Testing a Single-Item Visual Analogue Scale as a Proxy for
Cohesiveness in Group Psychotherapy
Matthew J. Hornsey, Sara Olsen,
and Fiona Kate Barlow
The University of Queensland
Tian P. S. Oei
The University of Queensland and Toowong
Private Hospital, Brisbane, Australia
Group cohesion is one factor that is widely suggested to be important in producing
clinically meaningful change in group psychotherapy. However, the construct has
proved difficult to define, and in an effort to capture the construct’s multidimen-
sionality, contemporary measures have become long, cognitively demanding, and
challenging for people with limited literacy. In response to this, we test a single-
item visual analogue scale that provides a simple, intuitive, time-efficient, and
user-friendly proxy for the cohesion construct. The Group Entitativity Measure–
Group Psychotherapy (GEM-GP) was validated in a clinical sample of individuals
who completed a group cognitive– behavioral therapy course for depression.
GEM-GP scores correlated highly with a lengthier, traditional measure of group
cohesion and were just as predictive of outcomes as a multi-item, traditional
measure. The GEM-GP is a valid, user-friendly, and brief proxy for the cohesion
construct in the group psychotherapy context.
Keywords: group psychotherapy, cohesion, visual analogue scale
There is good evidence that group psycho-
therapy is effective in producing significant and
clinically meaningful improvements in a variety
of clinical disorders (Jo ´nsson & Hougaard,
2009). The mechanisms of change within the
group context, however, remain unclear (Oei &
Browne, 2006). Cohesiveness is one factor that
is widely purported to be important in facilitat-
ing meaningful clinical change. It has been de-
scribed as the “bedrock of the group experi-
ence” (Butler & Fuhriman, 1983, p. 500) and
for Yalom (1995) is considered a central cura-
tive factor in group psychotherapy.
The assumption that cohesiveness is impor-
tant in facilitating clinically meaningful change
has been difficult to test, however, partly be-
cause the field has not rallied around one ac-
cepted measure of cohesiveness. One clear
trend has been the shift away from unidimen-
sional measures of cohesion to measures that
incorporate multiple dimensions. A brief sum-
mary of proposed multidimensional measures
of cohesion is provided by Hornsey, Dwyer, and
Oei (2007). What is clear from this summary is
that there is still a degree of confusion about
what the term cohesion is supposed to represent.
It has been variously operationalized as auton-
omy, affiliation, involvement, spontaneity, sup-
port, insight, variety, interest, involvement,
trust, cooperation, expressed caring, focus, in-
dividual attraction, perceived instrumental
value, support for risk-taking behaviors, bond-
ing, listening, empathy, self-disclosure, feed-
back, and process performance (e.g., Braaten,
1991; Budman et al., 1987; Silbergeld, Koenig,
Manderscheid, Meeker, & Hornung, 1975;
Stokes, Fuehrer, & Childs, 1983; see Burlin-
game, McClendon, & Alonso, 2011, for a recent
discussion). In short, just about anything that
This article was published Online First July 11, 2011.
Matthew J. Hornsey, Sara Olsen, and Fiona Kate Barlow,
School of Psychology, The University of Queensland, Bris-
bane, Queensland, Australia; Tian P. S. Oei, School of
Psychology, The University of Queensland and CBT Unit,
Toowong Private Hospital, Brisbane, Australia.
Professor Tian P. S. Oei was a full time academic at the
University of Queensland and had a small (maximum of 6
hours per week for about 45 weeks per year) where he
provided cognitive behavior therapy treatment at the
Toowong Private Hospital to patients. He received fees
directly from the patients and paid expenses from the fees
received.
Correspondence concerning this article should be ad-
dressed to Matthew J. Hornsey, School of Psychology, The
University of Queensland, Brisbane, Queensland 4072.
E-mail: m.hornsey@uq.edu.au
Group Dynamics: Theory, Research, and Practice © 2011 American Psychological Association
2012, Vol. 16, No. 1, 80 –90 1089-2699/11/$12.00 DOI: 10.1037/a0024545
80