dawson.pdf


Australasian Journal of
Educational Technology

2011, 27(1), 16-27

Using social network metrics to assess the effectiveness
of broad based admission practices
Shane Dawson, Leah Macfadyen
University of British Columbia

Lori Lockyer and David Mazzochi-Jones
University of Wollongong

An Outstanding Paper Award recipient, ascilite Sydney 2010 Conference

Notions of what it is to be knowledgeable and skilled in one‘s profession have evolved
in recent decades. For instance, medical practitioners are expected to think critically
and creatively, communicate effectively, and to be a professional and community
leader. While these attributes have always been well regarded, it is only relatively
recently that higher education institutions are actively incorporating these skills and
attributes into student admissions criteria. In parallel, methods of instruction and
course delivery have also changed over time with respect to these driving social
paradigms. Today‘s medical schools are expected to both select and develop students
in terms of these qualities through socially based pedagogical practices. This paper
investigates the admissions criteria that best predict student engagement in a social
learning environment and thus the related attributes such as communication,
creativity, and leadership. The paper frames this investigation in the scholarship
related to 21st century skills and achievement orientations.

Introduction

Since the early massification of education, the process for university student
admissions has long relied upon academic grades obtained through standardised
testing practices. Advocates of this approach make the assumption that grades
represent an essential measure of student preparedness and capacity for academic
success at university (Barro, 2001). However, detractors frequently cite the negative
impact of selection based purely on cognitive measures on student diversity and
minority representation (Espenshade & Chang, 2005). Regardless of the incentive,
there is recent renewed interest among higher education institutions in adopting more
broad-based admissions criteria. In essence, institutions are seeking criteria that act as
early indicators of learner skills in the non-cognitive realm that are commonly referred
to in university strategic statements about the qualities and attributes they will instil in
their graduates. These may include elements of leadership, communication,
innovation, and creativity.

In this paper we argue that admissions procedures that continue to rely heavily on
standardised test scores impact not only student (academic, ethnic and cultural)
diversity, but also dramatically affect the likelihood that admitted students will
display desirable attributes such as creativity, lateral thinking and engagement in the



Dawson, Macfadyen, Lockyer and Mazzochi-Jones 17

course learning community. In so doing, we begin by demonstrating that grades-based
reward and measurement schema are unlikely to encourage or select for the creativity,
communication, networking and leadership attributes that are so urgently needed in
the contemporary workforce. This argument is illustrated through analysis of student
data that demonstrates the predictive variables associated with engagement in a
learning community and course grades evolving from a broad-based admissions
process. In short, the paper uses academic analytics data to address questions about
the impact of broad-based admissions procedures on future student outcomes and
teaching and learning practice.

Student admissions

The higher education environment has historically made heavy use of extrinsic
motivators to reward and recognise achievement on learning related tasks. This is well
illustrated in the assignment of and recognition of grades for specific learning
deliverables. A further example can be found in the high stakes testing and
recruitment practices for student admission into high profile vocational disciplines,
such as medicine. In these instances, students are frequently streamed using selection
processes that give priority to measures of their previous performance on academic
tests. While admission processes to medical school may also incorporate an interview
and demonstration of a portfolio, students are first filtered using criteria such as prior
academic ability as evidenced through prior course grades and standardised test
scores (e.g. Graduate Australian Medical School Admissions Test – GAMSAT). In this
context, recruitment into a medical program is frequently weighted more heavily
towards standardised testing (Powis, Hamilton & McManus, 2007). These practices
persist even though there is little research-based evidence that prior grades have any
significant predictive power in relation to future academic performance (Groves,
Gordon & Ryan, 2007). The University of Queensland, for example, has recently
discontinued the inclusion of an interview stage for medical student selection in
preference for criteria relating solely to past grades and GAMSAT score.

In the higher education context this over-dependence on prior student grades may
result in the selection of students with what Carol Dweck (2000) has previously termed
a "high performance goal orientation". Dweck argues that two types of achievement
orientations exist and these orientations largely reflect student motivation. For
example, a learning goal orientation reflects a student’s intrinsic motivation to develop
new skills and knowledge. Conversely, a performance goal orientation relates to more
extrinsic sources of motivation, such as the confirmation of knowledge and skills
through the assignment of grades associated with various assessment tasks. Dweck
highlights the conflict between an education system that is ideally promoting a
learning oriented environment, whilst at the same time implementing practices that
serve to promote a performance based culture. Simply put, grades win and what is
assessed matters, and this leads directly to peer to peer competition, the antithesis of
social learning behaviour.

In this context, the prevalence of admissions practices that draw too heavily upon
student grades stresses a culture of competition and constant demonstration of
performance. In such instances, the past experience and culture of students entering
university life is one centred on direct competition with peers. Thus, based on this
prior experience there is limited motivation for students to quickly develop networks,
share resources, discuss and engage in learning activities. Furthermore, the skills



18 Australasian Journal of Educational Technology, 2011, 27(1)

required to commence these social network liaisons may be obstructed by their past
experience of competition and performance. In essence, the reliance on grades as
presumed evidence of cognitive capacity may be to the detriment of the various non-
cognitive dimensions of leadership, innovation, communication and creativity. These
are attributes and skills that are best fostered in a learning environment complemented
by a highly diverse student body, and frequent encounters with well-integrated
curriculum activities that draw on the strengths of all peers (McWilliam, 2008;
McWilliam & Dawson, 2008; Yang & Cheng, 2010). In short, student participation in
engaged interactions within a social learning framework.

Learning community – desirable attributes

To date there has been much written about the vast educational benefits associated
with a social learning pedagogy (see for example, Haythornthwaite, 2006; Lave, 1993;
Seely Brown, 2006). Collectively, the literature on social learning suggests that active
student participation within a diverse networked environment is critical for
developing understanding, facilitating knowledge and resource exchange, and
fostering 21st century skills and attributes such as creativity (Yang & Cheng, 2010),
leadership (Komives et al., 2009) and network agility (McWilliam, 2008). It cannot be
emphasised enough that these oft cited and highly desirable attributes are largely
developed through engaged social learning experiences. This raises an important
question: If the possession of these attributes and skills, or the ability to develop them,
are highly desirable, are these capacities adequately selected for and encouraged in
student admission criteria? This study sought to identify the types of admissions
criteria that would provide a lead indicator of student propensity for interacting
within a community-centred pedagogy. We might ask: What are the types of
admissions criteria that best predict student engagement in a learning community and
thus the related attributes of creativity, leadership and network agility?

Methodology

In light of the above discussion, this study aimed to identify in student admissions
data the potential predictors of student engagement in a learning community. In so
doing, the study also sought to validate prior research relating to standardised medical
test scores and future academic performance. Specifically, the study addressed the
following research questions:

• What is the relationship between student admissions criteria and engagement in
the learning community?

• Is there a significant relationship between prior student academic test scores and
future academic performance in a medical education program?

The existence of correlations between medical student cohort admission scores,
subsequent academic performance (end of year exam results) and social network data
was examined (social network data such as centrality measures are sound indicators of
the degree of student importance and level of engagement within the overall student
learning community). The following section outlines the School’s student admissions
process and how the learning community data was extracted and measured for further
statistical analysis and interpretation.



Dawson, Macfadyen, Lockyer and Mazzochi-Jones 19

Study overview

The study participants (N = 132) included all first year enrolled medical students at an
Australian university over a two-year period. Although the first year of study is more
traditional in terms of course delivery (on campus classes) there is extensive
curriculum integration with the institutional learning management system (LMS). The
medical curriculum has been designed to rapidly familiarise and engage new students
with the online resources and the associated learning activities. In this context, the
medical program can be seen to be a blended pedagogical model integrating on-
campus classes with online learning content and activities.

Admissions process

The graduate medical school’s admissions process is rigorous and detailed. Admission
is based on graduate level entry and demonstration of desirable skills and attributes.
Applicants must address a diversity of admissions criteria including prior course
grades (weighted GPA), GAMSAT score, and a portfolio demonstrating leadership
and experience. These criteria are used to shortlist the potential applicants for a final
interview stage. This stage involves a multiple mini-interview (MMI) in which each
interviewee attends a number of timed interview stations each conducted by two
trained assessors who provide a score for the applicant on their respective station (Eva,
Rosenfeld, Reiter & Norman, 2004). Questions at the interview stations are designed to
uncover attributes such as critical thinking, decision making, communication skills,
leadership, and knowledge of health care.

Aggregate scores for all criteria are then used as a cut off for offers to the program. The
diversity of criteria used in the admissions process provides a unique opportunity to
investigate which variables, if any, may act as significant indicators of future student
engagement and overall preparedness and capacity for the medical profession.

International applicants are considered under an alternate interview system. As such,
data for international students (representing 10% of the total combined cohort) have
been excluded from this study.

Measuring learning community engagement

To determine student engagement within the online learning community the study
extracted discussion forum data from the institutional LMS. The discussion forum data
(exceeding 2000 individual posts) was analysed after completion of the first year of
study for the 2007 and 2008 cohorts. The discussion forum is closely linked to the
learning materials and activities. For instance, content reflecting specific medical cases
is provided online for students to discuss collaboratively and decide on an appropriate
medical course of action. In this context, the discussion forum is designed as
collaborative medium for resource sharing and knowledge co-construction.

The forum data was extracted and analysed using the social networking software
SNAPP (Bakharia & Dawson, 2009) and Netdraw (Borgatti, 2002). The study adopted
the social network extraction tool, known as Social Networks Adapting Pedagogical
Practice (SNAPP) as a rapid means for exporting and analysing the established student
relationships emerging through the discussion activity. SNAPP was developed as a
“bookmarklet”, a web link that activates a javascript to extract the forum interaction
data for analysis and visualisation. The exported forum data was then analysed using



20 Australasian Journal of Educational Technology, 2011, 27(1)

Netdraw (Bakharia & Dawson, 2009; Borgatti, 2002), a third party social network
visualisation software program that allows calculation of network centrality properties
such as degrees, closeness, betweenness and eigenvector.

In a learning community network, the centrality measure degrees describes the
number of ties an individual student has established with other actors (students). A
high number of degrees indicates that an individual has established many
relationships that offer future opportunities for assistance with academic and social
needs. The centrality measure closeness is defined as the degree of relationships an
actor has formed with the entire network (Otte & Rousseau, 2002). A betweenness
measure commonly reflects an individual’s potential access to information as it flows
through the network. A student with a high betweenness score is well positioned to
control the flow of information and resources throughout a learning network.
Individuals with such positions in a network frequently act as bridges between
otherwise isolated or disparate clusters (Burt, 2004). Lastly, the eigenvector measure is
an indicator of a student’s overall importance within the broader network structure
(Faust, 1997). Briefly, eigenvector adopts a weighting system based on the centrality of
all the connected nodes as well as the strength of the individual ties. The assumption is
that the better connected nodes are rated more highly than their less well-connected
peers. Thus, eigenvector provides a good indication of the value of an individual
student’s contribution to the learning community (see Wasserman and Faust (1994) for
a more detailed overview of social network analysis).

Assessment

Student assessment scores were calculated from student performance on the final end
of first year written exam. This final exam is used to determine student progression
into second year of the medical course.

Statistical analyses

The admission, exam and community data were analysed using the software package
SPSS for Windows (vers 18.0) incorporating descriptive statistics and a simple non-
parametric correlation, to ascertain the degree of relationship between the investigated
variables. The descriptive statistics indicated that the sampled population was a non-
normal distribution with respect to the reported admissions criteria. While this was
anticipated as a result of the admissions screening processes, the subsequent statistical
analyses necessitated the use of non-parametric statistical tests. In this instance the
non-parametric test was Spearman’s rank correlation. This test is frequently utilised
when population samples demonstrate a non-normal distribution (Tabachnick &
Fidell, 2001). The reported statistical analyses were undertaken to investigate the
relationship between student admissions criteria, end of year exam results and student
engagement in the learning community. Learning community engagement was
determined using social network centrality measures.

Results
Participants

The study included all first year domestic medical students for the 2007 (n = 64) and
2008 (n= 68) calendar years (N=132). Females represented 56% (n = 74) and males 44%
(n = 58) of the sampled population. The mean age of participants was 26.8 (SD = 4.34),



Dawson, Macfadyen, Lockyer and Mazzochi-Jones 21

with male sub-population slightly older than females reporting a mean age of 27.2 (SD
= 4.17) and 26.5 (SD = 4.47) respectively.

Descriptive statistics

Various descriptive statistics were generated to provide a quick overview of student
admissions criteria scores and final exam results (Table 1). As the examined variables
were calculated from alternate scales, the raw data was firstly standardised as Z scores
to allow for assessment of covariance. While all sub-populations were comparable in
results, there were some identified areas of significant difference (Table 1). For
instance, a paired-sample t test illustrates a significant difference between the 2007 and
2008 cohort scores relating to the interview and exam results, t(130) = 6.58, p < 0.001
and t(130) = 2.97, p < 0.005 respectively. Similarly, a significant difference was
observed between female and male sub-populations relating to GPA, with females
reporting a small, yet significantly higher GPA score (t (130) = 3.30, p < 0.001).

Table 1: Mean and standard deviation for student
Admissions criteria and final exam results

All Female Male 2007 2008
GPA 5.98

(SD = 0.46)
6.10 *

(SD = 0.45)
5.84 *

(SD = 0.44)
6.00

(SD = 0.44)
5.97

(SD = 0.49)
GAMSAT 58.87

(SD = 4.89)
58.66

(SD = 5.16)
59.16

(SD = 4.55)
58.75

(SD = 3.71)
59.00

(SD = 5.80)
Portfolio 27.64

(SD = 7.37)
27.2

(SD = 7.74)
28.2

(SD = 6.9)
28.91

(SD = 7.56)
26.46

(SD = 7.03)
Interview 52.2

(SD = 6.20)
52.34

(SD = 6.47)
52.03

(SD = 5.89)
55.39 **

(SD = 4.67)
49.20 **

(SD = 6.00)
Final 1st Year
exam

71.14
(SD = 8.58)

71.4
(SD = 9.51)

71.78
(SD = 7.35)

73.67 **
(SD = 6.69)

69.32 **
(SD = 9.64)

* Indicates a significant difference between female and male sub-populations
** Indicates a significant difference between 2007 and 2008 cohorts

Social network analysis

The extraction of the discussion forum data and student developed relationships
afforded an opportunity to re-create the social network. In this instance the network,
and hence learning community engagement, is defined through the post and reply
activity in the course discussion forum. Table 2 illustrates the social network centrality
measures for the sampled population.

Table 2: Mean and standard deviation for student social network centrality measures

All Female Male 2007 2008
Degrees 6.95

(SD = 7.90)
7.62

(SD = 8.30)
6.10

(SD = 7.33)
5.23

(SD = 5.31)
8.57

(SD = 9.49)
Betweenness 38.29

(SD = 73.66)
43.43

(SD = 83.55)
31.73

(SD = 58.74)
36.42

(SD = 55.28)
40.04

(SD = 87.9)
Closeness 657.89

(SD = 834.00)
661.89

(SD = 791.61)
652.79

(SD = 892.20)
838.20

(SD = 888.19)
488.19

(SD = 221.25)
Eigenvector 0.066

(SD = 0.07)
0.072

(SD = 0.07)
0.059

(SD = 0.065)
0.065

(SD = 0.067)
0.067

(SD = 0.067)



22 Australasian Journal of Educational Technology, 2011, 27(1)

Admissions criteria and grades

The student admissions data were correlated against students’ end of first year
performance as determined by their final awarded grades. The results indicate that
prior student grades (GPA) and standardised tests (GAMSAT) are weak significant
predictors of student first year academic performance (r = 0.172, p < 0.05; r = 0.188, p <
0.05 respectively). Interview scores also significantly correlated with student academic
performance (r = 0.253, p < 0.05). In essence, a student’s previous test scores and
interviews are significant indicators of future academic performance. The observed
significant correlations also held true for each of the sub-populations – male, female,
and 2007 and 2008 cohorts.

Admissions criteria and learning community engagement

In an attempt to identify lead indicators of student engagement within the course
learning community, the social network centrality data were correlated against the
student admissions criteria. In this instance the significant correlations were identified
between interviews and closeness (r = 0.311, p < 0.05), and between interviews and
eigenvector scores (r = 0.152, p < 0.05). The results indicate that no significant
correlations exist between the SNA measures and GAMSAT and GPA criteria. Again,
the correlation trend was largely replicated within the various sub-populations (Table
3).

Table 3: Correlation between social network
centrality measures and student admissions criteria

Centrality
measure Interview Portfolio

Degrees
Betweenness
Closeness r = 0.311

All students

Eigenvector r = 0.152
Degrees
Betweenness
Closeness r = 0.296

Female

Eigenvector r = 0.114
Degrees
Betweenness
Closeness r = 0.308

Male

Eigenvector r = 0.171
Degrees
Betweenness
Closeness r = 0.198

2007 Cohort

Eigenvector
Degrees r = 0.296
Betweenness r = 0.267 r = 0.260
Closeness r = 0.333

2008 Cohort

Eigenvector r = 0.280 r = 0.263
Only variables of significance are listed (p < 0.05)
No significant correlations were observed between centrality measures
and GAMSAT and GPA criteria

Interestingly, the 2007 cohort demonstrated a significant correlation only between
interviews and closeness. Modifications to the discussion forum structure for the 2008
cohort may have contributed to the stronger observed correlations in the later cohort in



Dawson, Macfadyen, Lockyer and Mazzochi-Jones 23

comparison to their 2007 peers. Additionally, the use of student portfolios to
demonstrate desirable skills and attributes such as leadership was only demonstrated
to have a significant relationship with community engagement measures in the 2008
cohort. In sum, these results indicate that student prior grades (GPA and GAMSAT
scores) are not significant predictors of student capacity or willingness to engage in the
course learning community.

Discussion
This study aimed to investigate the potential relationship between the various medical
student admissions criteria, exam performance and engagement within the course
learning community. The analysis of the results was based on the assumption that
learning community engagement is an indicator that a student possesses the desirable
skills and attributes that are reflective of stated university graduate outcomes. While
this study is far from definitive, the preliminary results do indicate that medical
student admissions based on prior test scores will pre-select for a cohort that is
academically capable and largely prepared for university success. In this instance,
success is narrowly defined as a measure of future exam performance and thus, the
demonstration of discipline specific knowledge. The results of this initial study also
suggest that so-called “soft skills” such as creativity, communication, collaboration and
leadership are overlooked and unaccounted for in a grades-based admissions process.
Herein lies the rub.

Universities internationally are focussing greater attention and resources towards the
development of graduates who demonstrate the types of skills and attributes necessary
for the conceptual age. In Daniel Pink‘s (2005) terms, the new workforce will be
characterised by its capacity to re-contextualise existing products and services to
value-add in short time frames. Workers in the conceptual age who possess both high
concept and high touch aptitudes will be in greatest demand (Pink, 2005). Previously,
these sorts of activities (complex conceptualisations merged with design and
aesthetics) required skills that were commonly associated with the creative arts.
However, numerous studies have recently challenged these notions, demonstrating
that the skills and attributes frequently associated with creativity are essential for
broad social and economic enterprise (Cunningham, 2006; Florida, 2004; Haring-Smith,
2006; Hartley, 2004). As Mihaly Csikszentmihalyi asserts, creativity is “no longer a
luxury for the few, but... a necessity for all” (2006, p. xviii). The push and pull for the
rise of these new combinations and complementary skills for the application of
knowledge within an organisation is well summarised by Richard Florida and Jim
Goodnight (2005), who note that:

A company’s most important asset isn’t raw materials, transportation systems, or
political influence. It’s creative capital - simply put, an arsenal of creative thinkers
whose ideas can be turned into valuable products and services. (p. 125)

The importance of developing creative capacity in learners is further emphasised in
the recent European University Association’s report on creativity in higher education
(European University Association, 2007). In this report the EUA are emphatic in noting
that higher education must equip graduates with the necessary skills to cope and
prosper in a world that is characterised by constant change – a world that demands
adaptability, flexibility and a tolerance for uncertainty. The call for these forms of
creativity is also reflected in the Australian higher education sector. For instance,
McWilliam and Dawson (2008) noted that 75% of all Australian universities reference



24 Australasian Journal of Educational Technology, 2011, 27(1)

the term creativity in university strategic documentation such as graduate outcomes.
However, the capacity to readily foster such skills is not necessarily selected for, nor
promoted by a grade-based admissions process. While there are obvious workload
implications associated with broad-based admission processes influencing the use of
alternate metrics, it could be argued that these desirable skills and attributes can still
be cultivated through learning and teaching practice.

As alluded to earlier in this paper, social learning has received much attention in the
educational literature largely for its capacity to move curriculum emphasis from “what
we are learning to how we are learning” (Brown & Adler, 2008). A focus on how begins
to describe the types of student engagements teachers can inculcate, through well
designed activities and relevant content. In short, these types of learning engagements
are frequently described as learning communities or communities of practice. It is
through such ongoing social interactions that students foster skills such as leadership
(Wang, 2010), creativity (Burt, 2004; Dawson, 2010), communication, and critical
reasoning (Hakkarainen, Lipponen, & Jarvela, 2002). Few would argue the merits of
developing and implementing social learning activities. However, despite their best
intentions, teachers may be faced with unforseen challenges associated with student
motivations for collaboration. The results of this study suggest that some students are
more motivated and able to pursue collaborations with their learning community
earlier in the academic course than others. Medical students with high interview scores
were more inclined to participate and engage in the student learning community than
students with a low interview score.

This observation may reflect student motivation and goal alignment. For example,
early research by Johnson and Johnson (1995) has demonstrated that participation
within group activities is largely influenced by how well an individual’s goals and
motivations align with their fellow group members. The network analyses undertaken
in the present study reinforce this point. It is not merely the number of unique
relationships students have established (degrees) that correlate with interview scores.
It is how important those relationships are in relation to the majority of the
community. Interview scores therefore appear to select for students who are perceived
to value add to the student learning community. Further investigation is required to
discover whether it can be extrapolated from this idea that interview scores and social
network measures are an early indicator of leadership and creative capacity.

Although admission based on grades are cost-effective, countries such as Canada that
do not have a standardised testing process at either provincial or federal level are
increasingly concerned about grade inflation. Grades awarded at the secondary school
level impact future admission to university, and students are increasingly being
offered opportunities to obtain additional grade credit to ensure they submit a strong
and competitive admission application. For example, the Faculty of Arts at the
University of British Columbia has observed a 7% increase in the cut off grade for
admission to the Arts program (from 78% to 85%, 2002-2009). While this increase
reflects a growing applicant pool, it also illustrates the mounting competiveness for
achieving high grades for university admission. As a result the university and
associated faculties are recognising an opportunity to reduce the emphasis on grades,
to include more criteria that would effectively increase the diversity, skills and
attributes that reflect the ethos of the institution and its broad mandate.

Medical education has not been immune to the pressures for developing the more non-
cognitive personal qualities (Powis, 1994). The medical profession is a rapidly and



Dawson, Macfadyen, Lockyer and Mazzochi-Jones 25

constantly changing environment, in terms of the science, technology, governance, and
communication practices. Emergent medical professionals require skills associated
with communication, leadership, flexibility and a tolerance for change and uncertainty.
The call for more development and associated recruitment for these creative skills and
attributes within the medical profession is well promoted by Shee Lippell. For
instance, Lippell (2002) argued that while the traditional medical school curriculum
frequently suppresses the development of student creative skills, these are the very
same skills that are necessary for developing "the  'doctors of tomorrow' that society
requires" (p. 520). As in most professions, in addition to their discipline knowledge,
medical practitioners require the capacity to problem solve, collaborate, and
importantly communicate and translate information to a large diversity and range of
stakeholders (specialists to generalists to lay people). Such skills are not developed in
isolation, or through the emphasis on the admission of students on the basis of grades.
Discipline knowledge is not the sole criteria by which our future medical graduates
will be judged. The importance for fostering skills associated with safe and effective
doctors and greater integration of alternate metrics for medical admission is well
argued by McManus and Powis (2007) in noting that:

In medical school and beyond, most measures of competence assess knowledge,
whereas being a competent, safe and effective doctor probably depends to an equal
extent on behaviour, attitudes and approaches. Adequate knowledge is a necessary
part of good medicine, but is far from sufficient for competent practice. (p.118)

Given the need for, and obvious recognition of, the vast array of skills and attributes
associated with effective medical practice, it is surprising that medical schools such as
the University of Queensland have dismissed the relevance of interviews in favour of a
greater emphasis on knowledge based assessments for student admission (McManus
& Powis, 2007; University of Queensland, 2010).

The results of this study indicate that while grades-based admissions processes select a
student cohort who will perform well on future exams, they do not provide an
indication of other, desirable graduate qualities that reside in the more non-cognitive
domain. While there remains weighted emphasis on student assessment for
admissions practices there will be a parallel promotion towards more performance
base achievement orientations (Dweck, 2000). When performance on testing is the
basis for admission, savvy students will rapidly prioritise tasks to ensure they remain
highly competitive and ahead of the game.

References
Bakharia, A. & Dawson, S. (2009). Social Network Adapting Pedagogical Practice (SNAPP) (Version

1.5) [Bookmarklet]. Brisbane, Australia.
Barro, R. J. (2001). Economic viewpoint: Why college shouldn't dump the SAT. Business Week, 9

April, 20.
Borgatti, S. P. (2002). NetDraw: Graph visualization software. Harvard: Analytic Technologies.
Brown, J. S. & Adler, A. P. (2008). Minds on fire: Open education, the long tail, and learning 2.0.

EDUCAUSE Review, 43(1), 16-32. http://net.educause.edu/ir/library/pdf/ERM0811.pdf
Burt, R. (2004). Structural holes and good ideas. The American Journal of Sociology, 110(2), 349-399.
Csikszentmihalyi, M. (2006). Foreword: Developing creativity. In N. Jackson, M. Oliver, M. Shaw

& J. Wisdom (Eds.), Developing creativity in higher education: An imaginative curriculum (pp.
xviii-xx). London: Routledge.



26 Australasian Journal of Educational Technology, 2011, 27(1)

Cunningham, S. (2006). What price a creative economy? Platform Papers: Quarterly Essay on the
Performing Arts, No. 9. Sydney: Currency House.

Dawson, S. (2010). 'Seeing' the learning community: An exploration of the development of a
resource for monitoring online student networking. British Journal of Educational Technology,
41(5), 736-752.

Dweck, C. (2000). Self-theories: Their role in motivation, personality, and development. Philadelphia:
Psychology Press.

Espenshade, T. & Chang, C. (2005). The opportunity cost of admission preference at elite
universities. Social Science Quarterly, 86(2), 293-305.

European University Association (2007). Creativity in higher education. Brussels, Belgium:
European University Association. [verified 2 Feb 2011]
http://www.eua.be/typo3conf/ext/bzb_securelink/pushFile.php?cuid=400&file=fileadmin
/user_upload/files/Publications/Creativity_in_higher_education.pdf

Eva, K. W., Rosenfeld, J., Reiter, H. I. & Norman, G. R. (2004). An admissions OSCE: The
multiple mini-interview. Medical Education, 38(3), 314-326.

Faust, K. (1997). Centrality in affiliation networks. Social Networks, 19(2), 157-191.

Florida, R. (2004). The rise of the creative class: And how it's transforming work, leisure, community
and everyday life. New York, NY: Basic Books.

Florida, R. & Goodnight, J. (2005). Managing for creativity. Harvard Business Review, 83(7), 124-
131.

Groves, M. A., Gordon, J. & Ryan, G. (2007). Entry tests for graduate medical programs: Is it time
to re-think? The Medical Journal of Australia, 186(3), 120-123.

Hakkarainen, K., Lipponen, L. & Jarvela, S. (2002). Epistemology of inquiry and computer-
supported collaborative learning. In T. Koschmann, R. Hall & N. Miyake (Eds.), CSCL 2:
Carrying forward the conversation (pp. 129-156). Mahwah, New Jersey: Lawrence Erlbaum.

Haring-Smith, T. (2006). Creativity research review: Some lessons for higher education. Peer
Review, 8(2), 23-27.

Hartley, J. (2004). The value chain of meaning and the new economy. International Journal of
Cultural Studies, 7(1), 129-141.

Haythornthwaite, C. (2006). Learning and knowledge networks in interdisciplinary
collaborations. Journal of the American Society for Information Science and Technology, 57(8),
1079-1092.

Johnson, D. W. & Johnson, R. T. (1995). Social interdependence: Cooperative learning in
education. In B. B. Bunker & J. Z. Rubin (Eds.), Conflict, cooperation, and justice: Essays inspired
by the works of Morton Deutsch (pp. 205-251). San Francisco, CA: Jossey-Bass.

Komives, S. R., Longerbeam, S. D., Mainella, F., Osteen, L., Owen, J. E. & Wagner, W. (2009).
Leadership identity development: Challenges in applying a developmental model. Journal of
Leadership Education, 8(1), 11-47.

Lave, J. (1993). Situating learning in communities of practice. In L. B. Resnick, J. M. Levine & S.
D. Teasley (Eds.), Perspectives on socially shared cognition (pp. 17-36). Washington, DC:
American Psychological Association.

Lippell, S. (2002). Creativity and medical education. Medical Education, 36(6), 519-521.

McManus, C., & Powis, D. (2007). Testing medical school selection tests: Why is so little known
about what works in selecting medical students? Medical Journal of Australia, 186(3), 118-119.



Dawson, Macfadyen, Lockyer and Mazzochi-Jones 27

McWilliam, E. (2008). The creative workforce: How to launch young people into high flying futures.
Sydney: UNSW Press.

McWilliam, E. & Dawson, S. (2008). Teaching for creativity: Towards sustainable and replicable
pedagogical practice. Higher Education, 56(6), 633-643.

Otte, E. & Rousseau, R. (2002). Social network analysis: A powerful strategy, also for the
information sciences. Journal of Information Science, 28(6), 441-453.

Pink, D. H. (2005). A whole new mind. New York: Penguin.

Powis, D. (1994). Selecting medical students. Medical Education, 28, 443-469.

Powis, D., Hamilton, J. & McManus, I. C. (2007). Widening access by changing the criteria for
selecting medical students. Teaching and Teacher Education, 23(8), 1235-1245.

Seely Brown, J. (2006). New learning environments for the 21st century. Change, 6(5), 2-18.

Tabachnick, B. G. & Fidell, L. S. (2001). Using multivariate statistics (4th ed.). Boston: Allyn &
Bacon.

University of Queensland (2010). School of Medicine: Domestic student admissions. [verified 2
Feb 2011] http://www2.som.uq.edu.au/som/FutureStudents/MBBS/Admissions/
DomesticStudents/Pages/DomesticStudents.aspx

Wang, L. (2010). How social network position relates to knowledge building in online learning
communities. Frontiers of Education in China, 5(1), 4-25. [verified 2 Feb 2011]
http://www.springerlink.com/index/5R7712095761Q21T.pdf

Wasserman, S. & Faust, K. (1994). Social network analysis: Methods and applications. New York:
Cambridge University Press.

Yang, H.-L. & Cheng, H.-H. (2010). Creativity of student information system projects: From the
perspective of network embeddedness. Computers & Education, 54(1), 209-221.

This article received an Outstanding Paper Award at ascilite Sydney 2010 Conference,
gaining the additional recognition of republication in AJET (with minor changes
including a revised title). The reference for the Conference version is:

Dawson, S., Macfadyen, L., Lockyer, L. & Mazzochi-Jones, D. (2010). From neural
to social: Medical student admissions criteria and engagement in a social learning
environment. In Curriculum, technology & transformation for an unknown future.
Proceedings ascilite Sydney 2010. http://ascilite.org.au/conferences/sydney10/
Ascilite%20conference%20proceedings%202010/Dawson-full.pdf

Dr Shane Dawson, Director ISIT, Faculty of Arts, University of British Columbia,
Canada. Email: shane.dawson@ubc.ca
Leah Macfadyen, Science Centre for Learning and Teaching, University of British
Columbia, Canada. Email: leah.macfadyen@ubc.ca
Lori Lockyer, Faculty of Education, University of Wollongong, Australia.
Email: llockyer@uow.edu.au
David Mazzochi-Jones, Graduate School of Medicine, University of Wollongong,
Australia. Email: davidmj@uow.edu.au

Please cite as: Dawson, S., Macfadyen, L., Lockyer, L. & Mazzochi-Jones, D. (2011).
Using social network metrics to assess the effectiveness of broad based admission
practices. Australasian Journal of Educational Technology, 27(1), 16-27.
http://www.ascilite.org.au/ajet/ajet27/dawson.html