Content uploaded by William Bernet
Author content
All content in this area was uploaded by William Bernet on Mar 15, 2020
Content may be subject to copyright.
PAPER
PSYCHIATRY & BEHAVIORAL SCIENCE
William Bernet,
1
M.D.; Nilgun Gregory,
2‡
Ph.D.; Ronald P. Rohner,
3
Ph.D.; and Kathleen M. Reay,
4
Ph.D.
Measuring the Difference Between Parental
Alienation and Parental Estrangement: The
PARQ-Gap*
,†
ABSTRACT: Parental alienation (rejection of a parent without legitimate justification) and realistic estrangement (rejection of a parent for a
good reason) are generally accepted concepts among mental health and legal professionals. Alienated children, who were not abused, tend to
engage in splitting and lack ambivalence with respect to their parents; estranged children, who were maltreated, usually perceive their parents
in an ambivalent manner. The hypothesis of this study was that a psychological test—the Parental Acceptance–Rejection Questionnaire
(PARQ)—will help to distinguish severely alienated from nonalienated children. The PARQ, which was used to identify and quantify the
degree of splitting for each participant, was administered to 45 severely alienated children and 71 nonalienated children. The PARQ-Gap score
—the difference between each child’s PARQ: Father score and PARQ: Mother score—was introduced and defined in this research. Using a
PARQ-Gap score of 90 as a cut point, this test was 99% accurate in distinguishing severely alienated from nonalienated children. This research
presents a way to distinguish parental alienation from other reasons for contact refusal. The PARQ-Gap may be useful for both clinicians and
forensic practitioners in evaluating children of separating and divorced parents when there is a concern about the possible diagnosis of parental
alienation.
KEYWORDS: forensic child psychiatry, splitting, alienation, estrangement, Parental Acceptance–Rejection Questionnaire, PARQ-Gap
Contact refusal is a behavior that is sometimes manifested by
children of separating and divorced parents. When contact refu-
sal occurs, the child refuses to have a relationship with the
rejected parent. The child may strongly dislike and fear the
rejected parent. Contact refusal may be transitory and self-lim-
ited. In that circumstance, the family moves on and the problem
is not brought to the attention of a mental health professional.
However, if contact refusal is persistent, it indicates significant
problems within the family, which are apparent especially in the
relationship between the child and the rejected parent. Persistent
contact refusal is a serious problem in the family that should be
carefully assessed in order to determine the underlying cause.
The two most important reasons for contact refusal are
estrangement and alienation. Estrangement refers to a child’s
rejection of a parent for good cause, for example, because that
parent had a history of neglecting or abusing the child. On the
other hand, parental alienation (PA) refers to a child’s rejection
of a parent without a good reason. In PA, the child’s parents are
typically engaged in a high-conflict separation or divorce. The
child escapes the battle zone between the parents by gravitating
to one parent and persistently rejecting a relationship with the
second parent—even though the child and the rejected parent
previously enjoyed a healthy, mutually satisfying relationship.
When PA occurs, of course, the preferred parent has encouraged,
indoctrinated, and brainwashed the child to reject the alienated
parent.
It is important for clinicians to distinguish estrangement from
alienation because the treatment of the child and the family is
quite different, depending on whether the child’s rejection of the
parent is justified or driven by a false belief that the rejected par-
ent is unsafe to be with, unloving, or psychologically unavail-
able. Also, it is important for forensic practitioners to distinguish
estrangement from alienation because it bears on recommenda-
tions regarding the child’s parenting time arrangements.
After the parental alienation syndrome (PAS) was defined by
Gardner (1), clinicians and researchers proposed factors to
1
Department of Psychiatry and Behavioral Sciences, Vanderbilt University
School of Medicine, 1313 21st Avenue South, 209 Oxford House, Nashville,
TN 37232.
2
Vanderbilt Kennedy Center, Vanderbilt University, 110 Magnolia Circle,
Nashville, TN 37203.
3
Department of Human Development and Family Studies, Center for the
Study of Interpersonal Acceptance and Rejection, University of Connecticut,
Unit 1058, Storrs, CT 06269-1058.
4
International Institute for Parental Alienation Studies, Kelowna, British
Columbia, Canada.
‡
Present Address: Final International University, Toroslar Caddesi 6,
Girne, Cyprus.
Corresponding author: William Bernet, M.D. E-mail: william.bernet@
vumc.org
*Research was done at Vanderbilt University, Nashville, Tennessee. The
authors are members of the Parental Alienation Study Group.
†
Disclosures: Dr. Bernet receives royalties from Charles C Thomas Pub-
lisher. Dr. Gregory received support from the Turkish Scientific and Techno-
logical Research Council (TUBITAK). Dr. Rohner is a partner in Rohner
Research Publications, which publishes the Parental Acceptance–Rejection
Questionnaire. Dr. Reay was the founder and clinical director of the Family
Reflections Reunification Program.
Received 14 Oct. 2019; and in revised form 21 Jan. 2020; accepted 22
Jan. 2020.
1©2020 American Academy of Forensic Sciences
J Forensic Sci,2020
doi: 10.1111/1556-4029.14300
Available online at: onlinelibrary.wiley.com
consider and protocols for distinguishing estrangement from
alienation. These approaches, summarized below, focused on
features of alienated children, features of estranged children, the
parents’behavior, family history, record review, and psychologi-
cal testing. Of course, these factors and protocols are simply
components of a comprehensive child custody evaluation, which
consists of three main legs: forensic interviews, testing, and col-
lateral sources. Information about the children, parents, and fam-
ily history is derived in part from forensic interviews.
In a recent review of “Empirical Studies of Alienation,”Saini,
Johnston, Fidler, and Bala stated, “There is a virtual absence of
empirical studies on the differential diagnosis of alienation in
children from other conditions that share similar features with
parental alienation, especially realistic estrangement...”(2).
Although Saini et al. acknowledged that PA exists and that there
is ample empirical evidence to identify the types of behaviors
that alienating parents manifest, they highlighted the need for
tools to differentiate alienation from estrangement. The research
presented here explains an objective method for distinguishing
the mental state of severely alienated children from children who
are not alienated. This psychological test—the Parental Accep-
tance–Rejection Questionnaire (PARQ)—does not all by itself
differentiate alienation from estrangement; the PARQ identifies a
high level of splitting, which is one of the factors to consider in
distinguishing alienation from estrangement.
Features of Alienated Children
When Gardner introduced the concept of PAS in 1985, he
proposed that eight behavioral symptoms, taken together, identi-
fied PAS and no other psychological condition (1). These symp-
toms included: a campaign of denigration against the target
parent; frivolous rationalizations for the child’s criticism of the
target parent; lack of ambivalence; the “independent-thinker”
phenomenon; reflexive support of the alienating parent against
the target parent; absence of guilt over exploitation and mistreat-
ment of the target parent; borrowed scenarios; and spread of the
child’s animosity toward the target parent’s extended family.
According to Gardner, a child who manifested most or all of the
eight symptoms was likely to be experiencing PAS. Currently,
most writers—including the authors of this article—use the
phrase “parental alienation”rather than “parental alienation syn-
drome”because the latter phrase has become controversial for
some practitioners. In this context, we consider “parental alien-
ation”and “parental alienation syndrome”to be synonymous.
However, PA-detractors repeatedly criticize the use of “syn-
drome”—in writing and in court testimony—and it is not worth
the time and energy required to defend the use of that word.
Parental Behavior
After introducing the concept of PAS, Gardner addressed the
task of differentiating PAS from bona fide abuse or neglect. He
observed that competing parents made complementary accusations
about each other: “[One] parent has accused the other parent of
inducing PAS in the children. In response, the responding parent
accuses the other parent of abusing and neglecting the children”
(3). In addition to noting the symptoms presented by the child,
Gardner thought that the behaviors seen in the parents of alienated
children differed from the behaviors seen in the parents of abused
or neglected children. For example, he noticed that perpetrators of
abuse (alienating parents) are less likely than victims of the abuse
(alienated parents) to cooperate with a neutral examiner. Although
Gardner’s early comments on distinguishing alienation from bona
fide abuse were interesting and useful, they are now regarded by
many practitioners as being too simplistic.
Clarification of Terminology
While Gardner consistently distinguished PAS from “bona
fide abuse-neglect,”he did not use the term estrangement. The
distinction between alienation and estrangement was established
by Kelly and Johnston, who said, “Children who are realistically
estranged from one of their parents as a consequence of that par-
ent’s history of family violence, abuse, or neglect need to be
clearly distinguished from alienated children”(4). Although they
did not provide a method for accomplishing that task, Kelly and
Johnston did note, “Unlike alienated children, the estranged chil-
dren do not harbor unreasonable anger and/or fear.”
Differential Diagnosis
Several authors have presented sets of differential diagnoses
that address the task of differentiating estrangement from alien-
ation. All of these formulations were based on qualitative
research, such as clinical experience, and not validated through
quantitative research. Kelly and Johnston (4), for example, iden-
tified multiple reasons for children to resist visitation, including:
resistance rooted in normal developmental processes; resistance
related to high-conflict marriage and divorce; resistance due to a
parent’s parenting style; resistance related to alienating behaviors
by the parent with whom the child is aligned; resistance arising
from the child’s concern about an emotionally fragile custodial
parent; and, resistance prompted by a parent’s remarriage. Kelly
and Johnston also explained that children’s relationships to each
parent after separation and divorce can be conceptualized along
a continuum of positive to negative, with the following possibili-
ties: positive relationships with both parents, affinity with one
parent, allied children, estranged children, and alienated children.
Later, Bernet and Freeman (5) and Freeman (6) published a
differential diagnosis that should be considered in both clinical
and forensic evaluations when the presenting problem is contact
refusal. The differential included: the child’s normal preference;
a loyalty conflict; the child’s attempt at avoiding the conflict
between the parents; the worried child, for example, a child with
separation anxiety; the stubborn child, for example, a child with
oppositional defiant disorder; the abused child; delusional disor-
der shared by parent and child; and PA. These authors empha-
sized that selection of the most appropriate treatment depends on
correctly identifying the underlying cause of the contact refusal.
Evaluation Protocols
Likewise, the following protocols were also based on qualitative
rather than quantitative research. Lee and Olesen presented a
method for distinguishing alienated children from realistically
estranged children. The first step is to assess whether the child looks
alienated. That is, “We can look at the child’s rigidity; lack of
ambivalence; apparently trivial reasons given for the parental rejec-
tion; and the intense, unrelenting, negative portrayal of the parent”
(7). Here, Lee and Olesen were referring to the same clinical indica-
tors of PA that had been described by Gardner and restated by Kelly
and Johnston. The second step in Lee and Olesen’s model for differ-
entiating alienation from realistic estrangement is “the assessment
of both parents, the investigation of abuse allegations, and the
assessment of the child’s relationship to each parent.”
2JOURNAL OF FORENSIC SCIENCES
Three years later, Drozd and Olesen published an elaborate
decision tree for distinguishing abuse, alienation, and estrange-
ment (8). They used the same definitions of alienation and
estrangement that Kelly and Johnston (4) proposed. According
to Drozd and Olesen, there are multiple allegations and counter-
allegations in many litigating families, such that one parent may
allege domestic violence, and the second parent may allege
alienation against the parent who raised the issue of domestic
violence. Those authors introduced a model for dealing with
competing cross-allegations. They emphasized that if a child
does not have basically positive relationships with both parents,
the following hypotheses should be considered: normal develop-
mental processes (such as “affinity”and “alignment”); poor par-
enting (e.g., alienating behaviors); and abuse (e.g., child abuse,
substance abuse, and domestic violence). Drozd and Olesen
explained the features of the parents, such as the various types
of alienating behaviors and the typical patterns of domestic vio-
lence, but they did not address in depth how to distinguish an
alienated child from an estranged child.
Ellis described a three-step process for diagnosing PAS and
for distinguishing it from realistic estrangement (9). The three
steps were as follows: (i) determining if the refusal of contact
with the parent is extreme and the alienation is severe, (ii) deter-
mining if there is a basis for extreme fear and anger toward the
parent, and (iii) determining if the child meets at least 10 of the
15 criteria that she proposed. Ellis focused on the detailed
assessment of the child, consistent with Gardner’s eight criteria
for PAS, although she also discussed typical behaviors and per-
sonality traits of the alienating parent. Further, she identified 15
specific criteria—all of them behaviors and attitudes of the child
—for establishing the diagnosis of PAS. She recognized that
these criteria are tentative, but offered them as a first step toward
standardizing the evaluation of PAS.
Ellis’s second criterion was the child’s use of “the mechanism
of splitting to reduce ambiguity.”That is, the child views the
alienating parent as all good and denies that this parent has any
negative traits. Likewise, the child views the targeted parent as
all bad, and denies that this parent has any positive traits. Ellis
explained that splitting can be identified by interviewing the
child and asking questions which would elicit ambiguity. The
alienated child’s responses avoid ambiguity or ambivalence and
reflect a black-and-white perception of the parents.
Family History
The history of a child’s relationship with his or her parents is
an important consideration in distinguishing alienation from
estrangement. If, for example, the child had a long history of
conflict with the rejected parent, estrangement may be the under-
lying reason for a current pattern of contact refusal. However, if
the child previously had a healthy, loving relationship with the
rejected parent, alienation is more likely the reason for contact
refusal. This feature of PA was noted by Warshak when he pro-
posed a definition of pathological alienation: “A disturbance in
which children, usually in the context of sharing a parent’s nega-
tive attitudes, suffer unreasonable aversion to a person or per-
sons with whom they formerly enjoyed normal relations or with
whom they would normally develop affectionate relations”(10).
Gottlieb expressed the same opinion when she wrote: “Bonding
with the alienated parent prior to the alienation ... is the same
with all degrees of the PAS; the bonding was strong, healthy
and minimally problematic, if at all”(11). In addition, Clawar
and Rivlin emphasized, “The single most important element in
uncovering the content, intensity, and impact of programming-
and-brainwashing in children is researching the social history of
the children”(12).
Collateral Records
Johnston, Lee, Olesen, and Walters argued that collateral
records can be used to substantiate allegations of abuse by a par-
ent, such as child protective service reports, self-admissions, eye-
witness reports, expert testimony, medical records, police
reports, arrests, plea-bargains, and criminal convictions (13).
Those authors thought that such records often reliably differenti-
ate abuse from alienation. Of course, practitioners should look
for records of child maltreatment as well as evidence that the
child witnessed a hostile and abusive environment. If the child
has witnessed domestic violence, estrangement may be more
likely than alienation. However, children do not necessarily pre-
fer the less toxic parent. In some scenarios, the child may side
with the abusive parent and become another abuser of the
rejected parent. An abusive parent may persuade the child to
become alienated from the protective parent (14).
Features of Estranged Children
Many estranged children—in contrast to alienated children—
hope for a relationship with their abusive parents. In a commen-
tary on realistic estrangement, Fidler and Bala stated that
estrangement “may result from the trauma of witnessing domes-
tic violence or from experiencing physical abuse, sexual abuse,
or significantly inept or neglectful parenting by the rejected par-
ent”(15). They continued, “In these cases, children may exhibit
symptoms of post-traumatic stress disorder rather than a dispro-
portionate or unjustified reaction to their actual experience with
the rejected parent.”Fidler and Bala also commented that “even
abused children are likely to want to maintain a relationship with
their abusive parents.”A review of the literature found that
youth in foster care due to abuse or neglect typically yearn for
their abusive parent, while they also express feelings of gratitude
at being removed from the abusive home (16). Also, abused
children tend to engage in attachment-enhancing behaviors rather
than attachment-disrupting behaviors toward their abusive
parents (17).
Psychological Testing
Psychological testing is an accepted method in child custody
evaluations. The Minnesota Multiphasic Personality Inventory–2
(MMPI-2) is well-established and most often used in these eval-
uations. Siegel and Langford found that alienating mothers were
more likely to complete MMPI-2 questions in a defensive man-
ner, striving to appear as flawless as possible (18). Gordon, Stof-
fey, and Bottinelli found that parents who induced alienation in
their children manifested higher scores (in the clinical range) on
the MMPI-2 than control mothers and fathers (scores in the nor-
mal range), indicating primitive defenses such as splitting and
projective identification. The scores of target parents were
mostly similar to the scores of control parents (19).
There are also psychological tests of children that may help
distinguish alienation from estrangement. The Bricklin Percep-
tual Scales (BPS), which were developed specifically for use in
child custody evaluations, define and quantify children’s attach-
ment to and perceptions of their parents (20). Estranged children
are likely on the BPS to manifest ambivalence toward both
BERNET ET AL. .ALIENATION AND ESTRANGEMENTS 3
parents. Alienated children, on the other hand, are likely to see
the preferred parent as totally good and the rejected parent as
totally bad. The BPS consists of 64 questions, which pertain to
the child’s perception of the mother (32 questions) and the
child’s perception of the father (32 questions). Although Bricklin
did not use the term “splitting”in his discussion of the BPS, that
appears to be what he was measuring. Bricklin later said that
alienated children had a mind-made-up (MMU) configuration,
which occurred as part of a not-based-on-actual-interaction
(NBOAI) scenario (21). Bricklin found that MMU children rated
the preferred parent extremely or abnormally high (i.e., favor-
ably) and the rejected parent extremely or abnormally low (i.e.,
unfavorably) on the BPS. Although Otto et al. (22) criticized
some aspects of the BPS, this test appears to identify the psy-
chological mechanism of splitting.
Baker, Burkhard, and Albertson-Kelly introduced the Baker
Alienation Questionnaire (BAQ), which is intended to identify
alienated children using a paper-and-pencil measure that is short,
easy to administer, and easy to score objectively (23). The 28
items of the BAQ are designed to capture a child’s extreme
rejection of one parent and extreme idealization of the other.
The BAQ is administered to children who either select a
response from the choices provided (i.e., Not At All, A Little
Bit, Somewhat, Much, Very Much) or write an answer to an
open-ended question (e.g., What are some things you don’t like
about your mother?). In their pilot study, Baker et al. found that
children who had been court ordered for reunification therapy—
specifically for PA—consistently responded in a polarized fash-
ion in which one parent was denigrated and the other was ideal-
ized. Baker et al. found that the BAQ discriminated well
between alienated and nonalienated children.
More recently, Rowlands introduced the Rowlands' Parental
Alienation Scale (RPAS), which was a questionnaire for parents
designed to capture the manifestations of PA that had been
described in the literature (24). Six significant factors were
extracted from the original 42 items representing the eight tradi-
tional behavioral symptoms of PA. It is noteworthy that lack of
ambivalence or splitting was not one of the six factors in the
final version of the RPAS. It is easy to understand why that
occurred: The RPAS was administered to parents and was based
on the child’s behaviors that the parents personally observed.
However, splitting is a mental state of the child, which may not
be manifested in their visible behaviors. Although splitting is
readily elicited in a psychiatric or psychological interview, it
might not be recognized by outside observers such as parents or
teachers.
The Parental Acceptance–Rejection Questionnaire (PARQ),
featured in this article, is a 60-item questionnaire that children
complete regarding their perceptions of their mothers’and
fathers’accepting-rejecting behaviors (25) It was derived from
interpersonal acceptance–rejection theory (IPARTheory), an evi-
dence-based theory of socialization and lifespan development
that attempts to predict and explain major effects, causes, and
other correlates of parental acceptance and rejection worldwide
(26). The PARQ: Father refers to the child’s perceptions of the
father’s love-related behaviors; the PARQ: Mother refers to the
child’s perceptions of the mother’s love-related behaviors. Fur-
ther details about the PARQ are provided in the Methods sec-
tion. Here, though, it is important to note that there are
significant differences between the PARQ used in the current
research and the BPS and BAQ. The BPS and BAQ were devel-
oped specifically for use in child custody evaluations or related
forensic tasks. The PARQ, on the other hand, was developed as
part of IPARTheory and has been used hundreds of times to
assess children’s perceptions of their parents in a wide range of
clinical and research settings worldwide. Moreover, the BPS and
BAQ both have subjective components. The BPS requires
face-to-face interaction in which the evaluator reads the ques-
tions to the child; the BAQ requires the evaluator to rate the
hand-written prose responses of the child. The PARQ, in con-
trast, is administered with paper and pencil or on a computer,
and scoring is totally objective.
Significance of Splitting
As noted by Bricklin (21), Ellis (9), Gardner (1), Kelly and
Johnston (4), and Lee and Olesen (7), one of the most notable
features of alienated children is their lack of ambivalence toward
their parents, especially the rejected parent. It is normal for chil-
dren to perceive their parents in an ambivalent manner, recogniz-
ing their strong points as well as their weak points. It is not
normal for a child to perceive a parent in an all-or-none fashion,
totally good, or totally bad. Children who experience severe PA
almost always lack ambivalence toward the rejected parent, and
they usually manifest splitting. That is, they idealize the alienat-
ing parent and devalue the target parent.
Recently, Jaffe, Thakkar, and Piron related to how an alien-
ated child’s denial of ambivalence was expressed in an elaborate
case derived from “a sample of forensic child custody interviews
court ordered and conducted by the authors”(27). For example,
when asked whether she could say something positive about her
mother, the child said, “She’s not ugly. ... She doesn’t have
empathy, she can walk into a room full of people crying and not
feel anything. ... She knows how to use retail therapy.”The
authors summarized, “The expressed lack of ambivalence as
manifested by the alienated child serves as an observable defin-
ing characteristic of the presence of parental alienation.”The
extensive qualitative research regarding PA by Jaffe et al. com-
plements quantitative research regarding PA in this article.
The present study considers splitting to be a maladaptive men-
tal mechanism by which children protect themselves from the
uncomfortable feelings of cognitive dissonance. When parents
continually fight, children often find it difficult to maintain
affection for both parents at the same time. The children often
resolve the dissonance by the mechanism of splitting, that is,
gravitating to an enmeshed relationship with one parent and
strongly rejecting the other parent. In PA, the child’s rejection of
the target parent is far out of proportion to anything that parent
has done to justify the rejection. PA is maladaptive because the
child’s behavior is driven by a false belief that the rejected par-
ent is evil, dangerous, or not worthy of love. If a parent was
truly abusive or severely neglectful, the child’s rejection of that
parent would constitute realistic estrangement, not PA. While
alienated children perceive the rejected parent as evil, most
estranged children still perceive the abusive parent in an ambiva-
lent manner.
Measuring Splitting
The design of the current research was first described by Ber-
net (28). Later, Bernet, Gregory, Reay, and Rohner showed that
the PARQ clearly demonstrates the defensive splitting that often
occurs among alienated children (29). Bernet et al. administered
the PARQ: Father and PARQ: Mother to 116 children and ado-
lescents from the following family types: children from intact
families; children from divorced families (who continued to see
4JOURNAL OF FORENSIC SCIENCES
both parents on a regular basis); neglected children (who lived
with their mothers and rarely or never saw their fathers); chil-
dren who were alienated from their fathers; and children who
were alienated from their mothers. The authors found that (i)
neglected children manifested ambivalence toward the less pre-
ferred parent; (ii) alienated children tended to manifest splitting,
that is, the mean PARQ score for the preferred parent was
exceptionally low (perceived acceptance) and the mean PARQ
score for the alienated or target parent was exceptionally high
(perceived rejection); and (iii) the pattern of PARQ scores for
the neglected children differed significantly from the pattern of
PARQ scores for the alienated children.
Conclusions reached by Bernet et al. (29) were echoed by
Blagg and Godrey, using a different instrument, the Bene–
Anthony Family Relations Test (BAFRT), and a different popu-
lation, that is, children in the United Kingdom (30). The BAFRT
(developed by Eva Bene and E. James Anthony in the 1950s) is
a projective test that explores indirectly children’s perceptions of
their relationship with family members (31). In the BAFRT, chil-
dren are asked to place “postcards”with various positive and
negative messages into a collection of “mailboxes”representing
their family members. Blagg and Godfrey administered the
BAFRT to 17 neglected/emotionally abused children and 16
alienated children. They concluded that “children in the alienated
group who had not been abused or neglected by their target par-
ent expressed almost exclusively negative (hostile) feelings
toward them, while also expressing almost exclusively positive
(affectionate) feelings toward their preferred parent”(30). In
contrast, the neglected/emotionally abused children did not reject
their neglectful/emotionally abusive parents, but showed signs of
“dealizing their parents.”We should note, however, that the
BAFRT has been used less often since the mid-1970s because of
its questionable psychometric properties (32).
In this article, we analyze in a different manner the same data
reported in Bernet et al. (29). Here, we introduce the concept of
the PARQ-Gap, which is the absolute difference in participants’
PARQ: Father scores from their PARQ: Mother scores. The
hypothesis for this research was that there will be large and sig-
nificant PARQ-Gap differences between severely alienated chil-
dren on the one hand and nonalienated children on the other.
We predicted that PARQ-Gap differences would be so great that
it would be possible to identify a PARQ-Gap cut score that dis-
tinguishes severely alienated from nonalienated children with a
high degree of accuracy. We also expected that the PARQ-Gap
score would be helpful in distinguishing severely alienated and
nonalienated children in both clinical and forensic settings.
Review and approval of this study were obtained from the Van-
derbilt University Institutional Review Board.
Methods
Participants
As noted in Bernet et al. (29), 116 participants were recruited
for the following family types:
Children from intact families (n=35): a control group, that
is, children who lived together with both parents in one
household.
Children of divorced parents (n=20): a second control group,
that is, children whose parents were divorced or separated, but
the children continued to see both parents on a regular basis.
Neglected children (n=16): Children whose parents were
divorced or separated, but the children did not see both parents on
a regular basis. In this research, all the neglected children lived with
their mothers and rarely or never saw their fathers. The mothers
uniformly described the fathers as being unreliable in contacting
their children, uninterested in having a relationship with their chil-
dren, and neglectful by abandoning the family. Although we know
the children in this group were neglected by their fathers, we do
not know whether they were estranged from their fathers.
Alienated children (n=45): Children whose parents were
divorced or separated, and the children strongly rejected a relation-
ship with one of their parents. Of the 45 youth in the alienated
families, 24 were alienated from their fathers and 21 were alienated
from their mothers. Thus, the alienated families were divided into
the alienated-father families and the alienated-mother families.
Children from intact families, divorced families, and families
where the children were neglected by their fathers were recruited
through Research Match, a national health volunteer registry that
was created by several academic institutions and supported by
the U.S. National Institutes of Health as part of the Clinical
Translational Science Award program. Children alienated from
their fathers and children alienated from their mothers were
recruited from the Family Reflections Reunification Program, a
program in British Columbia, Canada, that specialized in the
treatment of PA (33). Other details about the methodology for
this research are described in Bernet et al. (29).
Measures
The Child Version of the Parental Acceptance–Rejection
Questionnaire: Mother and Father Forms (Child PARQ: Mother
and Child PARQ: Father) were administered. The Child PARQ
is a self-report questionnaire designed to assess children’s per-
ceptions of the degree to which they experience parental (ma-
ternal and paternal) acceptance or rejection (34). The measure
consists of four scales: (i) warmth and affection (or coldness
and lack of affection, when reverse scored); (ii) hostility and
aggression; (iii) indifference and neglect; and (iv) undifferenti-
ated rejection. Undifferentiated rejection refers to children’s
feelings that the parent does not really love them, want them,
appreciate them, or care about them in some other way without
necessarily having any objective indicator that the parent is
cold, aggressive, or neglecting. Collectively, the four scales
constitute an overall measure of perceived parental acceptance–
rejection. The mother and father versions of the measure are
identical except that one asks children to reflect on their moth-
ers’behavior, and the other asks children to reflect on their
fathers’behavior.
Sample items on the Mother version of the Child PARQ
include: My mother “lets me know she loves me”(warmth/affec-
tion), “yells at me when she is angry”(hostility/aggression),
“pays no attention to me”(indifference/neglect), and “does not
really love me”(undifferentiated rejection). Children respond to
items such as these on a 4-point Likert scale from (4) “almost
always true”through (1) “almost never true.”Scores on the
scales are summed after reverse scoring the entire warmth/affec-
tion scale to create a measure of perceived coldness and lack of
affection (a form of rejection), and after reverse scoring specified
items on the indifference/neglect scale. Possible scores on the
measure range from a low of 60 (revealing the perception of
extreme acceptance, probably unrealistic idealization of the par-
ent in most cases) through 240 (revealing the perception of pro-
found rejection). The PARQ is designed in such a way that its
midpoint of 150 reveals the perception of significantly more
rejection than acceptance. Scores between 140 and 150,
BERNET ET AL. .ALIENATION AND ESTRANGEMENTS 5
however, reveal the perception of serious rejection, though not
significantly more rejection than acceptance (25). On average, it
takes about 10–15 min to complete the PARQ.
Khaleque and Rohner summarized the reliability of the Child
PARQ (along with the Adult and Parent versions of the mea-
sure) in a meta-analysis of 51 studies worldwide (35). These
studies were based on 6898 respondents from every major ethnic
group in the United States (i.e., African Americans, Asian Amer-
icans, European Americans, and Hispanic Americans), as well as
respondents from Africa, the Caribbean, Europe, the Middle
East, and South Asia. Overall, the PARQ is known to have been
used with more than 150,000 children and adults worldwide; the
PARQ has been used extensively in research, in clinical settings,
in schools, by the courts, and in other applied contexts (26).
Results of research using the PARQ over the past 45 years
strongly suggest that the measure is reliable for research and for
clinical and applied purposes internationally as well as for use
among ethnic groups within the United States More specifically,
the mean weighted alpha coefficient for the Child PARQ was
0.89 (for the Adult PARQ it was 0.95, and for the Parent PARQ
it was 0.84). This evidence is especially compelling because no
study utilizing the PARQ anywhere in the world was found
where alpha coefficients were low and nonsignificant. Moreover,
there was no significant heterogeneity in effect sizes (alphas)
across the major geographic regions of the world or within the
American ethnic groups studied. Extensive evidence about the
convergent, discriminant, and construct validity of the PARQ is
provided by Rohner (25).
Multiple versions and forms of the PARQ are available. These
include the Child PARQ, Mother and Father versions, in a stan-
dard (60-item) as well as in a short (24-item) form. Additionally,
the Adult PARQ, Mother and Father versions, is also available
in a standard and short form, as is the Parent PARQ. The Early
Childhood PARQ, however, is available only in a short form.
The research reported in this paper is based solely on the
standard (60-item) form of the Child PARQ, Mother and Father
versions.
Results
The mean PARQ: Father scores and PARQ: Mother scores for
the 116 participants in the five groups were reported in Bernet
et al. (29). In summary, as shown in Fig. 1: for intact families,
the PARQ: Father and PARQ: Mother scores were almost
exactly equal; for divorced families, in which the children con-
tinued to see both parents on a regular basis, the PARQ: Father
and PARQ: Mother scores were also very close to each other; in
neglected families, in which the children of divorced parents
lived with their mothers and saw their fathers rarely or never,
the PARQ: Mother scores were significantly lower (revealing
greater perceived acceptance) than the PARQ: Father scores; for
the alienated-father families, PARQ: Mother scores were dramat-
ically lower (accepting) than PARQ: Father scores (rejecting);
and for the alienated-mother families, PARQ: Mother scores
were dramatically higher (rejecting) than PARQ: Father scores
(accepting). These scores indicate that alienated children have
extremely positive perceptions of the preferred parent and extre-
mely negative perceptions of the rejected parent. These data sup-
ported the hypothesis that alienated children tend to manifest the
psychological mechanism of splitting.
Measuring the PARQ-Gap
The current research extends the analysis of data previously
reported in Bernet et al. (29). We introduce here the concept
of the PARQ-Gap, which refers to the absolute difference in
scores between child’s responses on the PARQ: Father and
PARQ: Mother. In this study, the PARQ-Gap scores in intact
families and divorced families—where children had regular
contact with both parents—were very small. That is, children
in both family types perceived both parents as being loving
and accepting. The PARQ-Gap score in families where chil-
dren were neglected by their fathers was considerably greater
than in either intact or divorced families. The PARQ-Gap
score for alienated children, however, tended to be very large
FIG. 1–– Relation between perceived parental acceptance–rejection and family type. Lowest possible score on PARQ = 60 (very positive perception of par-
ent); highest possible score on PARQ = 240 (very negative perception of parent). Error bars: 95% CI. From Bernet et al. (29), used with permission.
6JOURNAL OF FORENSIC SCIENCES
because children perceived the preferred parent extremely posi-
tively, but the rejected parent extremely negatively. Table 1
indicates the Mean PARQ-Gap scores for the five family
types; Fig. 2 displays mean PARQ-Gap scores for the five
family types.
Results of statistical analysis shown in Table 1 reveal a signif-
icant difference between family types, as determined by one-way
ANOVA (F(4,111) =593.597, p<0.001). A Games-Howell
post hoc test revealed that the mean PARQ-Gap scores for all
family types were statistically different (p<0.001), except that
there was no significant difference between intact and divorced
family groups (p=0.241). Most importantly, though, the mean
PARQ-Gap scores for both the father-alienated group and the
mother-alienated group were significantly higher than that of the
neglected group (p<0.001). Table 2 shows the significant dif-
ferences among children neglected by their fathers, children
alienated from their fathers, and children alienated from their
mothers.
Using the PARQ-Gap to Help Identify Parental Alienation
Researchers and practitioners who use the PARQ to help iden-
tify possible PA should consider using a PARQ-Gap score of at
least 90 points—that is, the difference between scores on the
PARQ: Mother versus the PARQ: Father—as being an indicator
of PA. The proposed PARQ-Gap cut score of 90 is based on the
data displayed in Fig. 3, where the 90-point PARQ-Gap success-
fully distinguished children in the two groups (severely alienated
vs. nonalienated) with 99% accuracy. (Binary accuracy is the
proportion of true results [115 in this study] among the total
number of cases [116 in this study]). There was only one excep-
tion to this criterion in a sample of 116 children, 45 of whom
were alienated from either their mother or their father. The one
exception, a 12-year-old child severely alienated from the father,
had a PARQ-Gap score of 58 points (PARQ: Mother =67
points; PARQ: Father =125 points). In this case, the father was
not perceived to be all bad; in fact, he was still perceived to be
reasonably loving. This criterion—the PARQ-Gap score of at
least 90 points—is appropriate, however, only when the child
perceives the favored parent to be extremely (and probably unre-
alistically) accepting, as defined by PARQ scores between 60
and 70. Using the criterion of a 90-point PARQ-Gap then places
the disfavored parent in the seriously rejecting range of at least
140, and often higher. A PARQ-Gap score of 90 anywhere else
in the distribution of possible PARQ scores is likely to have dif-
ferent implications. For example, the meaning of scores is very
different in families where one parent scores 100 on the PARQ
and the other parent scores 190 (a 90-point PARQ-Gap). Scores
such as these reveal the fact that one parent is perceived to be
loving and accepting—though not idealistically perfect—whereas
the other parent is perceived to be seriously rejecting. This is a
common scenario in many psychologically neglecting and emo-
tionally abusive families.
Discussion
Clinicians and researchers have consistently observed for more
than 35 years that alienated children generally perceive their par-
ents in a manner consistent with splitting. That is, they tend to
perceive the preferred parent in a strongly positive manner and
TABLE 1–– Family types and the Mean PARQ-Gap Scores for each type,
with SD and 95% CI.
Family Type N
PARQ-Gap
MSD 95% CI
Intact families 35 7.54 5.65 1.87
Divorced families 20 11.35 6.86 3.01
Neglected families 16 45.31 13.42 6.58
Father alienated 24 167.21 13.89 5.56
Mother alienated 21 144.67 29.16 12.47
The PARQ-Gap is the difference between children’s responses on the
PARQ: Father and PARQ: Mother for each subject.
PARQ, Parental Acceptance–Rejection Questionnaire.
FIG. 2–– Relation between children's responses on mean PARQ-Gap and family type. The PARQ-Gap is the difference between the PARQ: Father and PARQ:
Mother for each subject. PARQ = Parental Acceptance–Rejection Questionnaire. Error bars: 95% CI.
BERNET ET AL. .ALIENATION AND ESTRANGEMENTS 7
the rejected or alienated parent in a strongly negative manner.
Likewise, clinicians and researchers have pointed out that chil-
dren who have been abused or neglected tend not to perceive
abusive or neglectful parents in such a negative manner. Abused
children typically have a somewhat negative perception of the
abusive parent, but they usually maintain a sense of ambivalence
toward that parent. That is, they dislike the abusive acts, but
they hold out hope that the abusive parent will reform and
become more loving (36). Thus, the difference between alienated
and neglected children is counterintuitive: The alienated child
(who was never abused by the rejected parent) frequently has a
more negative perception of the parent than does the neglected
child (who was actually maltreated by the rejected parent).
Empirical research presented here demonstrates how the
PARQ-Gap helps to distinguish severely alienated from non-
alienated (including neglected) children in an objective, quantita-
tive manner. The PARQ-Gap is defined as the absolute
difference between children’s scores on the PARQ: Mother and
the PARQ: Father. In this study, children in intact families have
very low PARQ-Gap scores. That is, they tend to perceive both
parents as being loving and accepting. Children in divorced
families, who continue to see both parents, also tend to have
low PARQ-Gap scores. That is, they also tend to perceive both
parents as being accepting, although not as positively as children
in intact families. These trends are typical for most intact and
divorced families where the children continue to have ongoing
contact with both parents (26). However, children in the current
study who were neglected by their fathers have moderately high
PARQ-Gap scores. That is, they continue to perceive their moth-
ers as being accepting, but they perceive their fathers to be mod-
erately rejecting. Children in this study who were alienated from
their fathers, on the other hand, tend to have extremely high
PARQ-Gap scores. They perceive their mothers as extremely
accepting and their fathers as extremely rejecting. Likewise, chil-
dren who were alienated from their mothers have extremely high
PARQ-Gap scores in that they perceive their mothers as extre-
mely rejecting and their fathers as extremely accepting. The 90-
point PARQ-Gap criterion used in this study distinguished
severely alienated children from neglected and other children
with 99% accuracy.
We should note that in this research we intentionally com-
pared neglected children (who experienced mild to moderate
maltreatment) with alienated children (who manifested a severe
degree of PA).That type of comparison is similar to the task
confronting child custody evaluators. In custody evaluations,
children might say that they refuse visitation and never want to
see their father again (consistent with a severe level of PA)
because their father failed to give them good food, confiscated
their cell phone, and always yelled at them (consistent with a
level of mild maltreatment). The custody evaluator could admin-
ister the PARQ to a child regarding each parent. This would
help clarify whether the more likely explanation for the child’s
contact refusal is PA or maltreatment by the father.
Although children who have been mildly to moderately mal-
treated are likely to maintain ambivalence toward the abusive
parent, children who have been very severely maltreated proba-
bly do not. Children who have been mildly to moderately
TABLE 2–– Statistically significant differences between the mean PARQ-Gap
scores of family types as determined by one-way ANOVA and Games-Howell
post hoc test.
Family Type 1 2 3 4 5
1. Intact families
2. Divorced families 2 >1,
p= 0.241
3. Neglected families 3 >1*** 3>2***
4. Father alienated 4 >1*** 4>2*** 4>3***
5. Mother alienated 5 >1*** 5>2*** 5>3*** 5<4***
The PARQ-Gap is the difference between children’s responses on the
PARQ: Father and PARQ: Mother for each subject.
PARQ, Parental Acceptance–Rejection Questionnaire.
***p<0.001.
FIG. 3–– Distribution of participants based on each individual's PARQ-Gap score. A cut score of 90 distinguishes severely alienated from nonalienated chil-
dren with 99% accuracy.
8JOURNAL OF FORENSIC SCIENCES
maltreated usually continue to hope that the abusive parent will
become nice again; children who have been very severely mal-
treated are likely to abandon hope that the abusive parent will
reform. Thus, the observations made in this article that mildly to
moderately abused children frequently maintain ambivalence
toward the abusive parent may not apply to children who have
been repeatedly and severely abused. Children who are estranged
due to a history of persistent, severe abuse are likely to have
very negative perceptions of the abusive parent, similar to the
perceptions of alienated children toward the rejected parent. The
PARQ-Gap appears to distinguish children who have been
mildly to moderately maltreated from children who are severely
alienated. It may not distinguish alienation from estrangement
due to very severe abuse. Furthermore, the PARQ-Gap score is
not “a test for parental alienation.”Rather, the PARQ-Gap score
may indicate a high degree of splitting, which is simply one of
the known features of severe PA.
Future research should clarify how children who have experi-
enced mild, moderate, and severe abuse respond to the PARQ
for the abusive parent versus the nonabusive parent. Future
research should also clarify how children who have experienced
mild, moderate, and severe levels of PA respond to the PARQ
for the preferred parent versus for the rejected parent. It will also
be relevant to explore whether the PARQ-Gap varies with chil-
dren’s age within the same family, since there have been anec-
dotal reports that the oldest children of several siblings tend to
be more intensely alienated compared with their younger broth-
ers and sisters. It would be interesting, also, to compare the
MMPI-2 results of the parents of alienated children with parents
of physically or sexually abused children.
This research has both strengths and limitations that should be
recognized. Strengths, for example, include the homogeneity of
the study groups, that is, all participants in the intact, divorced,
and neglected families came from the same pool of volunteers
by way of ResearchMatch. Likewise, all alienated respondents
came from the same PA-specific treatment program, that is, the
Family Reflections Reunification Program. Also, we consider it
a strength that the PARQ was not developed specifically for use
in child custody disputes or for identifying PA. In contrast, the
PARQ came out of an independent line of research—IPARThe-
ory—and has been used successfully in hundreds of studies only
occasionally having anything to do with divorce or child cus-
tody. The PARQ is also known to be reliable and valid for use
in international and multi-ethnic contexts. Since it is available
for use in 65 languages and dialects worldwide, it can be used
by many researchers and practitioners concerned with issues of
PA internationally.
The major weakness of this research involves the family type
that we characterized as neglected. There were 16 participants in
that group. In all cases, the children lived with their mothers and
rarely or never saw their fathers. Although their mothers reported
that those 16 children had been rejected, avoided, and aban-
doned by their fathers, none of the mothers said that the fathers
had been abusive. We therefore considered those children to be
mildly to moderately maltreated. Although it seems unlikely, it
is possible that a child characterized by the mother as neglected
by the father was actually alienated from the father. In future
research, the features of the control groups should be determined
with greater precision.
This research indicates that the PARQ-Gap might become use-
ful for clinical and forensic evaluations of children who may be
alienated or estranged. Of course, the PARQ should not be used
in isolation to determine whether a child is alienated or
estranged. When it is used during child custody evaluations, the
PARQ—like any psychological test—should be only one part of
a comprehensive psychiatric or psychological assessment of the
family. A comprehensive evaluation includes multiple inter-
views, meetings with collateral informants, review of records,
and teamwork with other professionals. However, we conclude
that both clinical and forensic practitioners should consider using
the PARQ as one component of a comprehensive evaluation
when they are concerned about the possible diagnosis of PA.
Acknowledgments
The authors appreciate the helpful suggestions of Richard A.
Warshak, Ph.D., in reviewing this manuscript.
References
1. Gardner RA. Recent trends in divorce and custody litigation. Acad
Forum 1985;29(2):3–7.
2. Saini M, Johnston JR, Fidler BJ, Bala N. Empirical studies of alienation.
In: Drozd L, Saini M, Olesen N, editors. Parenting plan evaluations:
applied research for the family court, 2nd edn. New York, NY: Oxford
University Press, 2016;374–430.
3. Gardner RA. Differentiating between parental alienation syndrome and
bona fide abuse-neglect. Am J Fam Ther 1999;27(2):97–107.
4. Kelly JB, Johnston JR. The alienated child: a reformulation of parental
alienation syndrome. Fam Court Rev 2001;39(3):249–66.
5. Bernet W, Freeman B. The psychosocial assessment of contact refusal.
In: Lorandos D, Bernet W, Sauber SR, editors. Parental alienation: the
handbook for mental health and legal professionals. Springfield, IL:
Charles C Thomas, 2013;47–73.
6. Freeman B. The psychosocial assessment of contact refusal. In: Lorandos
D, Bernet W, editors. Parental alienation –science and law. Springfield,
IL: Charles C Thomas, 2020;44–81.
7. Lee SM, Olesen NW. Assessing for alienation in child custody and
access evaluations. Fam Court Rev 2001;39(3):282–98.
8. Drozd LM, Olesen NW. Is it abuse, alienation, and/or estrangement? A
decision tree. J Child Custody 2004;1(3):65–106.
9. Ellis EM. A stepwise approach to evaluating children for parental alien-
ation syndrome. J Child Custody 2007;4(1–2):55–78.
10. Warshak RA. Bringing sense to parental alienation: a look at the dis-
putes and the evidence. Fam Law Q 2003;37(2):273–301.
11. Gottlieb LJ. The parental alienation syndrome: a family therapy and col-
laborative systems approach to amelioration. Springfield, IL: Charles C
Thomas, 2012;108.
12. Clawar SS, Rivlin BV. Children held hostage: identifying brainwashed
children, presenting a case, and crafting solutions, 2nd edn. Chicago, IL:
American Bar Association, 2013;321.
13. Johnston JR, Lee S, Olesen NW, Walters MG. Allegations and substan-
tiations of abuse in custody-disputing families. Fam Court Rev 2005;43
(2):283–94.
14. Beeble ML, Bybee D, Sullivan CM. Abusive men's use of children to
control their partners and ex-partners. Eur Psychol 2007;12(1):54–61.
15. Fidler BJ, Bala N. Children resisting postseparation contact with a par-
ent: concepts, controversies, and conundrums. Fam Court Rev 2010;48
(1):10–47.
16. Baker AJL, Creegan A, Quinones A, Roselle L. Foster children's views
of their birth parents: a review of the literature. Child Youth Serv Rev
2016;67:177–83.
17. Baker AJL, Miller S, Bernet W, Adebayo T. The assessment of the atti-
tudes and behaviors about physically abused children: a survey of men-
tal health professionals. J Child Fam Stud 2019;28(12):3401–11.
18. Siegel JC, Langford JS. MMPI-2 validity scales and suspected parental
alienation syndrome. Am J Forensic Psychol 1998;16(4):5–14.
19. Gordon RM, Stoffey R, Bottinelli J. MMPI-2 findings of primitive
defenses in alienating parents. Am J Fam Ther 2008;36(3):211–28.
20. Bricklin B. Bricklin perceptual scales. Furlong, PA: Village Publishing,
1984.
21. Bricklin B. The contribution of psychological tests to custody-relevant
evaluations. In: Galatzer-Levy RM, Kraus L, editors. The scientific basis
of child custody decisions. New York, NY: Wiley, 1999;120–56.
22. Otto RK, Edens JF, Barcus EH. The use of psychological testing in
child custody evaluations. Fam Concil Courts Rev 2000;38:312–40.
BERNET ET AL. .ALIENATION AND ESTRANGEMENTS 9
23. Baker AJL, Burkhard B, Albertson-Kelly J. Differentiating alienated
from not alienated children: a pilot study. J Divorce Remarriage 2012;53
(3):178–93.
24. Rowlands GA. Parental alienation: a measurement tool. J Divorce
Remarriage 2018;60:316–31.
25. Rohner RP. Parental Acceptance-Rejection Questionnaire (PARQ): test
manual. In: Rohner RP, Khaleque A, editors. Handbook for the study of
parental acceptance and rejection, 4th edn. Storrs, CT: Rohner Research
Publications, 2005;43–106.
26. Rohner RP.Introduction to interpersonal acceptance–rejection theory
(IPARTheory), methods, evidence and implications. https://csiar.uc
onn.edu/wp-content/uploads/sites/494/2019/11/Introduction-to-IPARThe
ory-for-web.pdf (accessed November 27, 2019).
27. Jaffe AM, Thakkar MJ, Piron P. Denial of ambivalence as a hallmark of
parental alienation. Cogent Psychol 2017;4(1):1327144.
28. Bernet W. Parental alienation, child psychological abuse, and parental
acceptance–rejection theory. In: Machado M, Machado F, editors. New
paths for acceptance: opening awareness in interpersonal acceptance–re-
jection. Boca Raton, FL: Brown Walker Press, 2015;15–23.
29. Bernet W, Gregory N, Reay KM, Rohner RP. An objective measure of
splitting in parental alienation: the parental acceptance-rejection ques-
tionnaire. J Forensic Sci 2018;63(3):776–83.
30. Blagg N, Godfrey E. Exploring parent–child relationships in alienated
versus neglected/emotionally abused children using the Bene-Anthony
Family Relations Test. Child Abuse Rev 2018;27(6):486–96.
31. Bene E, Anthony EJ. Manual for the family relations test. London,
U.K.: National Foundation for Educational Research in England and
Wales, 1957.
32. Parkin A. The Bene-Anthony Family Relations Test revisited: directions
in the assessment of children's perceptions of family relations. Br J Med
Psychol 2011;74(Pt 3):323–49.
33. Reay KM. Family reflections: a promising therapeutic program designed
to treat severely alienated children and their family system. Am J Fam
Ther 2015;43(2):197–207.
34. Rohner RP, Ali S. The Parental Acceptance-Rejection Questionnaire
(PARQ). In: Zeigler-Hill V, Shackelford T, editors. Encyclopedia of
personality and individual differences. New York, NY: Springer,
2016.
35. Khaleque A, Rohner RP. Reliability of measures assessing the relation
between perceived parental acceptance–rejection and psychological
adjustment: a meta-analysis of cross-cultural and intercultural studies. J
Cross Cult Psychol 2002;33(1):86–98.
36. Baker AJL, Schneiderman M. Bonded to the abuser: how victims make
sense of childhood abuse. Lanham, MD: Rowman & Littlefield, 2015.
10 JOURNAL OF FORENSIC SCIENCES