Journal of Research in Personality 45 (2011) 19–28

Contents lists available at ScienceDirect

Journal of Research in Personality

journal homepage: www.elsevier.com/locate/jrp

Young adult narcissism: A 20 year longitudinal study of the contribution

of parenting styles, preschool precursors of narcissism, and denial
Phebe Cramer
Department of Psychology, Williams College, Williamstown, MA 01267, United States

a r t i c l e i n f o a b s t r a c t

Article history: The role of parenting styles in the development of young adult narcissism is investigated with individuals
Available online 11 November 2010 from the Block and Block (1980) longitudinal study. At age 3, participants were assessed for the presence
of narcissism precursors, and mothers and fathers provided information about their parenting styles. At
Keywords: age 23, the presence of both healthy and maladaptive narcissism was assessed, along with the use of
Narcissism denial. The results showed that parenting styles had a direct effect on the development of healthy nar-
Parenting styles cissism, but the effect on the development of maladaptive narcissism depended on the child’s initial pro-
Longitudinal study
clivity towards narcissism. Also, the use of denial was positively associated with the presence of
Defense mechanisms
maladaptive narcissism, but not with healthy narcissism.
Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction expecting but not demanding approval and admiration from

others.
Narcissists are characterized as being excessively focused on For these reasons, significant relations between early child-
having their own needs met – especially the need for admiration. rearing styles and subsequent narcissism should be found. Parent-
As part of their egocentric focus, they often fail to form caring, last- ing that is responsive to the child’s needs is expected to predict
ing relationships with others, as has been demonstrated in the re- subsequent healthy, adaptive narcissism. Parenting that is unre-
search of Campbell and Foster (2002) and in the clinical writings of sponsive is expected to predict subsequent maladaptive narcis-
Kernberg (1998), although narcissists have an extreme need for sism. It is the purpose of the present study to investigate these
admiration from others. hypotheses.
The question arises as to the origin of the extreme need for
admiration. One common assumption is that the adult narcissist 1.1. Adult narcissism
was a spoiled child, having been overly indulged by parents who
offered excessive gratification. Having grown up with this back- In adults, narcissism may be adaptive or maladaptive (Pincus &
ground of indulgence, the individual continues to expect and to Lukowitsky, 2010; Raskin, Novacek, & Hogan, 1991; Russ, Shedler,
demand this kind of gratification as an adult (cf. Millon, 1990). Bradley, & Westen, 2008; Wink, 1992). Adaptive narcissism is char-
A different explanation for the origin of narcissism assumes that acterized by healthy ambitions, energy, creativity, and empathy,
narcissism results not from excessive gratification, but rather from supported by an underlying sense of self that is firm and cohesive
insufficient early gratification (Kernberg, 1975; Kohut, 1977; (Russ et al., 2008; Kohut, 1971; Pincus & Lukowitsky, 2010; Wink,
Miller, 1981). The adult who has experienced insufficient gratifica- 1992). Adaptive narcissists may be overly ambitious, but they have
tion as a child will be the one who demands excessive admiration sufficient interpersonal sensitivity so that they do not suffer the
from others, to compensate both for the lack of that psychological eventual rejection that is often experienced by maladaptive narcis-
support when a child and for the continuing expectation that needs sists. Wink (1991) has characterized this type of narcissism as
will not be met. ‘‘Autonomous’’. Although Autonomous narcissism and genuine
In contrast, the person who has experienced having his/her self-esteem share some characteristics, they also differ in some
physical and emotional needs adequately met as a child, has little important ways. The concept of Autonomous narcissism includes
anxiety about having needs met as an adult. From this point of high ambition, a preference to function without collaboration with
view, it is the individual who has experienced ample gratification others, and idiosyncratic thought processes (Wink, 1991), charac-
as a child who will go on to have a positive, healthy sense of self, teristics that are not an integral part of the concept of self-esteem.
In contrast to adaptive narcissism, maladaptive narcissism is
characterized by self-aggrandizement, power seeking and conde-
E-mail address: [email protected] scension (Raskin et al., 1991) in which an inflated sense of self

0092-6566/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jrp.2010.11.004
20 P. Cramer / Journal of Research in Personality 45 (2011) 19–28

masks underlying feelings of vulnerability and insecurity lan, & Cassidy, 1985). Likewise, parenting that conveys warmth
(Kernberg, 1975; Kohut, 1976). Maladaptive narcissism may be and affection, without being demanding (Indulgent) supports a
divided into two types – overt and covert (Wink, 1991). In the pres- sense of being a person who is respected by, and desirable to oth-
ent paper, we focus on overt or grandiose narcissism, characterized ers (George & Solomon, 1989).
by an open expression of grandiosity, self-confidence, and conde- This thesis, however, should be qualified by the assumption that
scension, in which there is willful manipulation and exploitation the effect of parenting style may differ at different developmental
of others (Wink, 1992; Ziegler-Hill, Clark, & Pickard, 2008). This periods. Whereas indulgence during infancy and very early child-
is the type of narcissism that is assessed by the self-report Narcis- hood is expected to support positive development, indulgence dur-
sistic Personality Inventory (NPI: Raskin & Terry, 1988). ing the adolescent period may result in an attitude of entitlement.
This is because the needs – both physical and emotional – of a very
1.2. Precursors of narcissism in childhood young child are different from those of a late adolescent. The infant
and young child are almost entirely dependent on the parent to
In school age children, clinical study has identified certain char- provide physical and emotional gratification, and require the sup-
acteristics that lead to a diagnosis of childhood narcissism. These port and encouragement of the parent in order to develop new
include an excessive need for attention and admiration, exhibition- capacities and skills, which is not true of the late adolescent. The
ism, impulsivity, aggression, and chronic violation of rules (Berren, parent who provides this support influences the development of
1998; Kernberg, 1998; Weise & Tuber, 2004). Although there is no ‘‘mastery skills, the capacity for affective involvement, and a sense
evidence that narcissism as a psychological disorder exists in of confidence . . . [In the preschool period] the child will be confi-
3 years old children, it is possible to identify precursors of narcis- dent, skilled and positive in dealing with peers and other tasks’’
sism at this young age. Based on Q-sort ratings provided by multi- (Sroufe, 1979, p. 837). Responsive parenting thus supports a posi-
ple observers of children’s behavior (Block & Block, 1980), Carlson tive model of the self, based on the parent’s belief that the child can
and Gjerde (2009) developed scales assessing five components, or acquire a level of competence appropriate to his/her developmen-
precursors, of narcissism in preschool children. These scales dem- tal level.
onstrated good reliability and were shown to predict an indepen- Parenting styles that do not provide this responsiveness and
dent measure of narcissism (10 CAQ items highly correlated with support for cognitive and physical development are likely to inter-
the NPI) at ages 14, 18, and 23. fere with positive development, with the child developing a sense
of inadequacy (Kernberg, 1998), and a model of the self that is
1.3. Parenting styles deemed unworthy of support (Bowlby, 1969; George & Solomon,
1989).
The child development literature has identified four different To expand on these ideas: What does it mean to be ‘‘indulged’’
styles of parenting (Baumrind, 1971). The Authoritarian/Autocratic as a 3-year-old? To be given as much food as desired, an excessive
parent tries to shape, control and evaluate the behavior of the child number of toys, too many hugs and kisses? What does the 3-year-
according to a set standard. Obedience to authority is stressed; or- old learn from this? Something like, although I am not able to pro-
ders are expected to be obeyed without explanation. Punitive mea- vide these things for myself, I believe that others will take care of
sures are favored to bring about compliance. This type of parenting me and meet my needs, because I am young and have not yet devel-
is demanding, but not responsive. The Authoritative/Responsive par- oped the capacities to provide these things for myself. For the late
ent provides clear standards for the child’s conduct, and uses rea- adolescent, parental ‘‘indulgence’’ is not required for survival.
soning and explanation to influence the child’s behavior. These When the late adolescent is indulged (excessive allowance, too
parents are assertive but not intrusive. They consider the child’s many fancy clothes, expensive car, no curfew, excessive non-con-
point of view, and assume that the child has rights. Discipline is tingent praise, etc.) s/he learns ‘‘I deserve to be given whatever I
supportive rather than punitive. This type of parenting is both want, despite my own ability to work and provide these things
demanding and responsive. The Indulgent/Permissive parent ex- for myself, because I am special and should be admired’’. The
presses affection easily, is lenient, finds it difficult to punish the presence of Indulgence in adolescence may result in maladaptive
child, and does not require mature behavior. This parent is more narcissism.
responsive than demanding. The Indifferent/Uninvolved parent is While the presence of indulgence in very early childhood may
neither demanding nor responsive. S/he expects the child to handle have positive results, the absence of indulgence may create a sense
problems alone, encourages the child to be independent of the par- of neediness and an expectation that others will not naturally be
ent, is not supportive, and expects the child to take responsibility responsive to those needs. To obtain this gratification, the child
for his/her own life (Baumrind, 1971, 1980, 1991a, 1991b). may develop compensatory defensive behaviors, including the
development of a ‘‘grandiose self’’ (Kernberg, 1998), in which the
1.4. Parenting styles and narcissism self is aggrandized, needs are met through wishful thinking, and
relations with others (who do not meet the child’s needs) are dis-
It is the thesis of the present paper that the type of parenting regarded (i.e., behaviors we characterize as narcissistic). The ab-
style the child experiences will influence the development of his/ sence of indulgence in late adolescence, when coupled with
her sense of self, which will in turn be related to the development minimal demandingness (Indifferent parenting) is also likely to
of narcissism. In this regard, the dimensions of ‘responsive’ and have negative effects, including antisocial behavior and illicit drug
‘demanding’ are especially important. Too much or too little use (Baumrind, 1991a; Kernberg, 1998) but is unlikely to produce a
Demandingness, without being Responsive to the child’s legitimate sense of personal inadequacy (Levy-Warren, 1998).
needs (i.e., Authoritarian or Indifferent) leaves the child feeling Similarly, it is suggested that whether ‘‘demandingness’’ is a po-
incompetent, either because the child is not allowed to develop sitive or negative factor in development depends on the develop-
his/her own skills, or because, without needed assistance and guid- mental level of the child. What are the effects of experiencing a
ance, the child is left vulnerable to mishap and trauma (Kernberg, demanding Authoritarian parental style as a 3-year-old?
1998). In contrast, Demandingness that at the same time recog- Authoritarian parents tend to be over-controlling, taking over the
nizes and is Responsive to the child’s needs – i.e., Authoritative child’s activities and requiring obedience to their way of doing
parenting – supports competent skill building and self-regulation, things (Baumrind, 1971). Excessiveness demandingness interferes
thus supporting autonomy (George & Solomon, 1989; Main, Kap- with the child developing his/her own skills and with recognizing
 P. Cramer / Journal of Research in Personality 45 (2011) 19–28 21

personal thoughts, feelings and skills – i.e., with the development Although denial is an immature defense that is characteristic of
of the self (Miller, 1981; Winnicott, 1965). The child remains young children, it sometimes continues to be used by older indi-
dependent on outside approval to maintain the false self, which viduals who exhibit evidence of psychopathology (e.g., Cramer,
may in turn be a source of pride and gratification for the parent. 1999; Vaillant, 1994). Because narcissism is characterized by
Such positive regard from the parent staves off feelings of vulner- immaturity (Morf & Rhodewalt, 2001), and because the narcissist
ability and worthlessness (Winnicott, 1965). is avoiding the recognition of a negative and vulnerable self repre-
When the late adolescent encounters the Authoritarian parent, sentation, the defense of denial may be expected to be associated
the attempt of the parent to take over or dictate the adolescent’s with the presence of maladaptive, but not healthy, narcissism in
behavior, creates doubts about personal competence, and retards young adults.
adolescent development (Baumrind, 1991a). This may result in
rebellion against the parent, or, if the parent lavishes praise and
1.5.1. Overview of study
approval on the adolescent for meeting the demands, it may con-
In this longitudinal study, we investigate the role of preschool
tribute to a feeling of specialness and importance, since adoles-
precursors of narcissism, parenting styles of mothers and fathers,
cents who follow the rules of the Authoritarian parent may be
and the defense of denial in predicting the presence of both mal-
‘‘over-idealized’’ and hence have an unrealistically grandiose
adaptive and healthy narcissism in young adulthood. Narcissism
picture of the self (Kernberg, 1975). This sense of specialness, how-
at both ages is assessed via expert observations. Parenting styles
ever, can only be maintained by the continuing admiration of oth-
are independently assessed at preschool age via an ipsative proce-
ers. Thus, throughout development, the reliance on the approval of
dure (Q-sort) that controls for response set bias and requires the
the Authoritarian parent may lead to a continuing requirement of
parents to focus on their treatment of the individual child, not as
admiration from others in order to sustain personal equanimity.
compared to treatment of other children, or to treatment by other
In contrast, too few demands by the parent, coupled with a lack
parents. Denial is assessed at young adulthood from the coding of
of responsiveness to needs (Indifferent parenting) will also inter-
narrative material. Despite specific predictions, two-tailed tests are
fere with the development of skills and competencies that would
used throughout.
support a positive sense of self. As suggested above, this may lead
to antisocial behavior.
There is some evidence supporting these assumptions regarding 1.6. Hypotheses
the relation of parenting style to narcissism. In a study of children
and adolescents, maladaptive narcissism, as assessed with a self- The development of adaptive and maladaptive narcissism at age
report child version of the NPI, was found to be positively related 23 will depend on the presence of early precursors of narcissism,
to a combined parent/child report of negative parenting styles, the use of denial, and the parenting style (responsive, demanding)
including lack of supervision, inconsistent discipline, and corporal that the child experiences. Indulgent parenting is characterized as
punishment (Barry, Frick, Adler, & Grafemen, 2007). Study of High responsive, but not demanding. Authoritarian parenting is
School students found that their concurrent report of their parents’ demanding but not responsive. Indifferent parenting is neither
psychological control and warmth was related to higher NPI scores, responsive nor demanding. Authoritative parenting is both
whereas parental monitoring was negatively related to NPI scores demanding and responsive.
(Horton, Bleau, & Drwecki, 2006). College students who completed
a self-report measure of maladaptive Narcissism and also provided Hyp. 1. The presence of early precursors of narcissism will pre-
retrospective reports of their parents’ parenting style showed that dict the presence of maladaptive narcissism at age 23,
Authoritarianism and Parental Disinterest were positively corre- but not healthy narcissism.
lated with maladaptive narcissism as assessed by the OMNI Hyp. 2. The use of Denial at age 23 will be associated with the
(O’Brien, 1987, 1988), whereas Authoritativeness was unrelated presence of maladaptive narcissism, but not healthy
to OMNI narcissism (Ramsey, Watson, Biderman, & Reeves, narcissism.
1996). For an older group (adults aged 18–52 years) self-reports Hyp. 3. The development of adaptive narcissism will be posi-
of current maladaptive narcissism, as assessed by the NPI, were tively predicted by parenting styles that are responsive
positively related to retrospective reports of parental ‘‘coldness’’ (Authoritative, Indulgent) and negatively predicted by
and ‘‘overevaluation’’ (Otway & Vignoles, 2006). styles that are non-responsive (Authoritarian,
The interpretation of these results is compromised by the fact Indifferent).
that in three of the studies both the narcissism measures and the Hyp. 4. The development of maladaptive narcissism will be pre-
report of parenting styles were based on concurrent self-report dicted by the combination of early precursors of narcis-
measures, and in two studies the report of parenting styles was sism and early parenting styles.
retrospective, raising the possibility in both cases that current nar- A. In combination with early precursors of narcissism, par-
cissism may have biased the report of parenting styles. Further, enting styles that are not responsive (Authoritarian,
these studies have focused on parenting styles during late child- Indifferent) will predict the development of maladaptive
hood and adolescence, the effects of which may differ from those narcissism. Parenting styles that are responsive (Author-
of infancy and very early childhood. itative, Indulgent) will be negative predictors of mal-
adaptive narcissism.
1.5. Narcissism and the defense of denial B. In combination with early precursors of narcissism, the
relation of parental demandingness to the development
Narcissism, itself, may be understood as a defensive process, the of age 23 narcissism will depend on the presence/
purpose of which is to defend the individual from the anxiety asso- absence of responsiveness.
ciated with a negative sense of self. By maintaining a façade of self-
importance, entitlement and condescension, the narcissist defends Thus, parenting styles that are demanding but not responsive
against recognition of self-doubt, vulnerability, and worthlessness. (Authoritarian) will predict the development of maladaptive nar-
If narcissism serves a defensive purpose (Morf & Rhodewalt, 1993, cissism. Parenting styles that are both demanding and responsive
2001; Carlson & Gjerde, 2009; Morrison, 1986), it is interesting to (Authoritative) will be negative predictors of maladaptive
consider its relation to other defensive processes, such as denial. narcissism.
22 P. Cramer / Journal of Research in Personality 45 (2011) 19–28

2. Method 2.2.3. Age 23 personality

On the basis of multiple observations, each participant was as-
2.1. Participants sessed with the California Adult Q-sort (CAQ; Block, 1961/1978) by
two or more judges at age 23. The judges were Ph.D. psychologists
The participants in this study come from the Block and Block and advanced graduate students in clinical or personality psychol-
Longitudinal Project (Block & Block, 1980). These individuals were ogy. The 100 CAQ items describe a wide range of cognitive, emo-
initially recruited into the study at age 3 from children at two nurs- tional, social and physical behaviors that can be used to
ery schools, one a university laboratory school, the other a parent characterize an individual. Each judge, working independently,
cooperative. The sample was assessed at multiple ages between 3 sorted the CAQ items into a forced-choice, nine-point distribution
(1968) and 23 (1988). The sample is heterogeneous with respect to for each participant, ranging from ‘‘extremely uncharacteristic’’
the social class and educational level of their parents, with a slight (rating of 1) to ‘‘extremely characteristic’’ (rating of 9). The Q-item
skew toward the middle class. The majority are Caucasian, with a scores were averaged across the raters to obtain final ratings for
smaller number of minority children. each item for each participant. These Q-item ratings were used
Narcissism precursor scores at both ages 3 and 4 (see below) for the present analyses.
and age 23 were available for 102 individuals. Of these, 89 had rat-
ings of parenting styles from mothers, 74 had parenting ratings 2.2.4. Age 23 narcissism
from fathers, and 90 had denial scores. Narcissism measures based on the CAQ are used here (Wink,
1991, 1992). Previously, nine expert judges sorted the 100 items of
the CAQ to represent a highly narcissistic person, based on the
2.2. Measures DSM-III criteria for a narcissistic personality disorder and their clin-
ical intuition. The alpha reliability of the aggregate judgments was
2.2.1. Childhood personality .91. The Q-ratings for each item (1–9) were then averaged across
The California Child Q-set (CCQ: Block & Block, 1980) was used judges, and the resulting value was used to create a CAQ template,
to assess childhood personality. The CCQ consists of 100 state- or prototype, measure of narcissism. The 13 CAQ items most charac-
ments regarding personality, cognition, and social characteristics teristic of this narcissism prototoype were then factor analyzed,
of children. Observers use these statements to describe each child, yielding three factors: Willfulness, Hypersensitivity, and Autonomy.
by means of a forced-choice distribution, such that only five items The three factor scores were then correlated with the total set of 100
can be rated as ‘‘extremely uncharacteristic’’ (rating of 1) and five CAQ items. Based on these correlations, Wink (1992) constructed
as ‘‘extremely characteristic’’ (rating of 9). Less extreme scores three new CAQ Narcissism scales: Willfulness (10 items), Hypersen-
(rating of 2–8) are allotted greater, but still specified, frequencies. sitivity (12 items), and Autonomy (11 items). Internal consistency of
Such ipsative scoring requires the rater to consider the target indi- the three scales ranged from .87 to .92 in the criterion sample of 105
vidual only in terms of his or her own unique personality, not in middle aged college women, and from .87 to .89 in a cross-validation
terms of how s/he compares to others of a similar age (see Ozer, sample of 350 community residents and college sophomores. In the
1993). present sample, the alpha values were: Willfulness, .89; Hypersensi-
At age 3, the observers who made the CCQ ratings were three tivity, .94; Autonomy, .80. It should be noted that each participant
nursery school teachers who had observed the child for at least can receive a score for each type of narcissism. Preliminary analyses
six months. At age 4, a different set of three nursery school teach- indicated that the Hypersensitivity scale was unrelated to any of the
ers again provided a Q-sort for each child. At ages 3 and 4, the inde- independent variables of the present study, and so this scale was re-
pendent Q-sort ratings of the three observers were averaged to moved from further study.
form a composite Q-sort evaluation for ages 3 and 4. The alpha Validity of the scales has been demonstrated through significant
of the CCQ items based on the correlation among nursery school correlations with established measures of narcissism and selected
teachers averaged .65 at both ages 3 and 4. These two composite personality inventory scales. For example, the Wilfullness scale cor-
Q-sorts were then averaged to form an overall (age 3 plus age 4; related positively with DSM-III ratings of narcissism, with the self-
age 3/4) preschool composite for each child. The average CCQ item report CPI Narcissism scale, and with measures of pathology
reliability of the composite was .91 (Carlson & Gjerde, 2009). These (Wink, 1992). This scale is characterized as ‘‘overt narcissism’’
Q-item ratings were used for the present analyses. (Wink, 1992). CAQ items include Characteristically pushes and tries
to stretch limits and rules; sees what s/he can get away with; Is self-
indulgent; tends to ‘‘spoil’’ or pamper himself or herself.
2.2.2. Preschool narcissism precursors In contrast, the Autonomy scale correlated positively with
Based on the age 3/4 CCQ ratings and theoretical expectations, inventory measures of self-assurance, confidence, empathy, and
Carlson and Gjerde (2009) developed a measure of preschool nar- energy, and did not correlate with measures of pathology (Wink,
cissism precursors with the current sample. Five subscales com- 1992). This scale is characterized as ‘‘healthy narcissism’’ (Wink,
prise this measure: Center of Attention (5 CCQ items), Activity (4 1992). CAQ items include Values own independence and autonomy;
CCQ items), Histrionic tendencies (4 CCQ items), Interpersonal Emphasizes his/her freedom to think and act without interference or
Antagonism (6 CCQ items), and Undercontrol of Impulse (4 CCQ help from others; Thinks and associates to ideas in unusual ways;
items). Scale items do not overlap. Examples include: ‘‘Tries to be Has high aspiration level for self; is ambitious; sets high personal goals.
the center of attention’’; ‘‘Is considerate and thoughtful of other Despite some similarity between Autonomous narcissism and trait
children (reversed)’’. The alpha values of these scales in the present self-esteem, this measure is unrelated to the Rosenberg Self-Es-
sample ranged from .70 to .94. These subscales are referred to as teem (RSE) scale in the present sample, r = .04.
‘precursors’ of adolescent and young adult narcissism, in that they
predicted an independent measure of narcissism at ages 14, 18, 2.2.5. Child-rearing styles
and 23; the correlations ranged from .21 to .42 (ps = .05–.001) The Child Rearing Practices Report (CRPR: Block, 2008; Block,
(see Carlson & Gjerde, 2009). Block, & Morrison, 1981) was used to assess four different styles of
The five subscales are added together to form an overall pre- child rearing: Authoritarian/Autocratic, Authoritative/Responsive,
school Narcissism Precursors score. The alpha value of this mea- Indulgent/Permissive, and Indifferent/Uninvolved. The CRPR is a
sure in the present sample is: a = .78. set of 91 items phrased in the first person form, tapping child-rearing
 P. Cramer / Journal of Research in Personality 45 (2011) 19–28 23

attitudes, values, behavior and goals. Parents are asked to arrange The validity of the coding approach has been demonstrated
the items as a Q-sort, with seven categories and a fixed, rectangular both through observational studies of children, adolescents, college
distribution of 13 items per category, ranging from ‘‘least descrip- students and psychiatric patients, in which defense mechanism
tive’’ to ‘‘most descriptive’’ of themselves. Parents were given de- scores differentiated between age or diagnostic groups (e.g., Cra-
tailed instructions for completing the Q-sort under the supervision mer, 1991a, 2003, 2004; Hibbard et al., 1994; Porcerelli et al.,
of research personnel. Mothers and father completed the Q-sorts 1998). Denial has also been shown to be positively related to the
independently when the child was 3 years old. personality variables of Anxiety (Cramer, 2002) and Big 5 Neurot-
To create prototypes for the four different styles of parenting – icism (Cramer, 2003), and to the Identity variables of Diffusion and
Authoritarian/Autocratic, Authoritative/Responsive, Indulgent/ Moratorium (Cramer, 1995), and to be negatively related to Loev-
Permissive, and Indifferent/Uninvolved – several doctoral level psy- inger’s Ego Level (Cramer, 1999) and Big 5 Agreeableness and Con-
chologists and advanced graduate students sorted the 91 items to scientiousness (Cramer, 2003) among others. Validity has also been
represent each of these styles of parenting. These sorts were com- demonstrated in empirical investigations in which experimental
posited to form a prototype for each style. Items that are highly manipulations that increased stress produced predicted changes
characteristic of the Authoritarian style include: Believes child should in defense mechanism use (Cramer, 1991b, 1998; Cramer & Gaul,
be seen and not heard; Does not allow child to get angry with me; Has 1988; Sandstrom & Cramer, 2003).
strict, well-established rules for child; Does not allow child to question
parental decisions. Items that are highly characteristic of the Author-
3. Results
itative style include: Respects child’s opinions, encourages expression;
Talks it over, reasons with child when he/she misbehaves; Future plans
3.1. Descriptive statistics
include child’s preferences; Trusts child to behave properly even when
not with him. Items that are highly characteristic of the Indulgent/
A comparison of the Narcissism Precursor scores at age 3/4 for
Permissive style include: Feels child should have time to think/day-
boys (Mean = 13.67, SD = 30.05) and girls (Mean = 7.28, SD =
dream/loaf; Lets child make many decisions for himself/herself; Child
25.39) indicated no significant difference, t(100) = 1.16, p = .25. A
should be comforted when scared; Finds it difficult to punish child.
comparison of Willfulness scores at age 23 for men (Mean = 34.06,
Items that are highly characteristic of the Indifferent style include:
SD = 10.80) and women (Mean = 28.52, SD = 10.36) showed that
Child expected to handle problems by self; Sometimes forgets promises
men scored significantly higher on Willfulness, t(100) = 2.64,
made to child; Encourages child to be independent of parent. Each par-
p < .009. Autonomous Narcissism did not differ for men (Mean =
ent’s score for each style of parenting was determined by correlating
35.00, SD = 13.97) and women (Mean = 35.85, SD = 9.58).
his/her own Q-sort with each prototype sort.
Scores for Child Rearing Practices are presented in Table 1. A ser-
There is considerable evidence for the test–retest reliability and
ies of t-tests indicated that Mothers’ and Fathers’ Child Rearing Prac-
the validity of the CRPR (Block et al., 1981; Dekovic, Janssens, &
tices scores did not differ for male and female children (see Table 1).
Gerris, 1991; Kochanska, Kuczynski, & Radke-Yarrow, 1989; Leahy,
However, overall, Mothers were more Authoritative, t(100) = 2.54,
1981; Mussen, Harris, Rutherford, & Keasey, 1970). Studies of the
p < .01, whereas Fathers were more Authoritarian, t(100) = 1.91,
CRPR have indicated consistency in item response across time
p = .059 and more Indifferent, t(100) = 5.30, p < .001. Mothers and
(age 3–12; age 7–16), with patterns of sameness or change being
Fathers did not differ on Indulgent/Permissive, t(100) = 1.25, p > .20.
similar for both mothers and fathers of girls and of boys (McNally,
For the total group, the mean Denial score = 5.39 (SD = 4.27).
Eisenberg, & Harris, 1991; Roberts, Block, & Block, 1984). Changes
Men and women did not differ, t(88) = .54.
that occurred were appropriate to the changing developmental le-
vel of the child. Research also demonstrated congruence between
the parent self-report CRPR and observational data on parenting 3.2. Correlations among the Narcissism measures
styles (Dekovic et al., 1991; Kochanska, 1990; Kochanska et al.,
1989). For a detailed description of the CRPR, see Block (2008). The age 3/4 Narcissism Precursor correlations with age 23
Narcissism are presented in Table 2. Age 3/4 Narcissism Precursors
2.2.6. Denial
The Defense Mechanism Manual (Cramer, 1991a) was em-
Table 1
ployed to assess the use of Denial in TAT stories told at age 23. Sub-
Child Rearing Practices Report: Means, SD.
jects told stories to six TAT pictures (Cards 2, 8GF, 8BM, 10, 15, and
18GF) representing males and females, individuals, dyads, a group, All Males Females t df p
and interior and exterior scenes. These stories were recorded and Mother Authoritarian
subsequently transcribed. Mean .40 .37 .43 1.67 87 .10
(SD) (.16) (.17) (.14)
Without knowledge of the subject’s gender or other informa-
tion, each of the 540 stories was scored by a trained coder for Mother Authoritative
Mean .63 .61 .64 .81 87 .42
the presence of Denial according to the Defense Mechanism
(SD) (.15) (.16) (.13)
Manual. To score Denial, there are seven categories representing
Mother Indulgent .51 .48 .53 1.45 87 .15
different aspects of the defense; each category is scored as many
(SD) (.16) (.18) (.14)
times as it occurs in each story. The scores were then summed over
Mother Indifferent .16 .16 .15 .42 87 .68
the six stories, yielding a total score for Denial. This measure of de-
(SD) (.12) (.11) (.12)
nial has been demonstrated in previous studies to have adequate
Father Authoritarian .36 .33 .38 1.11 72 .27
inter-rater reliability, with children, adolescents, adults, and psy-
(SD) (.18) (.19) (.17)
chiatric patients (e.g., Cramer, 1998, 2003; Cramer & Jones, 2008;
Father Authoritative .59 .59 .60 .03 72 .98
Hibbard et al., 1994; Porcerelli, Thomas, Hibbard, & Cogan, 1998;
(SD) (.17) (.19) (.15)
Sandstrom & Cramer, 2003). In the present study, a second coder
Father Indulgent .48 .46 .50 .84 72 .41
independently scored a random selection of 100 of the stories.
(SD) (.18) (.20) (.16)
The agreement between coders, based on intra-class correlation,
Father Indifferent .09 .07 .09 .76 72 .45
was .80, indicating adequate inter-rater reliability. Disagreements
(SD) (.14) (.16) (.13)
were resolved subsequently by discussion.
24 P. Cramer / Journal of Research in Personality 45 (2011) 19–28

Table 2
Intercorrelations: Narcissism, Child Rearing Practices and Denial.

1 2 3 4 5 6 7 8 9 10 11 12 13
*** !
1. Narcissism age 3 .38 .06 .01 .16 .19 .18 .08 .15 .08 .13 .21 .11
2. Willful. age 23 .41*** .19* .09 .06 -.06 .05 .08 .07 .05 .22* .29**
3. Hyper. age 23 .28** .09 .03 .10 .10 .01 .05 .01 .02 .12
4. Auton. age 23 .30** .40*** .38*** .04 .21! .28** .24* .18 .01
5. Mother Authoritarian .70*** .96*** .29** .41*** .27* .39*** .23* .17
6. Mother Authoritative .82*** .08 .46*** .48*** .52*** .18 .01
7. Mother Indulgent .16 .45*** .34** .46*** .20 .13
8. Mother Indifferent .19 .19 .14 .42*** .03
9. Father Authoritarian .79*** .97*** .46*** .02
10. Father Authoritative .85*** .56*** .16
11. Father Indulgent .40*** .02
12. Father Indifferent .15
13. Denial
!
p = .07.
*
p < .05.
**
p < .01.
***
p < .001.

were significantly related to age 23 Willfulness (r = .38) but not to Rearing Practices, and age 23 Denial. In the present section, the lat-
Autonomous Narcissism (r = .01). There was a negative relation ter three independent variables are included in hierarchical regres-
between Willfulness and Autonomous Narcissism, (r = .19, sion analyses, to determine the contribution of each to predicting
p < .05).1 age 23 Narcissism. In addition, the role of the interaction between
Childhood Narcissism Precursors and Child Rearing Practices in
3.3. Correlations of Father and Mother Child Rearing Practices predicting age 23 Narcissism is tested.
Because there is evidence that parenting styles of mothers and
In general, the correlations between Father’s and Mother’s Child fathers may have differential effects on the personality of children
Rearing Practices scores showed greatest similarity for the same (e.g., Baumrind, 1980; Gjerde, 1998; Vaughn, Block, & Block, 1988),
style of parenting (e.g., Father Authoritarian–Mother Authoritar- separate analyses are carried out for the four Mother Child Rearing
ian), varying from .41 to .48. The exception to this was Father Practices and for the four Father Child Rearing Practices measures.
Indulgent/Permissive Mother Authoritarian, r = .52; the correlation For Mother, this results in four separate regression analyses to pre-
between Mother Indulgent/Permissive and Father Indulgent/Per- dict Willful Narcissism and four analyses to predict Autonomous
missive = .46 (see Table 2). Narcissism at age 23. A similar plan is followed for the Father Child
Rearing Practices.
3.4. Correlations between Denial and Narcissism
3.7. Mother Child Rearing Practices, age 3/4 Narcissism Precursors and
Table 2 shows that the use of Denial at age 23 is positively cor- Denial predicting age 23 Willfulness Narcissism
related with the presence of Willful Narcissism, r = .29, p < .01. De-
nial is unrelated to healthy, Autonomous Narcissism (r = .01). A hierarchical multiple regression analysis was carried out, with
age 23 Willfulness as the dependent variable. At Step 1, Sex was
3.5. Correlations between Child Rearing Practices and Narcissism entered as a control variable. At Step 2, age 3/4 Narcissism Precur-
sors was entered. At Step 3, Denial was entered.2 At Step 4, Moth-
As may be seen in Table 2, the zero order correlations of Mother ers’ Parenting score was entered. At Step 5, the interaction between
and Father Child Rearing Practices with age 3/4 Narcissism Precur- age 3 Narcissism and Mothers’ Parenting score was entered. All vari-
sors are not significant. However, Child Rearing Practices were re- ables were centered prior to calculating interactions (Aiken & West,
lated to the presence of healthy Autonomous Narcissism at age 23. 1991).
For both parents, both Authoritative and Indulgent/Permissive It is predicted that age 3/4 Narcissism precursors, and age 23
styles of parenting positively predicted the presence of Autono- use of Denial, will be positively related to the presence of Willful-
mous Narcissism, whereas an Authoritarian style of parenting ness at age 23. It is further predicted that strong use of parenting
was a negative predictor of Autonomous Narcissism. Thus, consis- styles that are not responsive to the child’s needs – Authoritarian
tent with prediction, a more nurturant attitude on the part of the and Indifferent – when combined with an early proclivity towards
parents predicted healthy narcissism at age 23. On the other hand, narcissism, will increase the likelihood of the development of Will-
Indifferent Parenting on the part of the Father was a significant fulness by young adulthood. Also, it is predicted that weak use of
predictor of Willful Narcissism at age 23 (r = .22, p < .05). parenting styles that are responsive to the child’s needs – Author-
itative and Indulgent – when combined with an early proclivity to-
3.6. Age 23 Narcissism, as predicted by age 3/4 Narcissism Precursors, wards narcissism, will increase the likelihood of the development
Child Rearing Practices and Denial of Willful narcissism by young adulthood.
Table 3 shows that age 23 Willfulness was predicted by age 3/4
The previous zero order correlations show a variety of relations Narcissism Precursors (Beta = .40, p < .001), by Denial (Beta = .27,
among age 23 Narcissism, age 3/4 Narcissism Precursors, Child p < .01), and by the interaction between Mothers’ Authoritarian
Practices and age 3/4 Narcissism Precursors (Beta = .26, p < .01).
1
An analysis of simple slopes indicated that when Mother had a
In another study of community participants (Cramer & Jones, 2008), the
correlation between Willfulness and Autonomous narcissism was, r = .07. However,
2
Wink (1991) found positive relations between the two variables (rs = .24 and .28) in Preliminary hierarchical regression analyses indicated it made no difference if
samples that included college students. Denial was entered prior to Parenting Practices.
 P. Cramer / Journal of Research in Personality 45 (2011) 19–28 25

Table 3 Narcissism Precursors predicted age 23 Willfulness (Beta = .31,

Predicting Age 23 Narcissism from Narcissism Precursors Age 3/4, Mother (M) Child p < .004) (see Table 3). When Mother was high on Indulgent/Per-
Rearing Practices Report, Denial.
missive (1 SD above the mean) age 3/4 Narcissism Precursors did
Dependent variables not predict age 23 Willfulness, Beta = .07. When Mother was low
Willfulness Autonomous on Indulgent/Permissive (1 SD below the mean), age 3/4 Narcis-
Step 1
sism Precursors were a significant predictor of age 23 Willfulness,
Sex .13 .09 (Beta = .66, p < .001).
Step 2 A fourth hierarchical regression analysis indicated that Mothers’
Narcissism Precursors Age 3/4 .40*** .03 Indifferent score was unrelated to age 23 Willfulness, either as a
Step 3
main effect or in interaction (see Table 3).
Denial .27** .03
Step 4a
M Authoritarian .05 .33**
3.8. Mother Child Rearing Practices, age 3/4 Narcissism Precursors and
M Authoritative .06 .46***
M Indulgent/Permissive .08 .41*** Denial predicting age 23 Autonomous Narcissism
M indifferent .01 .04
Step 5a Four hierarchical regression analyses were used to determine
M Authoritarian  Narc Age 3/4 .26** .04 the predictors of Autonomous Narcissism. Age 3/4 Narcissism Pre-
M Authoritative  Narc Age 3/4 .36*** .01
M Indulgent/Permissive  Narc Age 3/4 .31** .06
cursors did not predict subsequent healthy Autonomous Narcis-
M Indifferent  Narc Age 3/4 .03 .08 sism. However, Mothers’ Child Rearing Practices were significant.

An Authoritarian style was a negative predictor of age 23 Autono-
p < .05.
**
p < .01.
mous Narcissism (Beta = .33, p < .005). Both Authoritative and
***
p < .001. Indulgent/Permissive Mother Child Rearing Practices were positive
a
Only one Mother Parenting variable was entered into each regression analysis. predictors of Autonomous Narcissism (Betas = .46 and .41,
Thus, 4 different regression analyses were conducted for each DV. ps < .001) (see Table 3). A fourth hierarchical regression analysis
indicated that Mothers’ Indifferent score was unrelated to age 23
high Authoritarian score (1 SD above the mean), age 3/4 Narcissism Autonomous Narcissism, either as a main effect or in interaction
Precursors predicted high Willfulness at age 23, Beta = .62, p < .001. (see Table 3).
When Mother was low Authoritarian (1 SD below the mean), age 3/
4 Narcissism Precursors were unrelated to Willfulness, Beta = .12 3.9. Father Child Rearing Practices, age 3/4 Narcissism Precursors, and
(see Fig. 1). Denial predicting age 23 Willful Narcissism
A second hierarchical regression analysis indicated that the inter-
action between Mothers’ Authoritative score and age 3/4 Narcissism A similar set of regression analyses were carried out to deter-
Precursors predicted age 23 Willfulness (Beta = .36, p < .001) (see mine the role of Father Child Rearing Practices in predicting age
Table 3). When Mother was high on Authoritative (1 SD above the 23 Narcissism. Separate analyses were conducted for each Father
mean), age 3/4 Narcissism Precursors did not predict age 23 Willful- parenting style.
ness, Beta = .07. When Mother was low on Authoritative (1 SD below Table 4 shows that age 3/4 Narcissism Precursors were a signif-
the mean), age 3/4 Narcissism Precursors were a significant predic- icant predictor of age 23 Willfulness (Beta = .38, p < .001). In addi-
tor of age 23 Willfulness, (Beta = .74, p < .001). tion, Willfulness was significantly predicted by Denial (Beta = .26,
A third hierarchical regression analysis indicated that the inter- p < .02). None of the Father Child Rearing Practices were
action between Mothers’ Indulgent/Permissive score and age 3/4 significant.

Interaction: Mother Authoritarian, Age 3/4 Narcissism Precursors

45

40

35
Age 23 Willful Narcissism

30

25

20

15
M High Authoritarian
M Low Authoritarian
10

0
0 0.5 1 1.5 2 2.5
Age 3/4 Narcissism Precursors (centered)

Fig. 1. Predicting Age 23 Willfulness Narcissism: interaction between Mother Authoritarianism (high = 1 SD above mean; low = 1 SD below mean) and Age 3/4 Narcissism
Precursors.
26 P. Cramer / Journal of Research in Personality 45 (2011) 19–28

Table 4 parenting based on the type of indiscriminate praise that is recom-

Predicting Age 23 Narcissism from Narcissism Precursors Age 3/4, Father (F) Child mended by the ‘‘self-esteem movement’’. The operational defini-
Rearing Practices Report, Denial.
tion of ‘‘Indulgence’’ in the present study is more similar to
Dependent variables ‘‘Permissive’’ and is characterized by providing emotional/physical
Willfulness Autonomous gratification rather than by providing indiscriminate praise.
Step 1
The findings also show that maladaptive narcissism in young
Sex .20 .03 adulthood was predicted by the presence of narcissistic precursors
Step 2 in preschool children, as had been demonstrated by Carlson and
Narcissism Precursors Age 3/4 .38*** .12 Gjerde (2009) using a different measure of young adult narcissism.
Step 3
However, the present study also demonstrates that this finding
Denial .26* .09
Step 4a should be interpreted in the context of mothers’ parenting styles,
F Authoritarian .08 .21! in relation to the developmental level of the child. Based on the re-
F Authoritative .06 .28* sults from the present study, and on those from previous research
F Indulgent/Permissive .10 .23!
(Barry et al., 2007; Baumrind, 1991a; Horton et al., 2006; Otway &
F Indifferent .09 .15
Step 5a
Vignoles, 2006; Ramsey et al., 1996) it appears that, in interaction
F Authoritarian  Narc Age 3/4 .01 .07 with the young child’s proclivity to narcissism, Responsiveness
F Authoritative  Narc Age 3/4 .04 .24! that is inappropriate to the child’s developmental level, being too
F Indulgent/Permissive  Narc Age 3/4 .01 .09 little in infancy/very young childhood (Authoritarian) or too great
F Indifferent  Narc Age 3/4 .09 .08
in adolescence (Indulgence), plus Demandingness that is inappro-
!
p < .10. priate to the child’s developmental level, being too great in in-
*
p < .05. fancy/very young childhood (Authoritarian) or too little in
**
p < .01.
*** adolescence (Indulgence), is related to maladaptive narcissism.
p < .001.
a
Only one father parenting variable was entered into each regression analysis. Demandingness and Responsiveness that are developmentally
Thus, four different regression analyses were conducted for each DV. appropriate (Authoritative) will predict adaptive, healthy narcis-
sism. The combination of low Demandingness and low Responsive-
3.10. Father Child Rearing Practices, age 3/4 Narcissism Precursors, ness (Indifference) in early childhood appears unrelated to age 23
and Denial predicting age 23 Autonomous Narcissism narcissism, but is related to the early presence of narcissism pre-
cursors, which then predict age 23 maladaptive narcissism. Paren-
There was one significant predictor of Autonomous Narcissism, tal indifference in adolescence has been found to be related to
and two predictors of borderline significance. Fathers’ Authorita- antisocial behavior.
tive Parenting score was a positive predictor of Autonomous Similar interactions between maternal parenting quality and
Narcissism (Beta = .28, p < .02). This result should be understood early difficult temperament have recently been reported by Pluess
in the context of the interaction between Father’s Authoritative and Belsky (2010). Infants with difficult temperament, as com-
score and age 3/4 Narcissism Precursors. When Father was high pared to those without difficult temperament, were more suscep-
on Authoritative (1 SD above the mean), age 3/4 Narcissism Precur- tible to the effects of negative early rearing when tested in late
sors predicted age 23 Autonomous Narcissism. (Beta = .36, p < .04). middle childhood. The present study shows that these effects can
When Father was low on Authoritative (1 SD below the mean), the extend to the period of young adulthood.
relation between age 3/4 Narcissism Precursors and age 23 Auton- There is weaker evidence for the relation between Father par-
omous Narcissism was insignificant, Beta = .10. enting and the development of Narcissism. The relations that were
Although not reaching statistical significance, the relation of found for Fathers replicated those found for Mothers – Authorita-
Father Indulgent/Permissive (positive, p < .06) and Father tive (significant) and Indulgent/Permissive (non-significant) styles
Authoritarian (negative, p < .09) with Autonomous Narcissism rep- positively predicted healthy, adaptive narcissism, with Authoritar-
licated the significant relations found for Mothers. Father Indiffer- ian being a (non-significant) negative predictor. However, Fathers’
ent was unrelated to Autonomous Narcissism, either as a main parenting was not related to maladaptive narcissism. Further, in
effect or in interaction, as was found for Mothers (see Table 4). contrast to the findings with Mother, Father’s style of parenting
did not interact with the child’s initial level of proclivity towards
4. Discussion narcissism.
In Carlson and Gjerde’s original study (2009), they suggested
The findings of the present research show that healthy, adaptive that the ‘‘defensive processes’’ associated with narcissism may
narcissism in young adulthood is predicted by early gratification of contribute to the continuity of narcissism from preschool to young
physical and psychological needs. The presence of healthy, Auton- adulthood (p. 575). The defense mechanism of denial has been
omous narcissism at age 23 was predicted by both mothers’ and shown to be predominant in the defense functioning of preschool
fathers’ use of Authoritative or Indulgent/Permissive parenting children (Cramer, 1991a, 2006). The present data indicate that that
styles, in which the parent is responsive to the child’s needs. In use of the immature defense of denial at age 23 is associated with
contrast, the use of an Authoritarian style, which is not responsive the presence of maladaptive narcissism. It is thus possible that de-
to the child’s needs, was a negative predictor of subsequent nial may play an important role for understanding the longitudinal
healthy narcissism, again for both mothers and fathers. continuity of maladaptive narcissism, by protecting the individual
Unexpectedly, Indifferent parenting, which is unresponsive to over time from recognizing the negative consequences of maladap-
the child’s needs, was not a negative predictor of adaptive narcis- tive narcissistic behavior. As expected, denial was not related to
sism, even though Indifferent parenting was significantly corre- healthy, adaptive narcissism.
lated with Authoritarian parenting. This suggests that it is the
combination of the demanding, controlling nature of the parent, 5. Limitations
without sufficient gratification, rather than the lack of gratification
alone, that interferes with the development of adaptive narcissism. The present research takes the point of view that innate tem-
It should be noted that the type of parenting characterized in perament (narcissism proclivity) will be influenced by type of
this study as ‘‘Indulgent/Permissive’’ should be differentiated from parenting, which determines if healthy or maladaptive narcissism
 P. Cramer / Journal of Research in Personality 45 (2011) 19–28 27

will develop. That is, it is assumed that it is the effect of parenting adaptive and maladaptive narcissism. Journal of Child and Family Studies, 16,
508–521.
on narcissism potential that is the causal direction for under-
Baumrind, D. (1971). Current patterns of parental authority. Developmental
standing subsequent narcissism. However, it is possible that the Psychology, 4(1, Pt. 2), 1–103.
preschool child’s narcissistic behaviors could influence parenting Baumrind, D. (1980). New directions in socialization research. American
style. This possibility seems less likely, insofar as there was no Psychologist, 35, 639–652.
Baumrind, D. (1991a). The influence of parenting style on adolescent competence
relation between age 3/4 narcissism precursors and age 3 parenting and substance use. Journal of Early Adolescence, 11, 56–95.
styles. Thus, it does not appear that the presence of narcissism Baumrind, D. (1991b). Parenting styles and adolescent development. In R. M. Lerner,
precursors at age 3/4 influenced the style of parenting adopted A. C. Peterson, & J. Brooks-Gunn (Eds.), Encyclopedia of adolescence
(pp. 758–772). New York: Garland.
by the parent. Berren, P. (1998). Narcissistic disorders in children. In P. Berren (Ed.), Narcissistic
Further, the study uses regression analyses that take into ac- disorders in children and adolescents (pp. 151–165). Northvale, NJ: Jason Aronson
count the characteristics of the child – i.e., narcissism precursors Inc..
Block, J. (1961/1978). The Q-sort method in personality assessment and psychiatric
– at the time that parenting was assessed. These analyses showed research. Palo Alto, CA: Consulting Psychologists Press.
that the longitudinal effect of parenting on young adult narcissism Block, J. (2008). The Q-sort in character appraisal: Encoding subjective impressions of
holds even after controlling for early child narcissism precursors, persons quantitatively. Washington, DC: American Psychological Association.
Block, J., & Block, J. H. (1980). The role of ego-control and ego-resiliency in the
providing indirect evidence that parenting has a subsequent effect organization of behavior. In W. A. Collins (Ed.). Minnesota symposia on child
on narcissism. Regression analyses do not rule out the possibility psychology (Vol. 13, pp. 39–101). Hillsdale, NJ: Erlbaum.
that different children may elicit different responses from parents, Block, J. H., Block, J., & Morrison, A. (1981). Parental agreement–disagreement on
child-rearing orientation and gender-related personality correlates in children.
but they do provide evidence that the relation between young
Child Development, 52, 965–974.
adult narcissism and parenting is not due solely to the effect of Bowlby, J. (1969). Attachment and loss. Vol. 1. Attachment. New York: Basic Books.
children on parenting behavior (Collins, Maccoby, Steinberg, Campbell, W. K., & Foster, C. A. (2002). Narcissism and commitment in romantic
Hetherington, & Bornstein, 2000). relationships: An investment model analysis. Personality and Social Psychology
Bulletin, 28, 484–494.
It is possible that the correlations between child-rearing styles Carlson, K. S., & Gjerde, P. F. (2009). Preschool personality antecedents of narcissism
and subsequent personality characteristics (narcissism) are due to in adolescence and young adulthood: A 20-year longitudinal study. Journal of
the effects of unmeasured third variables, such as genetically trans- Research in Personality, 43, 570–578.
Collins, W. A., Maccoby, E. E., Steinberg, L., Hetherington, E. M., & Bornstein, M. H.
mitted behavioral predispositions or sociocontextual factors that (2000). Contemporary research in parenting. American Psychologist, 55,
influence both child-rearing styles and children’s later characteris- 219–232.
tics. With regard to the present findings for young adult Willful Cramer, P. (1991a). The development of defense mechanisms: Theory, research and
assessment. New York: Spring-Verlag.
narcissism, an explanation based on intervening sociocontextual Cramer, P. (1991b). Anger and the use of defense mechanisms in college students.
factors is reduced by the finding that it is child-rearing styles at Journal of Personality, 59, 39–55.
age 3, in interaction with narcissism precursors at age 3/4, that Cramer, P. (1995). Identity, narcissism, and defense mechanisms in late
adolescence. Journal of Research in Personality, 29, 341–361.
predicts young adult Willful narcissism. That is, the child-rearing Cramer, P. (1998). Threat to gender representation: Identity and identification.
styles that are related to Willful narcissism function in tandem with Journal of Personality, 66, 233–247.
early childhood narcissism precursors, and continue to influence Cramer, P. (1999). Personality, personality disorders, and defense mechanisms.
Journal of Personality, 67, 535–554.
narcissism at young adulthood, regardless of other intervening
Cramer, P. (2002). Defense mechanisms, behavior, and affect in young adulthood.
factors over a 20 year period. Journal of Personality, 70, 103–126.
An important feature of the present investigation is that it uses Cramer, P. (2003). Personality change in later adulthood is predicted by defense
multiple assessment measures – observer ipsitive ratings and par- mechanisms use in early adulthood. Journal of Research in Personality, 37,
76–104.
ents’ ipsitive self-reports at age 3, plus observer ipsitive ratings and Cramer, P. (2004). Identity change in adulthood: The contribution of defense
coders’ scoring of participants’ narrative material at age 23. The age mechanisms and life experiences. Journal of Research in Personality, 38, 280–316.
23 data were provided independently of the data collected 20 years Cramer, P. (2006). Protecting the self: Defense mechanisms in action. New York:
Guilford Press.
previously. Thus, problems of measure overlap have been avoided, Cramer, P., & Gaul, R. (1988). The effect of success and failure on children’s use of
and ratings for each type of variable were made concurrently, not defense mechanisms. Journal of Personality, 56, 729–742.
retrospectively. However, like all self-report measures, it is possi- Cramer, P., & Jones, C. (2008). Narcissism, identification, and longitudinal change in
psychological health: Dynamic predictions. Journal of Research in Personality, 42,
ble that parents were not accurate in their reporting. Also, it may 1148–1159.
be important that the data collection for the present study began Dekovic, M., Janssens, J. M. A. M., & Gerris, J. R. M. (1991). Factor structure and
in 1968 and spans the time range to 1988. It is possible that results construct validity of the block child rearing practices report (CRPR).
Psychological Assessment: A Journal of Consulting and Clinical Psychology, 3,
with a more recent cohort would differ. 182–187.
Nevertheless, the results are generally consistent with earlier George, C., & Solomon, J. (1989). Internal working models of caregiving and security
studies that relied on self-report of narcissism and retrospective of attachment at age six. Infant Mental Health Journal, 10, 222–237.
Gjerde, P. F. (1998). Parental concordance on child rearing and the interactive
accounts of parenting, showing that parenting styles are important
emphases of parents: Sex-differentiated relationships during the preschool
determiners of the development of subsequent narcissism. Parent- years. Developmental Psychology, 24, 700–706.
ing was found to have a direct effect on the development of healthy Hibbard, S., Farmer, L., Wells, C., Difillipo, E., Barry, W., Korman, R., et al. (1994).
narcissism, and to interact with the preschool child’s initial pro- Validation of Cramer’s defense mechanism manual for the TAT. Journal of
Personality Assessment, 63, 197–210.
clivity towards narcissism in the development of maladaptive nar- Horton, R. S., Bleau, G., & Drwecki, B. (2006). Parenting narcissus: What are the links
cissism. Further, the present investigation expands earlier findings between parenting and narcisism? Journal of Personality, 74, 345–376.
and provides separate information regarding the effect of mother Kernberg, O. (1975). Borderline conditions and pathological narcissism. New York:
Aronson.
and father parenting. Importantly, it addresses the question of Kernberg, P. F. (1998). Developmental aspects of normal and pathological
the development of healthy narcissism, in addition to maladaptive narcissism. In E. F. Ronningstam (Ed.), Disorders of narcissism: Diagnostic,
narcissism. clinical and empirical implications (pp. 103–120). Washington, DC: American
Psychiatric Association.
Kochanska, G. (1990). Maternal beliefs as long-term predictors of mother–child
interaction and report. Child Development, 61, 1934–1943.
References Kochanska, G., Kuczynski, L., & Radke-Yarrow, M. (1989). Correspondence between
mothers’ self-reported and observed child-rearing practices. Child Development,
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting 60, 56–63.
interactions. Newbury Park: Sage Publications. Leahy, R. L. (1981). Parental practices and the development of moral judgment and
Barry, C. T., Frick, P. J., Adler, K. K., & Grafemen, S. J. (2007). The predictive utility of self-image disparity during adolescence. Developmental Psychology, 17,
narcissism among children and adolescents: Evidence of a distinction between 580–594.
28 P. Cramer / Journal of Research in Personality 45 (2011) 19–28

Levy-Warren, M. H. (1998). Adolescent narcissism. In P. Berren (Ed.), Narcissistic Porcerelli, J. H., Thomas, S., Hibbard, S., & Cogan, R. (1998). Defense mechanisms
disorders in children and adolescents (pp. 85–110). Northvale, NJ: Jason Aronson development in children, adolescents, and late adolescents. Journal of
Inc.. Personality Assessment, 71, 411–420.
Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and Ramsey, A., Watson, P. J., Biderman, M. D., & Reeves, A. L. (1996). Self-reported
adulthood: A move to the level of representation. In I. Bretherton & E. Waters narcissism and perceived parental permissiveness and authoritarianism. The
(Eds.), Growing points of attachment theory and research. Monographs of the Journal of Genetic Psychology, 157, 227–234.
Society for Research in Child Development, 50 (1–2), pp. 66–104 (Serial No. Raskin, R. N., Novacek, J., & Hogan, R. (1991). Narcissism, self-esteem, and defensive
209). self-enhancement. Journal of Personality, 59, 16–38.
McNally, S., Eisenberg, N., & Harris, J. D. (1991). Consistency and change in maternal Raskin, R., & Terry, H. (1988). A principal-components analysis of the narcissistic
child-rearing practices and values: A longitudinal study. Child Development, 62, personality inventory and further evidence of its construct validity: A review
190–198. and new findings. Journal of Personality and Social Psychology, 54, 890–902.
Miller, A. (1981). Prisoners of childhood. New York: Basic Books. Roberts, G. C., Block, J. H., & Block, J. (1984). Continuity and change in parents’ child-
Millon, T. (1990). The disorders of personality. In L. A. Pervin (Ed.), Handbook of rearing practices. Child Development, 55, 586–597.
personality. New York: Guilford. pp 339–370. Russ, E., Shedler, J., Bradley, R., & Westen, D. (2008). Refining the construct of
Morf, C., & Rhodewalt, F. (1993). Narcissism and self-evaluation maintenance: narcissistic personality disorder: Diagnostic criteria and subtypes. American
Explorations in object relations. Personality and Social Psychology Bulletin, 19, Journal of Psychiatry, 165, 1473–1481.
668–676. Sandstrom, M., & Cramer, P. (2003). Girls’ use of defense mechanisms following peer
Morf, C. C., & Rhodewalt, F. (2001). Unraveling the paradoxes of narcissism: A rejection. Journal of Personality, 71, 605–627.
dynamic self-regulatory processing model. Psychological Inquiry, 12, 177–196. Sroufe, L. A. (1979). The coherence of individual development: Early care,
Morrison, A. P. (1986). Essential papers on narcissism. New York: New York attachment and subsequent developmental issues. American Psychologist, 34,
University Press. 834–841.
Mussen, P., Harris, S., Rutherford, E., & Keasey, C. B. (1970). Honesty and altruism Vaillant, G. E. (1994). Ego mechanisms of defense and personality psychopathology.
among preadolescents. Developmental Psychology, 3, 169–194. Journal of Abnormal Psychology, 103, 44–50.
O’Brien, M. L. (1987). Examining the dimensionality of pathological narcissism: Vaughn, B. E., Block, J. H., & Block, J. (1988). Parental agreement on child rearing
Factor analysis and construct validity of the O’Brien Multiphasic Narcissism during early childhood and the psychological characteristics of adolescents.
Inventory. Psychological Reports, 62, 879–882. Child Development, 59, 1020–1033.
O’Brien, M. L. (1988). Further evidence of the validity of the O’Brien Multiphasic Weise, K. L., & Tuber, S. (2004). The self and object representations of narcissistically
Narcissism Inventory. Psychological Reports, 62, 878–882. disturbed children. Psychoanalytic Psychology, 21, 224–258.
Otway, L. J., & Vignoles, V. L. (2006). Narcissism and childhood recollections: A Wink, P. (1991). Two faces of narcissism. Journal of Personality and Social Psychology,
quantitative test of psychoanalytic predictions. Personality and Social Psychology 61, 590–597.
Bulletin, 32, 104–116. Wink, P. (1992). Three narcissism scales for the California Q-set. Journal of
Ozer, D. J. (1993). The Q-sort method and the study of personality development. In Personality Assessment, 58, 51–66.
D. C. Funder, R. D. Parke, C. Tomlinson-Keasey, & K. Widaman (Eds.), Study of Winnicott, D. W. (1965). Ego distortion in terms of true and false self. In The
lives through time (pp. 147–169). Washington, DC: American Psychological maturational processes and the facilitating environment: Studies in the theory of
Association. emotional development (pp. 140–152). New York: International Universities
Pincus, A. L., & Lukowitsky, M. R. (2010). Pathological narcissism and narcissistic Press.
personality disorder. Annual Review of Clinical Psychology, 6, 421–446. Ziegler-Hill, V., Clark, C. B., & Pickard, J. D. (2008). Narcissistic subtypes and
Pluess, M., & Belsky, J. (2010). Differential susceptibility to parenting and quality contingent self-esteem: Do all narcissists base their self-esteem on the same
child care. Developmental Psychology, 46, 379–390. domains? Journal of Personality, 76, 753–774.