10 International Journal of Cultic Studies ■ Vol. 1, No. 1, 2010
To do this, we concentrated in part on whether
the group-focussed treatment programme was
efficacious—and all the results consistently
show that it is. But at the same time, we used the
measure of Group Psychological Abuse (GPA)
developed by Chambers and colleagues to gauge
the perceived level of former group abuse. Now
it is also important to acknowledge that the
sample of ex-members that visits Wellspring is
self-selecting, and of course they are all ex-
members. It has so far proven very difficult in
this field—particularly when it comes to
psychometric tests—to get measures of
psychopathology from current members of
extremist groups. Maybe that day will come, but
understandably the groups rarely allow
researchers anywhere near their members with
such instruments!
So this research has this important design
characteristic, which always means that we are
looking to establish, post-hoc, whether there
appears to be a unique pattern of group-based
psychopathology for this particular sample. This
does not mean that all group members have this
pattern of harm. To quote Michael Langone,
Executive Director of the International Cultic
Studies Association, “some cults hurt some
people some of the time.” We can tell you about
some and not about others. About the others, we
can make no comment. We also cannot tell you
precisely, when we observe psychopathology
amongst ex-members of cults, how much if any
of this psychopathology predated the group
experience. However, again this does not render
meaningless what we do know and can discern
about the group-related aspects of the
psychopathology we have measured, in the same
way that we do not discount the trauma and
psychological effects of the earthquake or the
train crash just because the psychologist wasn’t
able to measure the victims beforehand to see
whether they had any of the same symptoms
before. At any rate, there are tests that we
employ that start to tease out that which might
not be group related and the same is true in the
analysis of disaster survivors. An equal playing
field is all that is called for here.
In 2003, in order to operationalise aspects of
Lifton and Social Identity Theory, I, together
with Paul Martin and Ron Burks, designed the
Extent of Group Identity Scale (EGIS). This is a
measure that builds on a measure I developed in
my doctoral research, which measures the extent
of group identification or social category
“accessibility,” to use Turner and Oakes’ term,
and the extent of “perceptual readiness,” to use
Bruner’s term (upon which Turner and Oakes
devised the notions of different levels of
accessibility). Put simply, EGIS asks former
members to think back and answer how much
they identified with the group when they were
still members. It asks, amongst other things,
How important was the group to you? How
much time did it take? How valued did it make
you feel? How unhappy would you have felt
about leaving when you were still a member?
This is a value-neutral measure of how
important the group was, and an attempt to
measure to what extent the group had become
dominant psychologically, in the way that the
Totalistic Identity Theory hypothesises. If the
GPA measures the group environment—the
milieu, then together these two measures are an
attempt to capture the category accessibility and
the category fit (extent of group-relevant stimuli
in the comparative context) that define, in Self-
Categorisation Theory, the psychological
salience of the social category or group.
We predicted that these two measures would be
positively related and would also demonstrate
positive relationships with the clinical measures.
In particular, we hypothesized that the extent of
group identity and the extent of group abuse
would predict levels of depression and anxiety,
and also, importantly, levels of dissociation. The
predicted relationship with dissociation is
particularly important as it relates to the
existential loss of a dominant group identity and
the associated loss of former group- and
personal-level identities that preceded that. All
in all, these measures should tell us how much
the group filled up the person to the exclusion of
other identities, and how much of a
psychological hole was left when the person was
no longer a member.
Key Findings
Now, in something of a conclusion, are some
statistics! This is not as bad as when I taught
statistics on a Friday afternoon or a Monday
To do this, we concentrated in part on whether
the group-focussed treatment programme was
efficacious—and all the results consistently
show that it is. But at the same time, we used the
measure of Group Psychological Abuse (GPA)
developed by Chambers and colleagues to gauge
the perceived level of former group abuse. Now
it is also important to acknowledge that the
sample of ex-members that visits Wellspring is
self-selecting, and of course they are all ex-
members. It has so far proven very difficult in
this field—particularly when it comes to
psychometric tests—to get measures of
psychopathology from current members of
extremist groups. Maybe that day will come, but
understandably the groups rarely allow
researchers anywhere near their members with
such instruments!
So this research has this important design
characteristic, which always means that we are
looking to establish, post-hoc, whether there
appears to be a unique pattern of group-based
psychopathology for this particular sample. This
does not mean that all group members have this
pattern of harm. To quote Michael Langone,
Executive Director of the International Cultic
Studies Association, “some cults hurt some
people some of the time.” We can tell you about
some and not about others. About the others, we
can make no comment. We also cannot tell you
precisely, when we observe psychopathology
amongst ex-members of cults, how much if any
of this psychopathology predated the group
experience. However, again this does not render
meaningless what we do know and can discern
about the group-related aspects of the
psychopathology we have measured, in the same
way that we do not discount the trauma and
psychological effects of the earthquake or the
train crash just because the psychologist wasn’t
able to measure the victims beforehand to see
whether they had any of the same symptoms
before. At any rate, there are tests that we
employ that start to tease out that which might
not be group related and the same is true in the
analysis of disaster survivors. An equal playing
field is all that is called for here.
In 2003, in order to operationalise aspects of
Lifton and Social Identity Theory, I, together
with Paul Martin and Ron Burks, designed the
Extent of Group Identity Scale (EGIS). This is a
measure that builds on a measure I developed in
my doctoral research, which measures the extent
of group identification or social category
“accessibility,” to use Turner and Oakes’ term,
and the extent of “perceptual readiness,” to use
Bruner’s term (upon which Turner and Oakes
devised the notions of different levels of
accessibility). Put simply, EGIS asks former
members to think back and answer how much
they identified with the group when they were
still members. It asks, amongst other things,
How important was the group to you? How
much time did it take? How valued did it make
you feel? How unhappy would you have felt
about leaving when you were still a member?
This is a value-neutral measure of how
important the group was, and an attempt to
measure to what extent the group had become
dominant psychologically, in the way that the
Totalistic Identity Theory hypothesises. If the
GPA measures the group environment—the
milieu, then together these two measures are an
attempt to capture the category accessibility and
the category fit (extent of group-relevant stimuli
in the comparative context) that define, in Self-
Categorisation Theory, the psychological
salience of the social category or group.
We predicted that these two measures would be
positively related and would also demonstrate
positive relationships with the clinical measures.
In particular, we hypothesized that the extent of
group identity and the extent of group abuse
would predict levels of depression and anxiety,
and also, importantly, levels of dissociation. The
predicted relationship with dissociation is
particularly important as it relates to the
existential loss of a dominant group identity and
the associated loss of former group- and
personal-level identities that preceded that. All
in all, these measures should tell us how much
the group filled up the person to the exclusion of
other identities, and how much of a
psychological hole was left when the person was
no longer a member.
Key Findings
Now, in something of a conclusion, are some
statistics! This is not as bad as when I taught
statistics on a Friday afternoon or a Monday
