1. The group comparison of OG versus FO
resulted in a significant difference (Hedge’s g
= 0.749, p < 0.001), representing a relatively
large effect size. The outcome of the student
connectedness and overall SOC in the OG
versus Hy condition resulted in a smaller
effect size, which was not statistically
significant (g = -.0158, p = 0.566). When
results from OL and Hy conditions were
aggregated, a significant difference between
them and OG contexts was observed (g =
0.459, p = 0.003).
Meta-Analytic Study: Student Connectedness across On-
ground, Fully Online, and Hybrid Contexts
Hannah L. Smith and An H. Dang
Concordia University, Portland, Oregon; CU Advisor: Reed M. Mueller, Ph.D.
Introduction Method
Using only quasi-experimental studies,
this meta-analysis was conducted to assess
student sense of classroom community
across three class contexts in higher
education settings: On-ground (OG), Hybrid
(Hy), and Fully Online (FO).
According to McMillan (1976), sense of
community (SOC) “is a feeling that members
have of belonging, a feeling that members
matter to one another and to the group, and
a shared faith that members’ needs will be
met through their commitment to be
together,” (as cited in McMillan & Chavis,
1986, p. 9). More recently, and specifically
applied to classroom settings, Rovai (2002)
used McMillan and Chavis’ model as a
foundation for the widely cited Classroom
Community Scale (CCS). The CSS has two
subscales, connectedness, which is of
primary interest in this study, and perceived
learning. The connectedness subscale
represents the feelings of connectedness,
cohesion, spirit, trust, and interdependence
that student's report when considering their
classroom experience (Rovai, 2002).
The connectedness subscale is also
related to the literature on the Social
Presence domain of the Community of
Inquiry (CoI) model (Garrison, Anderson, &
Archer, 2010). Social presence is understood
through three lenses: “participants
identifying with the community,
communicating purposefully in a trusting
environment, and developing interpersonal
relationships” (p. 7).
Given its relevance to both SOC and CoI
the CSS has been widely use as a measure in
may descriptive studies. What has not been
conducted is a meta-analysis of quasi-
experimental studies using the CSS to assess
on student connectedness among OG, FO,
and Hy classroom settings. It was our
hypothesis that OG would exceed FO and Hy
with regard to levels student connectedness.
This study utilized a meta-analytic research
approach in which an initial pool of studies
was secured using a pre-defined search
string across pre-defined academic
databases. Once the original pool of studies
was obtained, inclusion and exclusion
criteria were used to select studies
appropriate for the study. Coding and
analysis proceeded thereafter.
Inclusion/Exclusion: There were 40 studies
included after the preliminary pool, with
four duplicates. During the secondary review
phase, 18 articles were excluded because
they were non-experimental, descriptive, or
reported only qualitative data. In addition,
studies were also excluded if data reporting
was insufficient for analysis. Overall, there
were 6 studies that were included for the
final analysis with the connectedness
subscale, as well as one study (Haar &
Scanlan, 2012) that reported the
connectedness and learning outcomes
subscales as an aggregate score, without
reporting subscale results.
Conclusions
Students enrolled in on-ground courses
reported greater levels of connectedness and
overall SOC than those enrolled in fully
online courses; however, those enrolled in
Hybrid courses reported similar connected
levels as those enrolled in on-ground courses.
A small number of qualifying quasi-
experimental studies (and no experimental
studies) weakness our study; inadequate data
reporting within selected studies was an
additional limitation. Replication of this
analysis using other social presence scales is
an important area for future research.
References
Dawson, S. (2008). A study of the relationship between student social
networks and sense of community. Educational Technology &
Society,11(3), 224-238.
Drouin, M., & Vartanian, L. R. (2010). Students' feelings of and desire
for sense of community in face-to-face and online courses. The
Quarterly Review if Distance Education, 11(3), 147-159.
Garrison, D. R., Anderson, T., & Archer, W. (2010). The first decade of
the community of inquiry framework: A retrospective. The Internet
and Higher Education, 13(1-2), 5-9.
doi:10.1016/j.iheduc.2009.10.003
Haar, M., & Scanlan, C. (2012). Factors associated with sense of
community among allied health students. Journal of Allied Health,
41(3), 123-130.
McMillan, D.W. & Chavis, D.M. (1986). Sense of community: A
definition and theory. Journal of Community Psychology, 14 (1), 6-
23.
Ritter, C., Polnick, B., Fink II, R., & Oescher, J. (2010). Classroom
learning communities in educational leadership: A comparison study
of three delivery options. Internet and Higher Education, 13(1/2),
96-100.
Rovai, A. P. (2002). Development of an instrument to measure
classroom community. The Internet and Higher Education, 5(3),
197-211.
Wang, S.-K. (2008). The effects of a synchronous communication tool
(yahoo messenger) on online learners' sense of community and their
multimedia authoring skills. Journal of Interactive Online Learning,
7(1), 59-74.
Acknowledgements
I am grateful for the assistance from Reed
Mueller, whose guidance and contribution
helped make this project possible. I would
also like to thank An H. Dang for all of her
help.
For further information
Email: smith.19.hannah@gmail.com
Research advisor’s email: rmueller@cu-
portland.edu
Results
Study Name Comparison Statistics for each study
Hedge's g Standard error Variance Lower Limit Upper Limit Z -Value p-Value
Wang (2008)
OG (n=6) vs. FO
(n=14) 1.139 0.501 0.251 0.158 2.121 2.275 0.023
Ritter et al. (2010)
OG (n=33) vs.
FO (n=73) 1 0.219 0.048 0.57 1.43 4.56 0
Dawson (2008)
OG (n=372) vs.
FO (n=92) 0.924 0.12 0.014 0.689 1.159 7.691 0
Drouin & Vartanian
OG (n=79) vs.
FO (n=119) 0.676 0.149 0.022 0.385 0.967 4.554 0
Haar & Scanlan (2012)
OG (n=70) vs.
FO (n=128) 0.495 0.15 0.023 0.2 0.789 3.295 0.001
Rovai & Jordan (2004)
OG (n=24) vs.
FO (n=21) 0.414 0.297 0.088 -0.168 0.995 1.394 0.163
OG vs. FO Overall 0.749 0.189 0.036 0.38 1.119 3.974 0
Haar & Scanlan (2012)
OG (n=70) vs.
Hy (n=82) 0.543 0.165 0.027 0.22 0.867 3.296 0.001
Ritter et al. (2010)
OG (n=33) vs.
Hy (n=16) 0 0.3 0.09 -0.587 0.587 0 1
Rovai & Jordan (2004)
OG (n=24) vs.
Hy (n=23) -1.298 0.317 0.1 -1.918 -0.677 -4.099 0
OG vs. Hy Overall OG vs. Hy -0.158 0.275 0.075 -0.696 0.381 -0.573 0.566
OG vs FO/Hy Total 0.459 0.155 0.024 0.154 0.764 2.952 0.003