Document13/12/2016 8:42 PM
Dealing with Newcomers
Robert Kraut, Moira Burke& John Riedl
The challenges of dealing with newcomers .................................................................... 1
Problem 1: Recruiting newcomers .............................................................................. 3
Problem 2: Selecting the right newcomers .............................................................. 14
Problem 3: Keeping newcomers around ................................................................... 19
Problem 4: Teaching the newcomers the ropes ........................................................ 23
Problem 5: Protecting the group from newcomers ................................................... 28
References ..................................................................................................................... 33
The challenges of dealing with newcomers
In the face of inevitable turnover, every online community must incorporate successive
generations of newcomers to survive. Without replacing members who leave, a
community will eventually wither away. Newcomers can also be a source of innovation,
new ideas and work procedures or other resources that the group needs. However,
attracting newcomers and incorporating them into an existing community can be a
difficult endeavor. Newcomers have not yet developed the commitment to the group felt
by old-timers. As a result, they are very sensitive to the public image a community has
and to their own early experiences in it. As a result, they may not join or are likely to
leave in the face of even minor adversity. They have less motivation to be helpful to the
group or to display good organizational citizenship characteristic of many old-timers
(Organ & Ryan, 1995). In addition, for reasons of either ignorance or maliciousness, they
may behave in ways that can be harmful to the group. They do not yet know the norms
guiding behavior in the group and in their ignorance, may perform actions that offend
other group members or otherwise undercut the smooth functioning of the group. For
example, when participating in Wikipedia, the open-source encyclopedia, new editors
may fail to follow the policy of writing with a neutral point of view, or they may add
content that has already been determined by a consensus of more experienced editors to
belong in another article. Because they lack experience, when newcomers try to
participate, they imperil the work that other community members have already
performed. For example, they may introduce bugs in an open-source development
project, cause the (virtual) death of fellow guild members in an online role-playing game,
or ask redundant questions in discussion groups. Finally, their mere presence increases
diversity in the group and may in itself be off-putting to more experienced members of
the community, who prefer the people and routines they were familiar with.
When dealing with newcomers, online communities must solve five basic problems.
1. Recruitment: First, communities need to advertise to and recruit members, to
ensure a supply of newcomers for replenishment and growth.
2. Selection: Second, among those members who do show up, communities need to
decide which ones to select for membership so that the people who do join will
Page 1
Document13/12/2016 8:42 PM
ultimately be valuable to the community as a whole.
3. Retention: Third, both theory and experience suggest that newcomers’ ties to the
community are especially fragile. As a result, the community needs to engage in
tactics that keep potentially valuable newcomers around until they can develop
more permanent bonds to the community or learn how the group operates.
4. Socialization: Fourth, the group needs to socialize the newcomers, teaching them
how to behave in ways appropriate to the group. We discuss many techniques for
socializing members of an online community and encouraging them to behave
appropriately in Chapter N, on regulating behavior. In the current chapter we
focus on socialization strategies that are of particular relevance to newcomers.
5. Protection: Finally, throughout its interactions with prospective members, visitors
and newcomers in their early interactions, the community needs to protect itself
from the potentially damaging effects of the actions of those who either have little
knowledge of appropriate group behavior or little motivation to follow community
norms appropriately.
These problems reflect two perspectives— that of the newcomers and that of the online
community and its existing members. The recruiting process, for example, consists both
of the activities that potential members might perform in investigating different
communities and weighting alternatives and the activities that communities and their
members might perform in soliciting new recruits and interacting with them. In this
chapter we consider the perspective of the newcomer only in so far as it has implications
for how the online community needs to be designed to accommodate them. For example,
as we discuss below, newcomers are happier in a community and contribute more
effectively if they have a complete and accurate impression about the nature of the
community before they join it. To collect the information they need to form this
impression, they may lurk in the community, silent observing, or may try to participate,
to gauge the community’s reaction to them. This chapter doesn’t not address the moves
that newcomers should make in order to gain an accurate view of the community (e.g.,
see Burke, Kraut, & Joyce 200x), but concentrates on how the community should be
designed to provide the information newcomers need to make joining decision and to
respond to the common moves that newcomers are likely to make to form impressions.
For example, if communities publish FAQ (frequently asked question) pages to make
policies visible and allow outsiders to see archives of conversation among members,
enabling them to judge see the nature of the interactions, then prospective members
should gain a more accurate view of the community. However, some communities may
decide not to make these resources available to outsiders. For example, the managers of
cancer support groups hosted at the Association of Cancer Online Resources (acor.org)
believe that the privacy needs of current members outweigh the investigatory needs of
prospective members. Therefore, outsiders must register to become a member of a
support group before they can see any of the interactions that have occurred in them.
Researchers in both online and offline settings have identified an analogous set of stages
Page 2
Document13/12/2016 8:42 PM
that newcomers take on the path to becoming committed members of a group or
community. Levine and Moreland (1984), in discussing offline groups, use the terms
investigation, socialization and maintenance to describe the set of activities that
newcomers and groups engage in as they become increasingly committed to each other.
Individuals and groups go through an investigatory phase, in which the newcomers gather
information about the group to predict whether it will fit their needs, while the group uses
recruiting and selection processes to identify prospective members who would fit well
with the group. During the early stages of the socialization phase, after joining the
community, the central challenge for the community is to keep the newcomer around.
The relationship between the newcomer and community during the early socialization
phase is especially fragile, and even small problems may drive newcomers out.
However, as the newcomer becomes more committed to the community, helping
newcomers learn the norms of the community and how to behave becomes increasingly
important. In describing the progressive commitment of newcomers to online
communities, Preece and Shneiderman (2009) propose a ‘reader-to-leader’ funnel, in
which newcomers move from being readers, to contributors to collaborators and finally
leaders. Their transition from reader to contributor is analogous to Moreland and
Levine’s description of the transition of potential members from outsiders to
organizational members as they move through the investigatory to the socialization
phase.
Problem 1: Recruiting newcomers
In the face of turnover in their membership, online communities will inevitably die
without a constant supply of newcomers. Thus it is important to consider the processes by
which online communities advertise their existence and recruit newcomers. These
processes determine whether the community will have enough members to accomplish its
goals. In addition, the processes of recruitment may have direct consequences for later
problems that the community must solve, such as selection, retention and commitment.
Although there are many differences between an online community and a conventional
organization with employees, especially in terms of the formality of the recruiting and
acceptance process, we follow the tradition in research on employee recruiting and
consider recruiting to be “those organizational activities that (1) influence the number
and/or types of applicants who apply for a position and/or (2) affect whether a job offer is
accepted” (Breaugh, 1992, p. 4).
Many volunteer online communities do little active recruiting for new members. This
lack of attention to recruiting characterizes most Usenet groups and the open source
software development projects represented in SourceForge. This may be sufficient if
potential members’ random browsing of the web and word of mouth endorsements from
current members provides a sufficient supply of recruits to replace losses and meet the
community’s needs for growth. Active recruitment is likely to be necessary if a laissez
faire approach is insufficient. Some online communities explicitly recruit new members.
For example, nytimes.com provides links that allow readers to send by email an article
they read to other people or to share the article to the contacts on their social networking
sites. Many social networking sites, like Facebook and Linked-In, solicit new members
Page 3
Document13/12/2016 8:42 PM
by encouraging existing members to send invitations automatically to everyone on their
email lists. The social networking site tagged.com received thousands of complaints for
‘contact scraping,’ when, according to the company’s founder, a software glitch
accidentally sent invitations to people on members’ email list without the list owners’
permission or knowledge {Tugend, 2009 #4472}. Many smaller online groups, including
guilds in World of Warcraft and projects in Wikipedia, recruit new members informally
through their connections with existing members and more formally by posting openings
on relevant bulletin boards.
Design claim: Compared to laissez faire approaches, in which prospect members seek
out or stumble upon a community, active recruiting will lead to the community having
accesses to a larger pool of prospective members.
McGuire (1978) identified five components of persuasive communication: source,
message, channel, receiver and object of the persuasive attempt. Those attempting to
recruit members to an online community have moderate control over each of these
components. They can craft a message about the community that the public is likely to
see, both in terms of the information that is easily about the community on the
community’s homepage, on paid advertisements they place with search engine services or
social networking sites or information they post about the community themselves or
through surrogates on web forums, blogs and other online sites. Messages recruiting for
the community can come from highly credible sources, as when the eminent biologists E.
O. Wilson used his acceptance speech at the 2007 TED award to pitch participant in the
community creating the Encyclopedia of Life1or endorses it on the encyclopedia’s home
page (see Figure X), or from less credible sources. They can embed their message in text
messages, video on YouTube and related sites or use interpersonal word of mouth to get
their message across. They can target their message to receivers with particular
attributes or post it to general audiences.
Figure X: Credible endorsement for the Encyclopedia of life
at http://www.eol.org
Research on attitudes, attitude change and the influence of both interpersonal and mass
communication is a massive topic. Although we consider persuasion at several points in
this book, including our discussion of recruiting and socializing newcomers, encouraging
contributions and starting a community from scratch, a complete review is beyond the
scope of this book. Instead we sample some highlights in this book and refer the reader
to relevant reviews for more detail on persuasion in general (McGure, 1985; Childini &
Goldstein, 2004; Petty & Wegener, 1998).
1
http://www.ted.com/talks/e_o_wilson_on_saving_life_on_earth.html
Page 4
Document13/12/2016 8:42 PM
Recipients: Community managers can target the information they provide about the
community narrowly just to people are likely to be interested or more broadly.
Community mangers can advertise their site in public places, without much regard to the
people who are likely to attend, much as billboard advertisements indiscriminately target
anyone passing by. When World of Warcraft guilds post videos of raids to YouTube, for
example, they are following this strategy. Alternatively, they can target potential
members more directly, by advertising where potential members are especially likely to
visit, just a commercial advertisers place advertisements for products in the magazines
and TV shows that their target demographic groups are likely to view. Thus, for example,
guilds recruiting in World of Warcraft often make use of the recruiting forums especially
tailored to recruiting for Horde1 or Alliance2 players. Third, they can attempt to target
particular individuals who have already shown interest or competence in a domain
relevant to the community, just a charities use direct mail to target a potential
contributors who contributed in the past or contributed to a related cause or Google
targets web advertising based on a consumer’s search terms. Project leaders in Wikipedia
often use this technique by sending personalized invitations to non-members who have
previously edited articles within the project’s scope. Viral marketing or personalized
word of mouth, are an example of targeting, in which existing community members reach
out to potential members whom they identify as likely candidates, as when churches,
theaters, doctors and car mechanics all use word of mouth marketing from current patrons
to recruit new ones. Recently the car manufacture Hyundai has formalized this idea by
reposition their manufacturer’s rebate to customers as a payment for these new customers
to talk up their products to family, friends and neighbors. The research literature and
Google’s financial success strongly indicate that targeting advertisements are more
effective than indiscriminant ones
Design claim: Targeted appeals are more persuasive in influencing attitudes and
recruitment than are indiscriminant appeals, which do not target .
Modality: A long research tradition starting with Katz and Lazerfeld and Coleman
indicates that interpersonal appeals, in which the persuasion attempt comes from people
whom the target knows, are more effective at influencing attitudes and adoption than are
impersonal appeals from the mass media. Later in this chapter we will review relevant
research on viral marketing and interpersonal appeals. Interpersonal recruiting is more
effective than mass communication (Green et al, 2008).
One of the best sources of new members is people who already have part of their social
network in your community. By exploiting their prior relationships, the groups are more
likely to recruit new members who are knowledgeable about the community and fit with
its mission. These potential members will already have built-in mentors and a support
group should they join your community. New members recruiting via social networks
are more likely to stick around than are members recruited by other methods.
One of the most successful examples of this sort of viral marketing is Facebook “apps”,
1
2
http://forums.worldofwarcraft.com/board.html?forumId=7244844&sid=1
http://forums.worldofwarcraft.com/board.html?forumId=7244843&sid=1
Page 5
Document13/12/2016 8:42 PM
which often are designed so that with a single click a user can invite his friends to join
him in using the app. B.J. Fogg of Stanford taught a class in which students were
encouraged to build Facebook apps that used principles from psychology to draw in new
members virally. The result: in just 10 weeks the students had attracted an aggregate of
16 million users to their apps.
This sort of recruiting is most effective in environments in which bringing in a user’s
friends makes the community more valuable to him. For instance, one Facebook app lets
a user see the places his friends have traveled to, which if fun and engaging and valuable
if the user wants advice on travel in the future. The user is motivated to talk his friends
into using the app. Fogg and Eckles (2007) paper about “The Behavior Chain for
Online Participation” explains that these approaches leverage users’ credibility with their
friends in a way that is more powerful than any message directly from your community.
Design Claim: Encourage your users directly recruit their friends to join the community.
Even if it’s difficult to talk your users into recruiting their friends directly, make it easy
for them to indirectly increase their awareness of your community. For instance,
CostcoPhotoCenter is a free photo sharing web site that makes it easy for a user to share
photos with friends and relatives. To use CostcoPhotoCenter, the friends and relatives
have to create a (free) account – after which it is easy for them to share their own photos.
(Or to buy copies from Costco of the photos they see, but that’s a different motivation.)
Similarly, the New York Times makes it easy to share an article with friends via email.
The Times benefits both by the increased visibility along a persuasive social network, but
also by collecting the data: on their front page they list the “Top Emailed” articles of the
day. As with Costco and the Times, this sharing will be through channels outside of your
community – such as email – but may bring members into the community over time.
Design Claim: Make it easy for users to share content from your community site with
their friends, via easy email links, to increase the visibility of your community among
friends of your users.
Source. Malcom Gladwell in his book The Tipping Point described how marketers work
hard to get key influencers to use their products, because they know that many other
consumers will want to follow their lead. Researchers have been studying ways to
automatically discover these influencers by analyzing the graph structure of the social
network with computer algorithms. For instance, Domingos and Richardson (2001)
showed that by analyzing the way opinions appear to flow along a social network they
could choose users who would be the best people to market a product to. These users
were the ones who would most influence other users in the social network to use the
product. Kempe et al. (2003) showed that under more general models for disseminating
information through a social network a simple algorithm that maximized marginal gain at
each step could do a very good job of choosing a good set of users to market to.
Design Claim: Analyze the social network of your community to determine efficient ways
to disseminate information to your users.
Page 6
Document13/12/2016 8:42 PM
Message. Even though word of mouth recruitment can be very effective, impersonal
advertising also works (Assmus et al, 1984). These impersonal techniques can influence
targets’ beliefs about a stimulus, such as a consumer product, a health intervention,
political candidate or an online community, their affect towards the attitude object, and
the likelihood that targets with compatible beliefs and attitudes will act on them, for
example by attempting to join the community, buy a project or turn out to vote for a
candidate. (Roberts & Maccoby, 1985).
Despite the proven effectiveness of impersonal, mass-media advertising, its direct effects
are weaker than many advertisers would like and Orwellian critiques fear (see Klapper,
1960). According to Tellis’ recent review of almost 500 field experiments of advertising
effectiveness, for many mature brands the total amount of advertising a firm engages in
[i.e., its advertising weight] is not critical in influencing sales. “More than half the time,
increases in weight alone do not lead to an increase in sales. … [Moreover], neither do
decreases in weight lead to sales decline, at least in the short term. (2004, p 79). There are
many reasons why impersonal advertising has weaker direct effects that its practitioners
hope for — confirmatory biases, such as selective exposure, interpretation and retention,
miscomprehension of advertising messages and competing advertising from other
communities. Studies of selective exposure show that people are more likely to be
exposed to beliefs that they already agree with, in part because they affiliate with people
who are similar to themselves. People differentially interpret persuasive messages based
on their prior beliefs (Nickerson, 1998). For example, among people who watch
presidential debates, viewers strongly believe that the candidate whom they initially
supported won the debate (ref). However, the debates can increase viewers’ knowledge of
the issues and the ones they consider important in a candidate, and despite viewers biases,
can influence vote preference (Benoit et al, 2003). Advertising is may be especially
effective for new communities than established one, when information about the
community and confidence in it is low (e.g., the introduction of a new role-playing game
rather than a new release of World of Warcraft) (Tellis, 2004, Chapter 2).
Page 7
Document13/12/2016 8:42 PM
Figure X: Modified elaboration model
Adopted from Cacioppos & Petty (1985).
The best ways to construct a persuasive message depend on the degree of psychological
engagement that potential recruits have in their search for a community. There are cases
when a potential recruit is actively seeking a community a join, such as when a patient
with a life threatening disease is choosing an online support group. In these cases, the
potential recruits are likely to use effortful, systemic processing of the quality of the
information they are exposed to, the consistency of the arguments and the credibility of
the source. Psychologists (E.g., Petty; Chaiken; see Figure X) term this central
processing. For example, the patient may check to see whether the group is linked to a
credible national organization like the American Cancer Society, carefully read messages
posted to the group, see whether medical professionals are posting and compare the
advice they see in the group with the advice they are getting from their own physician.
In many other cases, however, people are not actively seeking a group or join or are
seeking one more casually. In cases such as these, people tend to form opinions based
on what psychologists call heuristic processing. They don’t think deeply and evaluate
carefully the information they are exposed to, but instead use many cognitive short cuts
or heuristics to form their opinions. When they are performing heuristic processing of
information and persuasive messages, they are likely to be influenced by very superficial
features of the attitude object or persuasive message (Petty; Chaiken).
Design claim: When potential recruits are relatively un involved in searching for and
selecting an online community, they will be influenced by superficial features of the
community and persuasive messages about it, but when they are highly involved in
choosing an appropriate online community important qualities of the communities,
factual information and logical argument will be more likely to influence their choice.
.
As we discussion in Chapter X, on starting online communities, if people see signs that
other people are active on the site, then they are more likely to join than if they think the
site is barren. Although this phenomenon could be caused by the thoughtful processes
implied in Chapter X, in which the presence of other people leads rational newcomers to
Page 8
Document13/12/2016 8:42 PM
Page 9
estimate that the site will enough membership to produce enough conversation, friendship
or other benefits to be an enjoyable place to hang out. However, joining after seeing
activity on the site may be caused by following what psychologists calls the social proof
heuristic (Cialdini & Goldstein, 2004), in which people simply believe that they should
do what others are doing, without effort calculate of benefits implied by the theory of
network externalities discussed in Chapter X. The social proof heuristic can be triggered
by minor changes in wording in persuasive messages urging people to join a site. For
example, an experiment by Goldstein and his colleagues have shows that changing the
wording of those ubiquitous hotel notices asking guests to reuse their towels (“Help save
the environment … by reusing your towels during your stay.”) by emphasizing social
proof (“Join your fellow guests in helping to save the environment.”) increased guests
reuse by 44% (Goldstein et al, 2008). Cialdini and his students also showed that,
ironically, standard signage that emphasized the environmental damage that visitors to a
national part had on the environment by stealing pieces of petrified wood actually
encourage people to steal compared to a sign that emphasized that stealing violated social
norms; 7.9% versus 1.7% of visitors of stole samples under the two conditions. Applying
this logic to the domain for recruiting for an online community, invitations with a
headline that implies ‘Join the crowd’ are likely to be much more effective than those that
say ‘We need you’.
Figure x: More (left) and less (right) effective recruiting pitches, based on social proof
Design claim: Emphasizing the number of people already participating in a community
rather than its need for more members will activate the social proof heuristic and cause
people to try out the site.
The familiarity heuristic is
another powerful one.
People tend to like people,
things and ideas they are
familiar with (Zajonc, ref).
It is for this reason in part
that name recognitions does
wonders in both consumer
products and politics.
Incumbents are
overwhelming reelected in
congressional and local
Figure X: Use of an ‘E-mail this’ link to increase familiarity with a
brand.
Document13/12/2016 8:42 PM
elections (Mann & Wolfinger, 1980) and movie stars regularly win elected offices for
which they have
little experience
or prior
qualifications.
The implication is
that managers of
online
communities
should simply get
the name of their
community in
front of the
relevant
Figure X: Recruiting for the MacMillan Cancer Support
population, to
Network
increase their
(http://www.whatnow.org.uk/)
liking of the
community and therefore likelihood of joining. The placement of email and tagging links
are many sites helps make the site familiar to others, increasing their liking for the brand
and their willingness to become members. (See Figure X).
Design claim: Placing the name of a community in front of people often will activate the
familiarity heuristic, their liking of the community and their willingness to ttry out the
site.
The halo effect is another cognitive heuristic, in which people perceive that a stimulus
that is good on one dimension will also be good on other, unrelated dimensions
(Thorndike, 1920). By prominently displaying features of a community that are easily
perceived as positive, managers will be better able to attract recruits than if they tried to
show features of a site that are perhaps more important but also more difficult to display
as positive. For example, a cancer support community might highlight the physical
attractiveness of its member, because it is difficult to demonstrate at a glance how
supportive the community actually is. (See Figure X).
Design claim: Emphasizing easily visible features of a site that make some aspect of it
seem attractive (e.g., the physical attractiveness of members) will active the halo effect
and cause recruits to increase their liking for other features of the site as well.
The role of fit. Insuring a good fit between the newcomers and the community is a major
challenge in the early phases of newcomer socialization, when communities are recruiting
prospective members, selecting among them and encouraging the desirable ones to stick
around, and newcomers are investigating communities and deciding which ones to
participate in. Communities’ decisions about the pools of prospective members to recruit
from, ways to select desirable members the information they make public and the early
experiences they provide newcomers all influence fit. A substantial amount of empirical
research shows that good things happen when newcomers to a group or organization
Page 10
Document13/12/2016 8:42 PM
Page 11
perceive themselves as having a good fit to it on a number of dimensions. These
dimensions include the interests, attitudes and values they and the group share and the
Raid-oriented
Socially-oriented
<Premonition> WotLK Cleared! Recruiting
hardcore players!!
<Damage Networks> - LF Warlock, SPriest,
HPriest, Ele Shaman, Paladin
We are Premonition --- a mature Alliance PvE
raiding guild on the Windrunner PST server. Please
visit us at http://premonitionwow.guildlaunch.com
to learn more. We are recruiting hard core, geared
players for WotLK. … We have completed Malygos
25 man/10 man, Sartharion 25 man/10 man, and
Naxxramas 25 man/10 man. …
… Damage Networks is currently looking to add a
few more people to our family. We're a PvE/PvP
guild that spends way too much time on ventrilo,
and not nearly enough time raiding. As a guild,
we've been around for a long time (completing MC
and BWL in their primes, and landing a top 20 US
Kel'Thuzad kill). We're not going anywhere.
Damage Networks began as a porn site, then an NS
team, and now it's a wow squad. So if you're
interested in more than just a raiding guild, we're
probably right for you.
Here’s some key information about the guild and
our expectations.
1. Raiding schedule is Sunday-Thursday 8:00pm 11:30pm PST (server time). You are REQUIRED to
attend 4/5 raids each week.
2. We have vent and EPGP.
3 . *Most* gear will be EPGP based, but there will
be some pieces that will go to the class/spec it will
suit the most to help progression of the guild. Very
few pieces will be decided in this manner but the
best interests of the guild and its progression will be
considered first and always. We’re a team and will
be most happy when we are downing new content
consistently. To maintain progression this is
necessary .
What are we looking for? Players are expected to
hold at least 75% raid attendance. Drama bombs,
Loot %%#!*s, ego-maniacs, self-centered retards,
incessant whiners, immature babies and idiots in
general should not apply (we have plenty of those
already, they give us ''flavor''). …
4. We’re very serious about having the most optimal
raid composition and most potential while raiding.
To maintain that we’ll require you to have:
a. PvE spec when you come to raids.
b. You will show up on time and ready and
you will have consumables to last the
entire raid.
…
g. You will know encounters having read
strategies, watched videos, read forums
before we engage those encounters.
Figure Error! No text of specified style in document.-1: Sample World of Warcraft
recruiting announcements at www.lookingforguild.net (downloaded 12/31/2008)
knowledge, skill and motivations they possess and that the organization desires. When
potential recruits see that they share a good fit with a group or organization, they are
more likely to be attracted to it, more likely to pursue membership in it, more likely to
join it if given a chance (e.g., by accepting a job offer), and, if they do join, are more
likely to be satisfied with their membership and to remain in the group or organization for
longer periods (Chapman et al). After joining, during the first few months of
Document13/12/2016 8:42 PM
membership, when newcomers perceive that they have a good fit with the group or
organization as a whole, with their work group, with their job or with their supervisor
they subsequently report having greater satisfaction with jobs that they take in the
organization and the people they work with, greater organizational commitment and
identification, lower intention to quit, less turnover and lower stress (Kristof-Brown).
While most of the research on the benefits of fit has been done in the context of job
candidates and employees in conventional organizations, the same phenomenon also
applies in the context of volunteers to voluntary organizations, which are similar in many
ways to membership in an online community. Just as people have a variety of motives to
participate in an online community, so too do volunteers to organizations like
BigBrothers/Big Sisters or the American Red Cross. Some may participate out of
altruism or because they share the organization’s goals. Others may participate for the
social experiences they get when working with similar others, to learn new skills or to
exercise skills and abilities that might otherwise go unpracticed, to obtain credentials or
experiences relevant to a new career or to make themselves feel good. When they
participate in volunteer activities that match the motivations they consider important, they
are more satisfied with their volunteerism and are more likely to volunteer again (Clary et
al, 1998). In addition, volunteer organizations can take advantage of this variability in
volunteers’ motivation to attract those with different motives by varying the wording in
their advertising. For example, brochures that highlight the way in which volunteering
helps one explore career options, develop a strong resume and network to make career
contacts were especially likely to appeal to recruits with career-related motivations (Clary
et al, 1998).
The importance of fit also applies to online communities. Consider the two recruiting
messages in Figure X. The one on the left describes a testosterone-driven raiding guild
seeking “hard core” player. It demands both time and particular types of equipment from
its members. In contrast the guild on the right is more laid back and socially oriented; it
seems to be seeking players with appropriate personalities and social skills (no drama
bombs, whiners or those who get angry over loot distribution). Imagine the problems that
would flare if Premonition ended up with a laid-back chatter or Damage Networks if a
hardcore warrior. Our own research shows that new recruits to World of Warcraft guilds
who prefer the style of play the guild values remain in their guild substantially longer
than players for whom there is a style of play mismatch (Choi et al). In addition, different
recruiting methods lead to different degrees of player-guild fit. It’s relatively easy to
identity the hard core players valued by Premonition by examining the players online
resume (e.g., the weapons they have accumulated) and by playing with them a little
while. In contrast, guilds can more easily identity the personality configurations valued
by Damage Networks by using referrals from existing members and by playing with the
new recruits during longer probationary periods.
Design claim: To increase the fit between a community and the members they recruit,
provide potential new members an accurate and complete picture of what the members’
experience will be once they join.
Page 12
Document13/12/2016 8:42 PM
Empirical research shows that conventional organizations that provide recruits accurate
and complete information about the organization help them to form accurate expectations
about it, which will influence their decisions to seek employment, to join if employment
is offered and to have realistic expectations once they become organizational members.
In the context of conventional organizations, this accurate information often is conveyed
in the form of realistic job previews, in which recruiters and other members of the
organization present both favorable and unfavorable job-related information to job
candidates (Rynes, 1991). Realistic job previews are associated with recruits’ lower
expectations about the nature of the job, lower levels of attrition from the recruitment
process, and lower turnover, if recruits are offered a job. Although the effects are small,
they are reliably larger if the realistic preview is delivered verbally than via a written
document or video (Philips, 1998). Realistic job previews seem to have beneficial effects
on recruiting success through two separable routes. First, they selectively attract people
who have characteristics, including motives, skills and attitudes that better fit the existing
community. Second, they lower expectations that newcomers have when they actually
join the community, reducing the chances that the reality newcomers experience once
they become organizational members will clash with unrealistic idealizations.
Online communities can use the methods typical of conventional organizations to give off
realistic information about the community — websites, online recruiting brochures and
other documents, videos, direct contact with recruiters or other formal representatives of
the community, and informal contact with community members. For example, when
soliciting new members, World of Warcraft guilds often post recruiting statements, with
information about their mission and style of play, at forums such as
www.lookingforguild.net or similar recruiting sites. (See Error! Reference source not
found. for examples.) The recruiting ad on the left describe the rigors of a high
testosterone, raid-oriend guild, while the one on the right describes a more laid-back,
socially oriented one.
Others post videos of their exploits on youtube.com, www.warcraftmovies.com and other
media sharing sites. One famous clip is the saga of Leeroy Jenkins, a dramatic
recreation by the guild Pals for Life of one of their encounters1. Leeroy Jenkins, one of
their members, misses the detailed planning the guild engaged in before raiding a
monster’s cave because he is away from his computer frying some chicken. Their plan is
ruined when Leeroy returns and rushes into the cave shouting his own name as a battle
cry. When the others follow him in, all die. While this video was originally produced as a
parody, to contrast the passion of play, with the calculated nerdiness some guilds engage
in, one of its consequences was that the guild was barraged by requests for membership
from applicants interested in their strategic style of play.
While these types of deliberate self-descriptions occur, they are rare compared to the
recruiting activities conducted in conventional organizations. More typical are the
frequently asked questions (FAQs) listed in many online communities, in which leaders
of the community post a mission statement or goals, but not realistic previews of life
inside (see http://www.faqs.org/ for an index to many FAQs for Usenet newsgroups and
1
http://www.youtube.com/watch?v=LkCNJRfSZBU
Page 13
Document13/12/2016 8:42 PM
Websites. For example, the FAQ for a stop smoking support group lists its policies
banning commercial posts, but provides little information about the frequency of this
practice.1 While comparatively few online communities explicitly create realistic
membership previews for recruiting purposes, the archival nature of the Internet means
that complete records of prior interactions among community members are available for
newcomers to examine and in this way get an unedited, realistic view of life in the
community.
Design claim: To increase the fit between a community and the members they recruit,
allow potential new members to see archives of communication within the community.
Even though online communities can use the same recruiting mechanisms as
conventional organizations, they have the potentially unique advantage that much of the
communication and production work in the community is archived automatically. While
in a conventional organization a new recruit must rely on the organization’s explicit
descriptions via brochure and similar sources or rely upon word of mouth form current
and former organizational members, with an online community they can see the
interactions on which these impressions are based. They see how members treat each on
various online forums and can see how team members work together in creating
Wikipedia articles by examining the histories and talk pages associated with each article.
However, not all communities provide public archives. In the online cancer support
group ACOR.org, readers cannot search or browse archived messages without a
subscription. Though subscribing is free, prospective subscribers are vetted by the list
owners, which delays the newcomer's opportunity to evaluate the group and his or her
expected fit. Furthermore, the archives are hidden from search engines. Though this
protects the privacy of existing members, it also reduces the likelihood that desirable
members will find the group.
Problem 2: Selecting the right newcomers
Newcomers to a community differ in the potential value they could bring if they were to
become full members. In an open-source software development community, for
example, some newcomers may be highly skilled software developers with deep
knowledge of the application domain, some may be novice developers or ignorant of the
domain, having little knowledge or skill to bring to the project, and others may have the
malicious intent of introducing bugs or Trojan horses. Analogously, in an online support
group for abused women, valuable members might be the survivors who have experience,
wisdom, and support to offer to others or women who are themselves currently the
victims of abuse. In contrast, spectators or stalking husbands are highly undesirable. How
do existing community members distinguish the potential quality of newcomers from the
limited information available when they first present themselves?
The problem of selecting applicants for membership is common to all groups, but it may
1
http://www.faqs.org/faqs/support/stop-smoking/compost/part1/
Page 14
Document13/12/2016 8:42 PM
be especially problematic in online communities because of the relative anonymity of the
interaction in them and the ease of creating new identities online. For example, one of
Wikipedia’s administrators who had presented himself as a professor with degrees in
theology and canon law was forced to resign when a magazine revealed that he had no
advanced degrees. He "used texts such as 'Catholicism for Dummies' to help him correct
articles on the penitential rite or transubstantiation"1. In eBay, legitimate sellers would
like to weed out scammers who sell used goods as new, or copies as originals, or those
who collude to inflate bids2.
In many real-life situations, individuals often have information relevant to a selection
decision but fail to reveal this during the selection process. Job candidates may have
poor work performance or criminal backgrounds they wish to conceal from potential
employers. This information asymmetry leads to what is known as “adverse selection,” in
both the insurance industry, where these types of problems were first identified, and in
economics, where theories of adverse selection have been elaborated. If an insurance
company sets a single premium for life insurance, then smokers, intravenous drug users
or others who have lifestyles leading to poor heath could differentially benefit from the
insurance. People with these poor health risks are more likely to apply for the insurance
than healthy consumers, and as a result, companies will end up disproportionately
insuring them. As in the case of health risks, it is sometime in people’s self-interest to
withhold private information when an online group is making a decision about accepting
them for membership. This phenomenon occurs in the online world of multi-players
games. For example, in World of Warcraft, players may apply to become a member of a
guild with the intention of staying only a short time, merely to “level-up” (i.e., gain
experience points) before moving to a superior guild (Ducheneaut, et al., 2007).
However, it is not in the guild’s interest to recruit players who have intentions to leave as
soon as they have accumulated the appropriate amount of experience.
In other cases, it is in individuals’ interest to reveal private information about themselves,
but the organization has difficulty distinguishing truthful from less truthful reports. In
conventional work organizations, it would be beneficial to strong applicants for a
software engineering position to be able to demonstrate their work ethic and skill at
software development to distinguish them from pretenders who are less dedicated or less
able. The same dilemma occurs for applicants for admissions to many online groups. For
example, applicants to an open-source software development site like Apache who are
very skillful at software-relevant tasks or who have been conscientious in the past in
turning in assignments on time would want to let the powerful people in the site know
this. However, less skillful and conscientious applicants would want to give off a similar
impression. As a result, when experienced software engineers tell of their software
experience in their day jobs, the community tends to ignore these self-reports as merely
"cheap talk" (Ducheneaut, 2005; Lakhani & Hippel, 2003). In World of Warcraft, many
guilds seek new players with a style of play and social relationships that fit with existing
group members. New recruits may honestly describe themselves for example, saying they
1
http://www.nysun.com/article/49955
http://www.truetex.com/ebayfraud.htm, http://reviews.ebay.com/Scam-Watch-The-Most-Popular-Scamson-Ebay_W0QQugidZ10000000000025238
2
Page 15
Document13/12/2016 8:42 PM
are extroverted and friendly and are seeking a relaxed style of play. However, guild
masters or recruiters have little way to distinguish recruits who will be difficult to get
along with from those who are easy to get along with. Economists often refer to these
cases as signaling problems, because one party is attempting to signal their true qualities
to perceivers using imperfect signals.
In the case of selection, decision-makers are looking for visible signals that are correlated
with the attribute they desire and which would be difficult for people without that
attribute to fake. In trying to select quality childcare for their children, parents should not
pay much attention to the cleanness of the reception area, which is an attribute that is
easy to fake and is often associated with lower child care quality, but should pay attention
to whether the center is for profit (often associated with poorer quality), whether it is part
of a national chain (often associated with high quality) or is regulated by the government
(often associated with higher quality) (Mocan, 2001). In open-source software
development projects, the community seems to rely more upon participation in technical
discussions and contributions of bug patches and software enhancements to make such
decisions. Communities that provide free email accounts or server space also operate on
the same principles of selection when they use CAPTCHAs to prevent bots from
automatically creating hundreds of accounts. A CAPTCHA (an acronym of "Completely
Automated Public Turing test to tell Computers and Humans Apart) is an automated test,
such as the challenge to recognize a distorted word presented against a cluttered
background, to differentiate people from computerized agents attempting to exploit
Internet services. These tests are difficult for machines to accomplish but easy for
humans, and therefore many Internet companies, including Yahoo, MSN and eBay use
this mechanism to identify whether an agent applying for an Internet service is a human
or a machine.
In the case of signals, similar principles apply, although one generally assumes that the
candidates for selection have the ability to invest in acquiring visible features that reliably
signal quality. Then organizations and other decision-makers can select among
candidates who have previously invested in resources that they cannot change during the
period of selection. For example, job candidates might invest in completing four years of
college at a good school to show their intelligence and their conscientiousness, even if the
education itself will not provide them skills that help them perform the job. Analogously,
players in the online game World of Warcraft, for example, can invest hours in playing
the game to acquire weapons and the steeds they use for transportation. In turn, guild
masters and others recruiting players for guilds can use candidates’ level, the weapons
they have already acquired, or their steeds to choose among candidates, because the
factors are reliable signals of past performance as well as being tools which the candidate
can use to carry out quests, if they become members.
To help select good community members, signaling theory suggests that designers of
online communities create entry barriers, diagnostic tasks that would be easy for
desirable members but difficult for undesirable ones. In many open-source software
(OSS) development projects, for example, potential members must first demonstrate their
competence and commitment to the group by offering bug fixes or small enhancements
Page 16
Document13/12/2016 8:42 PM
before they are given "committer" status, permission to commit (save) their own changes
to the software database (Krogh et al., 2003; Ducheneaut, 2005). In the FreeNet project,
only 8.4% of individuals who participated in the technical discussions were ever given
committer status and considered developers in the project (Krogh et al., 2003). Without
committer status, programmers must pass their modifications to more trusted members of
the group who then vet the software and decide whether to merge it with the existing
code base. Mere talk without code, such as describing one's offline technical
accomplishments, asking for tasks to work on, or proposing modifications did not lead to
committer status; potential members had to pass substantive contribution barriers to
become full members.
Another barrier is the credential check: Sermo.com, a discussion forum for physicians to
discuss medical decisions, asks potential members for their names and the zip code of
their primary practice, which it then cross-checks against a national physician database,
to ensure that its membership includes only physicians. Craigslist requires posters of
classified ads to enter their email address and then respond to an invitation sent to that
address before their ad goes public. The goal is similar to the goal for CAPTCHAs, to
prevent screen bots and other software agents from gaining membership in the
community. Pornographic websites that ask prospective members to provide a credit card
number or license do so to try to differentiate adults from minors (or give the illusion of
doing so), because adults are more likely to have a credit car or driver's license than
minors. To weed out spectators, established members of one online depression forum
have an unspoken practice of engaging newcomers in discussion about their symptoms
and treatment (Fussell, personal communication). Producing these medical terms is likely
to be easier for people who are clinically depressed than for those who aren't, although
obviously this is not a failsafe system, because anyone could potentially conduct an
online search to discover the names of antidepressants.
Design claim: Diagnostic barriers to membership allow communities to evaluate
newcomers and thereby help the community identify desirable members.
According to cognitive dissonance theory, difficult barriers have the added effect of
increasing commitment among members who have completed them: causing newcomers
to suffer a little before joining a group should increase their eventual commitment. The
theory of cognitive dissonance holds that if people have two ideas that are
psychologically inconsistent, they experience the negative drive state of cognitive
dissonance and try to find a way to reconcile the ideas, generally by changing one or both
to make them consonant (Festinger, 1957). This theory explains why we tend to change
our attitudes to be consistent with our behavior (e.g., Festinger & Carlsmith,1959). It has
also been used to explain people's initial feelings of connection to the groups they join. In
particular, experimental research by Aronson and Mills (1959) demonstrated that people
like groups more if they have to endure a severe initiation process to join them than if
they undergo a milder initiation (see also Gerard & Mathewson, 1966; Schopler &
Bateson, 1962). According to Aronson, people come to like things for which they
suffered, because this is the only way they can reconcile their views of themselves as
intelligent people with the actions they have performed (Aronson, 1996).
Page 17
Document13/12/2016 8:42 PM
Page 18
However, given the ease with which people can leave an online community (as opposed
to a military academy), severe initiations may drive away potentially valuable
contributors. Therefore online community designers should not instigate these types of
initiations unless there is a surplus of prospective members. However, the newcomers
who survive the initiation should have stronger loyalties than those who are invited in
without initiation. First, as selection theory predicts, the severe initiation acts as a filter,
preventing new recruits without the appropriate desire from joining. Second, as
dissonance theory predicts, these initiations should increase the commitment of those
subjected to them.
Initiation rituals imposed by online communication can range from non-existent to quite
severe. At one extreme, Usenet discussion groups impose no initiation at all. Newcomers
can read and post without any formal barrier. Wikipedia explicitly encourages gentle
treatment of new editors, with its "Don't bite the newcomer" policy [1]. At the other
extreme are game-playing groups like World of Warcraft and OSS projects that require
newcomers to go through a long period of initiation before they can become members.
Some guilds, for example, require the newcomer to play with the group for a month or
longer before the newcomer is allowed to become a regular member. In open-source
software projects, it is common practice for newcomers to offer "gifts" of code before
they are granted membership (von Krogh et al., 2003). While both of these activities
provide data by which existing group members can evaluate the newcomers and may
weed out the least motivated, the activities are also effortful actions that probably
increase the newcomers' loyalty to the group.
Percentage of users who added at least one
tag after entry
Average number of tages added post-entry, per
user, per day
Figure X: Effects of entry barriers on post-entry contributions to Movielens (from Drenner,
et al, 2008)
Drenner and her colleagues (2008) introduced effortful barriers to the online movie
recommender site, Movielens.org. All users to the site were asked to rate five movies, to
let the site know about their movie tastes. Some new registrants were in addition
Document13/12/2016 8:42 PM
required to apply 25 descriptive text tags to movies before they were allowed to register
(Tag-25), some had to apply five tags (Tag-5), some were simply shown the tagging
interface but not required to tag any movies (Tag-0), and some were not even shown the
tagging interface (Control). As expected, as the amount of work required in the
registration process increased, the proportion of registrants who completed the full
registration process declines, from 90% completion in the control condition, to 85% Tag0 condtion, 80% in the Tag-5 condition to 69% in the Tag-25 condition. More
interestingly, new users who were assigned to do more work as part of their registration
process were four times more likely to provide tags once they become members and
contributed over 10 times the number of tags as those in the control condition. The
effects were only partly the result of the type of selection we discussed previously. As the
light-colored bars in Figure X illustrate, the effects of the entry barriers were strong even
when considering all potential members exposed to them, not just to those who
surmounted them.
Design claim: Forcing newcomers to overcome an entry barrier before joining may
cause those who join to be more committed to the group and contribute more to it.
Problem 3: Keeping newcomers around
For newcomers to gain benefits from an online group and to eventually become
committed members who take on core responsibilities, they must stick around long
enough to learn the ropes, form relationships with other group members, and begin to
identify with the group as a whole. However, the research on online communities also
shows that these groups experience a substantial amount of turnover and that this
turnover is especially high among newcomers. For example, 68% of newcomers to
Usenet groups were never seen after their first post; in contrast, those who have
participated even once in the past are much more likely to return (Arguello et al., 2006).
Fifty-four percent of developers who registered to participate in the Perl open-source
development project never returned after posting a single message (Ducheneaut, 2005).
Forty-six percent of the members of guilds in the massive multiple-player game, World
of Warcraft, leave their group within one month, generally migrating to other groups
rather than abandoning the game itself (Ducheneaut et al., 2006).
What can a community designer do to keep newcomers around long enough for them to
start to understand the benefits they could derive from group membership and start to
learn how to behave appropriately in the group?
Investigation is a major reason that newcomers silently read posts (i.e., “lurk”) before
posting (Preece et al., 2004). At the same time, the group is evaluating whether the
newcomer would be a good fit, so the group needs the newcomer to do more than simply
lurk. Both the group and the individual require information exchange in order to estimate
the benefits they will bring to each other. This investigatory phase is an especially fragile
one for newcomers. During this early period, when they first encounter a group, they
have little commitment to it and often little data to make judgments about whether to
invest effort in finding out more or to explore alternatives. As a result, small amounts of
Page 19
Document13/12/2016 8:42 PM
either positive or negative evidence about how the group behaves and how it treats
members may have an especially large impact on whether they leave for good or return
again. Similarly, small amounts of either positive or negative evidence that the
prospective member has desirable attributes will impact how the group responds.
Initial positive interactions work to retain new members. Newcomers to Usenet groups
are more likely to come back for subsequent visits if others reply to them (Arguello et al.,
2006). These effects are stronger if the people conversing with the newcomer are
themselves old-timers who have been visible in the discussion in the recent past. The
effects were also stronger if the responses used more welcoming and inclusive language,
such as emoticon and first-person plural (“we”) pronoun, which are often used to indicate
solidarity between speaker and audience, rather than second-person (“you” pronouns),
which often indicate a us/them divide between a speaker and and the people being
addressed. Our analysis of initial interactions between newcomers and old-timers in
Wikipedia project shows a similar pattern. Wikiprojects are groups of editors who work
together on articles within a domain, like military history, sports or medicine1. New
members to a project who receive more communication from existing during the month
that they joined subsequently edit more on project pages (and in Wikipedia in general)
and stay active in the project for a longer period. Again, the effects vary with the nature
of the communication they receive. Personalized messages (e.g., comments about the
newcomers background or request to work on a particular tasks) lead to more powerful
effects than generic ones, in which the newcomer receives a standardized messages, such
a welcome-to-the-project template In an analogous results, Burke and her colleagues
showed that new Facebook members are likely to post more photos if others comment on
the initial photos they added (Burke et al, 2009). Lampe and Johnston (2005) found that
new Slashdot members whose first comment received a rating from other members
posted a second comment more quickly than new members whose comments weren't
rated. They also found that even newcomers who received negative ratings on their first
comments came back faster; they hypothesized that these newcomers returned quickly to
improve their records, or that they intentionally wrote inflammatory content which they
post more often
To retain new members, small-group socialization theory and the research on recruiting
reviewed previously suggest that newcomers should interact directly with experienced
members right away. The goal of the first contact should be to encourage initial positive
interactions as the newcomer is exploring the community and deciding whether to join or
not. Rather than isolating newcomers in a mandatory private sandbox or preventing them
from posting messages to a discussion group, they should be allowed to make themselves
known and interact with other community members.
Design claim: Newcomers who have friendly interactions with existing community
members will be more likely to stay in the community and contribute more.
Discussion forums often include a prominent "Introduction" thread, in which newcomers
are encouraged (or required) to post brief biographies. Newcomers to PGHDance.com, a
1
http://en.wikipedia.org/wiki/Wikipedia:WikiProject
Page 20
Document13/12/2016 8:42 PM
forum for Pittsburgh swing dancers, go to the "Hi, I'm ..." thread to describe their level of
dance experience, day job, and other cities where they've danced. The forum is a hybrid
community, in which many members socialize in person at local dances, but that
socialization is often hindered by loud music and a norm of not talking while dancing.
Thus, PGHDance allows members to get to know each other off the dance floor, and
provides a more neutral platform where dance skill is less salient. Veteran forum
members greet the newcomers and offer to dance with them at local events. Similarly,
WrongPlanet.net, a community for individuals with autism and other developmental
disabilities hosts a “Getting to know you” section, where newcomers describe their
hobbies and diagnoses. These introduction threads serve two purposes: First, they allow
the newcomers to move beyond the lurking stage and provide enough information to
invite interaction with other members, and second, the threads allow newcomers a safe
space to practice using the posting tools. As much research in social psychology shows,
when people self-disclosure others reciprocate and reveal information in exchange. This
mutual self-disclosure often leads to the strengthening of the relationship between the
pair (Collins & Miller, 1994).
Design claim. Encouraging newcomers to reveal themselves publicly in profiles or
‘introduction threads’ give existing group members a basis for conversation with
newcomers than therefore should increase interaction between old timers and
newcomers.
One way to accomplish initial positive interaction is to assign welcoming responsibilities
to old-timers. The host guide for one of the earliest online communites, The WELL’s
(Whole Earth 'Lectronic Link) Hosts’ Manual explains, moderators are responsible for
welcoming newcomers (Hoag, 1996)1: “Nobody likes to go into a conference for the first
time, post a response, then have it sit there without ever being acknowledged. Learning to
welcome, inspire and incorporate new visitors into the conversation is perhaps the most
important talent a host can acquire. At the very least, as host, you will want to keep an
eye out for postings by folks who have never responded in your conference before, and
acknowledge their participation.”
1
http://www.well.com/~confteam/hostmanual/
Page 21
Document13/12/2016 8:42 PM
For example, Wikipedia has a “Welcoming Committee1,” whose main activity is to greet
new editors, known as red users, because they have not yet made a personal page for
themselves and thus their usernames appear in red. Welcoming committee members
skim Wikipedia's account creation log and lists of contributions by newcomers, select
friendly text from a set of welcoming templates, and post the text to the user's talk page,
creating the page if necessary. Welcoming committee members encourage anonymous
contributors--identified by their IP addresses--to register, post links to tutorials, and offer
to answer questions.
Facebook uses a highly distributed welcoming mechanism. It has a feature that
encourages old time members to send request to people in their friend list to add new
members as a friend.
Design claim. Assigning these responsibilities of having friendly interactions with
newcomers to particular community members increases the frequency of these
interactions.
The previous examples from Wikipedia and The Well are prescriptive in a positive way,
assigning some people the
responsibilities for
welcoming newcomers
and giving the welcomers
some tools to make the
task easier. Another way
to encourage newcomers is
to discourage the hostility
that is often the result of
the interactions between
old-timer members of a
group and newcomers. As
the FAQ for the
Mozilla.org project, the
developer for the popular
Firefox web browser
notes. “Be kind to
newcomers. Newcomers
may be annoying. They
Figure X: Facebook’s feature encouraging friends to
ask the wrong questions,
welcome newcomers
including ones that seem
obvious (or whose answers
seem easy to find). But lots of valued contributors started out this way, and treating
newcomers kindly makes them more likely to turn into the valuable community members
1
http://en.wikipedia.org/wiki/Wikipedia:Welcoming_committee
Page 22
Document13/12/2016 8:42 PM
we all know and love (and cut some slack when they mess up).1” Similarly, policies for
experienced users answering questions in the forum for Ubuntu, a graphical user interface
for the Linux operating system, discourage experienced users from being rude to
newcomers. “Be considerate to the person asking the question. We were all a green user
at one point.. … If you wish to remind a user to use search tools or other resources when
they have asked a question you feel is basic or common, please be very polite. Any
replies for help that contain language disrespectful towards the user asking the question,
i.e. "STFU" [Shut the fuck up] or "RTFM" [Read the fucking manual] are unacceptable
and will not be tolerated.2” Wikipedia's DBTN (Don't Bite the Newcomer) policy
cautions old-timers that “New contributors are prospective 'members' and are therefore
our most valuable resource. We must treat newcomers with kindness and patience —
nothing scares potentially valuable contributors away faster than hostility or
elitism." (Emphasis in the original)
Design claim: To promote friendly initial interactions between newcomers and oldtimers, explicitly discourage hostility towards newcomers who make mistakes.
Problem 4: Teaching the newcomers the ropes
Different communities have standards and norms that shape and constrain the behavior of
their members. Some of these norms are broad and open to different interpretations. In
Wikipedia, for example, a series of guidelines and policies remind members to adopt a
neutral point of view in the articles they write [1] and that they should not use their
personal talkpages to discuss personal topics and promote relationships with other
Wikipedians [2]. Others are more narrowly targeted, such as the Wikipedia copyright
policy [3] or formatting guidelines [4]. Although many of the norms and behavioral
standards are explicitly described in Wikipedia, in many other online communities norms
that prescribe how members should behavior are implicit and must be learned by
observation.
Violation of the behavioral norms can be harmful to existing groups for
several reasons. First, norms are often set up because they are functional,
helping the group to achieve desired goals or to operate efficiently. Thus, the
norm in Wikipedia of citing sources helps the group achieve its goals of
becoming an authoritative encyclopedia. Similarly, norms to hold all
discussion via public distribution lists in open-source development projects
help the group capture its design rationale and supports coordination among a
distributed workforce. Violating these norms interferes with valued group
goals and smooth operation. Standards of behavior often form part of the
identity of the group’s members. Thus, when people act in non-normative
ways, the group may start to lose some of its defining features. Violations of
norms may also drive away old-timers. [point to Norms chapter here]
Socialization tactics. Organizational socialization theory identifies six dimensions that
1
2
http://www.mozilla.org/community/etiquette.html
http://ubuntuforums.org/index.php?page=policy
Page 23
Document13/12/2016 8:42 PM
Page 24
differentiate the techniques that organizations use to help newcomers get adjusted to the
organization and learn their place in it. Jones (1986) created a self-report scale
measuring these tactics, which has been extensively used in empirical research examining
antecedents and consequences of successful socialization of newcomers in organizations.
Table N provides an overview of the socialization tactics that van Maanen and Schein
identified. At the institutionalized end of these continua, newcomers are segregated from
contact with old-timers, are given uniform experiences, told the sequences of stages they
will go through as they progress though the organization and the time they will put in at
each stage, get training from experienced role model. They are encouraged to jettison
their distinctive personal characteristics in the process of trying to fit into the
organization. Students going though their PhD training, given strong mentorship and
clear expectations about requirements are at the institutionalized end of these dimensions.
Institutionalized
Collective: Newcomers go through a
common set of experiences designed
to produce standardized responses to
situations
Individualized
Individual: Each newcomer receives unique
training in isolation from others.
Formal: Newcomers are segregated
Informal: Newcomers receive on-the-job training
from other organizational members as to learn their roles.
they learn their roles.
Sequential: Newcomers are given a
clear sequence of experiences or
stages they will go through
Random: The sequence of stages isn't
communicated in advance.
Fixed: Newcomers are given a fixed Variable: The timing of role transitions is
timetable about when they will move variable.
through stages
Serial: Newcomers observe and get
Disjunctive: Newcomers nust develop their own
training from experienced role
definition of the situation and do not have more
models, who give newcomers a clear senior people to observe.
view of the experiences they will
encounter in the organization
Investiture: Newcomer receive
positive feedback confirming their
prior identity
1
Divestiture1: Newcomers confirms receive
negative feedback expressing organizational
disapproval of their piror identity.
Although van Maanen & Schien originally considered that divestiture (i.e., trying to strip away
newcomers’ prior identity) characterized institutionalized socialization, most subsequent research classifies
Document13/12/2016 8:42 PM
Table N. Six dimension of organizational socialization tactics (Adopted from van Maanen
& Schiene, 1979; Jones, 1986)..
Van Maanen and Schein hypothesized that these tactics would be used in very different
contexts (e.g., that collective socialization would be used in jobs where newcomers
needed to learn technique skills, while individual socialization would be used where
already existing organizational members were being prepared for promotions), and would
have different consequences (e.g., collective socialization would constrain innovation,
while individualistic socialization would allow for role innovation). However, most
subsequent research has not examined these boundary conditions. Instead, it has
developed a simpler picture of the effects of using these socialization tactics. Bauer et
al.'s meta-analysis summarized 70 separate studies of newcomer socialization to
organizations. The original research examined how the organizations’ use of these
socialization tactics shortly after newcomers joined were associated with the newcomers’
adjustment, commitment and performance in the organizations.
The literature reviewed by Bauer et al. shows that both active information-seeking by
newcomers and the use of a more institutional style of socialization, involving collective,
formal, serial, fixed, sequential and investiture socialization tactics led newcomers to do
their jobs better, to be more satisfied with their jobs, to become more committed to the
organization, and to leave the organization less. (See Figure X). The research suggests
these effects of the organizations’ use of institutionalized socialization tactics as well as
newcomers active information seeking has they effects by increasing newcomers
understanding how they should behave in the organization (role clarity), their beliefs that
they could do what was asked of them (self-efficacy) and their beliefs that others in the
organization accepted them (social acceptance).
Figure N. Antecedents and outcomes of newcomer adjustment during organizational
socialization. From Bauer et al (2007).
Although the evidence is strong that institutionalized socialization tactics are effective in
developing commitment and appropriate behavior in conventional organizations, they are
not the norm in online communities. However, socialization processes in most online
communities are informal and individualistic. For example, in Usenet groups, lurking
it as a type of individualized socialization.
Page 25
Document13/12/2016 8:42 PM
newcomers have no opportunities for formal mentorship, because their presence is
unknown to old-timers. Also consider Ducheneaut's description of the socialization of
newcomers to the Python open-source software development community (2004). Even in
this production-oriented environment, with defined workflows and sharp distinctions
among the social roles participating in the project, socialization is still informal, based on
trial and error. For example, although there is a progression of participation in this
community, with newcomers first participating in technical discussion and then
submitting bug fixes before obtaining committer status, this progression of roles is not
documented. When one new developer who was slowly making his way toward the core
of the community attempted to introduce a new module to the standard library used in
this project, he did not know the organizational routines he needed to engage in order to
make his contribution. A core member of the community eventually stepped in to offer
advice (i.e., provided mentorship), but mentoring was not a regular socialization tactic in
this community.
Rarely do online communities socialize and train newcomers as a group in isolation from
other members, provide mentors to give them guidance nor give them a clear timetable
about how to progress in the community. The primary exception to the pattern that
online communities generally adopt individualized socialization tactics is that many use a
variant of collective socialization, by providing standardized FAQs and help documents
to familiarize newcomers with how the community operates. Consider the community
that develops the popular Firefox web browsers. While the Mozilla foundation maintains
an extensive set of forums for communication among developers, except for FAQ-style
documents, we were unable to find forums or other resources for training new comers or
socializing them to the project. Newcomers who want to join this project must struggle
to make sense of how to contribute on their own.
According to the research on organization socialization, however, this type of
socialization experience is not effective. Instead,
newcomers should be exposed to a formal and
collective training experience, with a road map
providing clear and explicit information about roles
and sequence of responsibility, and a mentor to
provide guidance.
The multiplayer, fantasy European football (i.e.,
soccer) game hattrick.org provides an informative
exception to the training of newcomers. Each player is
the manager of a soccer team who competes against
Figure X: Hattrick.org’
other managers. The players must make decisions
Manager’s license
about such areas as acquiring, training and trading
players, game tactics, building stadiums or handling
finances. To encourage newcomers to read the full manual, new managers participate in
the Hattrick Manager License Challenge (aka Hattrick University), where they earn
money for studying the rules and answering challenge questions during their first weeks
in the community. If they successful answer all 24 challenge questions, they graduate and
Page 26
Document13/12/2016 8:42 PM
receive a manager’s license. To encourage the development of cohorts of players, new
players initially play in leagues composed of other new players; if players win their
division, they are promoted to the next level division the following season.
Design claim: By using more institutionalized socialization tactics, in which new
members are provided common training experience in their own environment and given a
roadmap about how to progress in the community, they are likely to become more
committed to the community, learn how to behave in it and contribute more.
Some communities have successfully deployed mentorship practices, both formal and
informal. Everything2 is a edited web publication where members write essays about
almost anything. The Everthing2 Mentor System matches new users with experienced
mentors who share interests and a time zone, who agree to log into the site at least once a
day and who are willing to answer the new members question and critique their work.
New sellers at eBay benefit from Trading Assistants, experienced and active eBay sellers
with at least 97% positive feedback. Trading Assistants assess whether an item is
saleable, plan starting prices and shipping methods for items, and communicate directly
with bidders. Newcomers benefit from the Trading Assistants' high reputation scores,
proficiency with seller tools, and familiarity with listing policies and best practices.
Newcomers search a directory for assistants who are geographically close and have
expertise in particular areas, such as estate liquidation or motor vehicles. Trading
Assistants themselves have training tutorials, guides to best practices and promotion, and
a discussion board. Help from Trading Assistants is not limited to newcomers; any busy
seller can outsource items to others in this way. However, unlike voluntary mentorship in
other communities, eBay's Trading Assistants negotiate fees with new sellers for their
services1.
Design claim: When old-timers provide newcomers formal mentorships the newcomers
will become more committed to the community, learn how to behave in it and contribute
more .
[check open-source communities to see if there's some semi-formal mentoring of new
developers going on; the more formal the better; John]
[how guilds help people get higher levels; emphasize teaching/mentorship; John]
[1] http://en.wikipedia.org/wiki/Wikipedia:Neutral_point_of_view
http://en.wikipedia.org/wiki/Wikipedia:What_Wikipedia_is_not#Wikipedia_is_not_a_blo
g.2C_webspace_provider.2C_social_networking.2C_or_memorial_site
[3] http://en.wikipedia.org/wiki/Wikipedia:Copyrights
[4] http://en.wikipedia.org/wiki/Wikipedia:Manual_of_Style
1
http://pages.ebay.com/tahub/index.html
Page 27
Document13/12/2016 8:42 PM
Problem 5: Protecting the group from newcomers
Although newcomers are essential to the survival of online communities, they also pose
real threats. Because newcomers have no history in the community, existing group
members do not know how much to trust them. Those established community members
need to ask whether to allow a new member of a guild in World of Warcraft to participate
in a high-stakes raid, to buy an expensive item from a new seller on eBay, or whether to
allow a new developer to commit code to the Apache server build. Empirical evidence
suggests that established members indeed distrust newcomers. For example, Resnick et
al. (2006) showed experimentally that buyers paid less for comparable items on eBay
when purchasing from newcomers (i.e., those with no prior transactions) than old-timers,
distrusting them because of their lack of history. Because newcomers have not yet
developed much commitment to the group and have not yet learned how the group
operates, it is rational for established group members to actively distrust them. Because
they don't yet identify with the group, they are less likely to have the best interests of the
group at heart in deciding courses of action. In addition, because they are relatively
unsophisticated in how the group operates, they may not have the skill or knowledge to
operate in the group's best interest, even if they cared to. For example, in Wikipedia,
newcomers (including those who have not registered and those who have not yet edited
extensively) are more likely to vandalize pages or offer changes that other, more
experienced Wikipedians will later delete (Adler, 2007). As a result of this lack of history
and potential lack of goodwill and relevant skills, groups need to protect themselves
against possible damage that newcomers can cause. Even if newcomers are not actively
behaving inappropriately, the mere fact that they are different from the old timers may
change the environment to make the community less desirable for old timers. An influx
of new members to social networking sites may change the culture for old-timers, such as
when MySpace transitioned from a promotion platform for small bands to a crowded
venue for any teenager to post mp3s.
So far, we have suggested that newcomers be socialized through friendly initial
interactions with old-timers, and that old-timers be explicitly discouraged from being
hostile to newcomers who make mistakes. . Yet this is not to say that newcomers should
receive carte blanche access to a group and its resources. Resources include both
people—other members' attention and support—and any artifacts the group produces,
such as wikis, collective movie ratings, or software. Should newcomers be isolated to
prevent annoying other members and damaging community artifacts, or should they be
allowed to ask questions, scribble on others' walls, delete code, and join raiding parties
right away? It depends on the kind of community and its goals.
The question of whether to deploy specific mechanisms depends on several factors,
including whether the community produces a group artifact with interdependent parts and
the consequences of newcomers’ mistakes on themselves and other members. Many
common protection mechanisms serve multiple purposes, not only preventing damage,
but also discouraging newcomers who would be a bad fit and socializing the right ones.
These mechanisms include sandboxes, newbie gardens, progressive access control,
throttles or quotas, and quick reversion tools.
Page 28
Document13/12/2016 8:42 PM
The sandbox achieves dual goals of protection and training. Sandboxes are safe, isolated
areas for exploration and skill development. Sandboxes take many forms, from text boxes
for practicing wiki syntax to virtual land parcels for practicing construction or simple
scripting, as in the virtual world Second Life (see Figure X). Content created in
sandboxes is removed regularly, protecting the community from clutter. All Wikipedia
editors have a personal sandbox by default, as well as access to communal sandboxes.
Typical policies, such as formatting guidelines or notifying other users before making
large changes, do not apply to the sandbox, although civility and copyright policies still
apply. Sandbox content is automatically cleaned every 12 hours, although other users
tend to overwrite content much faster (see Figure Y).
Figure X. Sandbox in Second Life (also check permissions. Image from
http://wiki.secondlife.com/wiki/Sandbox)
Page 29
Document13/12/2016 8:42 PM
Figure Y. Wikipedia Sandbox for practicing wiki formatting
Design claim: Provide sandboxes for newcomers to practice content creation without
cluttering shared spaces.
Similarly, multiplayer online games frequently contain "newbie gardens," safe spaces
where new players practice skills, thwarting players who intentionally harass or kill
unskilled avatars. Like the sandbox, a newbie garden allows exploration, but the goal is
to protect newcomers from their own mistakes, and reduce attrition from early negative
experiences. Newcomers in massively multiplayer online games like World of Warcraft
practice healing, basic avatar control, and farming until they are ready for combat.
Newcomers are not required to participate in these venues, and thus still have
opportunities to demonstrate their desirable attributes to the wider community, but
sandboxes and newbie gardens allow additional practice space.
Design claim: Provide safe spaces for newcomers to practice skills without fear of
harming themselves.
However, sequestering newcomers may contradict one of our previous design claims, that
of providing newcomers with an accurate and complete picture of what they can expect
upon joining. In fact, several 3D communities have had problems with the design of
sandboxes and newbie gardens, particularly when newcomers cannot tell if they’re
actually in the game. The newbie gardens appear similar enough to the full game that less
savvy newcomers may think they’re in the full game but are unable to complete the tasks
they expect, such as killing monsters. Eternal Lands, another MMORPG, stocked a
newbie island with two non-playing characters (NPCs, or scripts) that required
newcomers to accomplish particular goals before they could leave the island. However,
the newcomers did not read the text responses from the NPCs, and couldn’t leave the
island. Responding to user complaints, designers of the community replaced the scripted
Page 30
Document13/12/2016 8:42 PM
island with a larger island with “stuff to kill” and no required tasks for exit.1 Second Life
reported users of its “Orientation Island” couldn’t tell if the Exit link would take them to
the mainland or out of the game entirely. Linden Labs replaced the heavily scripted
Orientation Island with “Help Island,” where newcomers interact with volunteers and
exits are more clearly marked, and Linden Labs reports increased commitment metrics.2
Design claim: In virtual worlds with avatars, make it clear when newcomers are in a
training space, and make exit to the main world clear.
Another common protection mechanism is progressive access control, allowing
newcomers to participate in less critical tasks initially, and gradually allowing them to
take on more central tasks. This progression may simply be a suggestion in FAQs, or may
be a technical limit preventing newcomers from performing risky actions until they have
demonstrated competence with simpler ones. Lave and Wenger (1991) propose the
theory of legitimate peripheral participation, by which newcomers become more
experienced members through small but productive actions in the community. They
describe several apprenticeships, including for midwives, tailors, and butchers, and note
that in all of these cases learning encompasses more than just acquiring knowledge, but
also taking part in social practices and understanding the community’s activities,
artifacts, and language. By technically enforcing newcomers to learn through small,
peripheral, but helpful tasks, progressive access control serves as both a protection
mechanism and socialization tool.
Distributed Proofreaders, a community of volunteers proofreading digitized public
domain books, restricts unregistered users to proofing “Smooth Reading” pages, which
are nearly error-free. These pages have been reviewed in previous rounds and need light
checks by people reading for pleasure, rather than readers trained in the formal
procedures used by experienced checkers. A missed typo in one of these pages will have
little consequence to the community, and will likely be caught by another smooth reader.
Newly registered users may also edit “Beginners Only” pages, and cannot move to more
advanced pages until demonstrating mastery of community editing standards. “Beginners
Only” edits also trigger special reviews by second-level mentors, who email friendly
feedback, sensitive to the tenuous relationship a novice editor may have to the
community. One such review reads: “I have reviewed page 294 that you proofed for this
project. It looks like you’ve made a good start! For example, you joined the hyphenated
words, and you ‘clothed’ the em-dash at the end of a line. Nice job!”
1
http://www.devmaster.net/articles/mmorpg-postmortem/part2.php
2
http://vteamblog.com/2008/08/01/hi-is-the-new-black-officially/
Page 31
Document13/12/2016 8:42 PM
Figure X. Activity hub from Distributed Proofreaders showing “Beginners Only” pages.
Design claim: Allow newcomers to perform peripheral, less risky tasks that are of value
to the community initially, and increase access as they demonstrate mastery.
Throttles/quotas: Similar to progressive access control, but not intended as a situated
learning tool—instead, just prevent large-scale damage by newcomers who may be
spammers/trolls/phishers. Somewhat covered in selection section, but verifying newbie’s
identity. Not someone showing up once to abuse the group, phish others’ identities,
create free email accounts, etc. In that case protection is more at the selection level, but
also means potentially limiting access until trustworthiness is established (e.g.
rel=nofollow tags in blog comments, weighting algorithms at amazon, ebay? )
Facebook has several features designed to limit potentially abusive or annoying
behaviors, both by newcomers and more experienced members who may have a
compromised account. When it detects unusual speed or frequency of a behavior, such as
friending or wall posting, it sends a warning to the user.1
Similarly, many blog platforms, including Blogger and WordPress, while allowing
newcomers to post comments (subject to the blog owner's preferences), automatically
include the "rel=nofollow" attribute in links embedded in comments. This mechanism
directs search engines not to trust these links, preventing spam links from receiving
PageRank, and thus discouraging spammers from disguising links to their products within
blog comments. Slashdot also uses the nofollow attribute in some potentially
misbehaving users' comments, using heuristics based on the age of the user's account and
the user's karma2. These invisible mechanisms do not prevent newcomers from
participating; they simply limit the amount of damage a potentially malicious newcomer
could do.
1
2
http://www.new.facebook.com/help.php?page=421
http://en.wikipedia.org/wiki/Spam_in_blogs
Page 32
Document13/12/2016 8:42 PM
Design claim: Something about limiting repetitious behavior (overzealous newbie
looking like a spammer)
Finally, rollback mechanisms . . . Wikipedia rollback. Distributed Proofreaders allows
newbies to fix their own errors (if in the last 5 pages they proofed). Depends on
interdependence of tasks (note that OSS is designed to be modular so that changes can be
rolled back without hurting the rest of the repository). It is easy to revert to a prior
version of an article using MediaWiki's history mechanism. Compare this to OSS
development, with relatively high interdependence between contributions and a
somewhat more difficult reversion process through typical version control systems. More
importantly, there is simply less interdependence among the elements in a natural
language article than in a software program. Readers can interpret around a misspelling
introduced into a encyclopedia article, while the compiler cannot do this to a misspelled
variable name in a program. (note Reviewer’s comment about mechanisms in open vs.
closed-source software development. OSS is designed for legitimate peripheral
participation by being modular
Design claim: Rollback mechanisms . . .
References
Adler, B. T. (2007). A content-driven reputation system for the Wikipedia. Proceedings
of the 16th international conference on World Wide Web, 261-270.
Akerlof, G. A. (1970). The market for" lemons": Quality uncertainty and the market
mechanism. The Quarterly Journal of Economics, 84(3), 488-500.
Allen, D. G. (2006). Do organizational socialization tactics influence newcomer
embeddedness and turnover? Journal of Management, 32(2), 237-256.
Arguello, J., Butler, B. S., Joyce, L., Kraut, R., Ling, K. S., Rosé, C. P., et al. (2006).
Talk to me: Foundations for successful individual-group interactions in online
communities. In CHI 2006: Proceedings of the ACM conference on human-factors
in computing systems (pp. 959 - 968). New York: ACM Press.
Aronson, E., & Mills, J. (1959). The effect of severity of initiation on liking for a group.
Journal of Abnormal & Social Psychology, 59, 177-181.
Assmus, G., Farley, J., & Lehmann, D. (1984). How advertising affects sales: Metaanalysis of econometric results. Journal of Marketing Research, 65-74.
Bauer, T. N., Todd Bodner, {wrong format} Tucker, J. S., Erdogan, B., & Truxillo, D.
Page 33
Document13/12/2016 8:42 PM
M. (2007). Newcomer adjustment during organizational socialization: A metaanalytic review of antecedents, outcomes, and methods. Journal of Applied
Psychology, 92(3), 2007, Vol. 2092, No. 2003, 2707–2721.
Benoit, W., Hansen, G., & Verser, R. (2003). A meta-analysis of the effects of viewing
US presidential debates. Communication Monographs, 70(4), 335-350.
boyd, d. (2006). Identity production in a networked culture: Why youth heart MySpace.
American Association for the Advancement of Science. February 19, 2006.
http://www.danah.org/papers/AAAS2006.html
Burke, M., Marlow, C., & Lento, T. (2009). Feed me: motivating newcomer contribution
in social network sites CHI'09: Proceedings of the ACM Conference on Human
Factors in Computing Systems (pp. 945-954). New York, NY, USA: ACM
Butler, B., Sproull, L., Kiesler, S., & Kraut, R. (2007). Community effort in online
groups: Who does the work and why? In S. Weisband. (Ed.), Leadership at a
distance. (pp. 171-194). Hillsdale, NJ.: Lawrence Erlbaum Associates.
Butler, B., Sproull, L., Kiesler, S., & Kraut, R. (2007). Community effort in online
groups: Who does the work and why? In S. Weisband. (Ed.), Leadership at a
distance. (pp. 171-194). Hillsdale, NJ.: Lawrence Erlbaum Associates.
Cable, D. M., & Parsons, C. K. (2001). Socialization Tactics And Person-Organization
Fit. Personnel Psychology, 54(1), 1-23.
Cialdini, R. (2003). Crafting normative messages to protect the environment. Current
directions in psychological science, 12(4), 105-109.
Cialdini, R. B., & Goldstein, N. J. (2004). SOCIAL INFLUENCE: Compliance and
Conformity. Annual Review of Psychology, 55(1), 591-621.
Coleman, J., Katz, E., & Menzel, H. (1957). The diffusion of an innovation among
physicians. Sociometry, 20(4), 253-270.
Collins, N., & Miller, L. (1994). Self-disclosure and liking: A meta-analytic review.
Psychological Bulletin, 116(3), 457-475.
Dawkins, R. & Krebs, J. R. 1978: Animal signals: information or manipulation? in
Behavioural Ecology: an evolutionary approach 1st ed. (Krebs, J. R. & ,Davies,
N.B., eds) Blackwell: Oxford, pp 282-309.
Domingos, P. and Richardson, M. 2001. Mining the network value of customers. In
Proceedings of the Seventh ACM SIGKDD international Conference on Knowledge
Discovery and Data Mining (San Francisco, California, August 26 - 29, 2001). KDD
'01. ACM, New York, NY, 57-66. DOI= http://doi.acm.org/10.1145/502512.502525
Ducheneaut, N. (2005). Socialization in an open source software community: A sociotechnical analysis. Computer Supported Cooperative Work, 14, 323-368.
Ducheneaut, N., Yee, N., Nickell, E., & Moore, R. J. (2007). The life and death of online
gaming communities: A look at guilds in world of warcraft. In Proceedings of the
sigchi conference on human factors in Computing systems (pp. 839-4848). New
York, NY: ACM Press.
Page 34
Document13/12/2016 8:42 PM
Esteban, L., Gil, A., & Hernandez, J. (2001). Informative advertising and optimal
targeting in a monopoly. Journal of Industrial Economics, 161-180.
Festinger, L. (1957). A theory of cognitive dissonance: Stanford University Press.
Festinger, L., & Carlsmith, J. M. (1959). Cognitive consequences of forced compliance.
J Abnorm Psychol, 58(2), 203-210.
Fogg, B.J., Eckles, D.: The Behavior Chain for Online Participation: How Successful
Web Services Structure Persuasion. In: Persuasive Technology 2007. LNCS, pp.
199–209. Springer, Heidelberg (2007)
Fussell, S., & Setlock, L. (2003). Informal communication in an online volunteer
community: Implications for supporting virtual relationships. Unpublished
Manuscript, Human-computer Interaction Institute, Carnegie Mellon University,
USA.
Gerard, H. B. & Mathewson, G. C. (1966). The effects of severity of initiation on liking
for a group: a replication. Journal of Experimental Social Psychology, 2, 278-287.
Goldstein, N., Cialdini, R., & Griskevicius, V. (2008). A room with a viewpoint: Using
social norms to motivate environmental conservation in hotels. Journal of Consumer
Research, 35(3), 472-482.
Grammer, K., & Thornhill, R. (1994). Human (homo sapiens) facial attractiveness and
sexual selection: The role of symmetry and averageness. Journal of Comparative
Psychology, 108(3), 233-242.
Green, D., Gerber, A., & Nickerson, D. (2008). Getting out the vote in local elections:
Results from six door-to-door canvassing experiments. The Journal of Politics,
65(04), 1083-1096.
Guadagno, R., & Cialdini, R. (2005). Online persuasion and compliance: social
influence on the Internet and beyond The Social Net: Understanding human
behavior in cyberspace (pp. 91-114). New York: Oxford University Press.
Hoag, D. (1996, January). The well host manual. Version 4.4. Retrieved August 11,
2003, from http://www.well.com/~confteam/hostmanual/
Hoag, J. D., & Williams, G. A. (1996, Jan). The W@ELL Host Manual Available from
F:\0_articles\Hoag96-TheWELLHostsManual.mht
Hollenbeck, J. R. (2000). A structural approach to external and internal person-team fit.
Applied Psychology: An International Review, 49(3), 534-549.
Honeycutt, C. (2005). Hazing as a process of boundary maintenance in an online
community. Journal of Computer-Mediated Communication, 10(2), np.
Jastive. (2006, March 29). Eric raymond and the rtfm jerks. Retrieved Oct 22, 2007,
from http://www.libervis.com/blogs/15/Jastiv/eric_raymond_and_the_rtfm_jerks
Jones, G. R. (1986). Socialization tactics, self-efficacy, and newcomers' adjustments to
organizations. The Academy of Management Journal, 29(2), 262-279.
Page 35
Document13/12/2016 8:42 PM
Kempe, D., Kleinberg, J., and Tardos, É. 2003. Maximizing the spread of influence
through a social network. In Proceedings of the Ninth ACM SIGKDD international
Conference on Knowledge Discovery and Data Mining (Washington, D.C., August
24 - 27, 2003). KDD '03. ACM, New York, NY, 137-146. DOI=
http://doi.acm.org/10.1145/956750.956769
Kittur, A., Chi, E., Pendleton, B. A., Suh, B., & Mytkowicz, T. Power of the few vs.
Wisdom of the crowd: Wikipedia and the rise of the bourgeoisie. World Wide Web,
1(2), 19.
Kittur, A., Chi, E., Pendleton, B. A., Suh, B., & Mytkowicz, T. Power of the few vs.
Wisdom of the crowd: Wikipedia and the rise of the bourgeoisie. World Wide Web,
1(2), 19.
Klapper, J. (1960). The effects of mass communication: Free Press, Glencoe, Ill.
Kristof-Brown, A. L., Zimmerman, R. D., & Johnson, E. C. (2005). Consequences of
Individuals' Fit at Work: A Meta-Analysis of Person-Job, Person-Organization,
Person-Group, and Person-Supervisor Fit. Personnel Psychology, 58(2), 281-320.
Krogh, G. v., Spaeth, S., & Lakhani., K. R. (2003). Community, joining, and
specialization in open source software innovation: A case study. Research Policy
Special Issue On Open Source Software Development, 32(7), 1217-1241.
Lakhani, K. R., & Hippel, E. v. (2003). How open source software works!"free" user-touser assistance. Research Policy (Special Issue on Open Source Software
Development), 32, 923-943.
Lampe, C. and Johnston, E. Follow the (slash) dot: Effects of feedback on new members
in an online community. In Proc. GROUP 2005, ACM Press (2005), 11-20.
Lodewijkx, H. F. M., & Syroit, J. (2001). Affiliation during naturalistic severe and mild
initiations: Some further evidence against the severity-attraction hypothesis. Current
Research in Social Psychology, 6(7), 90-107.
Lodewijkx, H. F. M., & Syroit, J. E. M. M. (1997). Severity of initiation revisited: Does
severity of initiation increase attractiveness in real groups? European Journal of
Social Psychology, 27(3), 275-300.
Mann, T., & Wolfinger, R. (1980). Candidates and parties in congressional elections.
The American Political Science Review, 617-632.
Maynard Smith, J. 1994: Must reliable signals always be costly? Animal Behaviour 47,
1115-1120.
McGuire, W., Lindzey, G., & Aronson, E. (1985). Attitudes and attitude change.
Handbook of social psychology: Special fields and applications, 2, 233–346.
Mocan, N. H. (2001). Can consumers detect lemons? Information asymmetry in the
market for child care: SSRN.
Mockus, A., Fielding, R. T., & Herbsleb, J. D. (2002). Two case studies of open source
software development: Apache and mozilla. ACM Transactions on Software
Engineering and Methodology, 11(3), 309-346.
Page 36
Document13/12/2016 8:42 PM
Mockus, A., Fielding, R. T., & Herbsleb, J. D. (2002). Two case studies of open source
software development: Apache and mozilla. ACM Transactions on Software
Engineering and Methodology, 11(3), 309-346.
Mortensen, M., & Hinds, P. (2002). Fuzzy teams: Boundary disagreement in distributed
and collocated teams. In P. Hinds & S. Kiesler (Eds.), Distributed Work (pp. 284–
308). Cambridge, MA: MIT Press.
Nickerson, R. (1998). Confirmation bias: A ubiquitous phenomenon in many guises.
Review of General Psychology, 2, 175-220.
Organ, D. W., & Ryan, K. (1995). A meta-analytic review of attitudinal and
dispositional predictors of organizational citizenship behavior. Personnel
Psychology, 48(4), 775-802.
Petty, R., & Wegener, D. (1998). Attitude change: Multiple roles for persuasion
variables. The handbook of social psychology, 1, 323–390.
Phillips, J. (1998). Effects of Realistic Job Previews on Multiple Organizational
Outcomes: A Meta-Analysis. Academy Of Management Journal, 41, 673-691.
Preece, J., & Shneiderman, B. (2009). The Reader-to-Leader Framework: Motivating
technology-mediated social participation. AIS Transactions on Human-Computer
Interaction, 1(1), 13-32.
Resnick, P., Zeckhauser, R., Swanson, J., & Lockwood, K. (2006). The value of
reputation on eBay: A controlled experiment. Experimental Economics, 9(2), 79101.
Roberts, D., & Maccoby, N. (1985). Effects of mass communication. In G. Lindzey & E.
Aronson (Eds.), Handbook of social psychology (Vol. 2, pp. 539-598). New York:
Random House.
Saks, A. M., Uggerslev, K. L., & Fassina, N. E. (2007). Socialization tactics and
newcomer adjustment: A meta-analytic review and test of a model. Journal of
Vocational Behavior, 70, 413-446.
Schopler, J. & Bateson, N. (1962). A dependence interpretation of the effects of a severe
initiation. Journal of Personality, 30, 633-649.
Spence, A.M. (1973) Job Market Signaling, Quarterly Journal of Economics. 87:355374.
Swann, W. B., & Read, S. J. (1981). Acquiring self-knowledge: The search for feedback
that fits. Journal of Personality and Social Psychology, 41(6), 1119–1128.
Thorndike, E. (1920). A constant error in psychological ratings. Journal of Applied
Psychology, 4(1), 25–29.
Van Maanen, J., & Schein. (1979). Towards a theory of organizational socialization. In
B. M. Staw (Ed.), Research in organizational behavior (Vol. 1). Greenwich, CT: JAI
Press.
Page 37
Document13/12/2016 8:42 PM
WoWWiki (2008, Feb 28, 2008). Leeroy Jenkins (video) Retrieved Dec 31, 2008, from
http://www.wowwiki.com/Leeroy_Jenkins_(video)
Page 38
Document13/12/2016 8:42 PM
Page 39
Leftovers:
Theories
Design goals
S: Selection C: Commitment B: Behavior
1. Signaling
Theory
Contact
Barriers
2. Dissonance
Theory
(Theory
doesn't
address)
Barriers
(Theory doesn't
(Theory doesn't
address)
address)
Contact
Barriers
Contact
Welcoming
FAQs
(Theory doesn't
address, but
might be side
effects.)
3. Group
socialization
Theory (M&L)
Protection
(Theory doesn't (Theory doesn't Use signals to
address)
address)
show who to
trust
4. Organizational
socialization
Tactics Theory
(van Maanen &
Schein)
NA: Theory
assume
members
already
selected
Contact
Welcoming
FAQs
(Explicit
training, FAQs
& mentorship
increase
likelihood of
newcomer
learning how to
behave
appropriately )
5. Rewards &
deterrence
?
?
(Theory should
Impose
address this)
penalties
6. Theories of trust
Provide training
to show
newcomers how
to behave
Keep out those
you don't trust
Key: Pink=Moira, Blue=John, Yellow=Bob
(Remove this key and colors before sharing with C-Lab)
CHESS cancer group, moderator welcomes newbies. Acts more like a mentor
Barriers: [do MMOs make you do stupid things before you're accepted; find guilds that
haze?]
Document13/12/2016 8:42 PM
[hurdles in open source; perhaps groups are not relying on consistency theory]
Barriers to entry
Small-group socialization theory suggests that these barriers be low, so that the exchange
of information that allows the group and newcomers to evaluate each other can begin
early. Thus, barriers that prevent prospective members from investigating the group
reduce the likelihood that they will ever attempt to join. For example, in the online
cancer support group ACOR.org, readers cannot search or browse archived messages
without a subscription. Though subscribing is free, prospective subscribers are vetted by
the list owners, which delays the newcomer's opportunity to evaluate the group and his or
her expected fit. Furthermore, the archives are hidden from search engines. Though this
protects the privacy of existing members, it also reduces the likelihood that desirable
members will find the group.
While signaling theory and dissonance theory are unidirectional, with signaling theory
emphasizing the information that a test provides to the group about prospective members
and dissonance theory emphasizing changes in motivations of prospective members
toward the group, small-group socialization theory is bi-directional, simultaneously
considering the effects of barriers on choices that both the group and the individual make.
Barriers should be valuable only to the extent that they provide the group information
with which to realistically assess prospective members (akin to signaling theory) and for
prospective members to realistically assess benefits they might receive from the group if
they were to become members. Thus, while signaling theory would recommend that
groups impose tests on prospective members that only desirable members can easily pass,
socialization theory would recommend that these tests be conducted in the context of
realistic job previews, so potential members could select to join only communities where
they perceived a good fit with their interests, values and skills. [Add example from Bo's
research about matches between guilds and individuals on social/task-focus]
[1] http://en.wikipedia.org/wiki/Wikipedia:Please_do_not_bite_the_newcomers
[1] http://en.wikipedia.org/wiki/Wikipedia:Welcoming_committee
[2] http://www.well.com/~confteam/hostmanual
[3] http://www.libervis.com/blogs/15/Jastiv/eric_raymond_and_the_rtfm_jerks
Punishment:
While Wikipedia's "Don't bite the newcomer" policy poses a forgiving stance toward
newcomers' foibles, other communities show less leniency, warning or punishing
newcomers early to prevent future misbehavior. At JoBlo's Movie Emporium (check
Page 40
Document13/12/2016 8:42 PM
this) newcomers who post links to other communities in their first posts are treated with a
barrage of angry messages to their personal email accounts. Vandals in Wikipedia are
treated to a series of increasingly assertive warnings, ranging from a polite notice:
Welcome to Wikipedia. Although everyone is welcome to make constructive
contributions to Wikipedia, at least one of your recent edits, such as the one
you made to Article, did not appear to be constructive and has been reverted
or removed. Please use the sandbox for any test edits you would like to make,
and take a look at the welcome page to learn more about contributing to this
encyclopedia. Thank you.
to a threat of blocking:
This is the only warning you will receive for your disruptive edits.
If you vandalize Wikipedia again, as you did to Article, you will be blocked
from editing.
[need more examples of punishment directed at newcomers and what theory says about
this, or else refer readers to another chapter]
Page 41
Document13/12/2016 8:42 PM
Page 42
Candidate’s initial impression
of community & role
Organization’s information
provision
 Channel
 Amount
 Accuracy
Candidate’s information
gathering
Accuracy and completeness of
expectations about community &
candidate’s role in it
Candidate’s characteristics
compared to what
community wants/needs
Candidate’s needs/wants
compared to what
community offers
Candidate’s &
community’s selection
decisions
Fit between
community’s
wants/needs and
candidate’s attributes
Fit between
candidate’s
needs/wants what
community offers
Newcomer’s
satisfactory
performance in
community
Newcomer’s value
attainment
Mutual satisfaction
Retention
Figure Error! No text of specified style in document.-2: A Process Model of Formation
of Job Expectations and Their Influence on Recruiting Outcomes (adapted from Breaugh
& Starke, 2000).
Document13/12/2016 8:42 PM
Page 43
From Norms chapter:
There are two main ways to induce people to feel they are members of the same group
and to care about it. One approach is to recruit a homogeneous set of people who fit the
intended community profile. We discuss recruiting community members in the chapter on
newcomers, but raise it again here because recruiting is relevant to the promotion of
group identity as a means of regulating behavior. For instance, alt.hackers requires
members to perform a skilled task for entry into the group. This recruiting strategy
separates the “ingroup” from the “outgroup” and increases group identity (even though,
otherwise, this community is not very sociable).
Other communities, such as those for medical doctors, require members to prove they are
doctors. Many exclusive BitTorrent tracker sites (groups that provide private BitTorrent
seeds) require existing members to vouch for new members. Bad behavior on the part of
a new member (such as downloading much more than they upload) can result in sanctions
both to the new member and to the sponsor.
Still other communities filter new members by requiring them to be invited by existing
members, and limiting the invitations each existing member gets. Invitations to Google’s
exclusive Gmail Beta were so highly coveted that some users put them up for bid on
eBay (http://www.news.com/2100-1023_3-5203162.html). A number of Flickr photo
groups have requirements that users’ photos have awards or have been marked as
favorites by some number of other users. (Offline companies have started to use this
technique to recruit new employees as well; for example, Google has posted recruitment
billboards whose web address is hidden in the solution to a complex mathematical
question.
Include Drenner, S., Sen, S., & Terveen, L. (2008). Crafting the initial user experience to
achieve community goals Proceedings of the 2008 ACM conference on
Recommender systems (pp. 187-194): ACM New York, NY, USA.
Technique Description
(van Maanen
& Schein)
Sample of measurement items (from
Jones, 1986)
Collective
versus
individual
C13 This organization puts all
newcomers through the same set of
learning experiences.
C14 Most of my training has been
carried out apart from other newcomers.
(R)
Newcomers go through a common
set of experiences designed to
produce standardized responses to
situations, versus, each newcomer
receives unique training in isolation
from others.
Document13/12/2016 8:42 PM
Formal vs.
informal
Newcomers are segregated from
other organizational members as
they learn their jobs, versus,
newcomers receive on-the-job
training.
Page 44
FIl I have been through a set of training
experiences which are specifically
designed to give newcomers a thorough
knowledge of job-related skills.
F14 Much of my job knowledge has
been acquired informally on a trial and
error basis. (R)
Serial versus With serial tactics, newcomers
disjunctive
observe and get training from
experienced role models, who give
newcomers a clear view of the
experiences they will encounter in
the organization, while with
disjunctive tactics newcomers do not
have more senior people to observe.
SD2 I am gaining a clear understanding
of my role in this organization from
observing my senior colleagues.
SD3 I have received little guidance from
experienced organizational members as
to how I should perform my job. (R)
Fixed versus With fixed tactics, newcomers are
variable:
given a fixed timetable about how to
progress in the organization, while
with variable tactics, the timing of
role transitions is variable.
FV1 I can predict my future career path
in this organization by observing other
people's experiences.
FV4 I have little idea when to expect a
new job assignment or training exercise
in this organization. (R)
Sequential
versus
random:
With sequential socialization tactics,
newcomers are given a clear
sequence of steps leading to their
ultimate role, while with random
socialization the sequence of stages
isn't known in advance.
SRI There is a clear pattern in the way
one role leads to another or one job
assignment leads to another.
SR4 This organization does not put
newcomers through an identifiable
sequence of learning experiences. (R]
With investiture socialization
tactics, the organization confirms
newcomers' prior identity, while
with divestiture tactics the
organization tries to strip this prior
identity away.
ID1 I have been made to feel that my
skills and abilities are very important in
this organization.
ID3 I have had to change my attitudes
and values to be accepted in this
organization. (R)
Investiture
versus
divestiture:
Research on organizational recruitment makes a crisp distinction between recruits, who
are not members of an organization, and employees, who are. In many online
communities the distinction between non-member and member is more blurred. Theories
of socialization to groups (Levine & Moreland, 19xx) and to organizations (Van Maanen,
J., & Schein. 1979) are more likely to take into account the continuous nature of
commitment an individual might have to a group, organization or community or the
Document13/12/2016 8:42 PM
degree to which are they peripheral or central to it. For example, Levine and Moreland's
socialization theory (19xx) emphasizes separable phases in their relationship with a group
through which individuals can pass (investigation, socialization, maintenance,
resocialization, and remembrance). These phases are separated by four role transitions
(entry, acceptance, divergence, and exit). In understanding how to bring newcomers on
board, the first two of these phases -- investigation and socialization -- are most
important. In this section we concentrate on the investigatory phase, in which newcomers
and the groups assess each other. If successful, the investigatory phase ends in entry, in
which an outsider becomes a member of the community. In many formal conventional
groups and organizations, the distinction between outsider and group member is clear,
and is analogous to the recruit/employee distinction in the recruitment literature. Thus, an
applicant can become a member of a fraternity, enlist in the military, be admitted to a
university, or be hired by a company. All of these organizations have formal points at
which the member joined.
However, even in conventional groups and organizations, the distinction is sometimes an
ambiguous one and entry into the group or organization can be a matter of degree, rather
than a discrete event. When groups have little structure and diffuse boundaries (or low
"entitivity" in the jargon of social psychology), whether one is a member is a matter of
degree. Thus, one’s centrality to friendship cliques or even project groups at work can be
a matter of degree (Mortensen & Hinds, 2002). In academia, for example, those who hold
joint appointments and courtesy appointments in an academic department are often seen
as less legitimate members than those whose whole salary is paid by the department in
question and who have voting rights there. The same variability of the degree to which
boundaries are well defined exists in online communities. For example, membership in a
Usenet group may consist simply of reading posts occasionally, membership in a World
of Warcraft guild involves applying to and being accepted by the guild master and
existing guild members, and membership in Wikipedia projects is somewhere in between
[1].
Page 45