A Framework for Recommender Systems in Online Social Network

Proceedings of the 47th Hawaii International Conference on System Sciences - 2014
A Framework for Recommender Systems in Online Social Network
Recruiting: An Interdisciplinary Call to Arms
Ricardo Buettner
FOM University of Applied Sciences, Institute of Management & Information Systems, Germany
ricardo.buettner@fom.de
Abstract
I sketch an interdisciplinary framework for recommender systems searching online social networks for
future employees. In contrast to previous approaches
my framework covers the whole person-organization
environment (P-OE) fit, and so it also includes crucial
social components (e.g. workgroup fabric, and organizational culture). In a rudimentary way I show how
information extracted from online social networks can
be used to determine the P-OE fit. Due to a more
accurate way of calculating the P-OE fit my framework
facilitates higher quality recommendations. Furthermore my framework enables a better understanding of
the P-OE fit problem.
1. Introduction
Human Resource Management (HRM) has seen a
recent influx of electronic support tools and systems
[1]–[3]. In this paper I focus on e-recruiting as one of
the most important electronic-HRM (e-HRM) activities
[3]. Existing e-recruiting Recommender Systems (RS)
mainly focus on the problem of matching job demands
and candidates’ abilities. For the person-organizational
environment fit (P-OE fit)1 highly relevant social interactions between the potential employee and future
colleagues at work (e.g. [4]–[18]) are not sufficiently
taken into account. To predict such interactions various
inputs are required (e.g. the organizational culture, the
workgroup’s social communication styles and means,
a candidate’s personality and role behavior). A very
promising development in e-HRM is the incorporation
of Online Social Networks (OSN). During the last
years electronic recruiting in OSN (OSN-Recruiting)
– envisioned a quarter of a century ago by Macdonald
[19] – has emerged as a field of scientific research
1. The P-OE fit concerns all aspects of the fit between a job
candidate and a vacancy within the recruiting organization, cf. 2.
[20]–[22]. But even these approaches only address the
matching problem of job demands and a candidate’s
abilities (person-job fit), which has become known
as the expert finding problem. However, rating the
person-job fit is just one part of the whole P-OE
fit, e.g. [12,14]. Other crucial evaluation parts such
as the matching of the applicants’ personality to the
organizations’ culture or the candidates’ role behavior
to the workgroup fabric are largely ignored by both
Information Systems (IS) research as well as practical
e-recruiting solutions.
However, the underlying problems of these currently
widely ignored social matching phenomena receive
much more attention in sociology, psychology and their
related disciplines. Thus, incorporating sociological
studies such as Social Capital Theory [23]–[27] and
Social Network Analysis (SNA) [28]–[33], psychosocial studies on group dynamics [4]–[9], psychological
investigations of personality traits and Internet/OSN
use [34]–[46] as well as cultural influences on Internet/OSN use [47]–[50], combining human resource
management and occupational and vocational studies
[10,12]–[15,51]–[57] into IS research is promising.
In this conceptual paper I sketch an interdisciplinary
framework for OSN-Recruiting based on an extensive,
though by no means complete, review of the available
literature. Implementing this framework will facilitate
recommendations to be made to recruiters. Furthermore such a framework allows deeper interdisciplinary
investigations of the P-OE fit problem, which is longstanding and well-known in HRM.
The rest of the paper is organized as follows: Next I
study candidate evaluation and selection from a HRM
perspective in order to analyze the conditions for the
best P-OE fit. Then I show how relevant data can be
extracted from OSN, before I take a preparatory step
for designing OSN-Recruiting RS by mathematically
formalizing the P-OE fit. I conclude with a discussion
on the potential, limitations and the further applications
of my framework.
This work is a preprint version. Copyright is held by IEEE.
Proceedings of the 47th Hawaii International Conference on System Sciences - 2014
2. Candidate evaluation and selection
A review of current empirical work on e-HRM can
be found in [1], and the most recent theory-based erecruiting review has been provided by Maurer and
Cook [3]. In the latter study, the authors emphasized
the key role of the P-OE fit in order to recruit high
quality job applicants. Meta-analysis investigations
showed significant positive correlations between the POE fit and job performance, job satisfaction, organizational commitment and employee turnover [12]–[17].
The P-OE fit can be broken-down into sub-fits, with the
most appropriate consisting of three all but disjointed
sub-fits, which together almost cover the whole notion
of the P-OE fit [14]. These sub-fits are the PersonOrganization (P-O) fit (between candidate personality
and organizational culture) [12], the Person-Group (PG) fit (matching of individual and group roles and
interactions) [9] and the Person-Job (P-J) fit (between
a candidate’s skills, knowledge, and abilities and job
demands) [11]. An overview of investigations into
these and further P-OE sub-fits can be found in [14].
Thus we can formulate the following:
P-O fit
`
P-G fit
`
P-J fit
ù
P-OE fit .
(1)
The relative importance of these sub-fits depends on
the concrete vacancy. Kristof-Brown [51] empirically
showed that recruiters actually distinguish between
these sub-fits. Sekiguchia and Huber [58] recently
investigated the weighting of the P-O fit and the
P-J fit when hiring decision-makers to evaluate job
candidates. For instance the longer a recruiter expects
a candidate to stay within the organization the more
important the P-O fit becomes. The shorter the envisioned stay the more important the P-J fit becomes,
since skill and knowledge acquisition on the job become inefficient for relatively short stays within the
organization. For a formalization see Eqn. (2).
2.1. Person-organization fit
The P-O fit describes the macro-perspective compatibility between the employee’s personalities and the
organization’s culture they work in [14, pp. 285]. To
measure the P-O fit several tests have been developed,
e.g. [12]. The importance of such tests for hiring
organizations has been noted in [54, p. 21], and further
research has been surveyed in [13]–[15,51,55]. Early
studies of the P-O fit focused on the personalityclimate congruence, while current investigations emphasize the value congruence [12,13,56]. According to
these and other studies [10,55] employees are most
successful in organizations with a culture which is
compatible with their personalities. Hence measuring
the P-O fit requires (a) an analysis of the candidate’s
personality and (b) an investigation of the organizational culture:
Individual personality (part a) indicates a stable
pattern of behavior [55]. Numerous approaches for
modeling and measuring personality (traits) exist. The
Five Factor Model (FFM) refined by Costa and McCrae [59] has been most widely and successfully used
[60]. Within the FFM five personality traits (openness
to experience, conscientiousness, extraversion, agreeableness, and neuroticism) describe an individual’s
personality and several ways to measure these traits
have been developed. The inventories NEO-FFI-3 and
NEO-PI-R [59] are the most widely adopted.
To assess the organizational culture (part b) several
instruments have been developed and a current literature review of instruments for exploring and measuring
organizational culture is provided by Jung et al. [61] in
which the authors identify 79 established instruments,
of which 48 assess psychometric information (e.g. Hofstede’s Culture Measures [62] and the Organizational
Culture Profile (OCP) by O’Reilly et al. [63]).
2.2. Person-group fit
Interpersonal interactions within a workgroup (e.g.
subordinates, coworkers and supervisors) are in certain
circumstances more important to team performance
than technical job skills [7,8]. The P-G fit concerns
social interaction aspects of the matching of a job
candidate to the workgroup (meso-perspective). This
matching has two parts, a supplementary one (similar
qualities of group members) and a complementary one
(distinctive qualities or characteristics) [9]. Measuring
the P-G fit requires (a) an analysis of the candidate’s
social interaction characteristics and (b) an investigation of the group’s social fabric.
Individual interaction characteristics are partly determined by personality traits [6], and these have
been used to forecast team performance directly, cf.
[6,7]. Mount et al. [6] showed by conducting a metaanalysis that agreeableness, emotional stability, and
conscientiousness are positively correlated with team
performance. Kristof-Brown et al. [53] revealed that
teams are more attracted to an individual with an
extraversion score that is complementary to the groups’
extraversion score (i.e. high individual-low team or low
individual-high team levels).
Werbel and Johnson proposed a process to analyze
the workgroup [9, pp. 234]: (1) Investigate group interaction processes (e.g. time orientation, styles and
means of communication: e-mail, phone or meetings)
This work is a preprint version. Copyright is held by IEEE.
Proceedings of the 47th Hawaii International Conference on System Sciences - 2014
and furthermore identify critical group norms (e.g.
individualism or collectivism [62]). (2) The next step
is to investigate the performance of the team already in
place, as this allows the defining of the work role(s) to
be filled in order to improve team performance (complementary aspects). Critical task roles within a team
such as initiator-contributor, coordinator, elaborator,
information seeker or information giver, and critical
maintenance roles such as compromiser, encourager
or follower were delineated by Benne and Sheats
[4]; boundary spanning roles such as scout, ambassador, sentry and guard were identified by Ancona
and Caldwell [5]. (3) The last step is to calculate
the P-G fit using both the supplementary (1) and the
complementary (2) P-G fit.
Organization (macro)
Group (meso)
Person
(micro)
Established link
Potential new link
Figure 1: Macro-, meso- and microperspective on the P-OE fit in OSN-Recruiting
2.3. Person-job fit
The P-J fit generally describes the micro-perspective
matching between a candidate’s characteristics (such as
abilities or preferences) and job demands/supplies of
the recruiting organization. Edwards [11] emphasized
two basic conceptualizations of the P-J fit. The socalled demands-abilities fit compromises the matching
of a candidate’s Skills, Knowledge and Abilities (SKA)
on the one side and job requirements on the other side.
The often labeled needs-supplies or supplies-values fit
refers to the question of whether the employee’s needs,
desires or preferences are met by her job. The last type
of fit is mainly discussed within various theories of
adjustment, well-being, and satisfaction [14, pp. 284].
Since these two conceptualizations can be seen as two
sides of the same coin, I will only focus here on the
demands-ability fit.
To measure the P-J fit several tests have been
developed (e.g. P-J fit scale, Global Self-Report Measure [52]). Following [12, p. 8] the P-J fit should be
evaluated relative to the job tasks and not relative to
the employing organization. Hence, besides the most
widely used inventories – which are questionnairebased – direct examinations of a candidate’s CV, testimonials and letters of reference together with the job
description – provided by the recruiting organization –
can also be used to calculate the P-J fit.
In summary, I investigated in the above section the
conditions for the best fit between a job candidate and
the recruiting organization and identified the triad of
P-O fit, P-G fit and P-J fit as crucial. These fits correspond to the well-known macro-, meso- and microperspective of SNA (Fig. 1).
3. Information extraction from online social networks
In order to infer the job candidate’s personality traits
and SKA, the social fabric of the workgroup and the
organizational culture, the relevant data about OSN
users, their social environments and their interactions
have to be extracted. Recent work (e.g. [64]) shows
that the information stored in large-scale OSN (e.g.
LinkedIn) is more truthful compared to simple manually generated/maniplulated CVs.
3.1. Aspects of social network analysis
To address the problem of assessing the integration
of a person p in a social network (SN), connectivity and
centrality measures cf. [31,65,66] have been studied
intensively. One of the most basic measures to assess
the importance of a node is the degree centrality measure of Shaw [28]. The degree centrality of a node n is
defined as the number of neighboring nodes. The closeness centrality measure [29,30] focuses on the distance
between nodes. Nodes contained in many short paths
between nodes can efficiently propagate information
or considerably impede the flow of information. A
further classical notion of centrality is related to a
node’s ability to influence the flow of information. By
way of illustration I mention the betweenness centrality
measure [28,31], which is calculated as the number of
shortest paths a nodes is part of divided by the overall
number of shortest paths in the network. Milgram had
already demonstrated in 1967 that between almost any
two nodes in an SN there is a short path linking
This work is a preprint version. Copyright is held by IEEE.
Proceedings of the 47th Hawaii International Conference on System Sciences - 2014
them [23]. Furthermore he showed that short paths
can be found without too much effort. These results
have been found to also hold for OSN [67]. Although
computing the distance between users in large OSN
is an NP-hard problem, users can find short paths
connecting them to other users [68,69]. Granovetter
[24] showed that weak ties can be highly beneficial
as they provide connections to “distant clusters”. In
particular, weak ties are valuable for learning about
vacancies [24, p. 1371]. Burt [26] did not focus on
the strength relationships but on the absence of links
in an OSN. He coined the term structural hole to
describe missing connections between groups in an SN.
This notion proved to be highly influential and enabled
many important developments [27].
SNA has been applied to investigate individual performance as well as group performance, e.g. [70,71].
The results show correlations between certain centrality measures and job and group performance. Baldwin
et al. [70] reported that communication centrality was
linked to student’s grades (i.e. performance), Sparrowe
et al. [71] found that group performance was negatively
related to hindrance network density.
3.2. Online social networks
OSN2 such as Facebook, LinkedIn or Xing have
become a part of every day life.3 In turn OSN have already become an important tool for recruiters [22,73].
The scientific investigations of OSN [72]–[74] date
back about one and a half decade [75] and the focus
of the research is diverse, including investigations into
user groups, use of OSN, data privacy and protection,
social structures in OSN, graph theoretic properties of
OSN, comparisons between OSN and real-world SN.
Data generated by OSN users has drawn considerable
business interest, of which [76] provides an overview.
The value of social information stored in OSN, which
is key in my approach, has already been noted and
exploited by the RS community [77].
To transfer insights from SNA gained by studying
notions such as structural hole, degree and centrality
measures to OSN, only the underlying graph structure
of the OSN has to be known. To model weak and
strong links in OSN the notion of an activity network
has been introduced. To distinguish between a good
2. The literature distinguishes in particular between OSN and
Social Network Sites (SNS). The former takes a sociological perspective on electronic networks while the latter emphasizes the user
interaction with a website, cf. [72]. For my purposes both approaches
are relevant, however I will here only use the term OSN.
3. LinkedIn reports having more than 200.000.000 users as of
April 2013, Xing reports 13.000.000 users in December 2012 and
Facebook round about 1.000.000.000 as of April 2013.
friend and a mere acquaintance the frequency of interactions can be measured. Although the activity network
of Facebook, made up of the regularly interacting
users, is a priori of a different shape and form than
the whole network, Viswanath et al. [78] reported that
the activity network graph-theoretically resembles the
whole network. In OSN weak ties also play a crucial
role for the transfer of knowledge [79,80], in particular
for job searches (cf. [24]).
3.3. Personality extraction from online social
networks
Inferring personality traits from Internet/OSN use
is an emerging area of research. I will here highlight
some more relevant findings for our purposes by restricting our attention to the FFM. A good starting
point is the work of Ryan et al. [45] containing a
well researched state of the art overview on how
personality traits influence behavior on Facebook and
other OSN. Before going into details I remark that
the hypothesis that OSN profiles are created to escape
reality or to compensate for perceived personality flaws
has been found to be in general false for Facebook
users [42,43]. These findings support the idea of personality extractions from OSN. In fact, recruitment
and selection professionals regularly use OSN such
as LinkedIn and Facebook and they state that they
are able to draw conclusions on personality traits
from an applicant’s profile and network [22]. Gosling
et al. [46] showed that Facebook-based personality
impressions have some consensus and accuracy for all
FFM dimensions, with particularly strong consensus
for extraversion, but with the exception of accuracy
for neuroticism.
Ross et al. [38] studied user behavior on Facebook via self-reports and found a weak connection
between the personalities of the users and their behavior. Furthermore they reported that extraversion was
not significantly related to the number of Facebook
friends (in agreement with findings in [81]) nor was
the time spent online thus related. But they found that
individuals in the high Extraversion group reported
membership in significantly more Facebook groups.
Amichai-Hamburger and Vinitzky [39] replaced the
self-reports of Ross et al. [38] by more objective criteria and found a strong connection between personality
and Facebook user behavior. Tong et al. [37] found a
quartic relationship between the number of Facebook
friends of a user and perceptions of the individual’s
extraversion. Krämer and Winter [36] investigated the
relationship between self-reported (offline) personality
traits and (online) self-presentation in the German OSN
This work is a preprint version. Copyright is held by IEEE.
Proceedings of the 47th Hawaii International Conference on System Sciences - 2014
StudiVZ. Their content analysis of the respondents’
profiles showed that self-efficacy with regard to impression management (an aspect of extraversion) is
strongly related to the number of virtual friends, the
level of profile detail, and the style of the profile photo.
Extraverted users tended to present themselves in a less
restrained manner choosing less conservative pictures
of themselves. Recent studies (e.g. [40]–[42]) confirm
the strong link between extraversion and OSN usage.
Ross et al. [38] found a correlation between neuroticism and the usage of Facebook apps. Individuals
with a high neuroticism score were more likely to
declare the wall to be their favorite app whereas people
with a low score were more likely to prefer photos.
Individuals with a high score in agreeableness were
found to be more likely to be selected as a friend, cf.
[44]. Individuals that have high scores in openness to
experience and in neuroticism were found to be more
likely to be bloggers [35]. Karl et al. [48] analyzed the
relationships between personality differences and faux
pas posts in Facebook. Their results showed that those
who score high on conscientiousness, agreeableness
and emotional stability post significantly fewer faux
pas on their profile.
Interestingly the line of research assessing character
traits of an individual given the profile and information
about friends has been inverted. Denissen et al. [34]
found that the user profiles of an individual’s friends
can be used to infer character traits of the individual.
Results from [44] suggest that a friendship link between users is more likely if they have similar scores
in agreeableness, extraversion and openness.
3.4. Organizational culture extraction
Organizational culture can either be measured manually by conducting a questionnaire-based investigation
of the organizations’ members or it can be inferred
from their OSN use. The most common instrument for
manual evaluation is the Organizational Culture Profile
(OCP) developed by O’Reilly, Chatman and Caldwell
[63]. The OCP is a Q-sort instrument containing 54
value statements. The idea of an automated culture
extraction from OSN is supported by the fact that
the organizational culture is reflected by the members’
personalities and values [10,12,13,55,56]. The automation of an OSN-based culture evaluation is promising
since strong indications for culture dependent OSN
use have already been found (e.g. [47]–[50]). Yang
et al. [50] showed that culture traits can be inferred
from people’s social question and answer behavior in
OSN. Pfeil’s et al. [47] findings suggest that cultural
differences within the physical world also exist in the
virtual world. They show correlations between Hofstede’s [62] culture dimensions and use of Wikipedia
(e.g. a correlation between the masculinity index and
adding / clarifying information behavior). Callahan and
Herring [49] also observed a systematic cultural bias in
Wikipedia use. Karl et al. [48] analyzed faux pas posting in Facebook profiles of US and German students.
US students were more inclined than German students
to post faux pas to their Facebook profile. The authors
discussed possible cultural reasons (specifically that
US citizens tend to be more individualistic and lower
in uncertainty avoidance than Germans). In summary
there is a growing body of evidence to suggest that it
is possible to automatically infer organizational culture
traits from OSN.
3.5. Communication style extraction from online social networks
Investigating communication styles from Computer
Mediated Communications (CMC) is an emerging area
of research. In a first step it can be checked whether
an OSN profile is linked with further communication
tools such as Skype, ICQ, or Twitter. For instance,
if an OSN profile is linked to a twitter account, it
is possible to learn twitter use characteristics (e.g.
messages sent, following users, other users followed,
retweeting frequency and so on) [82]. Communication
intensity can be derived from blogging and tweeting
intensity or wall use in Facebook [38]) and time
orientation can be inferred from online times.
Business interests in this area have also begun to emerge. The importance of an individual’s
communication style for recommending items to
her has been noted and subsequently patented (US
2011/0054985A1) by Cisco. Social Behavior Analysis
(SBA) is the basis for two similar patterns granted to
Yahoo! that use SBA for generating recommendations
(US 2009/0164400A1 and US 2009/0164897A1).
3.6. Role extraction from online social networks
Since personality mainly predicts potential roles
of group members (e.g. [83]) we can infer these
roles from personality traits data extracted from OSN.
As an example consider the “initiator-contributor”role identified by Benne and Sheats [4] as a critical task role in every group. This role is characterized by entrepreneurial attributes (such as suggestions of new ideas and solutions to the group, (re-)
definition of group problems and goals, organizing
the group for the task ahead, handling difficulties the
This work is a preprint version. Copyright is held by IEEE.
Proceedings of the 47th Hawaii International Conference on System Sciences - 2014
group encounters, etc. [4, p. 43]). Zhao et al. [83] metaanalytically summarized the relationship of personality
traits and these entrepreneurial attributes. A broad
range of personality scales were categorized into a
set of constructs using the FFM of personality. Their
results showed that four of the five FFM dimensions
were associated with entrepreneurial intentions and
entrepreneurial performance. The ability to fill the
“information seeker” role [4] could be inferred for
instance from Facebook use [84] or from following
numerous Twitter users, while similarly the “harmonizer” role can be inferred from the use of the group
feature [84].
3.7. Extraction of skills, knowledge and abilities from online social networks
Measuring SKA has been a hot scientific topic
[11,12,51,52,58,85]. HR-XML is a current common
standard for expressing SKA in machine-readable format. Today HR-XML is extended and maintained by
a non-profit consortium and has also made its way
into practical applications in real world labor markets.
For instance, the German employment agency (BA)
has adopted this format and developed a specification
called HR-BA-XML. It is my firm view that it will
only be a matter of time before (business) OSN adopt a
generally accepted fully machine-readable standard for
uploading CVs, testimonials and letters of reference.
The problem of finding a person possessing certain
SKA (i.e. an expert) to solve a concrete problem or task
has scientifically been studied and is known under the
label expert finding or expert location. SN have proved
to be a fertile ground to search for experts. Search
strategies for expert finding in SN were evaluated in
[21] and the state of the art has been reviewed in [86].
3.8. Legal aspects of data extraction from
online social networks
In practice we cannot assume that all information
stored in OSN is freely available. In general accessibility of data is limited by user settings, by data
protection and privacy laws and the OSN terms and
conditions. Projecting current trends suggests that these
laws are only set to tighten. For the foreseeable future
however I believe that (future) laws do not render
OSN-Recruiting unfeasible. I foresee two possible approaches to maintaining recruiting RS searching OSN
data even under tightened rules and regulations. One
is that privacy-preserving data mining techniques [87]
can be applied, see [76] for a current state of the art
overview of data anonymisation techniques). Secondly,
micropayments can be made to OSN users and/or the
OSN provider for the right to extract data.
It is possible that current trends will reverse and
users of OSN will explicitly want their profiles to
be public to further business opportunities and career
advancement. It is plausible to conjecture that users of
business OSN (e.g. LinkedIn and Xing) will prefer to
follow the latter trend.
If and how personalities may be measured and
subsequently legally used in a recruiting process is a
complex legal question. Different laws (in particular
the ever changing labor and anti-discrimination laws)
and legal systems have given different answers which
vary by country and also over time, cf. [88].
4. Recommender systems for online social
network recruiting
The design of (OSN-)Recruiting RS is an emerging area, cf. [20,21,77,85]. But the existing RS only
address the P-J fit with the exception of Malinowski
et al. [85] who took both the P-J and the P-G fit into
account. However they completely ignored the P-O fit
and focused only “on a company-internal team staffing
scenario” [85, p. 431].
Here I consider the design of content-based and
collaborative RS for OSN-Recruiting. Content based
RS estimate the utility a user obtains from an item i
depending on other items the user has previously rated
which are similar to item i. Collaborative approaches
differ from the above by recommending items to a
user u based on ratings or recommendations by users
which are similar to user u. It is feasible to design
content-based recruiting RS searching OSN based on
a similarity notion between candidates and an assessment of selected candidates. Such collaborative RS can
be designed using recruiting strategies from organizations similar to the recruiting organization, and such
strategies can be obtained from (industrial) espionage
or less clandestinely by monitoring job ads and portals,
or, more in line with my topic, by monitoring OSN.
Given that content based and collaborative approaches
are possible hybrid RS are conceivable.
Before an RS can recommend candidates which
have not been assessed by the user the problem of
how a given candidate is evaluated has to be addressed. Formally such a utility function is given by
the P-OE fit, which can be represented by a function
FP OE . FP OE takes as inputs a vacancy V within
an organization and a candidate C and returns a real
number. This function may be decomposed into three
sub-functions: fP O , fP G and fP J (representing
the P-O, the P-G and the P-J fit) and an aggregation
This work is a preprint version. Copyright is held by IEEE.
Proceedings of the 47th Hawaii International Conference on System Sciences - 2014
function A. A combines these sub-functions by an
aggregation of the relative importance placed on the
sub-fits, cf. 2. We can thus formulate the following:
p
FP OE C, V
Macro level
Organizational
Culture*
Meso level
Group
Roles*
q ApfP O pC, V q, fP G pC, V q, fP J pC, V qq
P-O fit
p
q
p
supp
fP G fPcomp
G SIPC , SF W GV , fP G SIPC , SF W GV
qq
supp
where fPcomp
G , fP G are the complementary
respectively supplementary P-G fit. Likewise the
P-O fit only depends on the organization’s cultural
values VV and the candidate’s personality PC whereas
the P-J fit only depends on job requirements JRV and
a candidate’s skills, abilities and knowledge SKAC .
We hence reformulate equation (2) as
p q ApfP O pPC , VV q,
p
p
q
p
fP J pSKAC , JRV qq .
FP OE C, V
Job
Description
Integrated
Perspective
Hiring
Organization
(2)
Determining the aggregation function A constitutes
a Multi Criteria Decision Making (MCDM) problem.
One main challenge in MCDM is the above commensurateness problem, i.e. if, and if so how, a low
score in one criterion can be compensated by a high
score in another criterion. The Choquet and the Sugeno
integrals [89] are methods that are widely adopted
by the MCDM community to address this problem.
The RS community is well aware of multi-criterial
phenomena and has developed RS based on multiple
evaluation criteria [90].
Do note that the P-G fit only depends on certain
aspects of a vacancy. Indeed only the social fabric
of the workgroup SF W GV is relevant while for
instance required skills are completely irrelevant
for determining fP G . Similarly not all properties
of a candidate C matter for calculating fP G , as
only the candidate’s social interaction patterns SIPC
are required. We can hence formulate fP G pC, V q p
Group
Communication
styles*
Micro level
supp
fP G fPcomp
G SIPC , SF W GV , fP G SIPC , SF W GV
(3)
qq,
The challenges posed to the RS community compromise more than the development of RS estimating
FP OE pC, V q based on equation (3). Due to the high
complexity of modern work environments recruiting
decisions are made under uncertainty [57]. It appears
reasonable to use a structure that relies on the same
formal model of uncertainty for all three fits, which
might be phrased in quantitative (probabilistic), qualitative or linguistic (fuzzy) terms. Figure 2 graphically
summarizes the proposed framework.
5. Conclusions
I have developed an interdisciplinary framework for
OSN-Recruiting. Adopting a business HRM perspec-
P-G fit
Personality*
Potential
Roles*
Social context
Communication
style*
P-J fit
P-OE fit
Skills,
Knowledge
& Abilities*
Candidate
* Data inferable from OSN
Figure 2: An interdisciplinary P-OE fit
based framework for OSN-Recruiting
tive I identified the P-OE fit as crucial in the assessment of the suitability of job candidates to a given
vacancy. As shown, the P-OE fit can be broken down
into the P-O fit, the P-G fit and the P-J fit, cf. Eqn. (1).
In contrast to previous approaches I hence included
in my framework the macro-perspective (P-O fit) and
the meso-perspective (P-G fit). For the calculation of
all these sub-fits I identified the relevant inputs (e.g.
personality, organizational culture) and I showed that
these inputs can potentially be inferred from OSN with
the sole exception of the job description. Finally I took
a preparatory step for designing OSN-Recruiting RS
by mathematically formalizing the P-OE fit (Eqn. (3)).
This equation can then be used to forecast the P-OE
fit between an OSN user and a vacancy.
My main contributions are (a) illustrating a way to
improve existing recruiting RS and (b) providing a
platform for interdisciplinary research thereby enabling
further discoveries in various areas of research.
5.1. Limitations
The business potential of my framework for OSNRecruiting appears enormous, however there are several limitations: (i) There are sociopolitical, legal
and psychological barriers/reservations concerning disclosing, mining and processing of personal information. Privacy concerns can be addressed via privacypreserving data mining technologies or by market
mechanisms for data extraction from OSN in return for
rewards, cf. 3.8. (ii) Inferring social characteristics of
organizations, groups and individuals based on data extracted from OSN has not yet matured into a broad and
consistent area of research. However recent research
converges, for instance see 3.3 for extraversion. (iii)
My framework is limited by the models I used, such
This work is a preprint version. Copyright is held by IEEE.
Proceedings of the 47th Hawaii International Conference on System Sciences - 2014
as the P-OE fit model, the FFM, workgroup models
etc. However I used the most accepted and advanced
models to represent the state of the art, cf. 2. Further
developments of the underlying models can be used
to enhance my framework. (iv) Finally at present my
framework is curtailed due to a missing evaluation (e.g.
by prototypes, field studies, tests), cf. 5.2.
5.2. Future research
For a successful development and implementation of
an OSN-Recruiting RS based on my framework several interesting open interdisciplinary problems need
to be addressed. In particular I identify: (i) studying organizational culture extraction from OSN, (ii)
mining OSN for the social fabric of a workgroup
(group role detection, communication style extraction),
(iii) inferring personality traits based on OSN data,
(iv) improving models predicting on-the-job behavior
based on data extracted from OSN, cf. [48, p. 182], (v)
designing and implementing a fully machine-readable
standard for CVs, letters of references and testimonials,
(vi) developing mechanisms for trading OSN data,
(vii) implementing and evaluating of real-world OSNRecruiting RS, and (viii) studying of users of such RS.
Completing this research program does not only
supports recruiters but may also yield a better understanding of the P-OE fit problem. This deepened
understanding enables further contributions to sociology (group dynamics, SNA), psychology (personality
research, OSN use), computer science (data mining
and design of OSN/RS) and economics (organizational
behavior research).
Finally my framework may also be applied to other
HRM problems such as organizational development,
personnel development and retention management.
Acknowledgments
I would like to thank Juergen Landes for his support
on an earlier version of the paper. This research is
partly funded by the German Federal Ministry of Education and Research (BMBF, contract 03FH055PX2).
References
[1] S. Strohmeier, “Research in e-HRM: Review and implications,” HRMR, vol. 17, no. 1, pp. 19–37, 2007.
[2] I. Lee, “Modeling the benefit of e-recruiting process
integration,” DSS, vol. 51, no. 1, pp. 230–239, 2011.
[3] S. D. Maurer and D. P. Cook, “Using company web
sites to e-recruit qualified applicants: A job marketing
based review of theory-based research,” Comput Hum
Behav, vol. 27, no. 1, pp. 106–117, 2011.
[4] K. D. Benne and P. Sheats, “Functional Roles of
Group Members,” J Soc Issues, vol. 4, no. 2, pp.
41–49, May 1948.
[5] D. G. Ancona and D. F. Caldwell, “Beyond Task and
Maintenance: Defining External Functions in Groups,”
Group & Organization Studies, vol. 13, no. 4, pp. 468–
494, December 1988.
[6] M. K. Mount, M. R. Barrick, and G. L. Stewart,
“Five-Factor Model of personality and Performance
in Jobs Involving Interpersonal Interactions,” Human
Performance, vol. 11, no. 2-3, pp. 145–165, 1998.
[7] S. T. Bell, “Deep-level composition variables as predictors of team performance: A meta-analysis,” J. Appl.
Psychol., vol. 92, no. 3, pp. 595–615, May 2007.
[8] J. R. Mesmer-Magnus and L. A. DeChurch, “Information Sharing and Team Performance: A MetaAnalysis,” J. Appl. Psychol., vol. 94, no. 2, pp. 535–
546, 2009.
[9] J. D. Werbel and D. J. Johnson, “The Use of PersonGroup Fit for Employment Selection: A Missing Link
in Person-Environment Fit,” Hum. Resour. Manage.,
vol. 40, no. 3, pp. 227–240, 2001.
[10] V. R. Tom, “The Role of Personality and Organizational
Images in the Recruiting Process,” Organ. Behav. Hum.
Perform., vol. 6, no. 5, pp. 573–592, 1971.
[11] J. R. Edwards, Person-job fit: A conceptual integration, literature review, and methodological critique.
International review of industrial and organizational
psychology. Oxford, England: John Wiley & Sons,
1991, vol. 6, pp. 283–357.
[12] A. L. Kristof, “Person-organization fit: An integrative
review of its conceptualizations, measurement, and implications,” Pers Psych, vol. 49, no. 1, pp. 1–49, 1996.
[13] M. L. Verquer, T. A. Beehr, and S. H. Wagner, “A
meta-analysis of relations between person-organization
fit and work attitudes,” J. Vocat. Behav., vol. 63, no. 3,
pp. 473–489, 2003.
[14] A. L. Kristof-Brown, R. D. Zimmerman, and E. C.
Johnson, “Consequences of individuals fit at work:
A meta-analysis of person-job, person-organization,
person-group, and person-supervisor fit,” Pers Psych,
vol. 58, no. 2, pp. 281–342, 2005.
[15] B. J. Hoffman and D. J. Woehr, “A quantitative review
of the relationship between person-organization fit and
behavioral outcomes,” J. Vocat. Behav., vol. 68, no. 3,
pp. 389–399, 2006.
[16] A. Leung and S. Chaturvedi, “Linking the fits, fitting
the links: Connecting different types of PO fit to
attitudinal outcomes,” J. Vocat. Behav., vol. 79, no. 2,
pp. 391–402, 2011.
[17] A. Ramesh and M. J. Gelfand, “Will They Stay or Will
They Go? The Role of Job Embeddedness in Predicting
Turnover in Individualistic and Collectivistic Cultures,”
J. Appl. Psychol., vol. 95, no. 5, pp. 807–823, 2010.
[18] J. E. Sheridan, “Organizational Culture and Employee
Retention,” Acad Manage J, vol. 35, no. 5, pp.
1036–1056, 1992.
[19] S. Macdonald, “Headhunting in high technology,” Technovation, vol. 4, no. 3, pp. 233–245, 1986.
[20] T. Keim, “Extending the Applicability of Recommender
Systems: A Multilayer Framework for Matching
Human Resources,” in HICSS 2007 Proceedings,
2007, p. 169.
This work is a preprint version. Copyright is held by IEEE.
Proceedings of the 47th Hawaii International Conference on System Sciences - 2014
[21] J. Zhang and M. S. Ackerman, “Searching for Expertise
in Social Networks: A Simulation of Potential Strategies,” in GROUP ’05 Proc. ACM, 2005, pp. 71–80.
[22] R. Caers and V. Castelyns, “LinkedIn and Facebook in
Belgium: The Influences and Biases of Social Network
Sites in Recruitment and Selection Procedures,” Soc
Sci Comput Rev, vol. 29, no. 4, pp. 437–448, 2011.
[23] S. Milgram, “The small world problem,” Psychology
Today, vol. 2, no. 1, pp. 60–67, 1967.
[24] M. S. Granovetter, “The Strength of Weak Ties,” Am J
Sociol, vol. 78, no. 6, pp. 1360–1380, 1973.
[25] J. S. Coleman, “Social Capital in the Creation of
Human Capital,” Am J Sociol, vol. 94, no. Supplement:
Organizations and Institutions: Sociological and
Economic Approaches to the Analysis of Social
Structure, pp. 95–120, 1988.
[26] R. S. Burt, Structural holes: The social structure of
competition. Harvard University Press, 1992.
[27] G. Walker, B. Kogut, and W. Shan, “Social Capital,
Structural Holes and the Formation of an Industry
Network,” Organ Sci, vol. 8, no. 2, pp. 109–125, 1997.
[28] M. E. Shaw, “Group Structure and the Behavior of
Individuals in Small Groups,” J Psychol, vol. 38, no. 1,
pp. 139–149, 1954.
[29] M. A. Beauchamp, “An Improved Index of [Centrality,”
Behav. Sci., vol. 10, no. 2, pp. 161–163, April 1965.
[30] G. Sabidussi, “The centrality index of a graph,”
Psychometrika, vol. 31, no. 4, pp. 581–603, 1966.
[31] L. C. Freeman, “Centrality in Social Networks:
Conceptual Clarification,” Social Networks, vol. 1,
no. 3, pp. 215–239, 1978-1979.
[32] S. Wasserman and K. Faust, Social Network Analysis:
Methods and Applications.
Cambridge University
Press, 1994.
[33] J. P. Scott, Social Network Analysis: A Handbook.
Sage, 2007.
[34] J. J. A. Denissen, R. Geenen, M. Selfhout, and M. A. G.
van Aken, “Single-item Big Five Ratings in a Social
Network Design,” Eur J Personality, vol. 22, no. 1, pp.
37–54, 2008.
[35] R. E. Guadagno, B. M. Okdie, and C. A. Eno, “Who
blogs? Personality predictors of blogging,” Comput
Hum Behav, vol. 24, no. 5, pp. 1993–2004, 2008.
[36] N. Krämer and S. Winter, “Impression Management
2.0: The Relationship of Self-Esteem, Extraversion,
Self-Efficacy, and Self-Presentation Within Social Networking,” J Media Psych, vol. 20, no. 3, pp. 106–116,
2008.
[37] S. T. Tong, B. Van Der Heide, L. Langwell,
and J. B. Walther, “Too Much of a Good
Thing? The Relationship Between Number of Friends
and Interpersonal Impressions on Facebook,” JCMC,
vol. 13, no. 3, pp. 531–549, 2008.
[38] C. Ross, E. S. Orr, M. Sisic, J. M. Arseneault,
M. G. Simmering, and R. R. Orr, “Personality and
motivations associated with Facebook use,” Comput
Hum Behav, vol. 25, no. 2, pp. 578–586, 2009.
[39] Y. Amichai-Hamburger and G. Vinitzky, “Social network use and personality,” Comput Hum Behav, vol. 26,
no. 6, pp. 1289–1295, 2010.
[40] T. Correa, A. W. Hinsley, and H. G. de Zúñiga,
“Who interacts on the web?: The intersection of
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
users’ personality and social media use,” Comput Hum
Behav, vol. 26, no. 2, pp. 247–253, 2010.
E. Y. Ong, R. P. Ang, J. C. Ho, J. C. Lim, D. H. Goh,
C. S. Lee, and A. Y. Chua, “Narcissism, extraversion
and adolescents’ self-presentation on Facebook,” Pers
Individ Dif, vol. 50, no. 2, pp. 180–185, 2011.
S. D. Gosling, A. A. Augustine, S. Vazire, N. Holtzman, and S. Gaddis, “Manifestations of Personality
in Online Social Networks: Self-Reported FacebookRelated Behaviors and Observable Profile Information,”
Cyberpsychol Behav Soc Netw, vol. 14, no. 9, pp. 483–
488, 2011.
M. D. Back, J. M. Stopfer, S. Vazire, S. Gaddis,
S. C. Schmukle, B. Egloff, and S. D. Gosling, “Facebook Profiles Reflect Actual Personality, Not SelfIdealization,” Psychol Sci, vol. 21, no. 3, pp. 372–374,
2010.
M. Selfhout, W. Burk, S. Branje, J. Denissen, M. van
Aken, and W. Meeus, “Emerging Late Adolescent
Friendship Networks and Big Five Personality Traits:
A Social Network Approach,” J. Pers., vol. 78, no. 2,
pp. 509–538, 2010.
T. Ryan and S. Xenos, “Who uses Facebook? an
investigation into the relationship between the Big
Five, shyness, narcissism, loneliness, and Facebook
usage,” Comput Hum Behav, vol. 27, no. 5, pp.
1658–1664, 2011.
S. D. Gosling, S. Gaddis, and S. Vazire, “Personality
Impressions Based on Facebook Profiles,” in ICWSM
’07 Proc. AAAI, 2007, pp. 1–4.
U. Pfeil, P. Zaphiris, and C. S. Ang, “Cultural
Differences in Collaborative Authoring of Wikipedia,”
JCMC, vol. 12, no. 1, pp. 88–113, 2006.
K. Karl, J. Peluchette, and C. Schlaegel, “Who’s
Posting Facebook Faux Pas? A Cross-Cultural
Examination of Personality Differences,” Int J Select
Assess, vol. 18, no. 2, pp. 174–186, 2010.
E. S. Callahan and S. C. Herring, “Cultural bias
in Wikipedia content on famous persons,” JASIST,
vol. 62, no. 10, pp. 1899–1915, 2011.
J. Yang, M. R. Morris, J. Teevan, L. A. Adamic, and
M. S. Ackerman, “Culture Matters: A Survey Study of
Social Q&A Behavior,” in ICWSM ’11 Proc. AAAI,
2011, pp. 409–416.
A. L. Kristof-Brown, “Perceived applicant fit:
Distinguishing between recruiters’ perceptions of
person-job and person-organization fit,” Pers Psych,
vol. 53, no. 3, pp. 643–671, 2000.
M. Brkich, D. Jeffs, and S. A. Carless, “A Global
Self-Report Measure of Person-Job Fit,” Eur J Psychol
Assess, vol. 18, no. 1, pp. 43–51, 2002.
A. Kristof-Brown, M. R. Barrick, and C. Kay Stevens,
“When Opposites Attract: A Multi-Sample Demonstration of Complementary Person-Team Fit on
Extraversion,” J. Pers., vol. 73, no. 4, pp. 935–958,
2005.
D. M. Cable and C. K. Parsons, “Socialization tactics
and person-organization fit,” Pers Psych, vol. 54, no. 1,
pp. 1–23, 2001.
C. Anderson, S. E. Spataro, and F. J. Flynn, “Personality and Organizational Culture as Determinants of
Influence,” J. Appl. Psychol., vol. 93, no. 3, pp. 702–
710, 2008.
This work is a preprint version. Copyright is held by IEEE.
Proceedings of the 47th Hawaii International Conference on System Sciences - 2014
[56] J. R. Edwards and D. M. Cable, “The Value of Value
Congruence,” J. Appl. Psychol., vol. 94, no. 3, pp.
654–677, 2009.
[57] A. M. Spence, “Job Market Signaling,” Q J Econ,
vol. 87, no. 3, pp. 355–374, 1973.
[58] T. Sekiguchi and V. L. Huber, “The use of personorganization fit and person-job fit information in
making selection decisions,” Organ. Behav. Hum.
Decis. Process., vol. 116, no. 2, pp. 203–216, 2011.
[59] P. T. Costa and R. R. McCrae, Revised NEO personality inventory (NEO-PI-R) and the NEO Five-Factor
inventory (NEO-FFI): Professional manual. Odessa,
FL, USA: PAR, 1992.
[60] M. G. Rothstein and R. D. Goffin, “The use of
personality measures in personnel selection: What
does current research support?” HRMR, vol. 16, no. 2,
pp. 155–180, 2006.
[61] T. Jung, T. Scott, H. T. O. Davies, P. Bower,
D. Whalley, R. McNally, and R. Mannion, “Instruments
for Exploring Organizational Culture: A Review of
the Literature,” Public Adm. Rev., vol. 69, no. 6, pp.
1087–1096, 2009.
[62] G. H. Hofstede, Culture’s Consequences: International
Differences in Work-Related Values. Beverly Hills,
CA: Sage, 1984.
[63] C. A. O’Reilly III, J. Chatman, and D. F.
Caldwell, “People and Organizational Culture: A
Profile Comparison Approach to Assessing PersonOrganization Fit,” Acad Manage J, vol. 34, no. 3, pp.
487–516, 1991.
[64] J. Guillory and J. T. Hancock, “The Effect of Linkedin
on Deception in Resumes,” Cyberpsychol Behav Soc
Netw, vol. 15, no. 3, pp. 135–140, 2012.
[65] L. C. Freeman, “A Set of Measures of Centrality Based
on Betweenness,” Sociometry, vol. 40, no. 1, pp. 35–41,
1977.
[66] S. P. Borgatti and M. G. Everett, “A Graph-theoretic
perspective on centrality,” Social Networks, vol. 28,
no. 4, pp. 466–484, 2006.
[67] P. S. Dodds, R. Muhamad, and D. J. Watts, “An Experimental Study of Search in Global Social Networks,”
Science, vol. 301, no. 5634, pp. 827–829, 2003.
[68] L. Adamic and E. Adar, “How to search a social
network,” Social Networks, vol. 27, no. 3, pp. 187–203,
2005.
[69] D. J. Watts, P. S. Dodds, and M. E. J. Newman,
“Identity and Search in Social Networks,” Science,
vol. 296, no. 5571, pp. 1302–1305, 2002.
[70] T. T. Baldwin, M. D. Bedell, and J. L. Johnson,
“The Social Fabric of a Team-Based M.B.A.
Program: Network Effects on Student Satisfaction
and Performance,” Acad. Manage. J., vol. 40, no. 6,
pp. 1369–1397, 1997.
[71] R. T. Sparrowe, R. C. Liden, S. J. Wayne, and M. L.
Kraimer, “Social Networks and the Performance of
Individuals and Groups,” Acad. Manage. J., vol. 44,
no. 2, pp. 316–325, 2001.
[72] d. m. boyd and N. B. Ellison, “Social Network Sites:
Definition, History, and Scholarship,” JCMC, vol. 13,
no. 1, pp. 210–230, 2008.
[73] W. P. Smith and D. L. Kidder, “Youve been tagged!
(Then again, maybe not): Employers and Facebook,”
Bus. Horiz., vol. 53, no. 5, pp. 491–499, 2010.
[74] D. Rosen, G. A. Barnett, and J.-H. Kim, “Social
networks and online environments: when science and
practice co-evolve,” Social Network Analysis and
Mining, vol. 1, no. 1, pp. 27–42, 2011.
[75] L. Garton, C. Haythornthwaite, and B. Wellman,
“Studying Online Social Networks,” JCMC, vol. 3,
no. 1, 1997.
[76] F. Bonchi, C. Castillo, A. Gionis, and A. Jaimes,
“Social Network Analysis and Mining for Business
Applications,” ACM TIST, vol. 2, no. 3, 2011, Article
No. 22.
[77] D. Jannach, M. Zanker, A. Felfernig, and G. Friedrich,
Recommender Systems: An Introduction. Cambridge
University Press, 2010.
[78] B. Viswanath, A. Mislove, M. Cha, and P. K.
Gummadi, “On the evolution of user interaction in
Facebook,” in WOSN ’09 Proc., 2009, pp. 37–42.
[79] D. Z. Levin and R. Cross, “The Strength of Weak
Ties You Can Trust: The Mediating Role of Trust in
Effective Knowledge Transfer,” Management Science,
vol. 50, no. 11, pp. 1477–1490, 2004.
[80] J. Donath, “Signals in Social Supernets,” JCMC,
vol. 13, no. 1, pp. 231–251, 2007.
[81] T. Amiel and S. L. Sargent, “Individual differences in
Internet usage motives,” Comput Hum Behav, vol. 20,
no. 6, pp. 711–726, 2004.
[82] C. Wang and B. Huberman, “Long trend dynamics in
social media,” EPJ Data Science, vol. 1, no. 1, p. 2,
2012.
[83] H. Zhao, S. E. Seibert, and G. Lumpkin, “The Relationship of Personality to Entrepreneurial Intentions
and Performance: A Meta-Analytic Review,” Journal of
Management, vol. 36, no. 2, pp. 381–404, March 2010.
[84] D. Y. Wohn, C. Lampe, J. Vitak, and N. Ellison,
“Coordinating the ordinary: social information uses of
Facebook by adults,” in iConference ’11 Proc., 2011,
pp. 340–347.
[85] J. Malinowski, T. Weitzel, and T. Keim, “Decision
support for team staffing: An automated relational
recommendation approach,” DSS, vol. 45, no. 3, pp.
429–447, 2008.
[86] T. Lappas, K. Liu, and E. Terzi, “A Survey of
Algorithms and Systems for Expert Location in Social
Networks,” in Social Network Data Analytics, C. C.
Aggarwal, Ed. Springer, 2011, pp. 215–241.
[87] R. Agrawal and R. Srikant, “Privacy-Preserving Data
Mining,” in SIGMOD ’00 Proc. ACM, 2000, pp.
439–450.
[88] R. L. Hogler, C. Henle, and C. Bemus, “Internet
recruiting and employment discrimination: a legal
perspective,” HRMR, vol. 8, no. 2, pp. 149–164, 1998.
[89] M. Grabisch and C. Labreuche, “A decade of
application of the Choquet and Sugeno integrals
in multi-criteria decision aid,” Annals of Operations
Research, vol. 175, no. 1, pp. 247–286, 2010.
[90] G. Adomavicius, N. Manouselis, and Y. Kwon, “Multicriteria recommender systems,” in Recommender
Systems Handbook. Springer, 2011, pp. 769–803.
This work is a preprint version. Copyright is held by IEEE.