Italian Version of the Defense Style Questionnaire
Pietro San Martini, Paolo Roma, Sara Sarti, Vittorio Lingiardi, and Michael Bond
The Defense Style Questionnaire (DSQ) assesses defensive behavior by empirically evaluating conscious
derivatives of defense and coping mechanisms in everyday life. It was developed on the assumption that
defenses can be ordered along a maturity-immaturity
continuum and tend to group into clusters, or defensive styles. The original factor analytical study, by
Bond et al. (1983) identified four styles, called maladaptive, image-distorting, self-sacrifice, and adaptive styles. Successive studies only partially confirmed this factor structure. We present the factor
structure and the main psychometric features of the
Italian version of the questionnaire. The DSQ was
translated into Italian by the back-translation method
and administered to a sample of 294 men (mean age,
33.33 years) and 333 women (mean age, 32.38 years).
An exploratory factor analysis identified three factors
largely corresponding to Bond’s maladaptive, image-
distorting, and adaptive defensive styles and to analogous factors identified by other authors. Accordingly, three defense style scales were constructed,
containing respectively 37, 17, and 12 items. These
scales showed intercorrelations compatible with the
hierarchical model of defensive functioning at the
base of the questionnaire, acceptable, though ameliorable, test-retest reliabilities (r’s ⴝ .79, .63, and .81,
respectively) and, with the exception of the Adaptive
Style scale, sufficient internal consistencies (alphas:
.85, .72, .57). However, only the Maladaptive Style
scale, probably due to its greater length, showed values of reliability and internal consistency high enough
to warrant clinical use in its present form. Further
investigation is required to find new items that may
improve the reliability of the Image-Distorting and the
Adaptive Style scales.
© 2004 Elsevier Inc. All rights reserved.
T
of subjectivity, which necessarily distinguish
clinical opinion, in 1983 Bond et al12 proposed
the first version of the Defense Style Questionnaire (DSQ), a self-administered questionnaire
developed to assess defensive styles. The use of
a self-report questionnaire to assess defensive
behaviors seems to be in contrast to the classical
psychoanalytic concept of defenses as unconscious processes and, therefore, inaccessible to
the ego. In other words, if defenses are considered to be processes out of consciousness, how
can we measure them through a self-report instrument?
Integrating concepts like defense and coping
mechanisms partially helped to overcome this
contradiction. The authors of the DSQ coped
with this question in two steps. On one hand,
they point out that temporary failures of a defense can make individuals aware of the (more
or less automatic) ways they use to defend themselves from unacceptable or distressing feelings
and impulses. On the other hand, the conscious-
HE NOTION of ego defenses is unanimously
recognized as one of the most important psychoanalytic contributions.1 Sigmund Freud, the inventor of this construct,2 stated that defense mechanisms are “the cornerstone on which the whole
structure of psychoanalysis rests”3 (p. 16).
In the mid-1990s, the idea of defensive mechanisms was included in the DSM-IV4 among the
axes suggested for further studies.5-7 Defenses are
defined in the DSM-IV as “automatic psychological processes that protect the individual against
anxiety and from the awareness of internal and
external stressors”4 (p. 715). In this way, clinicians
are encouraged to consider this aspect of personality as an important part of diagnosis and treatment. Despite the importance and the usefulness of
this construct, an important question remains unanswered: Can we “measure” defensive mechanisms in a reliable way?
Among the instruments available today for evaluating defense mechanisms, one of the most interesting is
the Defense Mechanisms Rating Scales (DMRS).8,9
The DSM-IV Defensive Functioning Scale (DFS)
widely refers to the hierarchal structure of the DMRS.
Even though the DMRS has a good inter-rater reliability, in both the original10 and the Italian versions,11 it is
based on the assessment of video-/audiotaped interviews and is still difficult to use on a large level.
Moreover, its administration requires an accurate personal training course.
With the open intention of creating an instrument that is easy to handle and without elements
From the Department of Dynamic and Clinical Psychology,
University of Rome “La Sapienza,”Rome, Italy; and McGill
University and Sir Mortimer B. Davis Jewish General Hospital,
Montreal, Canada.
Address reprint requests to Pietro San Martini, Dipartimento
di Psicologia Dinamica e Clinica Via dei Marsi, 78, 00185
Roma, Italy. E-mail: [email protected]
© 2004 Elsevier Inc. All rights reserved.
0010-440X/04/4506-0013$30.00/0
doi:10.1016/j.comppsych.2004.07.012
Comprehensive Psychiatry, Vol. 45, No. 6 (November/December), 2004: pp 483-494
483
484
ness of their own defensive operations can be
accentuated from other people comments and
behaviors. Moreover, the habitual usage of some
defenses leaves a “print” in some of the individual’s usual attitudes. A sentence like “I’m able to
keep a problem out of my mind until I have time
to deal with it” could point to a habitual attitude,
a “forma mentis” that individuals use as a defense mechanism (suppression) to cope with a
stressor or a conflict. In this way, the DSQ is
created for stimulating individuals to express
their own ways of coping with conflicts and
stressors, and it is based on the clinical assumption that people can describe their past behavior
reliably, taking for granted that only a deep
clinical investigation can identify unconscious
processes at the very moment they happen. Thus,
we should remember that the DSQ evaluates
“conscious derivatives”12 of defensive mechanisms. Moreover, the aim of the DSQ is not just
to measure defense mechanisms separately, but
groups of defenses called defensive styles, as,
according to Bond,13 measuring individual defenses is not reliable.
The authors created this questionnaire starting
from the hierarchical model of Vaillant,14 which
distinguished four defensive levels based on a immaturity–maturity or pathology–mental health
continuum and, by using the empirical method of
factor analysis, they distinguished four groups or
defensive styles, which, even though they did not
correspond with the theoretical levels of Vaillant as
a whole, nevertheless could be arranged according
to its continuum. These styles are as follows: (1)
maladaptive style—composed of acting-out, regression, passive-aggressive behavior, withdrawal,
projection, and inhibition; individuals who endorse
this style show a basic incapability of coping with
internal or external stressors by using actions or
behaviors for their own advantage. (2) Image-distorting style—composed of omnipotence-devaluation, splitting, primitive idealization; individuals
who endorse this pattern basically split the selfimage and the image of others in extreme classes
avoiding a total view of reality that could prove to
be too much stressful. (3) Self-sacrifice style—
composed of defenses like pseudo-altruism and
reaction formation; people who use it tend to deny
their own needs and focus on being kind to others.
(4) Adaptive style—composed of sublimation, humor, and suppression; individuals who use it tend
SAN MARTINI ET AL
to be able to put distressing feelings out of their
minds by focusing on either the humor in situations, or the “silver linings,” or by channeling their
feelings into constructive, creative activities.
In 1986 Bond improved the original version of
the DSQ by increasing the consistency of its scales.
The final version is composed of 88 items linked to
25 defense mechanisms and of a “lie” scale. Each
defense mechanism is measured by a minimum of
one item (pseudo-altruism, sublimation, projective
identification, and fantasy) to a maximum of nine
items (projection).
Several studies performed in different countries
examined the dimensional structure of the questionnaire, with results only partially corresponding
to those originally found by Bond et al.12 The
Australian research of Andrews et al.,15 based on
271 normal individuals and 142 psychiatric patients, showed a three-factor solution. Two of the
factors were similar to the extreme factors, in terms
of Vaillant’s continuum, found by Bond, that is, an
immature factor and a mature factor. The third
factor included defenses belonging both to Bond’s
image-distorting and self-sacrifice styles.
Reister et al.16 analyzed the responses obtained
from a sample of 301 normal subjects with the
German version of the questionnaire and found
four defense factors that could be ordered on the
maturity-immaturity continuum and that, as they
state (a table of the factor loadings is not presented), had the same clinical meaning as the factors described by Bond.
Sammallhati and Alberg,17 in a Finnish sample
of 73 patients and 353 normal subjects, extracted
four factors. The first factor, labeled “immature
style“ included all the defenses of Bond’s maladaptive style with the addition of fantasy, somatization, consumption, and help-rejecting complaining.
The second factor was a ”mature style“ factor,
including task orientation, humor, and sublimation.
The third factor, labeled ”borderline style,“ included omnipotence-devaluation, isolation, denial,
splitting, and projective identification. The fourth
factor, labeled “neurotic style” included reaction
formation, undoing, pseudo-altruism, and primitive idealization. Despite differences in the labels,
on the whole, the nature of the factors extracted
shows strong similarities with Bond’s styles.
More recently, Rutherford et al.18 studied two
different versions of the questionnaire in a sample
of 215 patients undergoing methadone therapy.
ITALIAN VERSION OF THE DSQ
They used rather strict criteria to decide on the
number of factors, among them the criterion that
each retained factor should contain at least five
defenses with loadings ⱖ.30, and found two factors for both versions. The first factor included 16
defenses normally considered immature or neurotic, and the second depicted more mature defenses (task orientation, anticipation, reaction formation, suppression, and sublimation). As the latter
factor showed insufficient internal consistencies
(Cronbach’s alphas: .64 and .66), the authors concluded that the instrument could be properly used
to measure only one dimension of immature defenses.
Bonsack et al.,19 using a French version of the
questionnaire, found, both in their control group of 82
normal subjects and in their 140 psychiatric outpatients, four factors, ranging from immature to mature
defense styles. Although they state that, except for
minor differences, “. . .the psychometric features of the
French version are similar to the original scale. . .” a
thorough analysis of their factor structure shows remarkable differences in the content of the factors compared to Bond’s original scales.
The above studies support the notion that defenses can be assembled in clusters on a continuum
of maturity–immaturity and/or pathology–mental
health; however, they show considerable differences concerning the number of defensive clusters
to be distinguished and where to set their boundaries. All of them agree in showing a cluster of
maladaptive defenses, whereas the defenses placed
on the more mature levels on Vaillant’s continuum
present a rather inconstant segmentation.
The DSQ, in its original version or in a shortened
40-item version,20 has been used in the following different clinical areas: eating disorders,21-27 anxiety disorders,28-31 depressive disorders,32-34 personality
disorders,17,35-41 alexithymia,42-44 various problems
in adolescence.45-52 Furthermore, the DSQ has
been studied as a possible process and outcome
predictor of short-term psychotherapy.53,54
Within this context our aims are: (1) to confirm, through factor analysis, that the defensive
mechanisms, as measured by the Italian version
of the DSQ, gather in clusters that can be ordered on the maturity–immaturity and/or mental
health–mental pathology continuum proposed by
Vaillant; (2) to evaluate the internal and longitudinal reliability of the defensive style scales
constructed on the basis of these clusters; and (3)
485
to analyze the association between the defensive
styles and the personality dimensions assessed
by the Eysenck Personality Questionnaire-Revised (EPQ-R).55
METHODS
Subjects
Subjects were 628 volunteers between 17 and 66 years of
age: 294 males (mean age, 33.33; SD ⫽ 12.04) and 333 females
(mean age, 32.38; SD ⫽ 11.03). They were university students
(39%), civil servants and teachers (38%), and professionals
(14%), living in Rome (approximately 90%) and in other cities
and small communities of the same area. For the most part they
were informally recruited during 2 successive years on the
university campus, in the libraries and faculties of the University of Rome “La Sapienza,” and in the administration offices of
some banks, town halls, public schools, and other public departments in the region of Rome. On the whole, they may be
considered an opportunistic sample of the population living in
the urban and suburban area of Rome, with a prevalence of
middle-class people with high school and university education.
All participants provided written informed consent after complete description of the study.
Instruments
All respondents completed a booklet containing the items of
the DSQ and the items of a short form of the EPQ-R. The items
of the latter instrument were interspersed randomly among the
items of the DSQ.
The Defense Style Questionnaire. The 88-item version of
the DSQ56 was translated into Italian using the back-translation
method.57 The Italian and the original versions of the questionnaire and the distribution of the items among the 25 defense
mechanisms are displayed in Appendices I and II. A 9-point
Likert-type scale indicating degree of agreement or disagreement with the item content follows the items.
The Eysenck Personality Questionnaire-Revised. The short
form of the EPQ-R used in this study was developed by two of the
authors of the present report in two steps. On the basis of data from
a previous factor analytical study on the Italian version of the full
form of the questionnaire,58 for each of the four scales, extraversion,
neuroticism, psychoticism, and lie, the items that correlated most
strongly with the respective factors and were factorially pure were
selected. This 40-item form of the EPQ-R was successively administered to a new sample of 428 subjects, for the most part university
students. A principal component analysis followed by Varimax rotation of the first four components showed close correspondence
between factors of the scales; internal reliabilities were acceptable
and scale intercorrelations were similar to those reported for the
British short form.55
For the purpose of the present study, the EPQ-R questions
have been rephrased into assertions followed by 9-point Likertlike response scales identical to those of the DSQ.
RESULTS AND DISCUSSION
Descriptive Statistics and Internal Consistencies
Descriptive statistics and internal consistencies of the sets of items representing the 25
486
SAN MARTINI ET AL
Table 1. Descriptive Statistics and Internal Consistencies of the Defense Mechanisms
Men
Defense Mechanisms
Acting-out
Affiliation
Undoing
Anticipation
Passive-aggressive
Consumption
Denial
Fantasy
Reaction formation
Primitive idealization
Projective identification
Inhibition
Isolation
Help-rejecting complaining
Omnipotence
Task-orientation
Projection
Pseudo-altruism
Regression
Suppression
Withdrawal
Splitting
Somatization
Sublimation
Humor
Mean
3.52
2.79
2.67
4.86
2.74
1.94
1.80
4.52
2.80
3.14
.98
2.96
3.10
2.22
2.71
4.87
1.62
5.69
2.30
3.94
4.56
3.45
1.97
2.05
4.69
defenses and coping mechanisms are displayed
in Table 1 for males and females separately. As
expected, given the low number of items in each
set, internal consistencies are generally moderate, particularly so for the sets representing suppression and denial. An exception is the decidedly low internal consistency of consumption
(.16), a set of three items. This low value may be
at least partly due to one of the items of this set,
item 79 (“I take drugs, alcohol or medicine when
I am tense”) that, in our nonpsychiatric sample,
had a strongly positively skewed distribution
with low mean and standard deviation (M ⫽ 0.5;
SD ⫽ 1.67). Not surprisingly, this item had near
zero (r ⬍ .16) correlations with the remaining
two items.
We could not find in the literature measures of
internal consistency of the individual defense sets
for a direct comparison. We note, however, that
with the exception of consumption, our internal
consistencies are in line with the data of Bond et
al.,12 as the items of these a priori–constructed sets
were retained by Bond et al. if their item-total
correlations were significant at the .001 level, that
Women
SD
1.77
2.05
1.67
2.10
1.47
1.67
1.29
2.87
1.60
2.29
1.86
1.72
1.65
1.97
1.61
2.37
1.06
2.10
2.02
2.08
2.05
2.09
2.09
2.65
1.84
Mean
4.06
3.48
2.60
4.97
2.81
2.56
1.43
4.78
2.93
3.62
1.51
3.56
2.47
2.28
2.27
5.18
1.79
6.22
3.31
3.58
5.47
3.40
2.97
2.60
4.57
SD
Cronbach’s
Alpha
1.66
2.24
1.80
2.13
1.45
1.70
1.32
2.83
1.60
2.58
2.45
1.84
1.59
1.96
1.58
2.12
1.19
1.91
2.12
2.18
1.93
1.98
2.28
2.98
1.93
.69
.32
.36
.34
.32
.16
.27
.57
.43
.67
.41
.66
.71
.59
.68
.43
.26
.61
.48
.46
.54
is, considering the number of their subjects, if their
item-total correlations were above .22.
Factor Analysis
As a preliminary step, a maximum liklihood
confirmatory factor analysis was performed, to assess the goodness of fit of the four dimensional
model originally proposed by Bond et al.12 on the
basis of their exploratory factor analyses. Only the
14 defense sets representing the maladaptive, image-distorting, self-sacrifice, and adaptive styles in
the original study were included in the analysis,
which allowed factor components to be correlated.
For the analysis, the SEPATH module of the package STATISTICA (StatSoft, Tulsa, OK) was used.
Several measures were employed to estimate the
degree to which the model fit the data, among them
the chi-square statistic, the goodness-of-fit index
(GFI), the adjusted goodness-of-fit index (AGFI),
and the root mean square error of approximation
(RMSEA). The ␹2 value was 482.91 (df ⫽ 71; P ⫽
.000), the GFI was .890, the AGFI was . 838, and
the point estimate of the RMSEA was .102 with a
90% confidence interval between .094 and .111.
ITALIAN VERSION OF THE DSQ
Fig. 1
487
Scree plot of the eigenvalues of DSQ scores.
Overall, these indices showed an insufficient fit
and suggested that an amelioration of the model
was appropriate. For this purpose, we decided to
adopt an exploratory rather than confirmatory perspective in the subsequent analysis of the dimensional structure of the questionnaire.
A principal component analysis was performed on
the correlation matrix of all the 25 item sets. The
Kaiser-Meyer-Olkin Index of sampling adequacy was
.83, indicating that the variables shared sufficient common variance to justify factor analysis. According to
the scree-plot (Fig 1), it was clearly appropriate to
extract four factors, explaining 42.1% of the total variance. These four factors were rotated by the Quartimax
method (Table 2), the same rotation method used by
Bond et al.12and by the majority of the subsequent
studies. A Varimax rotation was also performed, with
substantially identical results and will not be further
considered. The rotated factor matrix is displayed in
Table 2. In the following description, an item will be
said to load saliently on a factor if it has its maximum
loading on that factor, provided that loading exceeds .39.
The first factor, explaining 19.81% of the variance,
is loaded saliently by 11 defenses, all of them usually
considered immature or neurotic. In terms of defense
content, it is similar to the first factor obtained by Bond
et al.12 with the previous form of the questionnaire,
sharing with it the following defenses: regression, projection, somatization, acting-out, passive-aggressive
Table 2. Rotated Factor Matrix
Factors
Defense Mechanisms
Regression
Acting-out
Projection
Somatization
Passive-aggressive
Withdrawal
Fantasy
Consumption
Help-rejecting complaining
Projective identification
Undoing
Denial
Omnipotence
Isolation of affect
Splitting
Anticipation
Task-orientation
Pseudo-altruism
Primitive idealisation
Sublimation
Suppression
Affiliation
Inhibition
Humor
Reaction formation
Explained variance: 42.07%
NOTE. Values below .25 are omitted.
1
.760
.690
.637
.618
.559
.541
.530
.515
.472
.461
.421
.334
⫺.253
2
3
⫺.316
.432
⫺.256
.257
.257
.250
.438
.295
.385
.667
.650
.647
.466
.308
.321
⫺.433
.288
.604
.598
.461
.415
.412
.401
.282
.466
19.81%
4
8.33%
.405
.348
7.23%
⫺257
.621
⫺.499
.493
6.74%
488
behavior, withdrawal, consumption, and projective
identification. The factor is loaded saliently also by
fantasy, which was not included in the 1983 version of
the questionnaire. This defense was found to belong to
the immature factor also by Andrews et al.15 and by
Sammallahti et al.17 Inhibition, which in Bond et al.12
belonged to the immature factor, in our sample loads
maximally on the fourth factor, but has a high secondary loading on the first factor as well. Help-rejecting
complaining and undoing also have salient loadings on
the first factor; however, the former with a secondary
loading on the second factor and the latter with similar
secondary loadings on the second and third factors. On
the whole, the content of this factor corresponds quite
closely to the concept of maladaptive style as defined
by Bond et al.12
The second factor, explaining 8.33% of the variance, is loaded saliently by denial, omnipotencedevaluation, isolation, and splitting, the same defenses constituting the borderline style factor
extracted by Sammallahti et al.17 This factor shares
omnipotence-devaluation and splitting with the image-distorting style found by Bond et al.12 Although this defensive style is often seen in borderline patients, we preferred to label it imagedistorting, rather than borderline, to use a term that
is less tied to a specific psychiatric diagnosis.
The third factor, explaining 7.23% of the variance,
is loaded saliently by anticipation, task-orientation,
pseudo-altruism, primitive idealization, sublimation,
and suppression. Humor, which loads maximally (negatively) on the forth factor, has a high (positive) secondary saturation on this factor as well. It is clearly a
cluster of adaptive defenses, although the presence of
items of pseudo-altruism and primitive idealization
requires some comment. The (single) item representing pseudo-altruism contains two different statements:
“I get satisfaction from helping others” and “If this
were taken away from me, I would get depressed.” We
think that our respondents could have focused their
attention predominantly on the first statement that was
in a way directly related to their actual situation, as
they were in fact just helping someone else doing a
research. In other words, this item might have been
perceived by our subjects, rather than a maladaptive
defensive behavior, as a normal expression of solidarity (see the DFS of the DSM-IV for differential definitions of pseudo-altruism and altruism).4 As for primitive idealization (which in the studies conducted by
Bond et al. in 198312 and by Sammallahti et al. in
199517 weighted significantly on more than one fac-
SAN MARTINI ET AL
Table 3. Correlations Between DSQ Factor Scores and
EPQ-R Scale Scores
Factor
Factor
Factor
Factor
1
2
3
4
Psychoticism
Neuroticism
Extraversion
.03
.33
⫺.14
.02
.78
.12
⫺.04
.07
⫺.21
.18
.36
⫺.49
NOTE. All values ⬎.07 are significant at .05 level
tor), we should consider that, depending on the situation, it may function as a healthy process, inspiring
progressive behavior. In the Italian, mainly Catholic,
cultural context, item 51 (“I always feel that someone
I know is like a guardian angel”) may be perceived as
reflecting ways of thinking and expressing oneself that
are more common and considered as more “normal”
than in the United States or even in Canada. A similar
interpretation might be proposed for the other item of
this defense (no. 58:“There is someone I know who
can do anything and who is absolutely fair and just”).
We decided to label this factor adaptive style.
The fourth factor, explaining 6.7% of the variance, is saliently loaded by the defenses of inhibition, reactive formation, and, negatively, humor, all
with secondary high loadings on the first and third
factors. Even though the factor has no clear markers, it may tentatively be interpreted as an “introverted” or “over-inhibited” style.
Correlations With the EPQ-R Scales
DSQ factor scores were calculated and correlated with the four EPQ-R scale scores (Table 3).
There was a high correlation between the immature
factor and neuroticism (r ⫽ .78), a moderate correlation between the mature factor and extraversion (r ⫽ .36), and a moderate positive correlation
between the image-distorting factor and psychoticism (r ⫽ .33). The fourth factor showed a moderate negative correlation (r ⫽ –.49) with extraversion, not unexpected considering that humor was
one of its three components. On the whole, the
pattern of correlations is compatible with the interpretation of the factors given above.
Psychometric Features of the Factorial Scales
Leaving aside the fourth factor that was composed only of (three) factorially complex defenses,
we constructed three factorial scales, composed of
all defenses loading saliently on the respective
factors. Descriptive statistics, internal consistencies (Cronbach’s alphas) gender differences, and
ITALIAN VERSION OF THE DSQ
489
Table 4. Descriptive Statistics, Internal Consistencies, and Significance of Gender Differences for the Three Style Scales
Maladaptive style
Image-distorting style
Adaptive style
Men
(n ⫽ 289)
Women
(n ⫽ 327)
No. of
Items
Mean
SD
Mean
SD
Cronbach’s
alpha
P for the
t Test
37
17
12
2.54
2.70
4.25
1.05
1.11
1.05
2.94
2.31
4.40
1.03
1.10
1.81
.85
.72
.57
.0000
.0000
.0040
test-retest correlations of these scales, evaluated on
the item level, are shown in Table 4. Retest was
measured after 6 months on a subsample of 47
subjects (21 men and 26 women, aged 20 to 58
years; mean, 34 years; SD ⫽ 11.3).
Internal consistencies were satisfactory for the
Maladaptive Style scale, which comprised 37
items, just sufficient for the Image-Distorting Style
scale, which comprised 17 items, and decidedly
low for the Adaptive Style scale, which comprised
12 items. The differences among these values were
likely to be at least partly due to the different scale
lengths. In fact, when evaluated on the basis of 37
items with the Spearman-Brown prophetic formula, the internal consistencies of the last two
scales became satisfactory (r ⫽ .81 for both scales)
and similar to that of the Maladaptive Style scale.
Test-retest correlations were adequate for the
Maladaptive Style scale (r ⫽ .79; P ⬍ .001) and
for the Adaptive Style scale (r ⫽ .81; P ⬍ .001),
and a bit low for the Image-Distorting Style scale
(r ⫽ .63; P ⬍ .001).
Compared to males, females scored significantly
higher than males in the maladaptive style and
significantly lower on the image-distorting and on
the mature styles.
Correlations Between the Factor Scales
Premise. According to the model, on the basis
of the questionnaire the defenses can be ordered
along a maturity–immaturity continuum. As a consequence, defense styles are not conceived as completely independent, that is, defense styles that are
close on the continuum are expected to be positively correlated and defense styles that are far on
the continuum are expected to be less (positively)
correlated, not correlated, or negatively correlated.
From this point of view, the orthogonal rotation of
the factors, which was performed following the
method of Bond et al.,12 should be interpreted as a
way of forcing the defenses into clusters that, if
meaningful, may be used as measuring scales.
There was a moderate positive correlation between the maladaptive and the image-distorting
styles (r ⫽ .44; P ⬍ .001), a weak positive correlation between the image-distorting and the adaptive styles (r ⫽ .21; P ⬍ .001), and a weak positive
correlation between the adaptive and maladaptive
styles (r ⫽ .18; P ⬍ .001). With the exception of
the last correlation, which is discrepant with the
negative correlation between the maladaptive and
mature styles reported by Bond et al.,12 on the
whole this pattern of scale intercorrelations is in
agreement with Bond’s data and compatible with
his model.
CONCLUSIONS
Three factors, identified as maladaptive, imagedistorting, and adaptive styles, emerged from our
data. These factors, despite some difference in the
labels used to describe them, were very similar to
three of the factors reported both by Bond et al.12
and by Sammallahti et al.17 and were correlated in
a predictable way with the EPQ-R scales. The
scales constructed on the basis of the factor analysis showed intercorrelations compatible with the
hierarchical model of defensive functioning at the
base of the questionnaire, acceptable test-retest
reliabilities, and internal consistencies with the exception of the consistency of the Adaptive Style
scale and the test-retest reliability of the ImageDistorting Style scale, that were lower than desirable. Only the Maladaptive Style scale, which,
compared to the other two scales, had a much
larger number of items, showed values of reliability and internal consistency high enough to warrant
use in its present form. Before investigating other
aspects of the validity of the questionnaire (e.g., its
concurrent and discriminant validity) it therefore
seems useful to try to improve these psychometric
faults of the shorter scales by finding new suitable
items and bring their lengths to the level of the
Maladaptive Style scale.
490
SAN MARTINI ET AL
APPENDIX I. The DSQ Items
Item
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Original Version
Italian Version
I get satisfaction from helping others and if this were
taken away from me I would get depressed.
People often call me a sulker.
I’m able to keep a problem out of my mind until I
have time to deal with it.
I’m always treated unfairly.
I work out my anxiety through doing something
constructive and creative like painting or
woodwork.
Once in a while I put off until tomorrow what I ought
to do today.
I keep getting into the same type of frustrating
situations and I don’t know why.
I’m able to laugh at myself pretty easily.
I act like a child when I’m frustrated.
I’m very shy about standing up for my rights with
people.
I am superior to most people I know.
Provo soddisfazione nell’aiutare gli altri e mi deprimerei se
mi si impedisse di farlo.
Le persone mi dicono spesso che sono scontroso.
Quando ho un problema sono capace di non pensarci fino a
quando non ho tempo per affrontarlo.
Mi trattano sempre ingiustamente
Gestisco l’ansia facendo cose costruttive e creative come,
ad esempio, dipingere o fare lavori con il legno.
People tend to mistreat me.
If someone mugged me and stole my money, I’d
rather he’d be helped than punished.
Once in a while I think of things too bad to talk
about.
Once in a while I laugh at a dirty joke.
People say I’m like an ostrich with my head buried in
the sand. In other words, I tend to ignore
unpleasant facts as if they didn’t exist.
I stop myself from going all out in a competition.
I often feel superior to people I’m with.
Someone is robbing me emotionally of all I’ve got.
I get angry sometimes.
I often am driven to act impulsively.
I’d rather starve than be forced to eat.
I ignore danger as if I were Superman.
I pride myself on my ability to cut people down to
size.
People tell me I have a persecution complex.
Sometimes when I am not feeling well I am cross.
I often act impulsively when something is bothering
me.
I get physically ill when things aren’t going well for
me.
I’m a very inhibited person.
I’m a real put-down artist.
I do not always tell the truth.
I withdraw from people when I feel hurt.
I often push myself so far that other people have to
set limits for me.
My friends see me as a clown.
I withdraw when I’m angry.
I tend to be on my guard with people who turn out
to be more friendly than I would have suspected.
Ogni tanto rimando a domani ciò che dovrei fare oggi.
Mi metto continuamente nello stesso tipo di situazioni
frustranti e non so perchè.
Mi è abbastanza facile ironizzare su me stesso.
Quando sono frustrato mi comporto come un bambino.
Sono molto timido net difendere i miei diritti con le
persone.
Sono superiore alla maggior parte delle persone che.
conosco.
La gente tende a maltrattarmi.
Se qualcuno mi scippasse rubandomi del denaro, preferirei
che fosse aiutato piuttosto che punito.
Talvolta mi capita di pensare a cose troppo brutte per
parlarne.
Talvolta mi capita di ridere ad una barzelletta sporca.
Mi dicono che metto la testa sotto la sabbia come gli
struzzi. In altre parole che, di fronte a fatti spiacevoli,
tendo a fare come se non esistessero.
Nelle competizioni, evito di espormi completamente
Spesso mi sento superiore alle persone che sono con me.
Qualcuno mi sta derubando di tutte le mie emozioni.
Qualche volta mi arrabbio.
Spesso mi sento spinto ad agire impulsivamente.
Preferirei morire di fame piuttosto che essere forzato a
mangiare.
Ignore il pericolo come se fossi Superman.
Sono orgoglioso della mia capacità di fare “abbassare la
cresta” alle persone.
La gente mi dice che ho il complesso di persecuzione.
Talvolta, quando non mi sento in salute, divento di cattivo
umore
Spesso, quando c’è qualcosa che mi disturba, reagisco
impulsivamente.
Mi ammalo fisicamente quando le cose non mi vanno benc
Sono una persona molto inibita.
Sono un artista nello “smontare” gli altri.
Non sempre dico la verità.
Mi ritraggo dalle persone quando mi sento offeso.
Spesso mi spingo cosı̀ in là che gli altri sono costretti a
frenarmi.
Gli amici mi considerano un “giocherellone”.
Quando sono arrabbiato tendo a ritrarmi in me stesso.
Tendo a stare in guardia con le persone che si rivelano più
amichevoli di quanto mi aspetti.
ITALIAN VERSION OF THE DSQ
491
APPENDIX I. The DSQ Items (Cont’d)
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
I’ve got special talents that allow me to go through
life with no problems.
Sometimes at elections I vote for men about whom I
know very little.
I’m often late for appointments.
I work more things out in my daydreams than in my
real life.
I’m very shy about approaching people.
I fear nothing.
Sometimes I think I’m an angel and other times I
think I’m a devil.
I would rather win than lose in a game.
I get very sarcastic when I’m angry.
I get openly aggressive when I feel hurt.
I believe in turning the other cheek when someone
hurts me.
I do not read every editorial in the newspaper every
day.
I withdraw when I’m sad.
I’m shy about sex.
I always feel that someone I know is like a guardian
angel.
My philosophy is “Hear no evil, do no evil, see no
evil.”
As far as I’m concerned, people are either good or
bad.
If my boss bugged me, I might make a mistake in
my work or work more slowly so as to get back at
him.
Everyone is against me.
I try to be nice to people I don’t like.
I would be very nervous if an airplane in which I was
flying lost an engine.
There is someone I know who can do anything and
who is absolutely fair and just.
I can keep the lid on my feelings if it would interfere
with what I’m doing if I were to let them out.
Some people are plotting to kill me.
I’m usually able to see the funny side of an
otherwise painful predicament.
I get a headache when I have to do something I
don’t like.
I often find myself being very nice to people who by
all rights I should be angry at.
There’s no such thing as “finding a little good in
everyone.” If you’re bad, you’re all bad.
We should never get angry at people we don’t like.
I am sure I get a raw deal from life.
I fall apart under stress.
When I know that I will have to face a difficult
situation like an exam or a job interview. I try to
image what it will be like and plan ways to cope
with it.
Doctors never really understand what is wrong with
me.
Ho doti particolari che mi permettono di affrontare la vita
senza problemi.
Qualche volta alle elezioni voto per uomini di cui so molto
poco.
Sono spesso in ritardo agli appuntamenti.
Elaboro molte più cose nelle mie fantasticherie piuttosto
che nella vita reale.
Sono molto timido nell’avvicinare le persone.
Non ho paura di niente.
Qualche volta mi sento angelo e qualche volta un diavolo.
Nei giochi preferisco vincere piuttosto che perdere.
Quando sono arrabbiato divento molto sarcastico.
Quando mi sento ferito divento apertamente aggressivo.
Quando qualcuno mi offende ritengo giusto porgere l’altra
guancia.
Non sempre leggo tutti gli articoli di fondo del giornale
Quando sono triste tendo a ritirarmi in me stesso.
Sono timido riguardo al sesso.
Ho sempre la sensazione che qualcuno che conosco sia
come un angelo custode.
La mia filosofia è “Non vedo, non sento, non parlo.”
Per quanto mi riguarda, le persone sono o buone o cattive.
Se il mio capo mi stesse troppo addosso, potrei fare
deliberatamente un errore nel lavoro oppure potrei
lavorare più lentamente per fargliela pagare.
Mi sono tutti contro.
Cerco di essere gentile con le persone che non mi
piacciono.
Mi innervosirei molto se l’aeroplano sul quale stessi
viaggiando perdesse un motore.
Conosco una persona che è in grado di fare qualsiasi cosa
ed è estremamente leale e giusta.
Sono in grado di tenere a freno i miei sentimenti se
esprimerli interferisse con ciò che sto facendo.
Alcune persone stanno congiurando per uccidermi.
Solitamente sono in grado di vedere il lato comico di
situazioni altrimenti penose.
Mi viene mal di testa quando debbo fare qualcosa che non
mi piace.
Spesso mi ritrovo ad essere molto gentile con persone nel
confronti delle quali avrei ogni ragione di essere
arrabbiato.
Non è vero che si possa “trovare un po’ di buono in
ognuno”: chi è cattivo è cattivo.
Non dovremmo mai arrabbiarci con le persone che non ci
piacciono.
Sono convinto di ricevere solo fregature.
Lo stress mi manda in pezzi.
Quando so di dover affrontare una situazione difficile come
un esame o un colloquio di lavoro, cerco di immaginare
come sarà e pianifico vari modi per affrontarla.
I medici non capiscono mai che cosa ho veramente.
492
SAN MARTINI ET AL
APPENDIX I. The DSQ Items (Cont’d)
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
When someone close to me dies, I don’t feel upset.
After I fight for my rights, I tend to apologize for my
assertiveness.
Most of what happens to me is not my
responsibility.
When I’m depressed or anxious, eating makes me
feel better.
Hard work makes me feel better.
My doctors are not able to help me really get over
my problems.
I’m often told that I don’t show my feelings.
I believe that people usually see more meaning in
films, plays, or books than is actually there.
I have habits or rituals which I feel compelled to do
or else something terrible will happen.
I take drugs, medicine or alcohol when I’m tense.
When I feel bad, I try to be with someone.
If I can predict that I’m going to be sad ahead of
time, I can cope better.
No matter how much I complain, I never get a
satisfactory response.
Often I find that I don’t feel anything when the
situation would seem to warrant strong emotions.
Sticking to the task at hand keeps me from feeling
depressed or anxious.
I smoke when I’m nervous.
If I were in a crisis, I would seek out another person
who had the same problem.
I cannot be blamed for what I do wrong.
If I have an aggressive thought, I feel the need to do
something to compensate for it.
La morte di qualcuno a me vicino non mi sconvolge.
Dopo essermi impegnato per far valere i miei diritti, tendo a
scusarmi per la mia assertività.
Non mi sento responsabile della maggior parte delle cose
che mi accadono.
Quando sono depresso o ansioso, mangiare mi fa sentire
meglio.
Lavorare sodo mi fa sentire meglio.
I dottori non sono in grado di aiutarmi veramente a
risolvere i miei problemi.
Mi dicono spesso che non mostro i miei sentimenti.
Credo che di solito le persone trovino nei film, nelle opere
teatrali o nei libri più senso di quanto realmente ce ne sia.
Ho delle abitudini o rituali che sento di dover rispettare,
altrimenti potrebbe accadere qualcosa di terribile.
Quando sono teso assumo droghe, farmaci o alcol.
Quando mi sento male cerco di stare con qualcuno.
Se posso prevedere in anticipo che una situazione mi
renderà triste, riesco ad affrontarla meglio.
Per quanto io possa lamentarmi, non ottengo mai una
risposta soddisfacente.
Spesso mi trovo a non provare nulla quando la situazione
sembrerebbe giustificare forti emozioni.
Allontano la depressione o l’ansia impegnandomi a fondo
in ciò che faccio.
Fumo quando sono nervoso.
Se fossi in crisi, cercherei qualcuno che avesse il mio stesso
problema.
Non è colpa mia se faccio qualcosa di sbagliato.
Se ho un pensiero aggressivo, sento il bisogno di fare
qualcosa per compensare.
ITALIAN VERSION OF THE DSQ
493
APPENDIX II. The DSQ Defenses and the Items That Represent Them
Defense Mechanisms
Item No.
Total
Items
Acting-out
Affiliation
Undoing
Anticipation
Passive aggressive
Consumption
Denial
Fantasy
Reaction formation
Primitive idealization
Projective identification
Inhibition
Isolation
Help-rejecting complaining
Omnipotence
Task-orientation
Projection
Pseudo-altruism
Regression
Suppression
Withdrawal
Splitting
Somatization
Sublimation
Humor
Le scale
7,21,27,33,46
80,86
71,78,88
68,81
2,22,39,45,54
73,79,85
16,42,52,57
40
13,47,56,63,65
51,58
19,
10,17,29,41,50
70,76,77,83
69,75,82
11,18,23,24,30,37
74,84
4,12,25,36,55,60,66,72,87
1
9,67
3,59
32,35,49
43,53,64
28,62
5
8,34,61
6,14,15,20,26,31,38,44,48,5 7
5
2
3
2
5
3
4
1
5
2
1
5
4
3
6
2
9
1
2
2
3
3
2
1
3
10
REFERENCES
1. Buckley P. Ego Defenses. A psychoanalytic perspective.
In: Conte HR, Plutchik R (eds). Ego Defenses: Theory and
Measurement., et al. New York, NY: Department of Psychiatry
of Albert Einstein College of Medicine of Yeshiva University,
1995:39-51.
2. Freud S. Le neuropsicosi da difesa (1894). Trans. Ital. in:
Sigmund Freud, Opere, vol. 1. Torino, Italy: Boringhieri, 1967.
3. Freud S. On the history of the psychoanalitic movement
(1914). In: Strachey J (ed and Trans). The Standard Edition of
the Complete Psychological Works of Sigmund Freud. Vol. 14.
London, UK: Hogarth Press, 1957:3-66.
4. American Psychiatric Association Committee on Nomenclature and Statistics. Diagnostic and Statistical Manual of
Mental Disorders. Ed. 4. Washington, DC: American Psychiatric Press, 1994.
5. Skodol AE, Perry JC. Should an axis for defense mechanism
be included in DSM-IV? Compr Psychiatry 1993;34:108-199.
6. Soldz S, Vaillant GE. A 50-year longitudinal study of
defense use among inner city men: a validation of the DSM-IV
Defense Axis. J Nerv Ment Dis 1998;186:104-111.
7. Perry JC, Hoglend P, Shear K, Vaillant GE, Horowitz M,
Kardos ME, et al. Field trial of a diagnostic axis for defense
mechanisms for DSM-IV. J Personal Disord 1998;12:56-68.
8. Perry JC. Defense Mechanism Rating Scales (DMRS).
Ed. 5. Cambridge, MA: J.C. Perry, 1990.
9. Lingiardi V, Madeddu M. I Meccanismi di Difesa. Nuova
edizione (contiene la versione italiana della Defense Mecha-
nism Rating Scale, DMRS). Milano, Italy: Raffaello Cortina
Editore, 2002.
10. Perry JC, Cooper S. An empirical study of defense mechanisms: I. Clinical interview and life vignettes ratings. Arch
Gen Psychiatry 1989;46444-452.
11. Lingiardi V, Lonati C, Fossati A, Vanzulli L, Maffei C.
Defense mechanisms and personality disorders. J Nerv Ment
Dis 1999;187:224-228.
12. Bond MP, Gardner ST, Christian J, Sigal C. Empirical
study of self-rated defense styles. Arch Gen Psychiatry 1983;
46:455-460.
13. Bond MP. The development and properties of the Defense Style Questionnaire. In: Conte HR, Plutchik R, et al (eds).
Ego Defenses: Theory and Measurement. New York, NY: Department of Psychiatry of Albert Einstein College of Medicine
of Yeshiva University, 1995:202-220.
14. Vaillant GE. Adaptation to Life. Boston, MA: Little
Brown, 1977.
15. Andrews G, Pollock C, Stewart G. The determination of
defense by questionnaire. Arch Gen Psychiatry 1989;46:455-460.
16. Reister G, Fellhauer RF, Franz M, Wirth T, Schellberg D,
Schepank H, et al. Psychometric measurements of ego defense
mechanisms: Correlation between questionnaire and expert rating. Psychother Psychosom Med Psychol 1993;43:15-20.
17. Sammallahti P, Aalberg V. Defense style in personality disorders—an empirical study. J Nerv Ment Dis 1995;183:516-521.
18. Rutherford MJ, Mcdermott P, Cacciola JS, Alterman AI,
494
Mulvaney F. A psychometric evaluation of the defense style questionnaire in methadone patients. J Personal Disord 1998;12:119-125.
19. Bonsack C, Despland JN, Spagnoli J. The French version
of the defense stile questionnaire. Psychother Psychosomat
1998;67:24-30.
20. Andrews G, Sing M, Bond M. The Defense Style Questionnaire. J Nerv Ment Dis 1993;181:246-256.
21. Steiger H, Van Der Feen J, Goldstein C, Leichner P.
Defense style and parental bonding in eating-disordered
women. Int J Eat Disord 1989;8:131-140.
22. Steiger H, Goldstein C, Mongrain M, Van Der Feen J.
Description of eating disordered, psychiatric, and normal
women along cognitive and psychodynamic dimensions. Int J
Eat Disord 1990;9:129-140.
23. Steiner H. Defense style in eating disorders. Int J Eat
Disord 1990;9:141-151.
24. Sullivan PF, Bulik CM, Hall S, Weltzin TE, Kaye WH.
Defense style in women with eating disorders. Int J Eat Disord
1994;16:251-256.
25. Sullivan PF, Bulik CM, Carter FA, Gendall KA, Joyce
PR. The significance of a prior history of anorexia in bulimia
nervosa. Int J Eat Disord 1996;20:253-261.
26. Tordjman S, Zittoun C, Ferrari P, Flament M, Jeammet P.
A comparative study of defense styles of bulimic, anorexic and
normal females. Isr J Psychiatry Relat Sci 1997;34:222-227.
27. Sarlio Lahteenkorva S, Rissanen A. Weight loss maintenance: determinants of long term success. Eat Weight Disord
1998;3:131-135.
28. Pollock C, Andrews G. Defense styles associated with
anxiety disorders. Am J Psychiatry 1989;146:1500-1502.
29. Andrews G. Anxiety, personality, and anxiety disorders.
Int Rev Psychiatry 1991;3:293-302.
30. Muris P, Merckelbach H. Defense style, trait anxiety, worry,
and bodily symptoms. Pers Individ Differ 1994;16:349-351.
31. Wastell CA. Defensive focus and the Defense Style
Questionnaire. J Nerv Ment Dis 1999;187:217-223.
32. Kneepkens RG, Oakley LD. Rapid improvement in the
defense style of depressed women and men. J Nerv Ment Dis
1996;184:358-361.
33. Lyoo IK, Gunderson JG, Phillips KA. Personality dimensions associated with depressive personality disorder. J Personal
Disord 1998;12:46-55.
34. Mullen LS, Blanco C, Vaughan SC, Vaughah R, Roose
SP. Defense mechanisms and personality in depression. Depression Anxiety 1999;10:168-174.
35. Bond M. Are “borderline defenses” specific for borderline personality disorders? J Personal Disord 1990;4:251-256.
36. Johnson JG, Bornstein RF, Krukonis AB. Defense styles
as predictors of personality disorder symptomatology. J Personal Disord 1992;6:408-416.
37. Bond M, Paris J, Zweig-Frank H. Defense style and borderline personality disorders. J Personal Disord 1994;8:28-31.
38. Sammallahti P, Aalberg V, Pentinsaari JP. Does defense
style vary with severity of mental disorder? An empirical assessment. Acta Psychiatr Scand 1994;90:290-294.
39. Paris J, Zweig-Frank H, Bond M, Guzder J. Defense
styles, hostility, and psychological risk factors in male patient
with personality disorders. J Nerv Ment Dis 1996;184:153-158.
40. Mulder RT, Joyce PR, Sullivan PF, Bulik CM, Carter FA.
SAN MARTINI ET AL
The relationship among three models of personality psychopathology: DSM-III-R personality disorder, TCI scores and DSQ
defenses. Psychol Med 1999;29:943-951.
41. Sinha BK, Watson DC. Predicting personality disorder
traits with the Defense Style Questionnaire in a normal sample.
J Personal Disord 1999;13:281-286.
42. Wise TN, Mann LS, Epstein S. Ego defensive styles and
alexithymia: a discriminant validation study. Psychother Psychosom 1991;56:141-145.
43. Parker JDA, Taylor GJ, Bagby RM. Alexithymia: relationship with ego defense and coping styles. Compr Psychiatry
1998;39:91-98;.
44. Kooiman CG, Spinhoven P, Trijsburg RW, Rooijmans
HGM. Perceived parental attitude, alexithymia and defense
style in psychiatric outpatients. Psychother Psychosom 1998;
67:81-87.
45. Muris P, Merckelbach H, Bogels S. Coping, defense, and
fear in college students. Pers Individ Differ 1995;18:301-304.
46. Steiner H, Feldman SS. Two approaches to the measurement of adaptive style: Comparison of normal, psychosomatically ill, and delinquent adolescents. J Am Acad Child Adolesc
Psychiatry 1995;34:180-190.
47. Erickson SJ, Feldman SS, Steiner H. Defense mechanism and adjustment in normal adolescents. Am J Psychiatry
1996;153:826-828.
48. Feldman SS, Araujo KB, Steiner H. Defense mechanisms in
adolescents as function of age, sex, and mental health status. J Am
Acad Child Adolesc Psychiatry 1996;35:1344-1354.
49. LI N, Zhang H, Zhao H, Tu X. Field test of self-rated
defense styles in 4,309 college students. Chin Ment Health J
1996;10:100-102.
50. Chan DW. Defense styles and psychological symptoms
among Chinese adolescents in Hong Kong. Soc Psychiatry
Psychiatr Epidemiol 1997;32:269-276.
51. Tuulio HA, Poikolainen K, Aalto ST, Lonnquist J. Psychological defense style in late adolescence and young adulthhood: a follow up study. J Am Acad Child Adolesc Psychiatry
1997;36:1148-1153.
52. Evans DW, Seaman JL. Developmental aspects of psychological defenses: their relation to self-complexity, self-perception, and symptomatology in adolescents. Child Psychiatry
Hum Dev 2000;30:237-254.
53. Piper EW, De Carufel FL, Szkrumelak N. Patient predictors of process and outcome in short-term psychotherapy.
J Nerv Ment Dis 1985;173:726-733.
54. Muris P, Merckelbach H. Defense style and behaviour
therapy outcome in a specific phobia. Psychol Med 1996;26:
635-639.
55. Eysenck SBG, Eysenck HJ, Barrett P. A revised version
of the psychoticism scale. Pers Individ Differ 1985;6:21-29.
56. Bond MP. Defense Style Questionnaire. In: Vaillant GE
(ed). Empirical Studies of Ego Mechanisms of Defense. Washington, DC: American Psychiatric Press, 1986.
57. Simonsen E, Mortensen E. Difficulties in translation of
personality scales. J Personal Disord 1990;4:290-296.
58. San Martini P, Mazzotti E, Setaro S. Factor structure and
psychometric features of the Italian version of for the EPQ-R.
Pers Individ Differ 1996;21:877-882.