An Internet Journal devoted to the Psychology of Personal Constructs

PDF version

Journal Main
Contents Vol 1



Kenneth W. Sewell*, Keith R. Cruise**

*Department of Psychology, University of North Texas, Denton, Texas, USA
**Department of Psychology and Philosophy, Sam Houston State University, Huntsville, Texas, USA


Criminal offending by adolescents is a serious problem. Often, chronic and serious adolescent offenders are conceptualized via a downward extension of the adult construct of “psychopathy”. Despite the obvious implications of social processes in psychopathic characteristics, most measurement tools fail to incorporate social processes in the assessment method of this construct. Thus, these tools often lead to classification without clear implications for treatment. The present paper presents the use of specialized repertory grids to assess psychopathy per se as well as other social construction in adolescent offenders. Two case examples are discussed, along with their grid results, to illustrate the potential utility of these methods. The paper concludes with a discussion of the implications and future of constructivist assessment of psychopathy.
Key Words: adolescents; psychopathy; repertory grid

Adolescent offending
Criminal activity by children and adolescents represents a significant social concern as represented by rising rates of serious criminal offenses committed by ever increasing groups of younger offenders (Hoge, 2000; Snyder & Sickmund, 1999). The response to juvenile offenses in much of the United States has been an increase in punitive sanctions and ‘get tough’ policies such as the increased use of waiver into adult court and mandatory sentencing.
It is estimated that a small percentage (approximately 15%) of offenders are responsible for a large proportion of serious criminal acts (approximately 50 to 60%) among adolescents (Cottle, Lee, & Heilbrun, 1999; Farrington, 1983; Moffit, 1993). These are the young offenders who progress to similar rates of offending as adults (see Loeber, 1991). This relatively small proportion of serious juvenile offenders have been described as a distinct group who begin offending at an early age and engage in a wide array of antisocial acts ranging from property crime to serious forms of aggression (Foote, 1997).
Psychopathy in adolescence
The construct of ‘psychopathy’ as a clinical syndrome has been applied as a discriminating variable in the identification and classification of children who will develop chronic offending careers (Lynam, 1996, 1997; Forth & Mailloux, 2000; Frick, Barry, & Boudin, 2000). In a review of the extant research regarding the predictive validity of psychopathy, a consistent association between psychopathy and aggressive behavior was demonstrated across 11 studies with an average correlation of .30 (range .19 to .67). This robust, consistent finding is moderated by the limited number of studies with prospective designs examining the association between psychopathy and later acts of violence (Edens, Skeem, Cruise, & Caufmann, 2000).
Despite the need for a viable means to identify and intervene with antisocial youth, caution is warranted in the direct downward extension of the psychopathy construct from adults (Edens, Skeem, Cruise, & Caufmann, 2000; Grisso, 1998). There are several important issues that remain to be fully understood.
First, the temporal stability of the construct in children and adolescents remains to be demonstrated. Can an adolescent merit a diagnosis of psychopathy at a particular time, and then not merit that diagnosis at a later time? Temporal stability in this context is a two-edged sword. On the one hand, a lack of temporal stability (e.g., if the same person could merit differing diagnoses over a short time period with no intervening explanation) would impede the ability to use the construct in a predictive manner. On the other hand, rigid temporal stability (e.g., if it were the case that ‘once a psychopath, always a psychopath’) suggests that psychopathy is a descriptive construct for which psychological interventions are not relevant. All too often, the latter of these extreme possibilities is assumed (that psychopathy is static and unchangeable). Nonetheless, there is a straightforward lack of evidence to suggest that the construct is even stable, let alone unchangeable. Clearly if psychopathy in adolescence is to prove useful, demonstration is needed that psychopathy is generally stable, measurement identifies a proportion of young offenders on a developmental trajectory of severe antisocial behavior, yet capable of changing under specifiable circumstances.
Second, it is inherently hazardous to apply to adolescents diagnostic criteria validated on adults without investigating potential developmental differences in their applicability. Examples of this hazard in the realm of psychopathy include the explicit downward extension of adult criteria of parasitic lifestyle, impulsivity, and irresponsibility. What does it mean to say that an adolescent does or does not lead a parasitic lifestyle (i.e., lives ‘off of’ the income and/or good will of others)? Surely it means something different for an adolescent than for an adult given that all adolescents are dependent on adults to some degree. Adolescence is generally considered a time of inherent impulsivity, irresponsibility, and so forth. In order for such traits to be considered diagnostically relevant, do they simply need to be extreme? Such traits in adults (even at moderate levels) imply a developmental failure; extremity of such traits in adolescents might be problematic, but they do not clearly imply developmental failure in analogous ways. Additionally, within the adolescent developmental period, normative developmental changes are expected that impact the adolescents’ cognitive capacities, emotional regulation, and complexities of social relationships. Extremity on a ‘psychopathic’ trait likely needs age-relevant criteria. The field awaits a comprehensive formulation of adolescent psychopathy that is both theoretically and developmentally sound.
Beyond the problems (discussed above) with extending the construct of adult psychopathy downward onto adolescents, constructivist critiques of current conceptions of ‘disorder’ offer some compelling reasons for concern (e.g., Raskin & Lewandowski, 2000). In general, constructivists view what are traditionally called ‘disorders’ as constellations of characteristics that are problematic not because of some flaw inherent to the individual, but rather because of difficulties that result in the person’s life (often social) processes. Given the social processes involved in ‘disorder’ (even what is defined as a disorder evolves via social negotiation), our conceptions of diagnoses should logically include implications for altering the social interactions and context of the troubled person. Thus, diagnoses are to be considered transitive starting points as opposed to conclusory end points. From this constructivist point of view, a diagnosis such as psychopathy is useful only insofar as it carries information about the ways in which the person construes the social world (including the self) and implications for how that construing (or, indeed the content and structure of the social world itself) might need revision to accommodate more adaptive functioning. The current conception of psychopathy falls short of this ideal, particularly in its situating the difficulties in the person’s psyche (or character) rather than in the person’s life (social) processes.
Despite these areas of ambiguity in the adolescent psychopathy literature and critiques from constructivists, the construct of adolescent psychopathy is receiving increased attention in the clinical and forensic literature with a predominant focus on classification and prediction without reference to developmental and treatment implications. Thus, the effective measurement of personality and behavioral patterns associated with psychopathy in adolescents is critical. Measurements that have implications for treatment are particularly desirable, given the need to transcend the negative and global assumptions often associated with the psychopathy label. Because psychopathy implies disrupted social relationships, and because forensic treatments are almost always offered in group (social) formats, measures that also address relevant social processes should prove most useful. Moreover, assessing this constellation in a manner that accounts for treatment and social processes would bring the construct of psychopathy into a more tenable relation with constructivist psychology.
Measurement of psychopathy in adolescence
Recently, a variety of measures have been developed that utilize interview and self-report methods to assess psychopathic characteristics in children and adolescents. The construction of such measures has relied on the accepted adult conceptualization of the clinical construct. The measures include the Psychopathy Checklist: Youth Version (PCL-YV; Forth, Kosson, & Hare, 2003) which is an interview-based clinician rating scale, and the Antisocial Process Screening Device (APSD; Frick & Hare, 2001) which is a self-report questionnaire designed to be completed by the adolescent and/or a parent or other informant (teacher, etc.).
None of the adolescent psychopathy measures currently in use addresses the needs identified above (treatment implications over labeling, social process information over characterological attribution, etc.). The present paper explores a psychopathy measurement technique based upon Kelly’s (1955) Role Construct Repertory Grid (or rep grid) that attempts to address some of these needs and align itself with a constructivist framework. The present work used two different repertory grids to assess psychopathic self-construction as well as the construction of peers and treaters in a group treatment context. This use of rep grid methods allows for a unique and novel approach to the evaluation of psychopathic characteristics by concurrent evaluation of self-report, construal of others, as well as peer ratings. To our knowledge, this approach has not been utilized previously in either adult or adolescent research on psychopathy. It is presented here in hopes that clinicians and researchers will apply this approach and/or similar methods and share their results in the professional literature. If this proves useful, assessment methods used with adolescents in clinical-forensic and treatment contexts can then reflect broad constructivist values of individual understanding over simple diagnoses.
The methodology will be described in detail and illustrated with two focused case examples. Finally, the potentials for the applicability of this methodology will be explored, including the need to incorporate personal-constructional and social-constructional levels when measuring psychopathy in adolescents.
Two separate grids were constructed to administer to adolescent male offenders in a group treatment context (a post-adjudication residential treatment facility).
Psychopathy Grid
The Psychopathy Grid was constructed to correspond to the concepts included in the traditional assessment of psychopathy. The criteria used in the PCL-YV were converted to bipolar constructs using labels understandable to the average teenager (Flesch-Kincaid reading level of grade 5.7). These psychopathic traits served as supplied constructs in the grid. Additionally, two other dimensions were included as supplied constructs: anger and agitation. These latter constructs were chosen because of the oft-asserted notion that psychopathy implies anger-proneness and (perhaps paradoxically) hypo-arousal. Group members rated themselves, each of their group treatment peers, and their therapist on 6-point scales anchored by these supplied construct poles (see Table 1). The size of the Psychopathy Grid was 14 (constructs) by 9 (elements, a number dependent upon group size).
Table 1: Constructs (with the psychopathic poles denoted by asterices) and elements comparing the Psychopathy Grid

supplied psychopathy constructs  
  insincere charmer* vs. sincere genuine
  never brags vs. brags all the time*
  thrill seeker* vs. not a thrill seeker
  honest vs. lies all the time*
  con artist* vs. not a con artist
  is sorry for hurting others vs. is not sorry for hurting others*
  emotions are not believable* vs. emotions are believable
  has concern for others vs. feels no concern for others*
  cannot control anger* vs. manages anger well
  makes realistic plans vs. missing or unrealistic goals*
  denies responsibility for crimes* vs. really accepts responsibility
  thinks things through vs. acts without thinking*
  cannot be counted on* vs. can be counted on
  done small time crimes vs. done lots of hard crimes*
additional supplied constructs  
  makes me feel edgy vs. makes me feel calm
  makes me really mad vs. never makes me mad
  peers in treatment group

There are a variety of indices that can easily be examined in the Psychopathy Grid. Several of these are briefly described below.
Psychopathic self-construction
The ratings on the 12 psychopathy constructs can be averaged and divided by the length of the scale to derive an index of the extent to which the psychopathic poles are attributed to the self. The potential range of this index is .17-1.0.
Psychopathic counterpart construction
Like the Psychopathic Self-Construction index discussed above, the ratings on the 12 psychopathy constructs can be averaged and divided by the length of the scale to derive an index of the extent to which the psychopathic poles are attributed to any element. Given that the other members of the treatment group (including the therapist) were used as elements, this index is calculable on any group member from the perspective of the examinee. The potential range of this index is .17-1.0.
Intensity score
The overall extent to which a person’s grid constructs intercorrelate has been referred to as the ‘intensity’ of the grid (Bannister & Fransella, 1966). This allows for a rough measure of the simplicity of the construct repertoire sampled (cf. Pierce, Sewell, & Cromwell, 1992).
Self/other distances
Pythagorean distance scores can be calculated between any two elements represented in the grid. Thus, the distance (or differentiation) between the self and every other element (from the examinee’s perspective) can be derived (Mitterer & Adams-Webber, 1988).
Construct intercorrelations
Simple correlations between constructs can be calculated to provide information regarding linkages between an examinee’s personal dimensions of meaning (Slater, 1972). Given the large number of correlations potentially generated, only specific pairs are examined based upon either their content-based interest (e.g. construct pairs that seem likely or unlikely to be related based upon semantic/theoretical content) or their extremity of relation (e.g., correlations approaching 1.0, -1.0, or 0)
Social Grid
The Social Grid was constructed as a tool to assess the general social construing of the adolescents. By including amongst the grid’s elements the self, some treatment group peers, and some family members (see Figure 2), a broad sampling of social construction relevant to the adolescent’s current life could be elicited. The adolescents rated each of the elements on a 6-point scale. The size of the Social Grid was 10 (elicited constructs) by 10 (elements).
Table 2: Construct Elicitation Method and Element Role Titles Comprising the Social Grid
Constructs 10 triadically elicited bipolar constructs with designated pole valence
Elements self
  mother or mother-figure
  father or father-figure
  therapist or counselor
  peer you respect
  peer you don’t respect
  peer who will complete the program
  peer who will not complete the program
  someone you trust
  self when graduating from the program

As with the Psychopathy Grid described above, there are a variety of indices that can easily be examined in the Social Grid. Several of these are briefly described below.
Construct content
Because the constructs in the Social Grid are elicited from the examinee, the semantic content of the constructs can be examined for thematic consistency, idiosyncrasies of terminology or oppositeness, and level of abstraction (e.g., physical versus emotional dimensions). Given that elicited constructs are seen by examinee’s as more personally meaningful than supplied constructs (Adams-Webber, 1970), reviewing the contents of such personal constructs can be informative regarding how the examinee experiences the world.
Sewell, Cromwell, Adams-Webber, and Mitterer (1991) developed a self-esteem measure for rating scale grids capable of utilizing the full sensitivity of the rating scale. Essentially, the self-esteem index is the proportion of positivity attributed to the self across all constructs. The self-esteem index approaches zero as the self is rated more negatively across constructs. Conversely, a person rating the self at the extreme positive pole of each construct would have a self-esteem index of 1.0. Although a variety of other methods have been developed to assess self-esteem via grids, this index was chosen because it does not rely on an ‘ideal self’ reference point (not present in the current grids), and it allows for comparisons across elicited constructs.
The ‘Self at Graduation’ was included as an element in the Social Grid. Thus, the extent to which that future state is idealized can be quantified by calculating the self-esteem index based upon the ‘Self at Graduation’ element.
Self/other distances
Self/other distances can be derived from the Social Grid just as described above in the Psychopathy Grid section.
Construct intercorrelations
Construct intercorrelations can be derived and inspected from the Social Grid just as described above in the Psychopathy Grid section.
This section will describe two adolescent offenders who were given the Psychopathy Grid and Social Grid within their shared group treatment context within a post-adjudication residential treatment center. Identifying details have been altered to protect the confidentiality of the participants. General descriptions of the two adolescents will be followed by a detailed presentation of results from their grid examinations to illustrate how this methodology can be informative within such adolescent forensic contexts.
Nathan, a 15 year-old boy, had been in the treatment program approximately 175 days at the time of the assessment. His adjudicated offenses were Burglary of a Habitation, and Unauthorized Use of a Vehicle. There was no reported drug or alcohol use in the referral. Nathan’s mother reported that Nathan’s behavior problems did not begin until he became an adolescent. She indicated she is unable to control or set expectations around his behavior. Other relevant family information includes that Nathan’s father was killed in an accident when Nathan was 7. Nathan seems to seek the approval of his older brother who himself has numerous juvenile court referrals. A pre-placement psychological evaluation of Nathan recommended a structured environment, noting risk areas involving educational problems, negative peers, and ‘personality’ problems.
Also a 15-year-old male, John had been in the treatment program for approximately 200 days at the time of assessment. He was adjudicated for charges of Burglary of a Vehicle and Theft of Greater Than $500. John’s file suggests a history of physical abuse and academic failure. John seemed to be generally cooperative with familial structures while living with his father but was reported to be ‘quiet, sneaky, and manipulative’ nonetheless, leading to his legal involvements and incarceration. After his incarceration, his father appears to have abandoned him altogether. No longer able to live with his father, John is facing the prospect of returning to live in another state with his mother in a blended family context; this is despite John’s removal from his mother’s custody as a young child due to physical abuse from John’s step-father. John’s pre-placement evaluation identified high-risk areas involving family, peers, education, and poor attitude. John is seen by the treatment staff as presenting a ‘tough guy’ image.
Psychopathy Grid results
Table 3 shows the Psychopathy Grid results for Nathan and John. John sees himself as more psychopathic than does Nathan, a configuration that is repeated in their views of each other (although they view each other as less psychopathic than their self-assessments portray). John’s psychopathy constructs are more intercorrelated than are Nathan’s, indicating that, for John, ‘badness’ is something of a monolithic concept. The average self/other distances for Nathan and John are roughly comparable; and John and Nathan each see the other as very different from himself.
Table 3: Psychopathy grid results for Nathan and John

Grid Indices Nathan John
psychopathy index .48 .81
rating of peer on psychopathy index .63 (John) .32 (Nathan)
intensity score 3581.14 44562.37
average self/other distances .67 .73
perceived distance from peer .87 (John) 1.05 (Nathan)

Although not a grid index per se, specific construct divergences can be inspected when two persons each include the other as elements in their grids. Table 4 shows the constructs on which Nathan and John diverged, from each of their separate perspectives. Nathan anticipates John as an explosive and irresponsible person around whom he is not particularly comfortable. John assigned himself and Nathan to opposite construct poles on all constructs except ‘done lots of hard crimes,’ ‘makes realistic plans,’ and ‘really accepts responsibility.’ Clearly, John sees Nathan as being unlike him in almost all important ways.
Table 4: Construct Divergence for Nathan and John

Nathan’s Perspective  
Self John
thrill-seeker (6) not a thrill-seeker (1)
manages anger well (1) cannot control anger (5)
really accepts responsibility (2) denies responsibility for crimes (5)
can be counted on (2) cannot be counted on (6)
makes me feel calm (2) makes me feel edgy (5)
Divergence between John and Nathan from John’s perspective  
  John assigned himself and Nathan to opposite construct poles on all constructs except ‘done lots of hard crimes,’ ‘makes realistic plans,’ and ‘really accepts responsibility.’

Finally, construct intercorrelations of note are seen in Table 5. Nathan seems to have an over-arching construct of dishonesty that, for him, implies many of the other psychopathic characteristics. Among other patterns noted, John shows an unusual separateness of his constructs of remorsefulness and responsibility-taking, likely indicating that he sees some value in ‘taking responsibility’ but does not value ‘feeling sorry’ for hurtful actions.
Social Grid results
Table 6 shows the constructs produced by Nathan and John during triadic elicitation. Looking at Nathan’s constructs, six out of ten elicited constructs are interpersonal and/or emotional; the others relate to accountability and motivation for change. Examining John’s constructs, a predominant theme of ‘either you are with me or you are against me’ emerges (with most social relationships evaluated in terms of their instrumental value).
Table 5: Selected construct intercorrelations

Youth Constructs Correlation
Nathan insincere charmer & lies all the time -.73
  honest & manages anger well -.91
  honest & can be counted on -.91
  denies responsibility for crimes & acts without thinking -.93
  denies responsibility for crimes & can’t be counted on .91
  thinks things through & can be counted on -.84
  makes me feel edgy & makes me mad .87
John insincere charmer & thrill seeker .90
  insincere charmer & missing or unrealistic goals -.90
  insincere charmer & acts without thinking -.90
  thrill seeker & acts without thinking -.92
  thrill seeker & cannot be counted on .92
  honest & makes realistic plans .97
  thinks things through & can be counted on -.98

Table 6: Elicited constructs in the Social Grid (preferred poles marked with +)

Nathan John
trusting+ vs. not reliable close+ vs. distant
friendly+ vs. rejecting won’t reoffend+ vs. will offend
stay out of trouble+ vs. getting into trouble trust+ vs. unhonest
denying problems vs. accepting problems+ handles problems+ vs. making problems
wants to change+ vs. doesn’t want to change angry vs. friendly+
have attitudes+ vs. doesn’t have an attitude helpful+ vs. unhelpful
friends+ vs. enemy calming+ vs. frustrating
supportive+ vs. not supportive friendly+ vs. foe
respectful+ vs. not respectful positive role model+ vs. negative role model
loving+ vs. not loving family+ vs. stranger

Table 7 shows the quantitative indices derived from the Social Grid for Nathan and John. Although Nathan has a predominantly positive view of self on the relational/emotional constructs, he is ambivalent about wanting to change in that he sees himself as still ‘having an attitude’ even after graduation. Nathan’s self/other distances show a fairly close association between himself and others in the grid (distance range of 2.83 to 9.0). The smallest distance is with ‘self at graduation’. Conversely, the element John saw as most distant from the self was ‘self at graduation’ (13.86). John’s element closest to the self was ‘someone you trust’ (3.74). All adults were rated as distant by John.
Table 7: Quantitative indices from the Social Grid

Grid Indices Nathan John
Self-esteem .77 .30
Self-graduate esteem .87 .97
Intensity score 2210.79 1885.12
Average self/other distance 5.25 8.80

Regarding Nathan’s construct intercorrelations, all constructs appear at least modestly correlated with ‘trusting vs. not reliable.’ A similar pattern is present for his construct of ‘loving vs. not loving.’ For John, ‘won’t reoffend’ shows a strong pattern of association with all other constructs. Not re-offending is strongly correlated with ‘trust’, ‘handling problems’, ‘friendly’, ‘helpful’, and ‘calming’. However, he rates himself on this construct as a 1 (definitely ‘will reoffend’). The discrepancy between what John wants to achieve and his perceived inability to get there seems related to his conception that reoffending, for him, is inevitable.
Utility of grid methods with adolescent offenders
Use of repertory grids in the present clinical context, as demonstrated by the case examples presented above, allowed for the constellation of personality characteristics often termed ‘psychopathic’ to be assessed and conceptualized in social context. Not only were we able to evaluate the extent to which the boys had adopted a psychopathic role, but also the ways in which particular characteristics are construed as making them similar to or different from their peers (including each other). Given the high frequency with which group treatment methods are used with adolescent offenders, and the extent to which social functioning is crucial to the development and maintenance of healthy as well as unhealthy adaptation, this ecologically sensitive technique of assessment seems particularly informative to the clinician wishing to intervene multisystemically.
The relation between behavioral problems and self-esteem/self-concept can be readily examined using the psychopathy and social grids with troubled adolescents. A clear example of this is John’s apparent conflict in his self-construction via the construct of ‘won’t reoffend versus will reoffend.’ From John’s perspective, the presence of a variety of desirable traits implies a strong likelihood that the person will not reoffend. John sees this relation as holding for virtually everyone except himself. Although he ascribes the same constellation of positive attributes to his projection of the self that will graduate from the treatment program, that new-and-improved self will still be destined to reoffend. Treaters would certainly want to help John unravel the rigidity of his self-construction as destined to reoffend regardless of how much he improves his functioning otherwise. This is just one example of how constructivist assessment such as the repertory grid can transcend the pigeonholing process inherent in many clinical assessment tools. This emphasis on generating assessment findings that bear clear treatment implications is seen by the present authors as the primary benefit to using such constructivist assessment tools with troubled adolescents.
Implications for the ‘psychopathy construct’
We believe that the psychopathy grid provides a unique method of measurement that has potential to make a substantial contribution to the measurement of this construct in adolescents. As previously mentioned, child and adolescent assessment involve the use of multi-informant, multi-method measurement strategies (Achenbach, 2000; Kamphaus & Frick, 1996). Current measures of adolescent psychopathy have relied on a traditional self-report format utilizing parallel forms across multiple informants (e.g., the APSD) or a combination of a semi-structured clinical interview incorporating collateral information (e.g., the PCL:YV). The APSD includes a self-report, parent, and teacher/probation officer form. The existing measures lack a contextual focus and place the clinician in the expert role of determining the presence and clinical severity of psychopathy characteristics.  
As utilized in the current study, the psychopathy grid served multiple functions in providing both a self-construal of psychopathic characteristics as well as an evaluation of peers who are involved in the treatment program. If used with larger groups, the ability to use multiple peer evaluations as collateral informant data provides a unique and novel contribution to the measurement literature in this area. Researchers have noted problems that arise when multiple informants produce inconsistent results (i.e., highly discrepant evaluations along the construct of interest). Such discrepancies have also been attributed to varied interactions between the rater and subject across different contexts, and use of open, context-free questions (Wright, Zakriski, & Drinkwater, 1999). The use of supplied constructs with a similar rating format provides a consistent methodology by which to solicit collateral information regarding psychopathic characteristics, allowing for a consistent and easily interpreted index of psychopathy across the self and peer informants. The ability to evaluate psychopathic self-construction within a specific context (i.e., the current treatment setting) allows treatment staff to generate clinically useful inferences regarding the youth engagement in the treatment process and identification with peers within the treatment context. Agreement and lack of agreement of peer informants on the psychopathy construct provides useful information in this context in that consistency/inconsistency may be related to psychopathic self-construction of the raters. Alternatively, inconsistency in ratings may indicate that the youth completing the grid does not find the supplied constructs (i.e., dimensions we perceive as prototypical indicators of psychopathy) to be useful discriminating constructs.
From a measurement perspective, the psychopathy grid provides three levels of data that are informative regarding the psychopathy construct. First, the psychopathic self- and counterpart-construction serve as global markers of the construct and how an individual views others along the same dimension. Second, the ability to make self-other comparisons on individual psychopathy characteristics (i.e., convergence and divergence on single constructs) is also informative. Traditionally, psychopathy has been defined along affective/interpersonal dimensions and behavioral dimensions (see Hare 1996, 1998). In addition to comparisons on the global index of psychopathy, similarities and differences can also be examined by combining items reflecting both the affective/interpersonal and behavioral dimensions.
Third and perhaps most important, the self/other distance scores provide a valuable and unique measurement of the construct within the current treatment context. Utilizing a combination of the psychopathy and social grid results, one could evaluate the role of the psychopathy construct in the peer identification and alienation process. The presence of psychopathy in adult offenders is viewed as disruptive to the treatment context (Ogloff, Wong, & Greenwood, 1990) and results in poor treatment response (Blackburn, 1993; Losel, 1998; and Serin, 1996). Although data regarding the treatment of adolescent psychopathy is lacking, a wealth of data exists regarding the importance of negative peer group identification in increasing or exacerbating juvenile delinquent or adolescent antisocial behavior (see Dishion & Patterson, 1999; Loeber & Farrington, 2000). By using treatment peers as elements in the grid, patterns of peer identification can be examined which could have a differential impact on treatment response. For example, patterns of strong identification with peers who are rated as psychopathic may suggest a negative treatment response and further exacerbation of antisocial tendencies. Peers within the treatment context who are viewed as less psychopathic may be characterized as weak and targeted within a treatment community. Thus, comparisons using self/other distances may allow for the implicit identification of treatment alliances and negative peer relationships.
Another related contribution is the ability to evaluate self/other constructions across multiple administrations of the psychopathy grid. Across multiple administrations, fluctuation or stability across self and peer ratings provides a unique way to examine the temporal stability of individual psychopathic characteristics. For example, early in the treatment process, it is quite possible that adolescents may adopt a psychopathic presentation, similar to the negative peer group identification, if there are perceived (i.e., respect from other delinquent youth) or real benefits (i.e., protection from harm) that could be gained from adopting this role. Variability across construct and element ratings could be indicative of ways in which the youth may be trying out various psychopathic characteristics and with whom the youth anticipates such characteristics as being valuable from a social perspective. Changes in self-construction over time could reflect the transient nature of this role adoption as the youth engages in the treatment process. However, stability in both self and peer ratings, either global via the psychopathy index, or specific to individual characteristics, could be suggestive of rigid adherence to maladaptive constructions and rejection from peers who also evaluate the youth as ‘psychopathic.’    
First of all, the methods described in the present paper need to be subjected to the tests of clinical applicability. The extent that clinicians find the treatment implications yielded by these tools to be useful, then the time and effort required to conduct the assessments will be invested. Although the present authors have attempted to describe a variety of datapoints (indices, etc.) derivable from these grids, the possibilities extend far beyond the scope of this chapter. Thus, clinicians familiar with grid methods are encouraged examine the broad structure of these assessments, and experiment with contents or analyses in addition to those discussed herein.
Secondly, the utility of these constructivist assessment techniques needs to be studied in relation to other measures of adolescent psychopathy. It is difficult to ascertain the potential value of a ‘convergent validity’ study using, for example, the psychopathy grid alongside the PCL-YV; given the lack of a clear gold standard, the meaning of minimal convergence would be unclear. However, the extent to which the two methods can predict useful functional aspects of the adolescent’s life can be meaningfully compared. If indices from the psychopathy grid and/or the social grid can predict recidivism (or community survival), social functioning, shifts in attitude toward others, and so forth, better than or in addition to the typical diagnostic assessment, then the value of the grid assessments will be incontrovertible. Such empirical studies, although difficult to perform, are needed to legitimize these potentially important tools.
Perhaps the greatest implication of the current methodology is the possibility for generating clinically useful hypotheses about the role of psychopathy in the treatment process. As previously mentioned, there is a complete absence of studies examining the role of psychopathy in the treatment of adolescents. We believe that the grid results allow for the generation of clinically meaningful hypotheses about the individual adolescent. For example, the degree to which an adolescent offender adheres to a psychopathic role in combination with high self-esteem suggests the possibility of treatment resistance. Under such circumstances, ensuring a treatment milieu that does not socially reinforce this self-construction would be crucial. Group treatment of such persons might actually be contraindicated, given the difficulty of managing the social reinforcers in such contexts. John’s grid result suggests another possibility. Adolescents like John seem to view themselves as psychopathic, distant from others (both within the treatment context as well as in broader social contexts), of low self-worth, and distant from some perceived ideal (e.g., the state requisite for successful completion of the treatment program). Thus, these adolescents are likely frustrated with the treatment process and have come to predict failure; this morbid prediction then becomes a self-fulfilling prophecy. With this latter constellation (as opposed to the high self-esteem psychopathic role-player discussed above), the value of a variety of social (e.g., group-based) interventions becomes apparent.
Additionally, repeated administrations of the two-grid methodology could provide useful information about treatment progress. Changes in the psychopathy index relates to the stability of the psychopathy construct and adherence (or lack of adherence) to the psychopathy role. The social constructions of peers along the psychopathy dimension could be monitored as an indicator of individual change. Finally, changes in the interaction between the psychopathy self-construction and patterns of positive and negative peer identification could be valuable data for the clinician.  
Clinicians will continue to measure ‘psychopathy’ in adolescents. Only with measures that are consistent with constructivist values and approaches will the great benefits of constructivist theory be brought to bear in this important domain. The present paper represents a first step toward introducing socially relevant, ecologically sensitive measures of psychopathic characteristics as a way to bring constructivist psychology to the adolescent forensic table.

Achenbach, T. M. (2000). Assessment of psychopathology. In A. Sameroff, M. Lewis, & S. Miller (Eds.). Handbook of developmental psychopathology (p. 41-56). New York: Kluwer Academic/Plenum Publishers.
Adams-Webber, J. (1970). Elicited versus provided constructs in repertory grid technique: A review. British Journal of Medical Psychology, 43, 349-354.
Bannister, D. & Fransella, F. (1966). A grid test of schizophrenic thought disorder. British Journal of Social and Clinical Psychology, 5, 95-102.
Blackburn, R. (1993). Clinical programs with psychopaths. In K. Howells & C. Hollin (Eds.), Clinical approaches to mentally disordered offenders (pp. 179-208). Chichester, England: Wiley.
Cottle, C. C., Lee, R. J., & Heilbrun, K. (1999). The prediction of criminal recidivism in juveniles: A meta-analysis. Paper presented at the annual conference of the American Psychological Association. Boston: MA.
Dishion, T. J. & Patterson, G. R. (1999). Model building in developmental psychopathology: A pragmatic approach to understanding and intervention. Journal of Clinical Child Psychology, 4, 502-512.
Edens, J. F., Skeem, J. L., Cruise, K. R., & Caufmann, E. (2000). Assessment of “juvenile psychopathy” and its association with violence: A critical review. Behavioral Sciences and the Law, 18, 1-28.
Farrington, D. P. (1983). Offending from 10 to 25 years of age. In K. Van Dusen & S. Mednick (Eds.). Prospective studies of crime and delinquency (p. 17-38). Boston: Kluwer-Nijhoff.
Foote, J. (1997, October). Expert panel issues report on serious and violent juvenile offenders. Washington DC: Office of Juvenile Justice and Delinquency Prevention.
Forth, A. E. & Mailloux, D. L. (2000). Psychopathy in youth: What do we know? In C. Gacono (Ed.). The clinical and forensic assessment of psychopathy (pp. 25-54). Mahweh, NJ: LEA.
Forth, A. E., Kosson, D., S., & Hare, R. D. (2003). The Psychopathy Checklist: Youth Version Technical Manual. Toronto: Multi-Health Systems.
Frick, P. J., & Hare, R. D., (2001). Antisocial Process Screening Device. Toronto: Multi-Health Systems.
Frick, P. J., Barry, C. T., & Boudin, S. D. (2000). Applying the concept of psychopathy to children: Implications for the assessment of antisocial youth. In C. Gacono (Ed.). The clinical and forensic assessment of psychopathy (pp. 3-24). Mahweh, NJ: LEA.
Grisso, T. (1998). Forensic evaluations of juveniles. Sarasota, FL: Professional Resources Press.
Hare, R. D. (1996). Psychopathy: A clinical construct whose time has come. Criminal Justice and Behavior, 23, 25-54.
Hare, R. D. (1998). Psychopathy, affect, and behavior. In D. Cooke, A. Forth, & R. Hare (Eds.), Psychopathy: Theory, research, and implications for society (pp. 105-138). Dordrecht, Netherlands: Kluwer Academic Publishers.
Hoge, R. D., (2001). The juvenile offender: Theory, research, and applications. Boston: Kluwer.
Kelly, G. A. (1955). The psychology of personal constructs (2 vols.). New York: Norton.
Loeber, R. (1991). Antisocial behavior: More enduring than changeable? Journal of the American Academy of Child and Adolescent Psychiatry, 30, 393-397.
Loeber, R. & Farrington, D. P. (2000). Young children who commit crime: Epidemiology, developmental origins, risk factors, early interventions, and policy implications. Development and Psychopathology, 12, 737-762.
Losel, F. (1998). Treatment and management of psychopaths. In D. Cooke, A. Forth, & R. Hare, (Eds.), Psychopathy: Theory, research, and implications for society (pp. 303-374). Dordrecht, Netherlands: Kluwer Academic Publishers.
Lynam, D. R. (1996). Early identification of chronic offenders: Who is the fledgling psychopath? Psychological Bulletin, 120, 209-234.
Lynam, D. R. (1997). Pursuing the psychopath: Capturing the fledgling psychopath in a nomological net. Journal of Abnormal Psychology, 106, 425-438.
Mitterer, J. & Adams-Webber, J. (1988). OMNIGRID: A program for the construction, administration, and analysis of repertory grids. In J. C. Mancuso & M. L. G. Shaw (Eds.), Cognition and personal structure: Computer access and analysis (pp. 91-105). New York: Praeger.
Moffit, T. E. (1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674-701.
Ogloff, J., Wong, S., & Greenwood A. (1990). Treating criminal psychopaths in a therapeutic community program. Behavioral Sciences and the Law, 8, 81-90.
Pierce, D. L., Sewell, K. W., & Cromwell, R. L. (1992). Schizophrenia and depression: Construing and constructing empirical research. In R. A. Neimeyer & G. J. Neimeyer (Eds.), Advances in personal construct psychology (Volume 2) (pp. 151-184). Greenwich, CT: JAI Press.
Raskin, J. D. & Lewandowski, A. M. (2000). The construction of disorder as human enterprise. In R. A. Neimeyer & J. D. Raskin (Eds.), Constructions of disorder: Meaning-making frameworks for psychotherapy (pp. 15-40). Washington, DC: American Psychological Association Press.
Sewell, K. W., Cromwell, R. L., Adams-Webber, J., & Mitterer, J. (1991). OMNIGRID-PC: A new development in computerized repertory grids. International Journal of Personal Construct Psychology, 4, 175-192.
Serin, R. (1995). Treatment responsivity in criminal psychopaths. Forum on Corrections Research, 7, 23-26.  
Slater, P. (1972). The measurement of consistency in repertory grids. British Journal of Psychiatry, 45, 45-51.
Snyder, H. N. & Sickmund, M. (1999). Juvenile offenders and victims: 1999 national report. National Center for Juvenile Justice.
Wright, J. C., Zakriski, A. L., & Drinkwater, M. (1999). Developmental psychopathology and the reciprocal patterning of behavior and environment: Distinctive situational and behavioral signatures of internalizing, externalizing, and mixed-syndrome children. Journal of Consulting and Clinical Psychology, 67, 95-107.

Kenneth W. Sewell, Ph.D., Professor and Director of Clinical Training at the University of North Texas in Denton, Texas. His current areas of focus include the study and treatment of persons who have experienced traumatic events, constructivist assessment methods, and the evaluation of competency to stand trial.
email: sewellk@unt.edu
Keith R. Cruise, Ph.D., MLS, Assistant Professor of Forensic-Clinical Psychology, Sam Houston State University, is now serving as the Research Coordinator and Assistant Clinical Professor of Public Health within the Louisiana State University Health Sciences Juvenile Justice Program. His current clinical and research interests include the assessment and treatment of juvenile offenders with a specific focus on assessing risk/needs and using this information to guide treatment. 


Sewell, K. W., Cruise, C. R. (2004). Adolescent psychopathy and repertory grids: preliminary data and focused case study. Personal Construct Theory & Practice, 1, 92-103
(Retrieved from http://www.pcp-net.org/journal/pctp04/sewell04.html)

Received: 14 Oct 2004 – Accepted: 13 Dec 2004 - Published: 30 Dec 2004

ISSN 1613-5091

Last update 30 December 2004