Journal of Personality Assessment, 92(5), 432–438, 2010

Copyright
C Taylor & Francis Group, LLC

ISSN: 0022-3891 print / 1532-7752 online
DOI: 10.1080/00223891.2010.497421

Convergent Validity of the Defense Mechanisms Manual and the

Defensive Functioning Scale
JOHN H. PORCERELLI,1 ROSEMARY COGAN,2 RAY KAMOO,3 AND KRISTEN MILLER4

1
Department of Family Medicine & Public Health Sciences, Wayne State University School of Medicine
2
Department of Psychology, Texas Tech University
3
Private Practice, Bloomfield Hills, Michigan
4
Department of Psychology, University of Detroit Mercy

We examined the convergent validity of Cramer’s Defense Mechanisms Manual (DMM; Cramer, 1991b) by comparing it to the Diagnostic
and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) Defensive Functioning Scale (DFS). A total of 60
low income urban women from a primary care medical facility responded to four Thematic Apperception Test (TAT; Murray, 1943) cards and
an interview of early memories and descriptions of significant others. We scored the TAT narratives with the DMM, and we coded the interview
narratives with the DFS. DMM Denial and Projection scales were negatively correlated with the DFS Overall Defensive Functioning scale (r =
–.28, p < .01 and r = –.22, p < .10, respectively) and were positively correlated with a DFS pathological composite score (r = .36, p < .01 and
r = .32, p < .05, respectively). These findings support the convergent validity of the DMM Denial and Projection scales.

Since Freud’s (1884/1965) introduction of the concept of de- of identity in adolescence and late adolescence, and its use is
fense mechanisms in his work, The Neuro-Psychoses of Defense, thought to lessen during young adulthood (Cramer, 2007). Re-
defense mechanisms have provided important information for sults from three recent studies have indicated that after young
the study of normal development, adaptation, and psychopathol- adulthood, identification no longer exceeds projection in relative
ogy (Porcerelli, Thomas, Hibbard, & Cogan, 1998). The concept use (Cramer, 2007; Cramer & Block, 1998; Cramer & Tracy,
of defense mechanisms is one of a few psychoanalytic constructs 2005).
that is part of the mainstream of American psychology (Cramer, A defense mechanisms scale was introduced in the DSM–IV
2001; Cooper, 1992) and psychiatry (American Psychiatric As- as the Defensive Functioning Scale (DFS) in 1994. Grounded
sociation, 1994). Mounting empirical evidence for the construct in the work of Vaillant (1977), Perry (1990a), and Perry and
of defense mechanisms has led to the inclusion of a provisional Hoglend (1998), the DFS scale includes seven levels (scales)
axis for the assessment of defensive functioning within the Di- of defenses ranging from a High Adaptive Level to a Psychotic
agnostic and Statistical Manual of Mental Disorders (4th ed. Level (High Adaptive Level, Mental Inhibitions Level, Minor
[DSM–IV]; American Psychiatric Association, 1994). In this Image Distorting Level, Disavowal Level, Major Image Distort-
study, we used the DSM–IV Defensive Functioning Scale (DFS) ing Level, Action Level, and Level of Defensive Dysregulation),
to assess the convergent validity of the Defense Mechanisms with three to eight individual defenses in each level. To use the
Manual (DMM; Cramer, 1991b), the most frequently used scor- scale, clinicians rate the predominant defense level of the in-
ing system for assessing defense mechanisms from Thematic dex patient. For calculating the Overall Defensive Functioning
Apperception Test (TAT; Murray, 1943) narratives. (ODF) score, each level is weighted, ranging from high adaptive
Cramer (1991b, 2006) has developed the DMM scoring sys- level (weighted score = 7) to level of defensive dysregulation
tem and carried out numerous investigations to test the hy- (weighted score = 1). The ODF is used for research purposes
pothesis that different defenses mechanisms become promi- as a score of defense mechanism maturity.
nent at different developmental stages. The DMM includes Using a multimethod, cross-sectional design, we assess the
three developmentally anchored defenses—denial, projection, convergent validity of the TAT-based DMM by comparing it with
and identification. Both cross-sectional (Cramer, 1987; Cramer the interview-based DSM–IV DFS. We used the ODF scale and
& Gaul, 1988; Porcerelli et al., 1998) and longitudinal studies a pathological composite scale of the DFS (described follow-
(Cramer, 1997) have supported the theory that denial, the least ing). We considered the relationship between the three DMM
cognitively complex of the three defenses, is most prominent defenses and six of the seven DFS levels (omitting the Psy-
in childhood, whereas projection, a more cognitively complex chotic Level defenses because they occurred rarely). Because
defense, becomes most prominent in early and middle ado- the DMM Denial is the least mature defense in Cramer’s (1991b)
lescence. Identification is used, in part, for the development system and the one with the greatest association with psy-
chopathology in adulthood, we hypothesized that Denial would
negatively and significantly correlate with ODF, the DFS matu-
Received July 30, 2009; Revised April 12, 2010. rity score. Given that the DMM and the ODF are derived from
Address correspondence to John H. Porcerelli, Wayne State Univer-
sity/Crittenton Family Medicine Residency Program, Department of Fam- different methods of assessment, we expected the magnitude of
ily Medicine & Public Health Sciences, Wayne State University School of this correlation to be of a moderate effect size (approximately
Medicine, 1135 W. University Drive, Suite 250, Rochester Hills, MI 48307– r = .30 according to criteria by Cohen, 1988). We hypothe-
1831; Email: [email protected] sized that DMM Projection, also an immature defense in the
432
DEFENSE MECHANISMS MANUAL AND DEFENSIVE FUNCTIONING SCALE 433

Cramer (1991b) system, but more developmentally mature and TABLE 1.—Demographic characteristics of the 60 women.a
less pathological than DMM Denial, would be negatively and
Variable N %
significantly correlated with ODF; and we expected the magni-
tude of this correlation to be of a small-to-moderate effect size Race
(approximately r = .25). We hypothesized that DMM Identi- African American 48 80
fication, the most mature of the three DMM defenses, would Caucasian 7 12
be positively and significantly correlated with ODF; and we Other 5 8
Marital status
expected the magnitude of this correlation to be of a small-to- Married 7 12
moderate effect size (approximately r = .20). Living with partner 4 7
To further assess the convergent validity of the DMM de- Single 25 42
fenses of Denial and Projection, we calculated a DFS patho- Separated/divorced 22 36
Widowed 2 3
logical composite score, described in the Measures section. We Income
hypothesized that DMM Denial and Projection would positively $0–$9.999 35 58
and significantly correlate with the DFS pathological composite $10,000–$19,999 15 25
score, and we expected the magnitude of these correlations to $20,000–$29,999 6 10
reach a moderate effect size of approximately r = .30. We also $30,000–$39,999 4 7
Education
hypothesized that DMM Denial would be significantly corre- Some high school 14 23
lated with the four lower level DFS scales (Minor Image Dis- High school graduate 19 32
torting Level, Disavowal Level, Major Image Distorting Level, Some college 25 42
and Action Level defenses) and DMM Projection would be sig- a
Age M = 36.5, SD = 10.4.
nificantly correlated with three of the four lower level defenses
(Minor Image Distorting Level, Disavowel Level, and Major
Image Distorting Level). We did not expect DMM Projection to
be correlated with Action Level defenses. Projection involves
the cognitive-affective process of attributing negative thoughts and (g) bizarre story or theme. Identification includes (a) em-
and feelings onto another, whereas with Action Level defenses, ulation of skills; (b) emulation of characteristics, qualities, or
action takes the place of cognitive-affective processes. Finally, attitudes; (c) regulation of motives or behavior; (d) self-esteem
we hypothesized that DMM Identification would be positively through affiliation; (e) work, delay of gratification; (f) role dif-
and significantly correlated with the DFS High Adaptive Level ferentiation; and (g) moralism. Each type of defense is scored
defenses and would be negatively and significantly correlated each time it is present in a story. The scores for each defense
with all five of the DFS pathological defense levels. are added together to make a total score for each of the three
defenses. In previous studies, the DMM defenses have shown
METHODS adequate interrater reliability with children, adolescents, and
Participants adults (e.g., Cramer, 1991b, 1997, 1998a, 1998b; Hibbard et al.,
1994; Porcerelli et al., 1998; Sandstrom & Cramer, 2003). In a
The participants were 60 urban women seeking medical recent meta-analysis, Meyer (2004) reported overall interrater
care at a university-based family medicine clinic in Detroit, reliability of .80 for the three DMM scales. The DMM defenses
Michigan. These women were the last 60 of 110 women partici- have demonstrated adequate levels of stability over a 3-year
pating in a larger study of women’s health. For details about the period (Cramer, 1998b).
larger sample, see Porcerelli, Cogan, Markova, Murdoch, and The validity of the DMM has been demonstrated through ob-
Porcerelli (2010). The majority of the women were single, low servational studies of children, adolescents, and college students
income, and African American (see Table 1). in which defense mechanism scores differentiated between age
groups or were related to personality variables in ways pre-
Measures dicted by theory (e.g., Blais, Conboy, Wilcox, & Normal, 1996;
DMM (Cramer, 1991b). The DMM was used to code the Cramer, 1997, 1999, 2001; Cramer, Blatt, & Ford, 1988; Cramer
use of defense mechanisms from the TAT (Murray, 1943) nar- & Block, 1998; Hibbard et al., 1994; Hibbard & Porcerelli,
ratives in response to Cards 2, 3BM, 4, and 5. These cards 1998; Porcerelli et al., 1998; Porcerelli et al., 2004; Sandstrom
represent a range of ages, relationships, and affects. The DMM & Cramer, 2003). The validity of the DMM has been supported
includes three developmentally anchored defense mechanisms: in several longitudinal studies (Cramer, 2002, 2008; Cramer &
Denial, Projection, and Identification. There are seven cate- Block, 1998; Cramer & Jones, 2007). The DMM has also dif-
gories representing aspects of each defense. Denial includes ferentiated between diagnostic groups (Cramer & Kelly, 2004;
(a) omission of major characters or objects; (b) misperception; Hibbard et al., 1994; Kim, Cogan, Carter, & Porcerelli, 2005).
(c) reversal; (d) statements of negation; (e) denial of reality; Following 15 months of intensive (inpatient) psychotherapy,
(f) overly maximizing the positive or minimizing the negative; overall DMM defense use decreased (Cramer & Blatt, 1990).
and (g) unexpected goodness, optimism, positiveness, or gen- Finally, the validity of the DMM system has also been sup-
tleness. Projection includes (a) attribution of hostile feelings or ported in experimental studies in which defense mechanism rat-
intentions or other normatively unusual feelings or intentions to ings are obtained before and after the introduction of a stressor
a character; (b) additions of ominous people, animals, objects, (e.g., a threat to the subject’s self-esteem). In three such studies
or qualities; (c) magical or autistic thinking; (d) concern for (Cramer, 1991a; Cramer & Gaul, 1988; Sandstrom & Cramer,
protection from external threat; (e) apprehensiveness of death, 2003), increases in defense use had been observed as a function
injury, or assault; (f) themes of pursuit, entrapment, and escape; of the experimental manipulation.
434 PORCERELLI, COGAN, KAMOO, MILLER

The DFS (American Psychiatric Association, 1994). The sive disorder scored higher on Mental Inhibition Level, Minor
DFS was developed as a proposed axis for the DSM to provide Image-Distorting Level, and Major Image-Distorting Level de-
both clinicians and researchers with a standard set of defense fenses than women who were not depressed (Porcerelli, Ol-
mechanisms, a set of definitions, and a method for aggregating son, Presniak, & Markova, 2009). Although the reasons for the
quantitative data for statistical analyses. The DFS is the outcome discrepancy in the findings of the two studies are not clear,
of many years of study by several researchers (Bond et al., 1989; differences may in part be due to differences in the two popula-
Bond & Vaillant, 1986; Cooper, 1992; Horowitz et al., 1992; tions (men and women vs. women; mixed ethnicity vs. African
Perry, 1990a, 1990b; Perry & Kardos, 1994; Vaillant, 1992). American women; psychotherapy vs. primary care clinic pa-
The scale includes 31 defenses organized into seven levels of tients). Women with major depression in the primary care sam-
defensive functioning: High Adaptive Level, Mental Inhibitions ple (Porcerelli et al., 2009) used more Mental Inhibition Level
Level, Minor Image-Distorting Level, Disavowal Level, Major defenses than any other level of defenses and thus may have
Image-Distorting Level, Action Level, and Level of Defensive had less severe psychopathology than women in the outpatient
Dysregulation. The clinician or researcher rates each defense psychiatric sample (DeFife & Hilsenroth, 2005).
each time it occurs during a clinical interview. The total number
of defenses within each level is multiplied by a weighted score
ranging from 1 to 7. Higher scores mean greater maturity and Procedures
less psychopathology. The sum of all the weighted level scores Women were recruited from the waiting room of the clinic. If
is then divided by the total number of defenses to obtain the they were interested in participating, they returned to the clinic
ODF score, a measure of the overall maturity of defenses. For at a later date to complete a series of self-report instruments and
this study, we combined four immature levels of defenses to an interview. The participants were first administered four cards
form a pathological/immature composite score: Minor Image- from the TAT (Murray, 1943). In the interview that followed, the
Distorting Level, Disavowal Level, Major Image-Distorting women were asked to provide 10 early memory narratives (earli-
Level, and Action Level. We did not use the Level of Defensive est, next earliest, mother, father, happiest, saddest, school, eating
Dysregulation because it rarely occurred in our sample. or being fed, warm/snug, transitional object) according to pro-
The interrater reliability of the ODF ranges from good to cedures by Fowler, Hilsenroth, and Handler (1995). We chose
excellent (Hilsenroth, Callahan, & Eudell, 2003; Lingiardi et al., the early memory procedure because it is experienced as a fairly
1999; Perry, 1990b, 2001; Perry et al., 1998) and scores were benign interview measure well tolerated by a primary care sam-
stable at 1 month (Perry, Hoglend, Shear, Vaillant, et al., 1998). ple and successful in eliciting defenses (Cousineau & Shedler,
The validity of the ODF has been supported by finding that 2006; Shedler, Karliner, & Katz, 2003; Shedler, Mayman, &
ODF scores are related to a self-report measure of social and Manis, 1993). The women were then asked to describe their
occupational functioning among outpatients (Hilsenroth et al., mother, father, and another significant other of their choosing,
2003). Lower ODF scores are associated with more Axis I and following the procedures of Blatt, Wein, Chevron, and Quinlan
Axis II symptoms, more interpersonal difficulties, and early (1979). Administration of the TAT and interview protocols were
treatment termination (Callahan & Hilsenroth, 2005; DeFife conducted by advanced graduate students in psychology, a doc-
& Hilsenroth, 2005; Hilsenroth et al., 2003; Hogland & Perry, toral level social worker, or a postdoctoral fellow in psychology.
1998; Lingiardi et al., 1999; Perry, 2001). Lower ODF scores are We coded the defenses from the DFS from the video-
associated with depressive symptoms in outpatients (DeFife & taped interviews. Two raters—J. Porcerelli and K. Miller—
Hilsenroth, 2005; Perry, Hoglend, Shear, Vaillant, et al., 1998) independently observed the interviews and rated the DFS. Our
and predicted less recovery from major depression at 6 months team agreed to this method of coding because the DFS was
(Hoglend & Perry, 1998). In summary, the ODF is an empirically developed as a clinical rating scale for clinicians conducting
grounded and clinically meaningful scale of defenses. face-to-face clinical evaluations. For both the DMM and the
The validity of the DFS levels has been supported by a study DFS, we discussed discrepancies in scoring, and we used a final
that found the expected correlations between the DFS levels and agreed on score for data analysis. The second rater (K. Miller),
clinician ratings of DSM–IV personality disorders (Blais et al., an advanced doctoral student in a psychodynamically oriented
1996). In a sample of adult outpatients, Cluster B personal- clinical psychology program, went through an extensive train-
ity disorders symptom scores were related to low-level defense ing period that included background readings on defense mech-
mechanisms (i.e., Action Level), whereas Cluster C personal- anisms theory and assessment and over 20 hr of coding practice
ity disorder symptom scores were associated with higher level interviews. With each of the practice interviews, detailed feed-
defenses (Lingiardi et al., 1999). Men who used mature de- back was given.
fenses had the best psychosocial and health functioning, men The TAT stories were videotape recorded and later transcribed
who used neurotic level defenses had intermediate psychosocial for coding purposes. This allowed the first author (J. Porcerelli)
and health functioning, and men who used primarily immature to be involved in the coding of the DMM without having been
defenses had the poorest psychosocial and health functioning biased by ratings from the DFS. All 240 TAT transcribed sto-
(Soldz & Vaillant, 1999). The levels we used in this study were ries were randomized and coded independently by the first
clusters and not the levels described in the DSM–IV. Adult out- (J. Porcerelli) and third (R. Kamoo) authors. Both raters have
patients with childhood sexual abuse had more Major Image- had more than 12 years of experience coding the DMM. The
Distorting Level defenses than other adult outpatients (Callahan raters were blind to all information about the participants. The
& Hilsenroth, 2005). In an outpatient sample, more depressive study was approved by the Michigan Department of Community
symptoms were related to lower scores on the Obsessional Level Health and Wayne State University Human Investigations Com-
Mental Inhibition defenses (DeFife & Hilsenroth, 2005). In a mittee of the Institutional Review Board. Participants received
study of primary care outpatients, women with a major depres- an $80 honorarium for their participation.
DEFENSE MECHANISMS MANUAL AND DEFENSIVE FUNCTIONING SCALE 435

TABLE 2.—Defense Mechanisms Manual (DMM) and Defensive Functioning Scale (DFS) scores.

M SD Range Skewness Kurtosis Interrater Reliabilitya

DMM
Denial 1.72 .96 0–5 0.84 0.77 0.53
Projection 2.17 1.64 0–6 0.51 −0.25 0.70
Identification 2.80 1.89 0–8 0.98 0.99 0.68
Total DMM defenses 11.45 4.67 4–22 0.66 −0.35 0.75
DFS
Overall Defensive Functioning 5.22 .59 3.67–6.50 −0.18 −0.18 0.80
Pathological Defenses 9.55 6.49 1–29 1.06 0.88 0.74b
High Adaptive Level 3.00 2.72 0–15 1.58 5.01 0.84
Mental Inhibitions Level 4.55 3.67 0–18 1.44 2.61 0.87
Minor Image Distorting Level 3.58 2.32 0–11 0.64 0.38 0.55
Disavowal Level 4.10 3.33 0–16 1.16 1.55 0.64
Major Image Distorting Level 1.13 1.94 0–8 2.08c 3.95 0.84
Action Level 0.62 1.26 0–6 2.75c 7.80c 0.94

Note. N = 60.
a
Intraclass correlation (1-way random effect).
b
Mean interrater relaibility of DFS Minor Image Distorting, Disavowal, Major Image Distorting, and Action levels.
c
Moderately non-normal.

RESULTS with DMM Identification. DMM Denial was positively corre-

Interrater reliability scores (intraclass correlation, one-way lated with the three most pathological DFS levels but was not
random effects) for the DMM and the DFS are shown in significantly correlated with the DFS Minor Image-Distorting
Table 2 and range from fair to excellent (Shrout & Fleiss, Level score. DMM Projection was positively correlated with two
1979). Individual intraclass correlations for the DFS levels were of the three pathological DFS levels (Minor Image-Distorting
High Adaptive Level, .84; Mental Inhibitions Level, .87; Mi- and Major Image-Distorting Levels) but was not significantly
nor Image-Distorting Level, .55, Disavowel Level, .64, Major correlated with DFS Disavowal level scores. Finally, DMM
Image-Distorting Level, .84; and Action Level, .94. The intra- Identification was not significantly correlated with any of the
class correlation for the ODF score was .80. Dysregulation level DFS level scores.
defenses (i.e., psychotic defenses) were not included in the anal-
yses because only three of these defenses were coded within DISCUSSION
the 60 interviews. As can also be seen in Table 2, skewness Despite numerous studies supporting the validity of Cramer’s
and kurtosis values show that the distributions are within nor- (1991b) DMM, this is the first study that compares the DMM
mal limits except for Major Image Distorting and Action Level with another validated measure of defense mechanisms. The
defenses, which are moderately non-normal (Curran, West, & DFS could be considered the “gold standard” of the assessment
Finch, 1996). of defenses, given its lineage and inclusion in the DSM–IV. Re-
The Pearson correlations of the DMM were Denial and Pro- sults from this study support the convergent validity of the im-
jection, r = .26, p = .05; Denial and Identification, r = –.02, mature DMM defenses of Denial and Projection. Not only was
p = .86; and Projection and Identification, r = .14, p = .30. Denial significantly related to a measure of defense maturity
The Pearson correlations between DMM defenses and total in the DFS, it was also significantly correlated with a com-
number of words in the TAT stories were r =.11, p = .38 for posite measure of pathological defenses in the DFS as well as
Denial; r = .14, p = .27 for Projection; and r = .31, p = .02
for Identification.
Our first hypothesis that ODF and DMM Denial would be TABLE 3.—Correlations between Cramer’s Defense Mechanisms Manual
negatively and significantly correlated was supported as shown (DMM) and Defensive Functioning Scale (DFS) scores.
in Table 3. The magnitude of the relationship approached a
moderate effect size. Our second hypothesis, that ODF and DMM
DMM Projection would be negatively and significantly cor- DSM–IV DFS Denial Projection Identification
related, showed a trend toward significance and approached a
small-to-moderate effect size. Our third hypothesis, that ODF ODF −.28∗∗∗ −.22∗ .02
and DMM Identification would be positively correlated, was not DFS Pathological Composite .36∗∗∗ .32∗∗ .04
supported. DFS levels
High Adaptive Level .12 −.05 .04
We hypothesized that the DFS pathological composite score Mental Inhibitions Level .11 .21 .12
would be positively correlated with the immature defenses of Minor Image Distorting Level .14 .29∗∗ .10
the DMM (Denial and Projection) and negatively correlated with Disavowal Level .32∗∗ .10 .08
DMM Identification. This hypothesis was partially supported in Major Image Distorting Level .26∗∗ .48∗∗∗ .01
that the DFS pathological composite score was positively cor- Action Level .27∗∗ −.05 −.15
related with DMM Denial and Projection, with moderate effect Note. DSM–IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.);
sizes in both cases, also shown in Table 3. However, the DFS ODF = Overall Defensive Functioning.
∗
pathological composite score was not significantly correlated p < .10. ∗∗ p < .05. ∗∗∗ p < .01.
436 PORCERELLI, COGAN, KAMOO, MILLER

Disavowal, Major Image-Distorting, and Action Level defense study is the hetero-method assessment of defense mechanisms,
scores. Most important, the correlations between the DMM and which is a conservative test of convergent validity.
the DFS were based on different methods of assessment, namely,
responses to the TAT and interview data, respectively. Although ACKNOWLEDGMENT
the divergence between TAT and interview data is not as great
Portions of this material were presented at meetings of the
as the difference between self-report and free response data,
Society for Personality Assessment, Chicago, Illinois, March
as described by Bornstein (2002), they do constitute different
2009.
methods of assessment.
These hetero-method findings also support the validity of
DMM Projection. DMM Projection correlated modestly with REFERENCES
ODF and significantly with the DFS pathological composite American Psychiatric Association. (1994). Diagnostic and statistical manual of
score and with Minor and Major Image-Distorting Level de- mental disorders (4th ed.). Washington, DC: Author.
fense scores, indicating that Projection is a pathological defense. Blais, M. A., Conboy, C. A., Wilcox, N., & Normal, D. K. (1996). An empirical
DMM Projection was not significantly correlated with DFS Ac- study of the DSM-IV Defensive Functioning Scale in personality disordered
patients. Comprehensive Psychiatry, 37, 435–440.
tion Level defenses, an expected finding. However, DFS Action
Blatt, S. J., Wein, S., Chevron, E. S., & Quinlan, D. M. (1979). Parental rep-
Level defenses were significantly correlated with DMM Denial. resentations and depression in normal young adults. Journal of Abnormal
Denial serves to defend against the consequences of action and Psychology, 88, 388–397.
is necessary for the success of Action Level defenses (Cramer Bond, M., Perry, J. C., Gautier, M., Goldenberg, M., Oppenheimer, J., & Simand,
& Kelly, 2004). An unexpected finding is the lack of a signifi- J. (1989). Validating the self-report of defense styles. Journal of Personality
cant correlation between DMM Projection and DFS Disavowal Disorders, 3, 101–112.
level. This is particularly surprising given that DFS Disavowal Bond, M., & Vaillant, G. E. (1986). An empirical study of the relationship
is composed of three defenses: Denial, Projection, and Rational- between diagnosis and defense style. Archives of General Psychiatry, 43,
ization. After considerable reflection, we have no explanation 285–288.
for this finding. Bornstein, R. F. (2002). A process dissociation approach to objective-projective
test score interrelationships. Journal of Personality Assessment, 78, 47–68.
Another surprising finding is the lack of a relationship be-
Callahan, K. L., & Hilsenroth, M. J. (2005). Childhood sexual abuse and adult
tween DMM Identification, the most mature DMM defense, defensive functioning. Journal of Nervous & Mental Disease, 193, 473–479.
and any of the DFS scales. Cramer (2006) suggested that al- Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd
though the DMM defenses are hierarchically ordered, the scale ed.). Hillsdale, NJ: Lawrence Erlbaum.
was originally designed to study defenses in children and does Cooper, S. (1992). The empirical study of defensive processes: A review. In J.
not include defenses more characteristic of adulthood (i.e., the W. Barron, M. N. Eagle, & D. L. Wolitzky (Eds.), Interface of psychoanalysis
scale has a low maturity ceiling). Another possibility is that and psychology (pp. 327–346). Washington DC: American Psychological
identification as conceptualized in the DMM represents more Association.
of a cognitive capacity than a defense mechanism in the adult Cousineau, T. M., & Shedler, J. (2006). Predicting physical health: Implicity
years. Although developmental studies have found that identi- mental health measures versus self-report scales. Journal of Nervous and
Mental Disease, 194, 427–432.
fication is a more mature defense mechanism than denial and
Cramer, P. (1987). The development of defense mechanisms. Journal of Per-
projection, it is not considered a mature defense in adults, who sonality, 55, 597–614.
have essentially consolidated their identities. Our findings show Cramer, P. (1991a). Anger and the use of defense mechanisms in college stu-
that DMM Identification is not related to a DFS measure of the dents. Journal of Personality, 59, 39–55.
maturity of defenses (ODF), the DFS higher level defenses, or Cramer, P. (1991b). The development of defense mechanisms: Theory, research
the DFS lower level defenses in adults. and assessment. New York, NY: Springer-Verlag.
The limitations of the study include a modest sample size and Cramer, P. (1997). Evidence for change in children’s use of defense mechanisms.
the inclusion of only one gender. Although a low income urban Journal of Personality, 65, 233–247.
community sample might be considered a limitation of the study, Cramer, P. (1998a). Defensiveness and defense mechanisms. Journal of Person-
it could also be considered a strength because this population is ality, 66, 879–893.
Cramer, P. (1998b). Freshman to senior year: A follow-up study of identity,
typically underrepresented in research. Because of the limited
narcissism and defense mechanisms. Journal of Research in Personality, 32,
sample size, we were not able to consider correlations between 156–172.
each of the seven DFS levels and the DMM defenses. However, Cramer, P. (1999). Personality, personality disorders, and defense mechanisms.
we have included the means and standard deviations for all of Journal of Personality, 67, 535–554.
the DMM and DFS defenses, as well as correlations between Cramer, P. (2001). Defense mechanisms in psychology today: Further processes
them, in the Appendix.1 The most important strength of the for adaptation. American Psychologist, 55, 637–646.
Cramer, P. (2002). Defense mechanisms, behavior, and affect in adulthood.
Journal of Perosnality, 70, 103–126.
Cramer, P. (2006). Protecting the self: Defense mechanisms in action. New
1 In the Appendix, the reader can note that DMM and DFS Denial were York, NY: Guilford.
not significantly correlated, nor were DMM and DFS Projection. However Cramer, P. (2007). Longitudinal study of defense mechanisms: Late childhood
DMM Denial was significantly correlated with DFS Projection. The lack of to late adolescence. Journal of Personality, 75, 1–24.
convergence for individual defenses may be due in part to differences in their Cramer, P. (2008). Identification and the development of competence: A 44 year
definitions. It is also likely that different response processes associated with TAT longitudinal study from late adolescence to late middle age. Psychology &
stimuli and interview prompts (i.e., for early memories and parental description) Aging, 23, 410–421.
elicited different individual defenses but within similar levels. These findings Cramer, P., & Blatt, S. J. (1990). Use of the TAT to measure change in de-
support the validity of DFS and DMM defense levels versus individual defenses fense mechanisms following intensive psychotherapy. Journal of Personality
as shown in Table 3. Assessment, 54, 236–251.
DEFENSE MECHANISMS MANUAL AND DEFENSIVE FUNCTIONING SCALE 437

Cramer, P., Blatt, S. J., & Ford, R. Q. (1988). Defense mechanisms in the Meyer, G. J. (2004). The reliability and validity of the Rorschach and TAT
anaclitic and introjective personality configuration. Journal of Consulting compared to other psychological and medical procedures: An analysis of
and Clinical Psychology, 56, 610–616. systematically gathered evidence. In M. Hilsenroth & D. Segal (Eds.) &
Cramer, P., & Block, J. (1998). Preschool antecedents of defense mechanism M. Hersen (Ed.-in-Chief), Personality assessment: Volume 2. Comprehensive
use in young adults. Journal of Personality and Social Psychology, 74, 159– handbook of psychological assessment (pp. 315–342). Hoboken, NJ: Wiley.
169. Murray, H. A. (1943). Thematic Apperception Test. Cambridge, MA: Harvard
Cramer, P., & Gaul, R. (1988). The effects of success and failure on children’s University Press.
use of defense mechanisms. Journal of Personality, 56, 729–742. Perry, J. C. (1990a). The Defense Mechanism Rating Scales (5th ed.). Cam-
Cramer, P., & Jones, C. J. (2007). Defense mechanisms predict differential bridge, MA: Cambridge Hospital.
lifespan change in self-control and self-acceptance. Journal of Research in Perry, J. C. (1990b). Psychological defense mechanisms and the study of af-
Personality, 41, 841–855. fective and anxiety disorders. In J. D. Maser & R. C. Cloninger (Eds.),
Cramer, P., & Kelly, F. D. (2004). Defense mechanisms in adolescent conduct Comorbidity of mood and anxiety disorders (pp 545–562). Washington, DC:
disorder and adjustment reaction. Journal of Nervous and Mental Disease, American Psychiatric Association.
192, 139–145. Perry, J. C. (2001). A pilot study of defenses in adults with personality disorders
Cramer, P., & Tracy, A. (2005). The pathway from child personality to adult entering therapy. Journal of Nervous & Mental Disease, 189, 651–660.
adjustment: The road is not straight. Journal of Research in Personality, 39, Perry, J. C., & Hoglend, P. (1998). Convergent and discriminant validity of
369–394. overall defensive functioning. Journal of Nervous & Mental Disease, 186,
Curran, P. J., West, S. G., & Finch, J. F. (1996). The robustness of test statis- 529–535.
tics to nonnormality and specification error in confirmatory factor analysis. Perry, J. C., Hoglend, P., Shear, K., Vaillant, G. E., Horowitz, M., Kardos, M. E.,
Psychological Methods, 1, 16–29. . . . Kagan, D. (1998). Field trial of a diagnostic axis for defense mechanisms
DeFife, J. A., & Hilsenroth, M. J. (2005). Clinical utility of the Defensive for DSM-IV. Journal of Personality Disorders, 12, 56–68.
Functioning Scale in the assessment of depression. Journal of Nervous & Perry, J. C., & Kardos, M. (1994). Ego defenses: Theory and practice. New
Mental Disease, 193, 176–182. York, NY: Wiley.
Fowler, J. C., Hilsenroth, M. J., & Handler, L. (1995). Early memories: An Porcerelli, J. H., Cogan, R., Kamoo, R., & Leitman, S. (2004). Defense mech-
exploration of theoretically derived queries and their clinical utility. Bulletin anisms and self-reported violence toward partners and strangers. Journal of
of the Menninger Clinic, 59, 79–98. Personality Assessment, 82, 317–320.
Freud, S. (1965). The neuro-psychoses of defense. In J. Strachey (Ed. and Porcerelli, J. H., Cogan, R., Markova, T., Murdoch, W., & Porcerelli, M. (2010).
Trans.), The standard edition of the complete psychological works of Sig- Abuse, outpatient costs and utilization, and psychiatric symptoms in urban
mund Freud (Vol. 3, pp 41–61). London, England: Hogarth. (Original work women on Medicaid. Journal of the American Board of Family Medicine, 23,
published 1894) 363–370.
Hibbard, S., Farmer, L., Wells, C., Difillipo, E., Barry, W., Korman, R., & Sloan, Porcerelli, J. H., Olsen, T. R., Presniak, M. D., & Markova, T. (2009). Defense
P. (1994). Validation of Cramer’s defense mechanisms manual for the TAT. mechanisms and major depressive disorder in African American women.
Journal of Personality Assessment, 63, 197–210. Journal of Nervous & Mental Disease, 197, 736–741.
Hibbard, S., & Porcerelli, J. H. (1998). Further validation for the Cramer Porcerelli, J. H., Thomas, S., Hibbard, S., & Cogan, R. (1998). Defense mecha-
Defense Mechanism Manual. Journal of Personality Assessment, 70, 460– nisms development in children, adolescents, and late adolescents. Journal of
483. Personality Assessment, 71, 411–420.
Hilsenroth, M. J., Callahan, K. L., & Eudell, E. M. (2003). Further reliabil- Sandstrom, M., & Cramer, P. (2003). Girls’ use of defense mechanisms follow-
ity, convergent and discriminant validity of Overall Defensive Functioning. ing peer rejection. Journal of Personality, 71, 605–627.
Journal of Nervous & Mental Disease, 191, 730–737. Shedler, J., Karliner, R., & Katz, E. (2003). Cloning the clinician: A method for
Hoglend, P., & Perry, J. C. (1998). Defensive functioning predicts improvement assessing illusory mental health. Journal of Clinical psychology, 59, 635–650.
in Major Depressive Episodes. Journal of Nervous & Mental Disease, 186, Shedler, J., Mayman, M., & Manis, M. (1993). The illusion of mental health.
238–243. American Psychologist, 48, 1117–1131.
Horowitz, M. J., Cooper, S., Fridhandler, B., Perry, J. C., Bond, M., & Vaillant, G. Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlation: Uses in assessing
(1992). Control processes and defense mechanisms. Journal of Psychotherapy rater reliability. Psychological Bulletin, 86, 420–428.
Practice and Research, 1, 324–326. Soldz, S., & Vaillant, G. E. (1999). The Big Five personality traits and the life
Kim, M., Cogan, R., Carter, S., & Porcerelli, J. H. (2005). Defense mechanisms course: A 45-year longitudinal study. Journal of Research in Personality, 33,
and self-reported violence toward strangers. Bulletin of the Menninger Clinic, 208–232.
69, 305–312. Vaillant, G. E. (1977). Adaptation to Life. Cambridge, MA: Harvard University
Lingiardi, V., Lonati, C., Delucchi, F., Fossati, A., Vanzulli, L., & Maffei, C. Press.
(1999). Defense mechanisms and personality disorders. Journal of Nervous Vaillant, G. E. (1992). Ego mechanisms of defense: A guide for clinicians and
& Mental Disease, 187, 224–228. researchers. Washington, DC: American Psychiatric Press.
438 PORCERELLI, COGAN, KAMOO, MILLER

APPENDIX.—Correlations between Defense Mechanisms Manual (DMM) and Defensive Functioning Scale Defenses (DFS) with DFS means and standard
deviations.

DMM

M SD Denial Projection Identification

High Adaptive Level

Anticipation 0.13 0.34 .17 .11 −.09
Affiliation 0.70 0.91 −.02 −.15 .10
Altruisim 0.12 0.37 .19 .27∗ −.09
Humor 1.27 1.67 −.06 −.l9 .06
Self-assertion 0.20 0.51 −.02 −.02 −.10
Self-observation 0.25 0.70 .28∗ .32∗∗ .08
Sublimation 0.12 0.37 .09 .02 .06
Suppression 0.22 0.49 .20 −.15 −.08
Mental Inhibition Level
Displacement 0.80 0.99 .15 .19 .12
Dissociation 1.18 2.12 .13 .03 .15
Intellectualization 0.58 0.96 .05 −.14 −.04
Isolation of affect 0.77 0.91 .02 .18 .00
Reaction formation 0.40 0.78 .02 .24 .07
Repression 0.75 1.11 −.13 .19 .10
Undoing 0.07 0.25 .15 −.07 −.01
Minor Image Distorting Level
Devaluation 1.63 2.00 .12 .24 .16
Idealization 1.78 1.38 .08 .12 −.05
Omnipotence 0.17 0.42 −.05 .08 −.04
Disavowal Level
Denial 1.47 1.56 .17 .11 .11
Projection 0.37 0.61 .38∗∗ .04 −.01
Rationalization 2.27 2.31 .24 .05 .04
Major Image Distorting Level
Autistic fantasy 0.02 0.13 .18 .23 −.12
Projective identification 0.25 0.75 .03 .32∗∗ .02
Splitting of self/others 0.87 1.56 .29∗ .42∗∗∗ .01
Action Level
Acting out 0.32 0.98 .13 −.04 −.03
Apathetic withdrawal 0.08 0.33 .02 .01 −.13
Help-rejecting complaining 0.02 0.13 .18 .07 −.05
Passive aggressive 0.20 0.63 .29∗ −.05 −.18

Note. N = 60.
∗
p < .05. ∗∗ p < .01. ∗∗∗ p < .001.