.0 Mod

Note. Mod � moderation; NA � not available; Med � medium.
a Moderation level was determined by the extent to which the message board moderator blocked or labeled
potentially triggering posts. In high-moderation boards, potentially triggering or disruptive posts were blocked
entirely. In medium-moderated message boards, triggering or otherwise disruptive posts were identified and
labeled. Low-moderation message boards took no steps to identify or block posts of any sort.

410 WHITLOCK, POWERS, AND ECKENRODE

compassed virtually all message board exchanges and served, along with
all specific themes they encompassed, as a coding tool (e.g., loneliness,
anger, and dysphoria were all themes in the motivation category). Posts left
uncoded were those with little capacity to illuminate understanding of
self-injurious behavior, such as idle discussion about a current event or
activity. The resulting 11 areas were as follows: (a) informal support and
exchange, (b) motivation for self-injury, (c) concealment of self-injurio

us

behavior (e.g., anxiety about exposure, methods for concealment of cuts
and scars), (d) addiction language (e.g., days self-injury free, difficulty
stopping), (e) formal help seeking and treatment, (f) sharing techniques, (g)
links to other mental health or behavioral conditions known to be associ-
ated with self-injurious behavior, (h) references to popular culture, (i)
perceptions of non-self-injurers reactions to self-injurious behavior, (j)
perception of self and behavior (e.g., self-worth, lovability, dissociation),
and (k) venting or apologizing. These areas contained a total of 70 themes
into which nearly all posts could be categorized.

Coding of posts. A content analysis was conducted on all original and
follow-up responses to the original post in 10 message boards over a
2-month period. A time frame of 2 months was selected to ensure adequate
breadth in content areas. A total of 3,219 posts were examined during this
period. Both original posts and responses to posted messages were coded.
Because not all posts contained content relating to the coding scheme, the
total number of posts examined exceeds the total number of posts to which
a code was assigned (2,942). (The criterion for leaving postings uncoded is
described later). However, if posts contained more than one thematic
reference, they were assigned multiple codes. Therefore, totals in the
Results section can add up to more than 2,942.

Four coders were trained by Janis L. Whitlock and the study coordinator
to observe and code message boards during this time period. This was
accomplished by having each coder and the principal investigator code the
same posts and discuss code assignment. Once training was completed,
each coder logged on one to two times per week during the study period to
identify and code new posts. To establish intercoder reliability, each coder
team was given three randomly selected threads to code. Codes for each
individual coder were then compared to establish agreement within each
team. The teams were in full agreement about codes for 96% and 97% of
the postings, respectively.

Results

Message board prevalence, dates of inception, and colisting
with other topics. The search procedure described previously
revealed 406 boards. As shown in Table 2, examination of the 140
boards available through the Internet service providing historical
information shows an increase from 1998 to 2000 and then a
generally stable trend over time, with peaks in 2000 and in 2002.
Although caution is warranted when interpreting these data, be-

cause public familiarity and dedicated server space with message
boards have also increased over time, these findings indicate that
message boards as a cultural tool for self-injury grew dramatically
between 1998 and 2000 and that interest in both establishing and
participating in self-injury specific groups over the past 5 years has
been sustained.

Examination of these same message boards for colisting with
other topics reveals links with a number of topics known to be
associated with self-injury in the literature. Table 3 shows the
extent to which self-injury message boards were cross-listed with
other topics. Although listed most often alone (56%), 44% of the
message boards were cross-listed with one or more other topics.
When linked to another topic, self-injury occurs most often with
depression (32%) and eating disorders (17%). Although less fre-
quent, it often occurs with message boards dedicated to discussio

ns

of dissociative identity and multiple personality disorders (10%);
bipolar disorder (9.2%); sexual abuse (7%); obsessive– compulsive
disorders (7%); addiction (2.8%); anxiety disorder (2.8%); lesbian,
gay, bisexual, and transgender issues (2.8%); and autism (1.4%).

Message board membership analysis. The 10 most popular
message boards were selected for membership analysis. Table 1
shows the averages for self-description: member age, age range,
and percentage of females. It also shows the percentage of mem-
bers with public blogs, the percentage of registered members who
have never posted (potential viewers), percentage of posters with
more than 100 posts, and moderation level. The average self-
described age of members ranged from 16.4 to 23.9 years for each
message board, although there was large variation in ages repre-
sented and the mean tends to be negatively skewed. The majority
of message boards had a mean stated age of 18 years; 80% of the
members described themselves as being between the ages of 14
and 20; 31% of all members described themselves as being 15 or
16 years old. In all 10 boards examined, posters describing them-
selves as female were more likely to be registered and to partici-
pate actively. The number of members in each message board
varied dramatically from 70 to 6,656. Membership data from one
of the sites were not available because members post in any of a
wide number of forums, not all of which relate to self-injury.

Table 3
Conditions and Behaviors With Which Self-Injury Message
Boards Are Cross-Listed

Cross-listed condition/behavior
Frequency
(n � 140)

% of
Total

Depression 45 32.0
Eating disorders 24 17.0
Dissociative identity disorder/multiple

personalities 14 10.0
Bipolar 13 9.2
Obsessive-compulsive disorder 10 7.1
Sexual abuse 10 7.1
Posttraumatic stress disorder 7 5.0
Anxiety 4 2.9
LGBTQ 2 1.4
Mood disorders 4 2.8
Addiction 4 2.8
Autism 2 1.4
No other disorder specified 78 56.0

Note. LGBTQ � lesbian, gay, bisexual, transgender, and questioning.

Table 2
Number of Self-Injury Message Boards From One Internet
Service Established by Year

Year No. boards Total membership

1998 1 93
1999 7 949
2000 26 2,831
2001 25 703
2002 28 1,611
2003 19 952
2004 24 806
2005 38 1,698

Total 168 9,643

411SPECIAL SECTION: VIRTUAL CUTTING EDGE

The number of message board members with blogs ranged from
10% to 30%. Preliminary examination of blog content revealed
few substantive differences from message board posts other than
the fact that blogs often provided more detailed accounts of events
and feelings. There was considerable variation in the extent to
which members posted. In four of the boards, more than half of the
membership had never posted a single comment. The majority of
those who did participate posted 2 to 10 messages, although each
message board had a group of high posters with more than 100
posts.

One notable characteristic of message board membership was
that virtually all members have images associated with their online
identity. Characters, such as “Emily the Strange,” digital images
made by posters of scars and bleeding wounds, and other graphic
and bloody icons were frequently included with posted text. Also,
lines from songs, poetry, or books were included in signature lines
and were often expressions of sentiment. One example is the lyric
by the music artist Eminem: “Sometimes I even cut myself to see
how much it bleeds. It’s like adrenaline, the pain is such a sudden
rush for me.”

Message board content. A total of 3,219 posts were examined
from six message boards over a 2-month period; 2,942 were
assigned a code. Although posts could be assigned multiple codes,
most were not. Tables 4 and 5 list the total number of times a post
was assigned a particular content code and the proportion of all
posts examined that fell into this content area. As shown in Table
4, the vast majority of all posts examined fell into 1 of 11 broad
thematic categories. Several of the most common categories con-
tained a variety of subthemes (see Table 5).

Providing informal support to other posters was the most com-
mon type of exchange on the message boards, occurring in 28.3%
of all posts (see Table 4). Comments such as “We’re glad you’ve
come here” and “Just relax and try to breathe deeply and slowly”
were very common. Although less frequent, exchanges around
support were often linked with apologizing behavior (“I’m so sorry
to lay this on you”) and confessions of self-loathing (“I hate myself
for doing this”). Discussion of events that triggered a self-injurious
episode occurred in 19.5% of the posts (see Table 4). Conflict with
important others constituted the primary trigger (34.8% of all

trigger-related posts) followed by depression (24.8%) and stress
(10.3%; see Table 5).

The next largest thematic category discussed involved conceal-
ment of the practice and its effects (primarily scarring), accounting
for 9.1% of all posts examined (see Table 4). Comments focused
largely on anxiety about being discovered, how to manage scars,
and the extent to which posters had to be dishonest to maintain
secrecy (see Table 5).

Posts about the perceived addictiveness of self-injury were
almost as common (8.9%) (see Table 4). Typical examples include
“It just haunts me and I don’t think I’ll ever get away from it” and
“I may try and quit but even if I succeed, I’ll always dream of
razorblades and blood.” Observations starting with the phrase
“I’ve been cutting free for [length of time]” accounted for almost
half of posts coded with addiction elements. References to stable
patterns of self-injury, tendency to minimize the problem, similar-
ities to other drugs, multiple quit attempts, the need to self-injure
more or more deeply because of increased tolerance for effects of
self-injury, and relapse after quitting were all areas discussed
within this category (see Table 5). Often this discussion co-
occurred with comments regarding frustration at parents, caregiv-
ers, and others, who, according to the posters, do not appreciate the
addictive nature of the behavior.

Discussion of formal help seeking from physical or mental
health professionals occurred in 7.1% of all posts (see Table 4).
Attitudes toward and experiences with treatment were largely
positive (44.1% of all help-seeking posts reflected positive atti-
tudes toward therapy), and discussion of experiences related to

Table 4
Dominant Thematic Categories in Message Board Content

Category
No. of category

occurrences

% Posts
examined

(N � 3,219)a

Informal provision of support for others 913 28.3
Motivation/triggers 629 19.5
Concealment issues 292 9.1
Addiction elements 288 8.9
Formal help seeking, treatment 229 7.1
Requesting, sharing techniques 200 6.2
Link to other mental health conditions 153 4.7
References to popular culture 137 4.2
Interpretation of other’s perceptions 85 2.6
Perceptions of self 70 2.1
Venting or apologizing behavior 61 2.9
Uncategorized 277 8.6

a Individual posts could be assigned multiple codes; total percentages will
not equal 100%.

Table 5
Breakdown of Most Common Themesa

Category/themes
Total no. of
occurrences

Percentage of
primary
category

Triggers
Conflict with important others 212 34.8
Perceived depression 151 24.8
School or work stress 63 10.3
Loneliness 44 7.2
Sexual abuse/rape 22 3.6
Other 117 19.2

Concealment issues
Anxiety about concealment 149 51.8
Managing scars 109 37.3
Acknowledgment of dishonesty 29 10.9

Addictive elements
“I’ve been cutting free for . . .” 120 41.7
Stable pattern/cannot control urge 52 18.1
Minimize problem 28 9.7
Liken to other drugs 27 9.3
Multiple attempts to quit 27 9.4
Increased tolerance, need more over time 24 8.4
Relapse 10 3.4

Help seeking and treatment
Positive attitudes toward therapy 101 44.1
In therapy 60 26.2
Negative attitude toward therapy 43 18.7
References to medications 25 10.9

a Provision of informal support, although the dominant theme, is not
included in this table because specific categories of support were not
tracked.

412 WHITLOCK, POWERS, AND ECKENRODE

therapy accounted for a sizable portion of formal help-seeking
posts as well (26.2%). Negative attitudes and active discourage-
ment from seeking therapy were also evident in 18.7% of the
formal help-seeking posts. References to specific medications oc-
curred in about 1 in 10 cases.

The sixth most common category of discussion for all age
groups related to sharing details about techniques use to self-
injury, accounting for 6.2% of all posts. These exchanges were
generally either descriptions of specific self-injury techniques or
requests for specific technique information. The following conver-
sation exemplifies this type of exchange:

Poster 1: Does anyone know how to cut deep without having it sting
and bleed too much?

Poster 2: I use box cutter blades. You have to pull the skin really tight
and press the blade down really hard. You can also use a tourniquet
to make it bleed more.

Poster 3: I’ve found that if you press your blade against the skin at the
depth you want the cut to be and draw the blade really fast it doesn’t
hurt and there is blood galore. Be careful, though, ‘cause you can go
very deep without meaning to.

Poster 1: Okay, I’ll get a Stanley blade ‘cause I hear that it will cut
right to the bone with no hassle. But I’ll be careful if I do use a
tourniquet and I won’t cut that deep.

Mental health conditions empirically linked to self-injurious
behavior, such as depression, eating disorders, suicidality, and
sexual abuse, were cited as either linked with or as a trigger for
self-injury in 4.7% of all posts (see Table 4). References to pop
culture (e.g., music, movies, books, celebrities, and characters with
special significance to posters) appeared in just over 4% of the
posts. Discussion and interpretation of the perceptions of others
came up in 2.6% of the posts. Recognition of the pain their
behavior caused or might cause others, in conjunction with diffi-
culty stopping a behavior once a pattern was formed, was often
linked to the need for secrecy and feelings of shame. Similarly,
perceptions of self, usually negative, were shared in just over 2%
of all posts. Remaining posts were rants, aimless venting, usually
frustration or anger, or apologies for sharing.

Study 2: Variations in the Use of Message Boards Among
Self-Injurers

This study was intended to explore the correlations between
content areas raised for discussion. It focuses on exchange in four
broad areas: (a) help seeking and disclosure, (b) technique sharing,
(c) comorbidity, and (d) attitudes toward self and other posters.
These areas were selected to explore the role of message boards
exchange in spreading information about self-injurious practices
and influencing help-seeking behavior.

Method

Identification and sampling. The unit of analysis for this study was
the individual self-injury message board poster. Individual posters were
selected from five low- and one medium-moderation self-injury mes-
sage boards included in the previous study (A, B, D, E, F, and G in
Table 1).

Because the extent to which posters sought and shared self-injury
techniques was of interest, low-moderation boards were selected as
recruitment sites to ensure that the content of message posts would not
be blocked. One medium-moderation board was selected to permit a
balanced sample by age, because the low-moderation boards did not
contain an adequate number of active posters for which age was
available. Each site allowed for all posts from individual posters to be
searched readily.

Individual posters were identified by looking through threads and iden-
tifying individuals in the age brackets of interest with more than 50 posts.
Then 20 in each of three self-described age groups (13–15, 16 –18, 19 –22
years) were selected at random, for a total sample size of 60. Once
individual posters were identified, 50 posts from each individual were
selected at random from posts made from July 2004 to January 2005. A
6-month window was chosen to ensure that each poster monitored would
have adequate time to make multiple posts and engage in a variety of
interactions.

Measures. As in Study 1, the content analysis of posts was conducted
using a set of binary (present–absent) codes. These codes were similar to
those used in Study 1 but were focused on six specific types of exchange:
(a) soliciting and sharing techniques, (b) attitudes toward and disclosed use
of formal support, (c) solicitation and provision of informal support, (d)
disclosure to nontherapeutic others (e.g., family and friends), (e) disclosure
of other mental health conditions, and (f) disclosed self-concept. A total of
17 measures were used to indicate each of these broad conceptual domains
(Table 6). Although many of these measures are similar to those used in
Study 1, Study 2 measures were designed to focus on a specific type of
exchange and to permit individual-level analysis of correlations between
variables. Self-represented demographics, including gender, age, and total
number of posts, were also recorded.

Coding of posts. A total of 3,000 individual posts were coded (50 per
each 60 individuals). The actual number of posts per individual ranged
from 60 to several hundred. Because coders could query for all threads
posted by individual posters, they randomly selected 50 by identifying the
total number of pages and threads any one poster made and dividing by 50.
The resulting number was used to systematically identify threads included
in the analysis. For example, if a poster made 100 posts, coders coded
every two posts. If the resulting interval inadvertently identified a redun-
dant thread (one already coded but that surfaced later), the next thread was
used. Posts were not coded when they did not contain content relevant to
the study objectives. Six coders were assigned individual posters to track
and code. To establish intercoder reliability, 40% of the observations
(1,200 posts) were independently coded by three pairs of coders. Each pair
of coders coded 400 posts in common. No two pairs coded the same posts.
Agreement was assessed by calculating the proportion of posts each
individual in the pair coded the same. The average intercoder agreement
across all three pairs was 93% (range � 90 –96%).

Results

To assess the relationship between soliciting and sharing self-
injurious practices and informal and formal help seeking, Spear-
man correlations were conducted between all measures. Results
are shown in Table 6.

The analysis revealed two trends in the correlations: one in
which more positive exchanges were correlated and one in which
negative exchanges were correlated. For example, offering infor-
mal support was correlated with disclosing that someone knows,
suggesting formal treatment, seeking advice on stopping, and
disclosing a history of trauma. Similarly, seeking advice on stop-
ping and harm reduction were correlated with each other and with
seeking advice about disclosure and sharing positive remarks
about oneself. Encouraging formal treatment of self-injurious be-

413SPECIAL SECTION: VIRTUAL CUTTING EDGE

T
ab

le
6

S
p
ea

rm
a
n
’s

C
o
rr

el
a
ti

o
n
s

fo
r

A
ll

S
tu

d
y

V
a
ri

a
b
le

s
(N

�
6
0

in
d
iv

id
u
a
ls

;
3
,0

0
0

p
o
st

s)

M
ea

su
re

1
2

3
4

5
6

7
8

9
10

11
12

13
14

15
16

17

T
ec

hn
iq

ue
s

1.
S

ha
re

te
ch

ni
qu

es

—

2.
R

eq
ue

st
te

ch
ni

qu
es

.2
31

—

In
fo

r

m
al

su
pp

or
t

3.
O

ff
er

in
fo

rm
al
su
pp
or
t

.0
75

�
.2

04
—

4.
S

ee
k

ad
vi

ce
on

st
op

pi
ng

.0
98

.0
88

.2
99

*
—

5.
S

ee
k
ad
vi
ce
on

re
du

ci
ng

ha
rm

�
.0

42
.0

85
.1

08
.3

19
*

—

D
is

cl
os

ur
e

at
ti

tu
de

6.
S

ee
k
ad
vi

ce
ab

ou
t

di
sc

lo
su

re
�

.0
84

�
.2

18
.0

06
.2

28
.2

81
*

—
7.

E
nc

ou
ra

ge
di

sc
lo

su
re

.1
03

.0
48

.1
87

.1
12

.0
09

.1
63

—
8.

D
is

co
ur

ag
e

di
sc
lo
su

re
.3

32
**

.0
09

.1
26

.3
14

*
.2

43
.1

60
.1

34
—

F
or

m
al
su
pp
or
t

9

.
D

is
cl

os
e

tr
ea

t

m
en

t
.1

11
�

.1
45

.0
82

.1
50

.1
14

.0
02

�
.0

88
.0

73
—

10
.

S
ug

ge
st

fo
rm

al
tr

ea
tm

en
t

�
.0

25
�

.2
28

.5
29

*

*
.1

29
�

.1
56

.1
28

.3
34

**
.2

39
.1

08
—

D
is
cl
os
ur
e

st
at

us

11
.

D
is
cl
os

e
th

at
so

m
eo

ne
kn

ow
s

.3
15

*
�

.0
79

.2
89

*
.1

91
.0

04
.0

56
.1

25
.1

97
.1

99
.1

69
—

A
ss

oc
ia

te
d

co
nd

it
io

ns

12
.

S
ui

ci
da

li
ty

.1
10

�
.1

76
.1

45
.1

49
.1

63
.0

31
�

.0
87

.1
67

.4
57

**
�

.1
90

.2
99
*
—

13
.

H
is

to
ry

of
tr

au
m

a
.0

55
.1

05
�

.2
86

*
.0

22
.0

08
.0

99
�

.0
82

.1
06

.1
57

�
.0

24
�

.0
13

.0
05

—
14

.
D

ia
gn

os
ed

di
so

rd
er

.1
64

.1
20

�
.0

32
.1

08
.0

96
.0

82
�

.0
58

.1
20

.4
78

**
.1

25
�
.0
48

.1
42

.2
34

—
15

.
D

is
pa

ra
gi

ng
co

m
m

en
ts

ab
ou

t
se

lf
.1

85
.1

33
�

.0
68

.0
61

.1
07

.3
19

*
�

.1
84

�
.0

59
.0

92
.0

08
.0

70
.1

71
.1

39
.2

61
*

—
16

.
P

os
it

iv
e

co
m

m
en

ts
ab

ou
t

se
lf

.1
83

�
.0

87
�

.1
27

.2
91

*
.2

07
.2

37
.1

99
.3

01
*

.1
32

.1
82

.1
44

.1
54

.0
29

.1
03

.2
69

*
—

17
.

D
is

pa
ra

gi
ng

co
m
m
en
ts
ab
ou
t

ot
he

rs
.0

29
�

.0
10

�
.1

44
.1

98
�

.0
31

�
.1

14
.2

93
*

.2
37

�
.0

03
.1

00
.0

66
.1

93
�

.1
02

�
.1

43
�

.2
41

.0
98
—

*
p

�
.0

5.
**

p
�

.0
1.

414 WHITLOCK, POWERS, AND ECKENRODE

havior, offering informal support, and encouraging disclosure to
others were interrelated as well.

A somewhat more negative cluster of exchange linked discus-
sion of technique sharing with negative attitudes toward disclo-
sure. However, discouraging disclosure was also associated both
with positive views of self and with seeking advice on stopping. In
addition, disclosing that one was in treatment was correlated with
disclosing another diagnosed disorder and suicidality. Individuals
who disclosed a history of trauma were likely to offer informal
support, and those who disclosed suicide-related behaviors were
more likely than others to disclose that someone knows about their
self-injurious behavior.

Discussion

Our findings confirm that Internet message boards provide a
powerful vehicle for bringing together self-injurious adolescents.
Although the message boards examined for these studies may not
be representative of all self-injury message boards, they do provide
a snapshot of content and exchange common in those with high
activity. We found that in the last 5 years hundreds of message
boards specifically designed to provide a safe forum for self-
injurious individuals have come into existence. Many of these are
populated by individuals who identify themselves as females be-
tween 14 and 20 years of age. Although the strong preponderance
of females may not accurately reflect the gender breakdown of
self-injurious behavior in the general population (Whitlock, Eck-
enrode, & Silverman, in press), it may reflect the tendency for
females to solicit more informal and formal help and social support
compared with males (Fuhrer, Stansfield, Chemali, & Shipley,
1999; Saunders, Resnick, Hoberman, & Blum, 1994).

Once online, message board members are able to post or pas-
sively observe a wide variety of anonymous exchanges. Just less
than half of all message boards we investigated were cross-listed
with conditions known to be comorbid with self-injurious behav-
ior, such as depression, eating disorders, and suicide. Of all types
of online dialogue, the giving and receiving of informal support
and discussion of proximal life events that trigger self-injury are
most common. Posters also share casual and sometimes very
personal information related to the addictive qualities of their
practice, their fears relating to disclosure, experiences with ther-
apy, how they self-injure, and other related health concerns. Al-
though our findings are generally consistent with existing literature
on self-injurious behavior, the correlations documented among
informal support, encouraging disclosure, and advising formal
treatment suggest that online interactions may be providing self-
injurers support and meaning outside the clinical setting.

What the self-injurious adolescents in our study appeared to do
online is what most people who trust each other do in conversa-
tion: exchange support, share personal stories about daily life
events, and voice opinions and ideas. Because the anonymity of
the Internet inspires the most personal and trusting of exchange
between individuals with little or no previous relationship, online
sharing may encourage greater and more truthful disclosures
(McKenna & Bargh, 2000), especially among self-injurers, many
of whom suffer from depressive symptomatology (Ybarra et al.,
2005). For adolescents, this support may be particularly valuable,
because healthy social and emotional development hinges on their
ability to establish caring, meaningful relationships, to find accep-

tance and belonging in social groups, and to establish interpersonal
intimacy (Reis & Shaver, 1988; Sullivan, 1953). These develop-
mental tasks can be especially difficult for young people struggling
with intense shame, isolation, and distress, particularly when the
source and outcome of these feelings must be kept hidden. The
assurance of online anonymity may contribute to identity construc-
tion by providing opportunities for adolescents with marginalized
or nonmainstream proclivities to experiment with different social
roles and selves (McKenna & Bargh, 2000; Turkle, 1995).

The less positive side of our findings suggests that participation
in self-injury message boards may also expose vulnerable adoles-
cents to a subculture in which self-injury is normalized and en-
couraged. For example, issues related to concealment of self-
injurious behavior, identified as a dominant theme here, may make
self-injury message boards particularly potent agents of self-injury,
because sharing techniques and motives can take place anony-
mously. In light of evidence here and elsewhere that self-injurious
behavior may possess addictive qualities (see Yates, 2004, for
review), the adolescent drive to belong and the satisfaction that
comes with associating with a community of similar others may
inadvertently feed a fundamentally self-destructive behavior for
some participants. The correlation documented here between shar-
ing injurious techniques and discouraging disclosure lends support
to this possibility.

Although not impossible for individual self-injurers to have
gathered before the advent of the Internet, easy access to a virtual
subculture of like-minded others may reinforce the behavior for a
much larger number of youth. The tendency for self-injurious
behavior to follow epidemic-like patterns in institutional settings
such as hospitals and detention facilities (Matthews, 1968; Ross &
McKay, 1979; Rosen & Walsh, 1989; Taiminen, Kallio-
Soukainen, Nokso-Koivisto, Kaljonen, & Helenius, 1998) suggests
that the behavior may be socially contagious in other settings and,
therefore, through the Internet as well. As Brumberg (1992) has
argued for eating disorders, discussion of techniques and the
perceived benefits of self-injury may add potentially lethal behav-
iors to the repertoire of established self-injurers, not yet committed
message board members, and even nonparticipating message board
viewers who are exploring identity options. Indeed, some message
boards contain links to pro-self-injury Web sites where Internet
users can purchase self-injury paraphernalia such as bracelets or
clothing that signify self-injury status and cutting clubs have been
rumored to be a growing form of friendship ritual (Booth, 2004).
Some self-injury Web sites host forums specifically dedicated to
sharing new self-injury techniques. Discussion of technique shar-
ing, triggers, negative attitudes toward formal and informal help
seeking, and the pleasures and pains of self-injury addiction may
influence behavioral choices outside of the virtual realm that are
later brought back, shared, and used to ensure support and mem-
bership. It may also make some youth targets for individuals who
falsely pose as supporters to accomplish other, less benevolent
aims. For vulnerable adolescents, the difficulty of ending a strat-
egy for coping with distress (self-injury) and leaving a needed
source of support (individual or collective members of the virtual
community) may stifle the desire to find alternate ways of coping
with stress. Moreover, the low-sense of self-worth common among
self-injurers may expose them to damaging online relationships.

This study supports the findings of other scholars of Internet and
development (Suzuki & Calzo, 2004) by suggesting that electronic

415SPECIAL SECTION: VIRTUAL CUTTING EDGE

forums provide a rich data source for studying issues pertinent to
marginalized subgroups of the adolescent population that are hard
to identify and reach. Observation of interaction as it unfolds,
rather than in retrospective self-reports, eliminates biases relating
to self-report and recall and allows study of actual transactions
between individuals. Moreover, because self-injury is typically a
private, secretive behavior, the Internet provides a unique opportunity
to study exchange between members of a group rarely assembled
outside of a clinical setting. As such, it may provide a valuable means
of accessing information and perspectives useful in clinical settings.
Indeed, self-injury message boards may provide a vehicle for admin-
istering a Web-based intervention to reach self-injurers.

Our methods have limits. It was not possible to observe all 406
message boards identified in this study. The 10 selected for anal-
ysis, although the most active at the time of the study, may not be
generalizable to all self-injury message boards. Message boards,
like all communities, possess their own culture and character and
are governed by subtle and overt norms and mores. Rules posted
and enforced by the moderator is one example of this. How these
affect participant self-selection into specific message board com-
munities and the content of their exchange is unclear. Moreover,
the absence of demographic data on individual Internet users
prohibits many important and useful analyses.

In conclusion, this research has several implications for those
interested in the nexus between adolescents, the Internet, and
self-injury. Like other social environments (e.g., schools, families,
or neighborhoods), the Internet plays a powerful role in shaping
opportunities for adaptive and maladaptive social interaction.
What occurs in the virtual world, however, is largely invisible to
adults and treatment providers who have a vested interest in
adolescents and self-injurious behavior. Although it is neither
possible nor, perhaps, desirable to monitor all adolescent Internet
use, particularly as youth age and become increasingly indepen-
dent, it is very important for adults to know something about what
adolescents, particularly vulnerable adolescents, encounter in the
virtual communities they inhabit. The data presented here are
exploratory and would be strengthened by tracking a larger num-
ber of individual posters over a longer period of time to prospec-
tively capture variability over time within individual and within
and between age groups. Additionally, future research in this area
should seek to verify the accuracy of what adolescent users share
online, to better understand the unique role the Internet may play
in affecting off-line behavior, and to clarify the role played by
Internet use in spreading or deepening self-injurious practices
among adolescents.

References

Alderman, T. (1997). The scarred soul: Understanding and ending self-
inflicted violence. Oakland, CA: New Harbinger.

Arnett, J. J. (1999). Adolescent storm and stress, reconsidered. American
Psychologist, 54, 317–326.

Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for
interpersonal attachments as a fundamental human motivation. Psycho-
logical Bulletin, 117, 497–529.

Becker, H. J. (2000). Who’s wired and who’s not: Children’s access to and
use of computer technology. Future of Children, 10, 44 –75.

Bishop, J. A., & Inderbitzen, H. M. (1995). Peer acceptance of friendship:
An investigation of the relation to self-esteem. Journal of Early Ado-
lescence, 15, 476 – 489.

Booth, S. (2004, February). Cutting clubs: What’s the latest and most
shocking new “friendship” ritual? Teen People looks at how a growing
number of kids are bonding with their peers by slicing themselves with
razor blades. Teen People, 7, 98.

Briere, J., & Gil, E. (1998). Self-mutilation in clinical and general popu-
lation samples: Prevalence, correlates, and functions. American Journal
of Orthopsychiatry, 68, 609 – 620.

Brodsky, B. S., Cloitre, M., & Dulit, R. A. (1995). Relationship of
dissociation to self-mutilation and childhood abuse in borderline per-
sonality disorder. American Journal of Psychiatry, 152, 1788 –1792.

Brumberg, J. J. (1992). From psychiatric syndrome to “communicable”
disease: The case of anorexia nervosa. In C. Rosenberg & J. Golden
(Eds.), Framing disease (pp. 134 –154). Piscataway, NJ: Rutgers Uni-
versity Press.

Claes, L., Vandereycken, W., & Vertommen, H. (2003). Eating-disordered
patients with and without self-injurious behaviours: A comparison of psy-
chopathological features. European Eating Disorders Review, 11, 379 –396.

Compas, B. E. (1987). Coping with stress during childhood and adoles-
cence. Psychological Bulletin, 101, 393– 403.

Connors, R. (2000). Self-injury: Psychotherapy with people who engage in
self-inflicted violence. Northvale, NJ: Jason Aronson.

Conterio, K., & Lader, W. (1998). Bodily harm: The breakthrough healing
program for self injurers. New York: Hyperion Press.

Deiter, P. J., Nicholls, S. S., & Pearlman, L. A. (2000). Self-injury and self
capacities: Assisting an individual in crisis. Journal of Clinical Psychol-
ogy, 56, 1173–1191.

Dubo, E. D., Zanarini, M. C., Lewis, R. E., & Williams, A. A. (1997).
Childhood antecedents of self-destructiveness in borderline personality
disorder. Canadian Journal of Psychiatry, 42, 63– 69.

Dulit, R. A., Fyer, M. R., Leon, A. C., Brodsky, B. S., & Frances, A. J.
(1994). Clinical correlates of self-mutilation in borderline personality
disorder. American Journal of Psychiatry, 151, 1305–1311.

Favaro, A., & Santonastaso, P. (2002). The spectrum of self-injurious
behavior in eating disorders. Eating Disorders, 10, 215–225.

Favazza, A. R. (1992). Repetitive self-mutilation. Psychiatric Annals, 22,
60 – 63.

Favazza, A. R. (1996). Bodies under siege: Self-mutilation and body
modification in culture and psychiatry (2nd ed.). Baltimore, MD: Johns
Hopkins University Press.

Favazza, A. R. (1998). The coming age of self-mutilation. Journal of
Nervous and Mental Disease, 186 (5), 259 –268.

Favazza, A. R. (1999). Self-mutilation. In D. G. Jacobs (Ed.), The Harvard
Medical School guide to suicide assessment and intervention (pp. 125–
145). San Francisco: Jossey-Bass.

Favazza, A. R., & Conterio, K. (1989). Female habitual self-mutilators.
Acta Psychiatrica Scandinavica, 79, 283–289.

Favazza, A. R., DeRosear, L., & Conterio, K. (1989). Self-mutilation and
eating disorders. Suicide & Life-Threatening Behavior, 19, 352–361.

Fuhrer, R., Stansfield, S. A., Chemali, J., & Shipley, M. J. (1999). Gender,
social relations and mental health: Prospective findings from an occu-
pational cohort. Social Science & Medicine, 48, 77– 87.

Galley, M. (2003, December 3). Student self-harm: Silent school crisis.
Education Week, p. 2.

Gardner, D. L., & Cowdry, R. W. (1985). Suicidal and parasuicidal
behavior in borderline personality disorder. Psychiatric Clinics of North
America, 8, 389 – 403.

Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory.
Strategies for qualitative research. Chicago: Aldine.

Gratz, K. L., Conrad, S. D., & Roemer, L. (2002). Risk factors for
deliberate self-harm among college students. American Journal of Or-
thopsychiatry, 72, 128 –140.

Gross, E. F. (2004). Adolescent Internet use: What we expect, what teens
report. Journal of Applied Developmental Psychology, 25, 633– 649.

416 WHITLOCK, POWERS, AND ECKENRODE

Gross, E. F., Juvonen, J., & Gable, S. L. (2002). Internet use and well-being
in adolescence. Journal of Social Issues, 58, 75–90.

Hartup, W. W. (1996). The company they keep: Friendships and their
developmental significance. Child Development, 67, 1–13.

Hawton, K., Fagg, J., Simkin, S., Bale, E., & Bond, A. (2000). Deliberate
self-harm in Oxford, 1985–1995. Journal of Adolescence, 23, 47–55.

Heitner, E. (2002). The relationship between use of the Internet and social
development. Unpublished doctoral dissertation, Pace University.

Holmes, A., & Nadelson, C. A. (Eds.). (2000). Cutting the pain away:
Understanding self mutilation. Philadelphia: Chelsea House.

Kokaliari, E. (2005). Deliberate self-injury: An investigation of the prev-
alence and psychosocial meanings in a non-clinical female college
population. Unpublished doctoral dissertation, Smith College, School
for Social Work.

Kraut, R., Kiesler, S., Boneva, B., Cummings, J., Helgeson, V., & Crawford,
A. (2002). Internet paradox revisited. Journal of Social Issues, 51, 49 –74.

Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Mukopadhyay, T., &
Scherlis, W. (1998). Internet paradox: A social technology that reduces
social involvement and psychological well-being? American Psycholo-
gist, 53, 1017–1031.

Lenhart, A., Madden, M., & Hitlin, P. (2005, July 27). Teens and technol-
ogy: Youth are leading the transition to a fully wired and mobile nation.
Retrieved October 13, 2005, from http://www.pewInternet.org/pdfs/
PIP_Teens_Tech_July2005web

Lenhart, A., Rainie, L., & Lewis, O. (2001). Teenage life online: The rise
of the instant-message generation and the Internet’s impact on friend-
ships and family relationships. Washington, DC: Pew Internet & Amer-
ican Life Project.

Maczewski, M. (2002). Exploring identities through the Internet: Youth
experiences online. Child & Youth Care Forum, 31, 111–129.

Martin, G., Rozanes, P., Pearce, C., & Allison, S. (1995). Adolescent
suicide, depression and family dysfunction. Acta Psychiatrica Scandi-
navica, 92, 336 –344.

Matthews, P. C. (1968). Epidemic self-injury in an adolescent unit. Inter-
national Journal of Social Psychiatry, 14, 125–133.

McKenna, K. Y. A., & Bargh, J. A. (2000). Plan 9 from Cyberspace: The
implications of the Internet for personality and social psychology. Per-
sonality and Social Psychology Review, 4, 57–75.

McKenna, K. Y. A., & Green, A. S. (2002). Virtual group dynamics.
Group Dynamics: Theory, Research and Practice, 6, 116 –127.

McKenna, K. Y. A., Green, A., & Gleason, M. (2002). Relationship
formation on the Internet: What’s the big attraction? Journal of Social
Issues, 58, 9 –31.

Merten, D. E. (1996). Visibility and vulnerability: Responses to rejection
by nonaggressive junior high school boys. Journal of Early Adolescence,
16, 5–26.

Moffit, T. E. (1993). Adolescence-limited and life-course-persistent anti-
social behavior: A developmental taxonomy. Psychological Review,
100, 674 –701.

Muehlenkamp, J. J., & Gutierrez, P. M. (2004). An investigation of
differences between self-injurious behavior and suicide attempts in a
sample of adolescents. Suicide & Life-Threatening Behavior, 34, 12–24.

Nie, N. H., & Erbring, L. (2000). Internet and society: A preliminary report.
Stanford, CA: Stanford Institute for the Quantitative Study of Society.

Nixon, M. K., Cloutier, P. F., & Aggarwal, S. (2002). Affect regulation and
addictive aspects of repetitive self-injury in hospitalized adolescents.
Journal of the American Academy of Child & Adolescent Psychiatry, 41,
1333–1341.

Petersen, A. C., Kennedy, R. E., & Sullivan, P. (1991). Coping with
adolescence. In M. E. Colton & S. Gore (Eds.), Adolescent stress (pp.
93–110). New York: Aldine.

Reis, H. T., & Shaver, P. (1988). Intimacy as an interpersonal process. In
S. Duck (Ed.), Handbook of personal relationships: Theory, research,
and interventions (pp. 367–389). Chichester, UK: Wiley.

Roberts, D. F., Foehr, U. G., & Rideout, V. (2005). Generation M: Media
in the lives of 8 –18 year-olds. Washington, DC: Henry J. Kaiser Family
Foundation.

Rosen, P. M., & Walsh, B. W. (1989). Patterns of contagion in self
mutilation epidemics. American Journal of Psychiatry, 146, 656 – 658.

Ross, R. R., & McKay, H. B. (1979). Self mutilation. Lexington, MA:
Lexington Books.

Ross, S., & Heath, N. (2002). A study of the frequency of self-mutilation
in a community sample of adolescents. Journal of Youth and Adoles-
cence, 31, 66 –77.

Ross, S., & Heath, N. L. (2003). Two models of adolescent self-mutilation.
Suicide & Life-Threatening Behavior, 33, 277–287.

Sansone, R. A., & Levitt, J. L. (2002). Self-harm behaviors among those
with eating disorders: An overview. Eating Disorders, 10, 205–213.

Saunders, S. N., Resnick, M. D., Hoberman, H. M., & Blum, R. W. (1994).
Formal help seeking behavior of adolescents identifying themselves as
having mental health problems. Journal of American Academy of Child
and Adolescent Psychiatry, 33, 718 –728.

Strong, M. (1998). A bright red scream: Self-mutilation and the language
of pain. New York: Viking.

Subrahmanyam, K., Greenfield, P. M., & Tynes, B. (2004). Constructing
sexuality and identity in an online teen chat room. Journal of Applied
Developmental Psychology, 25, 651– 666.

Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York:
Norton.

Suzuki, L. K., & Calzo, J. P. (2004). The search for peer advice in
cyberspace: An examination of online teen bulletin boards about health
and sexuality. Applied Developmental Psychology, 25, 685– 698.

Taiminen, T. J., Kallio-Soukainen, K., Nokso-Kovisto, H., Kaljonen, A., &
Helenius, H. (1998). Contagion of deliberate self-harm among adoles-
cent inpatients. Journal of the American Academy of Child & Adolescent
Psychiatry, 37, 211–217.

Tantam, D., & Whittaker, J. (1993). Self-wounding and personality disor-
der. In P. Tyrer & G. Stein (Eds.), Personality disorder reviewed (pp.
191–224). London: American Psychiatric Press.

Turkle, S. (1995). Life on the screen: Identity in the age of the Internet.
New York: Simon & Schuster.

Tynes, B., Reynolds, L., & Greenfield, P. M. (2004). Adolescence, race,
and ethnicity on the Internet: A comparison of discourse in monitored vs.
unmonitored chat rooms. Journal of Applied Developmental Psychology,
25, 667– 684.

Walsh, B. W., & Rosen, P. M. (1988). Self-mutilation: Theory, research
and treatment. New York: Guilford Press.

Wartella, E., Caplovitz, A. G., & Lee, J. H. (2004). From baby Einstein to
leapfrog, from doom to The Sims, from instant messaging to Internet
chat rooms: Public interest in the role of interactive media in children’s
lives. Social Policy Report, 18, 3–16.

Welsh, P. (2004). Look students’ scars point to emotional pain. USA
Today, p. 11a.

Whitlock, J. L., Eckenrode, J. J., & Silverman, D. (in press). Self-injurious
behavior in a college population. Pediatrics.

Yates, T. M. (2004). The developmental psychopathology of self-injurious
behavior: Compensatory regulation in posttraumatic adaptation. Clinical
Psychological Review, 24, 35–74.

Ybarra, M. L., Alexander, C., & Mitchell, K. J. (2005). Depressive symp-
tomatology, youth Internet use, and online interactions: A national
survey. Journal of Adolescent Health, 36, 9 –18.

Young People and Self Harm: A National Inquiry. (2004). What do we
already know? Prevalence, risk factors, and models of intervention.
Retrieved June 29, 2004, from www.selfharmuk.org

Received April 1, 2005
Revision received December 15, 2005

Accepted December 16, 2005 �

417SPECIAL SECTION: VIRTUAL CUTTING EDGE