The UIC college student weblogs at Xanga.com: A content analysis
Prepared for delivery at the
6th International and Interdisciplinary Conference of the
Association of Internet Researchers, Internet Generations 6.0, October 5-9, 2005
Francisco Seoane Pérez
Department of Communication
University of Illinois at Chicago
1140 Behavioral Sciences Building
1007 W. Harrison St., MC 132
Chicago, IL 60607-7137
The UIC college student weblogs at Xanga.com: A content analysis
This paper presents the results of a content analysis of 237 weblogs authored by college
students from the University of Illinois at Chicago (UIC) hosted at Xanga.com. The sample
is statistically representative of the 650 online diaries that UIC students had registered in
the named website at the time of the sample selection (April 2005). The coding protocol,
adapted from the one used by Huffaker (2004) on his thesis on gender similarities and
differences among teenage bloggers, looked at how the weblog authors introduce
themselves, the kind of issues they discuss in their diaries, and the type of resources they
use (images, videos, music, emoticons) to communicate. The feedback from the readers
was also considered, as well as the integration of the individual weblogs into broader
networks of weblogs (“blogrings”) within the Xanga blogging community. The results
show that there are as many males as female UIC bloggers at Xanga, although a majority of
them are of Asian origin. Seven in ten blog authors identified themselves by their first
name, and half of them included their full name in their profiles. The average blog writer
was 21 years old. Six in ten blogs did not include any image in their posts. Two in ten blogs
had been abandoned at the time of sample selection. The average of comments per post was
3.4. The most frequently discussed topics were friends (61% of blogs), school (60%),
nightlife (39%) and romance (37%).
Going online is an everyday activity for most college students in the United States
(Jones, 2002). And there is increasing evidence that the act of reading blogs is taking part
of their time on the Internet. A recent Gallup survey (Saad, March 11, 2005) showed that
almost half of Internet users between 18 and 29 years old (the age range of the majority of
college students) read blogs. Online personal diaries, the type of weblog favored by the
younger generations, have become part of many college students’ media diet. This
exploratory research focuses on the life blogs created by the students of one particular
university, the University of Illinois at Chicago, in one of the most popular weblog hosts,
Our initial aim was to offer a representative description of the form and content of
these online diaries that attract so much of the college students’ attention. But, if
determining the population blogs is still a challenge for those scholars doing research on the
topic, identifying the population of the college student-authored blogs in the United States
seems to be as a difficult endeavor as the first.
As long as we don’t find the proper criteria to study the college student blogs in a
representative manner, we decided to concentrate our efforts in the detailed description of
the blogs authored by the students at our university, the University of Illinois at Chicago.
During a previous qualitative research that involved in-depth interviews with students about
their Internet use, we knew about the popularity of Xanga among UIC students. This
website, operated by New York based Xanga.com, Inc., is the fourth blog host in unique
visitors in the United States (Comscore Networks, p. 5) and one of the three leaders of the
blog-hosting business along with BlogSpot and LiveJournal (Henning, 2005). Jeffrey
Henning, chief operating officer of Perseus Development Corporation and author of a
recent survey on the blog population in the United States, estimates the number of Xanga
accounts in 7.5 million 1 .
LiveJournal and Xanga are specialized in community-based blogging. Users can
join groups of blogs (“communities” in LiveJournal, “blogrings” in Xanga) that share a
common feature of interest. For the purposes of our research, we were interested in those
bloggers who become members of communities identified with the name of the university
they belong. The UIC “blogring” at Xanga had 648 members in April 2005, about six times
more members than the UIC “community” at LiveJournal.
Xanga is thus, to our knowledge, the most popular community-based blog-host
among UIC students, and that’s why we decided to describe the form and content of the
blogs that belong to the UIC blogring at Xanga. A representative randomly selected sample
of 237 blogs was selected among this virtual community to perform a content analysis. We
looked at how bloggers identify themselves, recording their demographic data. We also
paid attention to the design features of the blogs, their frequency of use, and the most
popular topics discussed on them, among other aspects.
The present paper intends to provide a systematic depiction of the personal blogs
authored by the students of the largest institution of higher education in the Chicago area by
analyzing the diaries publicized in their favored blog host. Although the results of this
study will only be representative of a particular community of blogs, we believe they may
provide some valuable insights about the blogging activity of college students in the United
Personal communication with author, September 28, 2005. Xanga did not respond to our inquiries regarding
information about their users’ demographics. This silence contrasts with the publicity of data offered by
LiveJournal on its statistics page, which is updated every 24 hours: http://www.livejournal.com/stats.bml
2. Literature Review
2.1. Placing personal diary weblogs into the blogosphere
In the first quarter of 2005, about 10 million blogs were created (Henning, 2005).
This means that in only four months, the blogosphere grew as much as it had grown from
2000 to 2004. At the end of 2005, the blogosphere is expected to harbor 53 million voices.
Personal blogs, those in which the author publicizes accounts of his/her own life
experiences, may constitute the most important subtype of weblogs. Although every
measure related to the size of such an elusive population as the blogosphere must be taken
with caution, there are several bits of evidence that help support the claim of the
prominence of online diaries over other types of blogs, such as filters or k-logs.
Blogs owe much of their popularity to the easiness with which they can be created.
The browser-based software offered for free by the blog hosts that mushroomed since 1999
was the killer-application that explains their adoption by so many Internet users. According
to data gathered by ComScore Networks (2005), of the top 5 blog hosts in number of
unique visitors during the first quarter of 2005 in the United States, three (Livejournal,
Xanga and AOL Journals) are specialized in offering online personal journals. The other
two blog hosts in this top 5, Blogger and TypePad, are not strictly oriented to personal
bloggers, but may have thousands, even millions of users who keep a life blog.
MSN Spaces, a social networking site that introduced a blog publishing platform in
December 2004, experienced the fastest grow in unique audience during the first half of
2005. According to Nielsen/NetRatings, MSN Spaces grew a 947% since January 2005
(311,000 unique visitors) to July 2005 (3,257,000 unique visitors). Blogger (with a 45%
increase) and Xanga (with a 45% increase) were, respectively, the third and fourth fastest
growing blog networks in the six first months of 2005 (ZDNet Research, August 15, 2005).
Considering the number of accounts (registered users), Blogger, LiveJournal and
Xanga are the leaders in the blog hosting market: “At the end of the first quarter of 2005,
each had between 6.6 and 8.2 million accounts” (Henning, 2005). MSN, with 4.5 million
accounts, is the identified by Henning as the “challenger.”
In their analysis of 357 English written blogs randomly selected from the results
offered by a blog tracking site (blo.gs), Herring et al. found that “the personal journal is the
most popular type in every demographic category” (2004, p. 5). The result is even more
eloquent considering that the blogs hosted by Xanga and LiveJournal were excluded from
In order to be confident about the prominence of personal diaries on the
blogosphere, it would be necessary to know how many of the blogs hosted by Blogger and
TypePad are life diaries. However, the sustained popularity of LifeJournal and Xanga, as
well as the exponential growth of MSN Spaces accounts, demonstrate at least the relevance
and vigor of personal blogs in today’s blogosphere.
Although the success of blogs as media watch-dogs during the US Presidential
Campaign of 2004 has focused the attention on their role on politics and journalism, most
weblogs don’t treat public affairs. A great deal of the blogosphere is formed by accounts of
personal lives written by high school and college students that are not intended to impact on
large audiences, but to be shared with friends or random visitors.
Most bloggers are young and female. According to Henning (2005), who conducted
a survey of 10,000 randomly blogs selected among the twenty leading blog-hosting
services, 94.3 percent of bloggers are under 30 years old. More than half (58.3 percent) are,
in fact, teenagers. Female bloggers outnumber males in a proportion of 2 to 1 (68.1 percent
versus 31.9 percent). The portrait offered by Henning may be the one that gets closer to
reality, considering the size of his sample.
A survey by the Pew Internet Project revealed that, by the end of 2004, 8 million
adults in the United States had created a blog (Rainie, 2005). The number of bloggers
would have been considerably higher if the survey had asked teenagers. Three quarters of
LiveJournal users, one of the leading blog hosts, are between 16 and 24 years old (Kumar,
2005). It is interesting to note that personal bloggers may be younger than non-personal
bloggers. LiveJournal and TypePad, two services offered by the company Six Apart,
exemplify this dichotomy. LiveJournal users are 70 percent female and 70 percent under 21
years. TypePad users are evenly split between both sexes, and the average age is set in the
early 30s (Perez, May 27, 2005).
The blog readership triplicates the blog production. In response to the referred Pew
survey, 27% of Internet users in the U.S. said they read blogs. That translates into 32
million Americans who read blogs regularly. It is interesting to observe, however, that more
than half Internet users don’t know what a blog is (Rainie, 2005).
There seems to be a blogging divide in which age and technical competence may be
two of the most determining factors. As noted above, a Gallup Survey released on March
2005 showed that 44 percent of Internet users aged 18 to 29 read blogs. The readership
among younger teenagers may be even higher. ComScore Networks (2005) concluded in its
study that blog readers are more likely to live in wealthier households, be younger and
enjoy a high-speed connection to the Web than the average Internet user.
Blogging is a business of millions who spread messages to only a few. Rather than
many-to-many, the blogging communication paradigm should be renamed many-to-few. It’s
true that there are several successful blogs that attract thousands of readers daily, but the
304,000 unique visitors that one of the most widely read weblogs, Gawker,
(www.gawker.com) attracted in April 2005, pale in comparison to the 29.8 million visitors
of the New York Times online version during that same month (Bialik, May 26, 2005).
The preliminary results of an online survey conducted by Cameron A. Marlow, a
graduate student from the Media Lab at the Massachusetts Institute of Technology (MIT),
show that 46 percent of the 3,200 bloggers who had responded to his questionnaire by June
2005 said they receive 25 or fewer visitors on a typical day. 22 percent said they attract
between 26 and 100 readers on any given day, and less than 1 percent reported they have
250 to 1,000 visitors per day (Chronicle of Higher Education, June 23, 2005).
In sum, most of the blogging activity occurs below the radar. Most bloggers are not
media watch-dogs who report on professional journalists’ malpractices, but teenage girls
(and, by extension, high school and college students of both sexes) who exchange their life
experiences through their online diaries.
2.2. Research on personal blogs
2.2.1. Personal blogs and their predecessors
The feature that distinguishes a blog from any other type of website is the reversed
chronological order of its text entries or posts, with the newest of them always at the top.
Personal journal blogs, also known as “lifelogs” (Van Dijck, 2003), are only one of the
several categories of blogs identified by the literature.
In our research we have adopted the taxonomy proposed by Herring et al. (2005),
which is itself a modified version of Blood’s classification (2002). Herring et al.
categorized five types of blogs according to their purpose:
• Filter blogs. The author comments on “external, typically public events” (Herring et
al. 2005, p. 7).
• Personal journals. The blogger reports about events in his/her life and shares
his/her state of mind and intimate reflections.
• K-logs. The author focuses around a topic, project or product, usually -but not
necessarily- related to technology.
• Mixed. The functions of two or more of the first three types are combined. The
blogger may mix personal accounts with comments on public affairs and
discussions about a single topic.
• Other. The blog may be used as a vehicle for artistic expression (e.g. author’s
poetry drafts), as an archive for class notes or song lyrics, and as a conversation
board, among other uses. (Herring et al., 2005, p. 26, note 10).
Herring et al. (2004) contend that the filter blog type has been erroneously identified
by the media, the academia and the elite bloggers (those who belong to the so-called A-list
of blogs) as the epitome of the genre. Personal journals, which they found to be the most
popular type of blogs, have been overlooked. Two characteristics of lifelogs have
contributed to their underestimation: their authorship ––mostly are written by women, who
tend to be less socially valued than men–– and their content ––personal issues are deemed
as less serious than other kinds of accounts, such as technical or public affairs–– (Herring et
al. 2004, p. 12).
Filter blogs are characterized by their external connectedness. They consist mainly
on lists of links to external resources that are commented by the blogger. However, this is
not the defining feature of most blogs. In their content analysis of 203 randomly selected
blogs, Herring et al. discovered that, contrary to what is commonly thought, blogs do not
include many links (fewer than 1/3 of entries contained links), are not very interactive (the
mean for comments per entry was 0.3) and are not oriented to external events (only 36.1
percent linked to news sites). Therefore, in Herring et al.’s opinion, the filter blog created
by software developer David Winer in 1996 should not be credited as the clearest
predecessor of today’s blogs. Online journals, in which since the mid 90s women and
teenage girls recorded their life experiences, may constitute better candidates to be the
parents of contemporary blogs, specially the personal ones.
The killer-application that allowed for a massive adoption of blogs, a free browse-
based software that enables users to edit blogs without any knowledge about html language,
was the defining feature offered by BlogSpot, LiveJournal and Xanga, all launched in 1999,
and today’s leaders in the blog-hosting market.
[Chart 1. Blog Population Growth here]
2.2.2. Life blogs and personal (paper-based) diaries
The public availability of personal blogs is an apparent contradiction with the
intimate and private condition of their paper-based predecessors. But traditional diaries may
not be as private as commonly thought. After a review of the basic research on the history
of personal diaries, Van Dijck (2003) and Serfati (2004) agree that the myth of privacy as
one of constitutive features of traditional journals needs to be debunked. In the 19th Century
friends would exchange their personal diaries, and those young girls who used to keep
spiritual journals had to show them to their parents or guardians (Serfati, 2004, p. 84). Van
Dijck (2003) contends that privacy is more an effect than an intrinsic feature of personal
diaries. The presence of an addressee (either imaginary or real) is a sign of their
communicative nature. The diary is always written for someone else. Even when nobody
else than the author reads it, the author who becomes a reader of his/her literary production
years later becomes a different reader (at least, he/she is older) from the original author.
Van Dijck (2003) also combats two more popular attributes of paper diaries: their
supposed uniformity and single-authorship. In the light of history, paper diaries should not
be identified solely with personal and intimate confessions; some diaries were records of
journeys or notebooks for future projects, and some were meant to be publicly used. They
shouldn’t be identified with individual authorship either. Van Dijck cites religious
congregations and explorers as writers of collective diaries that were used as “communal
means of expressing and remembering” (2003, p. 3). “The genre has been dialogic rather
than monologic”, writes the author (Van Dijck, 2003, p. 3).
Serfati ascribes the popularity of blogging in America to the philosophical tradition
of Transcendentalism, which encouraged each individual to find a personal relation with
the Universe and God. One of the best ways to achieve this goal was through writing about
one’s everyday life. Ralph Waldo Emerson, the leader of this philosophical movement,
used to keep a journal himself (Serfati, 2004, pp. 46-47).
2.2.3. The community dimension of personal blogging
Paper diaries in the past could have a communitarian dimension or not. But personal
blogging today is almost always an activity that relies on the inputs provided by a
community of readers. As Van Dijck observes, “blogging, besides being an act of self-
disclosure, is also a ritual of exchange: bloggers expect to be signalled and perhaps to be
responded to” (2003, p. 10).
This may be the reason why LiveJournal and Xanga, two of the most signified hosts
of online personal diaries, offer a community-based blogging. Authors may become
members of groups interested in a particular subject (e.g. poker) or self-identified for their
belonging to a place (e.g. Chicago) or an institution (e.g. University of Illinois). Bloggers
can subscribe to their favorite blogs within the same host and post and receive comments
from other registered users. The successful strategic shift of MSN Spaces, originally a
social networking site that saw its number of accounts grow after incorporating a blogging
feature, is another sign of the relevance of community to personal blogging.
Serfati sets the “co-production” as one of the structural features of personal blogs:
“Where traditional diaries were written for an implied, ideal reader, online diaries explicitly
search for an audience and in so doing, turn themselves into a collaborative project”
(Serfati, 2004, p. 39). “Diary writers”, continues this author, “create themselves as the
central characters in a fictional theatre populated by a large supporting cast of minor
characters and of readers” (Serfati, 2004, p. 40). To Scheidt (2004), who studied the
implied audience of adolescent weblogs, “it is clear that weblogs create loci of audience
and author interaction” (p. 20).
That interaction with readers, Serfati argues, is crucial to the construction of the
blogger’s identity: “Each response to an entrance gives diarists confirmation of their own
existence” (p. 40). With whom does this interaction occur? Mostly with friends, but
probably also with strangers. In her ethnographic about teenage girl blogs, Bortree (2005)
found that teen bloggers write for their acquaintances, but also “appeared to be aware to
some degree that others, beyond their close circle of friends, may be reading their posts” (p.
32). They imagine this larger audience is composed by other teenage readers. “The teens”,
writes Bortree, “did not always appear to be aware that this mass audience might include
parents, teachers and others outside the world of teenagers” (p. 32).
In any case, the imagined audience by teen bloggers is dual (their friends and other
unknown teens), so this leads them to keep a continuously re-negotiated balance between
the intimate confessions shared with their pals and the exaggerated claims aimed to impress
the broader teenage audience (Bortree, 2005).
2.2.4. Who writes personal blogs and why?
Determining the demographics of bloggers is not an easy task. Content analyses as
the practiced by Herring et al. (2004 and 2005) face the difficulty of gathering a true
representative sample of the blogosphere. Blog search engines, the devices used to get the
samples, are biased in several ways: they don’t search among the whole blogosphere, and
their searching algorithm may prioritize a certain kind of blogs over others. Online surveys
conducted among samples recruited through e-mails and posts on blogs and other websites
(Viegas, 2004; Carl, 2003; Maxwell, 2005) are limited by their use of nonprobability
samples. Telephone surveys among adults, such as the ones conducted by the Pew Internet
Project (Rainie, 2005) or Gallup (Saad, March 11, 2005), ignore teenagers, the
demographic group that has embraced blogging with more enthusiasm.
If we the data provided by Henning (2005) on his survey of 10,000 blogs on twenty
leading blog-hosting services as the closest to the reality of the blogosphere, it can be said
that blogging is essentially a female and young phenomenon (more than two thirds of
bloggers are females, and teens comprise over half of the total population). This data ends
with the relative gender parity of blogging showed by previous studies.
[Table 1. Blog Creator Demographics here]
The current female dominance in blogging creation reverses the former male
dominance in personal websites. A study by Dominick (1999) revealed that 87 percent of
the sample of 317 personal websites of his study had been created by males. Interestingly,
this analysis advanced the inclination of females to include more personal information than
males, a trend that was later confirmed with blogs, as the data by Herring et al. (2004)
showed. Yet the act of blogging may render more similarities than differences between the
two sexes. In his study of teenage blogs, Huffacker (2004) encountered a high disclosure of
personal information among males and females. Males were more likely to reveal their
homosexuality and their location than females, although it turned out that girls linked more
to personal pages.
Carl (2003) compared the demographics of the respondents to her online survey of
blog authors with the demographics of American Internet users. She found that bloggers are
younger than the average Internet user. Latinos and African-Americans happened to be
infra-represented in the blogging community.
In her ambitious Senior Honor’s Thesis, Maxwell (2005) studied the uses and
gratifications of online diarists. In a response to an online survey, blog authors said they
saw their diaries as a tool that enabled them to keep in touch with others. They used their
blogs to vent emotions and archive self-experiences. Maxwell concluded that online diaries
were “a management tool for erasing the tension between connecting with others and
expressing self” (2005, p. 63).
2.3. Research questions
Our objective in this research was to describe systematically the content of the blogs
authored by UIC students at Xanga. The following research questions guided the collection
of data and helped design the coding protocol used in the content analysis:
• What are the demographic characteristics of bloggers (gender, age, race)?
• How do bloggers identify and introduce themselves? Do they use their true full
names or do they use nicknames? Do they upload personal photos?
• What kind of resources do they use to communicate? Do they use text, pictures,
videos, music, and emoticons?
• What are the students’ patterns of use of blogs? Are the weblogs updated daily or
weekly? Is there any kind of feedback in the form of comments?
• What are the topics bloggers write about? E.g. school, work, romance, family,
• What is the breadth of their social network? How many blogs they link to? To how
many blogrings (virtual communities within Xanga) do they belong?
The method employed to answer the research questions is one of the most
traditional tools in communication research: content analysis. We adapted the coding book
used by Huffaker (2004) in his thesis about gender similarities and differences among
teenage bloggers, tailoring it to meet Xanga’s particular features.
The coding protocol took into account the following categories: the name used by
the author, the type of title of the blog (e.g. a phrase, a realistic name), the author’s gender,
age and race, the longevity of the blog, the use of photos, music, videos and emoticons, the
frequency with which the blog was updated, the number of comments it received, and the
topics discussed by the author (school, work, economy, romance, friends, family, travel,
sports, nightlife, religion, politics, sex, celebrities, music, movies, books, games).
The population of our study was composed by the 648 blogs that belong to the “UIC
- University of Illinois at Chicago) “blogring” at Xanga 2 at the time we accessed the
website to perform our study (in April 2005). In order to work with a 95 percent confidence
interval, we randomly sampled 241 blogs. We printed a list of the entire user names listed
in that blogring and consulted a table of random numbers to determine which blogs to pick,
as if our list was a phone book and each username was a phone number. Once the
usernames that comprised the sample were known, we visited each of them on April 14,
2005, and saved a copy of the posts that appeared on the main page in PDF format, so that
we could have a fixed photography to perform the coding. This blog captions were our
units of analysis. Due to unforeseen technical problems, four blogs could not be coded, so
the final sample was reduced to 237 diaries. A graduate student was trained to share the
coding task with the author. The inter-coder reliability coefficient, calculated using Holsti’s
formula, was 91 3 .
4.1. Online identification
UIC bloggers reveal a high amount of personal information. This disclosure does
not happen at the first level (most bloggers don’t use their real names on their usernames or
on the title of their blogs), but at a second level, in the profile page included in each blog,
where a 77% reveal their first name and a 47% discover their full name. In the first level of
identification, the username, seven in ten bloggers choose an online handle (e.g. Toxic
Avenger), although a significant minority (22%) used their real name to build a nickname
as their username (e.g. Roger87).
There are two basic ways in which UIC bloggers put a title to their blogs. They may
keep the same name they used for their username (that is, an online handle, 39.8%), or they
may use a phrase or adagio such as “The early bird catches the worm” in 42.1% of the
The author would like to express his enduring gratitute to Evan P. Venie, graduate student at UIC’s
Department of Communication, who not only shared with us the time-consuming task of coding hundreds of
blogs, but also provided good suggestions that helped improve the definitive coding book.
Xanga offers users the chance to indicate their country. Only half of the members of
the UIC blogring (51.7%) decided to show it, and most of them identified as Americans
(94.2%). The belonging to a blogring that receives its name from a university with a known
placement (Chicago) makes the location category less relevant. That may explain why only
half of bloggers made explicit their place of residence (54.4%). Of those who did it, most
are, understandably, from Illinois (93.4%).
Seven in ten UIC blog authors used a personal photography to identify themselves.
This image was a self-potrait in most cases (74.2%), although some preferred to appear
with a friend or romantic partner (13.5%) or with several friends in a collective photo
(9.2%). One in ten chose not to use any image at all, and two in ten opted to present
themselves with avatars such as popular cartoons, images of objects or animals.
The two sexes were almost equally represented in the sample: 48.4% of blog authors
were males, 51.1% females. Three in four users disclosed their birth date. Half of the
sample was born between the years 1984 and 1985, what tells us that, on average, users are
21 years old (M=21, SD=1.06). Almost eight in ten UIC bloggers registered on Xanga
during the years 2003 and 2004.
The coders were able to identify the race of 70% of the blog authors that comprise
the sample 4 . UIC bloggers are mostly Asian (70.7%), being whites (18.1%) the second
group. Surprisingly, African Americans are not more numerous (1.6%) than American
The identification of race was performed looking at the bloggers’ personal photos. In some cases it was
useful to look at the blogrings the author belonged to (e.g. Proud Philipinos). The categories employed to
clasify race and ethnicity were the same used by the U.S. Census Bureau in its 2000 census.
Indians or Alaska Natives (1.1%). Hispanics (of whatever race) can be considered as the
third group in importance, comprising a 6.5% of the sample.
4.3. Interactivity and community features
Almost 70% of the bloggers who belong to the UIC blogring show a way of contact
with them. IM (Instant Messaging) is the first mean of contact, with half of users showing it
(55.3%). A 36.3% of users give an e-mail and a 28.3% a website. Of those who link to a
website, 32.4 do it to their own blog at Xanga. One in four link to a true homepage, while a
35.3 link to “other” sites, mostly to MySpace.
It is interesting to note that, although Xanga is a community-oriented blog host, only
half of the bloggers (52.7%) link to other blogs. Among those who do it, the average
number of blogs linked is 23.49.
On average, UIC Xanga bloggers belong to 3 blogrings (communities) besides their
UIC blogring. Only a 9% of our sample of bloggers had UIC as their unique blogging
community 5 .
4.4. Blog appearance
Although the blogring we studied (UIC - University of Illinois at Chicago) is not the only one that bears the
name of the university, the rest of UIC blogrings have a more reduced membership: UIC Sucks has 11 users,
UIC Crew has 9 bloggers, and UIC Bowling Crew or Music Crew gives shelter to 5 writers. Only a 13% of
the members of the UIC main blogring belong to any of these other UIC communities.
Three in four users don’t use any particular image on the background. Of those who
do use an image on the background (25%), environment images (28.8%) and cartoons
(21.2%) are the most popular.
The use of music or video clips on the background is marginal. Only a 5% of the
UIC bloggers welcome their readers with a song, and fewer with a video clip (1.7%).
Xanga allows their users to include lists songs and video clips to be played, but the
percentage of users who do that is anecdotic: 1.7% for music, 2.5 for videos.
Only a 40% of users posts images on their blogs. So weblogs are still essentially
textual. However, we noticed that among those who post images there are some heavy users
who upload dozens of pictures per post. This is possible thanks to the space offered by
some image hosting services like Photobucket 6 , which was found to be one the preferred
among UIC bloggers.
The most popular types of images were: self-images (24.5%), images of/with
friends (22.4%), commodities (13%), images of/with romantic partner (9%) and family
images (7%). By “commodities” we understand those objects of desire (usually commercial
ones) for which the blogger expresses a high regard. The most pictured commodities were:
automobiles (32%), apparel/clothes (27%), food and drinks (17%), and electronics (15%).
[Chart 2. Most Popular Images in Posts here]
[Chart 3. Most Pictured Commodities]
The sample was almost evenly split over the use of emoticons: 48.1% did, 51.9%
did not. Among those who used them, the emoticon average per post was 0.7.
4.5. Frequency of blogging and blog abandonment
Most users (76.8%) post five posts per page. There is no compulsive blogging. Only
8 percent of users post more than one post per day. 60% of users post more than once in a
week, and some do it everyday. However, it is interesting to note that a 40% wait more than
one week to post on their blog.
Over two in ten of the blogs (22%) had been abandoned at the time of the sample
selection (April, 2005). As Huffacker (2004) did, we considered a blog abandoned when
there were no posts dated within the three previous months.
Most bloggers have comments in their posts (92.7%). The average of comments per
post is 3.4 (M=3.4, SD=4.5). Most users use “e-props” too (93.1%). The average of e-props
per post is 6.1 (M=6.1., SD=8.1). “E-props” are virtual marks with which the commentator
congratulates the writer for his/her post. The users who comment have the chance to reward
the writer with two, one or none. Since the default option is two e-props, it is no surprising
that the number of e-props doubles the number of comments.
It is interesting to note, however, that 12% of the blogs received no comments or e-
props at all. The frequency of comments and e-props follows a skewed curve to the right.
That is, the more comments and e-props the user has, the more extraordinary he/she is.
Most bloggers have few comments and e-props.
[Chart 4. Average Number of Comments per Post here]
4.7. Most popular topics
The friendships formed at college and the hardships of student life are issues that are
present in six of every ten blogs. The nightlife experiences were the third most recurrent
topic (39% of the sample). Romantic and family relations were talked about in 37 and 33
percent of the blogs, respectively. The expression of music preferences and the reports
about part-time jobs were found in one in four diaries. Accounts related to travel, movies
and commodities appear in 20% of the blogs, followed closely by commentaries about
sports (17%). Other topics present in the diaries were: discussions about celebrities (12%),
reflections on religion (11%), and commentaries about books and games (9% each).
Remarks about politics (6%) or explicit references to sexual matters (6%) were not very
common, and the same can be said about reports of personal economic problems (5%).
[Chart 5. Most popular topics here]
4.8. Authorship and blog type
The majority of the blogs hosted by Xanga in its UIC blogroll are individually
authored (97.9). According to the taxonomy proposed by Herring et al. (2004), most of the
blogs analyzed in this study fall within the category of personal diaries (94%). A 3% are of
mixed purpose (a mix of diaries and k-logs, such as the car collector who talks about his
life and his desired cars). K-logs represent a 2% of the sample. They are, in essence, virtual
boards in which the blogger informs about the activities of the organization he/she belongs
4.9. Gender differences
The sample of blogs turned out to be almost evenly divided by genre. Of the 220
bloggers whose sex were identified by the coders, 48.6% (107) were males, and 52.4%
(113) were females. A chi-square test was run to assess if the differences between the
blogging activities of the two sexes were statistically significant. Two basic divergences
were found. First, males tend to disclose more personal information than females. Males are
more likely to reveal their full name [X² (1, N=108) = 6.53, p<.008], their country [X² (1,
N=119) = 11.33, p<.001] and their location [X² (1, N=126) = 4.42, p<.024]. Second,
females write more about emotions, and they do it in a more emotional way than males.
They were more prone to use emoticons (in a proportion of 3 to 1) than males [X² (1,
N=109) = 18.66, p<.000], and they are significantly more inclined to discuss romantic
relationships than their masculine counterparts [X² (1, N=84) = 4.75, p<.020].
Although religion was a marginal topic of discussion, the test revealed that males
were more likely to talk about faith than females [X² (1, N=25) = 6.16, p<.011].
[Table 2. Contingency Table for Gender Differences among Bloggers here]
5.1. Interpretation of results
Personal journals are, indeed, personal. Visiting the online diary of someone
unknown produces some kind of voyeuristic sensation. One can see family pictures, photos
of boyfriends/girlfriends, and images of students having fun in private or public parties. As
some authors have noted before (Chandler, 1998), blogs ––and, by extension, personal
websites and social networking sites— resemble the teenagers’ bedrooms. Blogs are the
online version of the walls papered with posters of popular idols and the cork-boards full of
pictures with friends (Williams, August 28, 2005).
College students may be not so different from teenagers in their revelation of
personal information. Our sample of UIC bloggers follows the tendency Huffacker (2004)
found on teen blog authors: they tend to disclose a great amount of private details about
themselves. Seven in ten UIC blog writers indicated their first name, and half of them their
full name. In some cases the kind of personal details exposed was striking: we saw bloggers
who posted their class schedules, even their exam qualifications. Sometimes, by the very
fact of documenting their life, they allow any reader to know their daily routines, such as
the train routes between their homes and the campus. This candid disclosure of personal
identifying information has been a matter of social concern, because teens and young adults
unconsciously expose themselves to sexual criminals and cyber stalkers. A recent news
story qualified teen personal blogs as “a child predator’s dream” (Loesch, July 25, 2005).
Huffacker (2004) suggested that parents and teachers, but also the blog-hosts themselves,
should encourage awareness about the potential dangers of displaying so much identifying
information online. It is still to be seen if these warnings make blog authors conceal their
identities, or at least not to be so explicit in their daily life accounts. In our study we found
that females were less likely to reveal their full name and location. This attitude may
respond to their awareness of those dangers.
Our sample of UIC bloggers showed parity in gender, but a great dominance of
Asian students, who authored 70 percent of the journals. Hat Nim Choi (2003), a
Psychology student from Haverford University who compared the demographics of Xanga
and Livejournal, attributes the Asian dominance to the way Xanga membership is
promoted. Seeing themselves as a minority, those Asian-Americans who are at Xanga
invite their friends to join the network, so they can interact with other people who share
their same cultural identity (Choi, 2003). It sounds like a plausible explanation, but should
be considered as a hypothesis to be tested in further studies.
The UIC blogs were found to be essentially textual: 60 percent contained no images
in their posts. However, the increasing popularity of digital cameras and image-hosting
services may change this tendency in the future.
The consideration of blogs as frequently updated websites may have to be re-
thought. In our sample we found that four in ten bloggers posted less than once per week.
For a significant portion of bloggers, posting is not a daily, or even weekly activity. But this
does not necessarily mean they have abandoned their journal. Opening a blog is the easiest
way of having a website. These users might only be looking for a space in the web, and not
a daily journal. It should be noted that a minority of our sampled bloggers introduced a link
to “my website” that in 32.4 of the cases led to the Xanga site itself.
The revelation of personal information is not the unique characteristic that college
students share with teenagers. Instant Messaging (IM) was the most frequent medium of
contact offered by UIC bloggers, comprising a 55.3% of those who offered any contact. It
turned out to be more popular than e-mail (36.3%), so college students may also deserve to
be named the “IM generation”, just as a Pew Internet study qualified American online
teenagers (Lenhart et al., 2001).
5.2. Limitations and suggestions for future research
The generalizability of this study, circumscribed to a single virtual community of
college bloggers, is one the limitations of this research, that cannot claim (and did not
intend to do so) that all college blogs are like the ones authored by the members of the UIC
Xanga blogring. The coding book employed in this study, an adaptation of the original
created by Huffacker (2004), could be improved by introducing a more precise way to
record the updating frequency of the diaries. Measuring the interval of update between
entries, as Herring et al. did (2004), could be a solution. During the coding work we noticed
that some bloggers would include reflections about their physical appearance (e.g. their
weight) and observations about heir frame of mind and mood. A future content analysis of
personal blogs should code for the presence or absence of these health and psychological
In this research we have found that college student blogs resemble in many ways the
diaries authored by teenagers. But we also have found a contradiction that could serve as
the starting point of future studies: although Xanga is a community-based blog-host, only
half of the bloggers in our research linked to other blogs within Xanga. In the light of these
results, we advance the following idea, whose plausibility may be tested in a further study:
there seems to be two ways of blogging, a stand-alone blogging, in which the author seeks
self-expression without becoming part of a broader community, and a networked blogging,
in which the activity of documenting one’s life is inseparable of the activity of connecting
and interacting with friends. The apparent success of the newly redesigned MySpace, a
social networking site that added blogging as one of its services, could represent some
evidence of this latter trend.
Bialik, C. (May 26, 2005). Measuring the impact of blogs requires more than counting. The
Wall Street Journal.
Blood, R. (2002). Weblogs: A history and perspective. In Blood, R., et al., We've Got Blog:
How Weblogs Are Changing Our Culture, Cambridge, MA: Perseus Publishing, pp. 7-16.
Bortree, D.S. (2005). Presentation of self on the Web: An ethnographic study of teenage
girls’ weblogs. Education, Communication & Information, 5 (1), 25-39.
Carl, C.R. (2003). Bloggers and their blogs: A depiction of the users and usage of
weblogs on the World Wide Web. Master’s Thesis. Georgetown University. Washington,
DC. Retrieved March 7, 2005, from http://cct.georgetown.edu/thesis/ChristineCarl.pdf
ComScore Networks (2005). Behaviors of the blogosphere: Understanding the size,
composition and activities of weblog readers.
Chandler, D. (1998). Personal homepages and the construction of identities on the Web.
Choi, H. N. (2003). LiveJournal and Xanga Survey.
Chronicle of Higher Education (June 23, 2005). So many bloggers, so few readers. The
Chronicle: Wired Campus Weblog.
Van Dijck, J. (2004). Composing the self: Of diaries and lifelogs. Fiberculture Journal 3.
Dominick, J. R. (1999). Who do you think you are? Personal webs and self-presentation in
the World Wide Web. Journalism & Mass Communication Quarterly 76 (4), 646-658.
Greenspan, R. (June 23, 2003). Blogging by the numbers. CyberAtlas.
Gurak, L., Antonijevic, S., Johnson, L., Ratliff, C., Reyman, J. (2004). Introduction:
Weblogs, rhetoric, community, and culture. In: L. Gurak, S. Antonijevic, L. Johnson, C.
Ratliff, & J. Reyman (Eds.), Into the blogosphere: Rhetoric, community, and culture of
Henning, J. (2003). The blogging iceberg - Of 4.12 million hosted weblogs, most little seen,
quickly abandoned. Perseus Development Corporation White Papers.
Henning, J. (2005). The blogging geyser - 31.6 million hosted blogs, growing to 53.4
million by year end. Perseus Development Corporation White Papers.
Herring, S.C., Kouper, I., Scheidt, L. A., Wright, E. L. (2004). Women and children last:
The discursive construction of weblogs. In: L. Gurak, S. Antonijevic, L. Johnson, C. In
Ratliff, & J. Reyman (Eds.), Into the blogosphere: Rhetoric, community, and culture of
Herring, S.C., Scheidt, L.A., Bonus, S., and Writght, E. (2005). Weblogs as a bridging
Huffaker, D. A. (2004). Gender similarities and differences in online identity and language
use among teenage bloggers. Master’s Thesis. Georgetown University. Washington, DC.
Jones, S. G. (2002). The internet goes to college: How students are living in the future with
today's technology. Pew Internet & American Life Project.
Kaye, B.K. (2005). It’s a blog, blog world: Users and uses of weblogs. Atlantic Journal of
Communication, 13 (2), 73-95.
Kumar, R., Novak, J., Raghavan, P., & Tomkins, A. (2004). Structure and evolution of
blogspace. Communications of the ACM, 47(12), 35-39.
Lenhart, A., Rainie, L., Lewis, O. (2001). Teenage life online: The rise of the instant-
message generation and the Internet's impact on friendships and family relationships. Pew
Internet & American Life Project. http://www.pewinternet.org/pdfs/PIP_Teens_Report.pdf
Lenhart, A., Horrigan, J., Fallows, D. (2004). Content creation online. Pew Internet &
American Life Project. http://www.pewinternet.org/pdfs/PIP_Content_Creation_Report.pdf
Lenhart, A., Madden, M., Hitlin, P. (2005). Teens and technology. Pew Internet &
American Life Project.
Loesch, G. (July 25, 2005). Warnings issued for popular website: Kids, teens reveal
personal information. WPXI News. http://www.wpxi.com/news/4754969/detail.html
Maxwell, R.L. (2005). Online lives? - Personal diaries on the web. Senior Honors Thesis.
The Ohio State University. http://hdl.handle.net/1811/412
Perez, J.C. (May 27, 2005). The battle for the blogosphere. The Industry Standard.
Rainie, L. (2005). The state of blogging (Data Memo). Pew Internet & American Life
Saad, Lydia (March 11, 2005). Blogs not yet in the media big leagues. Gallup Poll News
Scheidt, L.A. (2004). Adolescent diary weblogs and the unseen audience. Paper presented
at the Digital Generations Conference, London, July, 2004.
Serfaty, Viviane (2004). The mirror and the veil: An overview of American online diaries
and blogs. Amsterdam/New York: Rodopi.
Spooner, T. (2001). Asian-Americans and the Internet: The young and the connected. Pew
Internet and American Life Project. Retrieved July 9, 2005, from
Viégas, F. (2004). Blog survey: Expectations of privacy and accountability.
Williams, A. (August 28, 2005). Do you MySpace? The New York Times.
ZDNet Research (August 15, 2005). Fastest-growing blogs: MSN Spaces, Fark, Blogger,
Xanga. IT Facts/ZDNet. http://blogs.zdnet.com/ITFacts/?p=8692
Chart 1. Blog Population Growth
Blog Population Growth
Number of Blogs
Years per Quarter
Henning, J. (2005). The blogging geyser - 31.6 million hosted blogs, growing to 53.4 million by year end.
Perseus Development Corporation White Papers. http://www.perseus.com/blogsurvey/geyser.html
Table 1. Blog Creator Demographics
Study Males Females Dominant age group
Henning, 2003 44% 56% 52.8% under 20 years
Carl, 2003 47% 53% 80% between 18 and 34 years
Herring et al, 2004 52% 48% 39% under 20 years
Viegas, 2004 63% 36% 46.3% between 21 and 30 years
Pew Internet Project, 2005 57% 43% 48% under 30 years
Henning, 2005 31.9% 68.1% 58.3% under 20 years
Maxwell, 2005 14% 86% 64.6% between 19 and 22 years
Chart 2. Most Popular Images in Posts
Most Popular Images
Percentage of Blogs
Self-images Images of/with friends Commodities Images of/with romantic Family images
Chart 3. Most Pictured Commodities
Most Pictured Commodities
Percentage of Total Pictured Commodities
Automobiles Apparel/clothes Food and drinks Electronics Other
Chart 4. Average Number of Comments per Post
Mean = 3,4132
Std. Dev. = 4,54027
0 N = 237
0,00 10,00 20,00 30,00 40,00
Percentage of Blogs
Chart 5. Most Popular Topics
Most Common Topics in Blogs
Table 2. Contingency Table for Gender Differences among Bloggers
CATEGORY TOTAL MALE FEMALE X² df p
(N=220) (N=107) (N=113)
Author reveals full name 49.1% (108) 28.2% (62) 20.9% (46) 6.53 1 p<.008
Author shows country 54.3%(119) 32% (70) 22.4% (49) 11.33 1 P<.001
Author shows location 57.3% (126) 31.4% (69) 25.9% (57) 4.42 1 P<.024
Author uses emoticons 49.5% (109) 16.8% (37) 32.7% (72) 18.66 1 p<.0001
Author writes about romance 38.2% (84) 15% (33) 23.2% (51) 4.75 1 p<.020
Author writes about religion 11.4% (25) 8.2% (18) 3.2% (7) 6.16 1 p<.011