adaptive behavior, mental retardation, and the death penalty
TRANSCRIPT
This article was downloaded by: [University of Alberta]On: 15 October 2014, At: 16:24Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK
Journal of Forensic Psychology PracticePublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/wfpp20
Adaptive Behavior, Mental Retardation, and the DeathPenaltyKay B. Stevens EdD a & J. Randall Price PhD ba Texas Christian University, School of Education, Alice Neeley Special Education Institute ,201G Bailey Building, Fort Worth, TX, 76129 E-mail:b Forensic Psychology Associates of Texas , 1221 Abrams Road, #109, Richardson, TX, 75081E-mail:Published online: 11 Oct 2008.
To cite this article: Kay B. Stevens EdD & J. Randall Price PhD (2006) Adaptive Behavior, Mental Retardation, and the DeathPenalty, Journal of Forensic Psychology Practice, 6:3, 1-29, DOI: 10.1300/J158v06n03_01
To link to this article: http://dx.doi.org/10.1300/J158v06n03_01
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
AREA REVIEW
Adaptive Behavior, Mental Retardation,and the Death Penalty
Kay B. Stevens, EdDJ. Randall Price, PhD
ABSTRACT. The article focuses on the assessment of adaptive behav-ior as a variable in the diagnosis of mental retardation in defendantscharged with capital murder. A brief history of the development of adap-tive behavior assessment as a major component in the determination ofmental retardation is provided. Current adaptive behavior measurementinstruments used in the assessment of mental retardation in adults are de-scribed and analyzed in terms of their response formats, informant op-tions, and psychometric properties. Issues, appropriate practices, andpotential problems related to the assessment of adaptive behavior in death
Kay B. Stevens is affiliated with Texas Christian University, School of Education, Al-ice Neeley Special Education Institute, 201G Bailey Building, Fort Worth, TX 76129(E-mail: [email protected]).
J. Randall Price is affiliated with Forensic Psychology Associates of Texas, 1221Abrams Road, #109, Richardson, TX 75081 (E-mail: [email protected]).
Address correspondence to: Kay B. Stevens, Texas Christian University, School ofEducation, Alice Neeley Special Education Institute, 201G Bailey Building, FortWorth, TX 76129 (E-mail: [email protected]).
Journal of Forensic Psychology Practice, Vol. 6(3) 2006Available online at http://jfpp.haworthpress.com
© 2006 by The Haworth Press, Inc. All rights reserved.doi:10.1300/J158v06n03_01 1
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
penalty cases are discussed. Recommendations for current forensic evalu-ations as well as future research in the area of adaptive behavior assess-ment in death penalty cases are described. doi:10.1300/J158v06n03_01[Article copies available for a fee from The Haworth Document Delivery Ser-vice: 1-800-HAWORTH. E-mail address: <[email protected]>Website: <http://www. HaworthPress.com> © 2006 by The Haworth Press, Inc.All rights reserved.]
KEYWORDS. Adaptive behavior, mental retardation, death penalty,assessment
Two U.S. Supreme Court decisions focus on the issue of mental re-tardation and the death penalty. In Penry V. Lynaugh (1989), the U.S.Supreme Court held that the sentencing jury had not been given ade-quate opportunity to consider the mitigating evidence of Penry’s mentalretardation and its relevance to his culpability. While falling short ofbanning the execution of individuals with mental retardation, the deci-sion in Penry acknowledged the notion that mental retardation lessensthe culpability of a criminal defendant. In Atkins v. Virginia (2002), theU.S. Supreme Court expanded the Penry decision and banned the exe-cution of individuals with mental retardation by ruling it a violation ofthe Eighth Amendment, which prohibits cruel and unusual punishment.
According to the majority opinion, the Atkins decision banning theexecution of individuals with mental retardation was based primarily onan evolving national consensus as evidenced by state legislation trendsand opinion polls. Also, the Atkins court held that the execution of indi-viduals with mental retardation did not contribute to either the deterrenceor retribution goals of the criminal justice system and that individualswith mental retardation are categorically less culpable than offenderswithout mental retardation. The lessened culpability of individualswith mental retardation is related to several functional limitations thatinclude the reduced ability to: (a) understand and process information;(b) communicate; (c) abstract from mistakes and learn from experience;(d) engage in logical reasoning; (e) control impulses; (f) understand thereactions of others; and (g) lead rather than follow.
The U.S. Supreme Court left psychological procedures and legal pro-cesses related to Atkins decisions up to the States resulting in potentiallydiffering diagnostic criteria across jurisdictions. Most diagnostic crite-ria for mental retardation require significantly subaverage intelli-gence, adaptive behavior deficits, and an onset prior to age 18. Each of
2 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
these components requires careful assessment and consideration in im-portant decisions related to the lives of individuals with mental retarda-tion. The required assessment of adaptive behavior for determiningmental retardation in death penalty cases will be the focus of this article.
A BRIEF HISTORY OF ADAPTIVE BEHAVIOR DEFINITIONS
The American Association on Mental Retardation (AAMR) and theAmerican Psychiatric Association promulgated the two most frequentlyused diagnostic references for mental retardation. The AAMR is thelargest professional organization in the world devoted to the study ofmental retardation and since 1921 has published 10 editions of the textMental Retardation, the latest edition being published in 2002. In 1952,the American Psychiatric Association began publishing what became aseries of periodically updated manuals referred to as the Diagnostic andStatistical Manual of Mental Disorders (DSM) (APA, 1952). The mostcurrent DSM is the Diagnostic and Statistical Manual of Mental Disor-ders, Fourth Edition, Text Revision (DSM-IV-TR) (APA, 2000).
The AAMR began defining mental retardation in 1908. Over the past94 years, ten versions of the definition have been published. In 1959,two significant changes were made to the definition including settingthe cut-off score on IQ tests to more than one standard deviation belowthe mean and considering adaptive behavior as a component for diagno-ses. At that time, adaptive behavior was broadly defined and measuredglobally. According to the AAMR, adaptive behavior measurementoriginally was included in the definition of mental retardation, at least inpart, to decrease the number of individuals incorrectly identified byclarifying the nature of the condition and decreasing dependence on IQscores (AAMR, 2002). In 1973, the cut-off score on IQ tests was reset attwo or more standard deviations below the mean existing concurrentlywith deficits in adaptive behavior. For the first time, adaptive behaviorbecame a requirement rather than just a consideration in the diagnosis ofmental retardation, even though IQ scores continued to have the great-est weight. In 1983, the standard score cut-off for diagnosis of mentalretardation on IQ tests was placed at 70 to 75 to include consideration ofthe standard error of measurement. Concurrent impairments in adaptivebehavior remained a requirement of the 1983 diagnostic definition(AAMR, 2002).
The 1992 definition, which remains in broad use today, conceptual-ized adaptive behavior as consisting of ten specified skill areas:
Area Review 3
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
Mental retardation refers to substantial limitation in present func-tioning. It is characterized by significantly subaverage intellectualfunctioning, existing concurrently with related limitations in twoor more of the following applicable adaptive skills areas: commu-nication, self-care, home living, social skills, community use,self-direction, health and safety, functional academics, leisure,and work. Mental retardation manifests before age 18. (AAMR,2002, p. 22)
Both the 1983 and the 1992 definitions remain in use today in manystates. The latest definition, published in 2002, re-conceptualized adap-tive behavior into three broad clusters:
Mental retardation is a disability characterized by significant limi-tations both in intellectual function and in adaptive behavior as ex-pressed in conceptual, social, and practical adaptive skills. Thisdisability originates before age 18. (AAMR, 2002, p. 8)
Deficient adaptive behavior criterion was defined as two or morestandard deviations below the mean on one of the three clusters or on acomposite score encompassing all the clusters (AAMR, 2002).
Like the AAMR, and in many ways based on the AAMR, the DSMgradually has amended and expanded the definition and assessment rec-ommendations related to adaptive behavior. For example, in the originalDSM (APA, 1952), the term “adaptive behavior” was not used. Instead,the following guidelines were provided: the “degree of defect is estimatedfrom other factors than merely psychological test scores, namely, con-sideration of cultural, physical and emotional determinants, as well asschool, vocational, and social effectiveness” (APA, 1952, p. 24). By thepublication of the DSM-III (APA, 1980), a global definition of adaptivebehavior remained but the use of standardized adaptive behavior ratingscales was discouraged because the existing measurement tools wereconsidered unreliable and invalid for measuring the construct without astrong and necessary emphasis on clinical judgment. Seven years fol-lowing the publication of the DSM-III, the DSM-III-R (APA, 1987)listed specific areas of adaptive behavior and endorsed the limited useof adaptive behavior rating scales combined with clinical judgment.Currently, the DSM-IV-TR (2000) contains a somewhat expanded de-scription of adaptive behavior:
4 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
The essential feature of mental retardation is significantly subaveragegeneral intellectual functioning (Criterion A) that is accompanied bysignificant limitations in adaptive functioning in at least two of thefollowing skill areas: communication, self-care, home living, so-cial/interpersonal skills, use of community resources, self-direction,functional academic skills, work, leisure, health, and safety (Crite-rion B). The onset must occur before age 18 years. (Criterion C)(DSM-IV-TR, 2000, p. 41)
Since 1959 when the definition of mental retardation first includedthe assessment of adaptive behavior, a consensus on the empirical fac-tor structure of adaptive behavior has been elusive (Harrison, 1987;McGrew & Bruinicks, 1989). Nevertheless, in 1992, two deficient skillareas out of 10 specified areas became the AAMR’s adaptive behaviorcriteria for diagnosing mental retardation (Luckasson et al., 1992). TheAmerican Psychiatric Association followed suit and adopted the sameadaptive behavior criteria in the DSM-IV (APA, 1994). The selection of10 adaptive domains was based on subjectively determined organiza-tion rather than empirical research (Heal & Tasse, 1999). In fact, neverhave factor analytic studies identified the 10 domains (Spreat, 1999). Intwo scathing critiques, the 1992 AAMR guidelines for the assessmentof adaptive behavior functioning were labeled inappropriate for deci-sion-making (MacMillan, Gresham, & Siperstein, 1993, 1995). Severalspecific criticisms were offered. First, and most importantly, no adap-tive behavior rating scale measured all 10 adaptive behavior skills in apsychometrically valid and reliable manner, and no empirical basis wasprovided for the 2 out of 10 criteria. Second, the construct and measure-ment of adaptive behavior easily detected deficiencies in individualswith moderate to severe mental retardation, but the deficiencies weremore difficult to discriminate in higher functioning individuals. Third,developmental considerations were not incorporated. For example, achild cannot have adaptive limitations in the adaptive skill area of work,even though work is a specified area. Fourth, to be meaningful, criticsbelieved that adaptive behavior assessment should be conducted only inrelation to the environment to which the individual must adapt.
In 2002, the AAMR asserted that results from factor analytic studiesindicated an “emerging consensus” concerning the underlying structureof the adaptive behavior construct. Three factor clusters were identified:conceptual, social, and practical skills. The conceptual skill cluster in-cluded: language, reading, writing, money concepts, and self-direction.The social skill cluster included: interpersonal, responsibility, self-es-
Area Review 5
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
teem, gullibility, naiveté, rule compliance, law compliance, and victim-ization avoidance. The practical skill cluster included: activities of dailyliving, instrumental activities of daily living, occupational skills, andsafe environmental maintenance. One additional factor emerged butwas not included in the AAMR 2002 diagnostic criteria–motor or physi-cal competence (Schalock, 1999a).
Considering the changing definitions presented in this brief history, itis interesting to note that the diagnoses of moderate to profound mentalretardation are rarely determined by a detailed measurement of adaptivebehavior. Limited and deficient adaptive behaviors clearly accompanythese levels of disability and typically are not difficult to recognize re-gardless of the diagnostic criteria used. Individuals with moderate toprofound retardation are benefited most by adaptive behavior measuresin the area of individual program development and progress monitoringrather than diagnosis. Determining mild mental retardation appears todrive the search for precise definitions and assessment processes due toits diagnostic dependence on adaptive behavior measures. Individualswith IQ scores in the borderline to mild range of mental retardation of-ten demonstrate variable levels of adaptive behavior abilities makingtheir diagnosis difficult. These ranges are those at issue in determiningmental retardation in capital murder defendants.
THE ASSESSMENT OF ADAPTIVE BEHAVIOR AND REVIEWOF SPECIFIC ADAPTIVE BEHAVIOR RATING SCALES
Traditionally, the assessment of adaptive behaviors consists of inter-views or questionnaires completed by informants (e.g., parents, teach-ers, caregivers) who have considerable knowledge of the target person.Informants are asked to subjectively rate the target person’s typicaladaptive behavior in settings common to the target person, such as athome, school, or work. Adaptive behavior rating scales are intended tomeasure a target person’s actual performance in the natural environ-ment rather than the person’s ability to perform certain adaptive behav-iors. Rating systems include response options such as always, some-times, or never performs the behavior when needed. Informant ratingsof adaptive behavior are subjective estimations of typical performance.For example, an adaptive behavior rating scale might ask the informantto rate the target person’s frequency and independence of finding rest-rooms in public places.
6 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
The psychometric data upon which adaptive behavior rating scalescores are based must conform to the same rigor as other norm-refer-enced standardized tests (American Education Research Association(AERA), American Psychological Association (APA), National Coun-cil on Measurement in Education (NCME), 1999). For example, inter-nal consistency and inter-rater reliability coefficients should meet aminimum of .90 when used for important decision-making (Kelly,1927; Nunnally & Bernstein, 1994; Salvia & Ysselkyke, 1998). Evi-dence of validity is required for all standardized psychometric instru-ments including adaptive behavior rating scales (AERA, APA, NCME,1999). Both the Specialty Guidelines for Forensic Psychologists (1991,Section VI. A.) and the APA Ethical Principles of Psychologists andCode of Conduct (2002, Section 9.02) direct forensic examiners toconsider these psychometric qualities when selecting an assessmentinstrument. The remainder of this section will review adaptive behaviorassessment instruments commonly used with adults and offer conclu-sions regarding their appropriateness.
The major resource that guided our selection of the adaptive behaviorrating scales reviewed below was the excellent review of adaptive be-havior by Reschley, Myers, and Hartel (2002) who described five out ofmore than 200 published adaptive behavior assessment instruments thathave a strong research and practice base. We address three of these fiveinstruments considered to be “adaptive behavior scales with well knownproperties” (Reschley et al., 2002, p. 166). The instruments reviewed inthis article are the: (a) Vineland Adaptive Behavior Scale; (b) AdaptiveBehavior Assessment System-II; and (c) Scales of Independent Behav-ior-Revised. The remaining two instruments identified by Reschly et al.are the Adaptive Behavior Scale-Residential and Community, SecondEdition (ABS-RC:2) and the Battelle Developmental Inventory. Thesetwo instruments were judged to be inappropriate for use in diagnosingmental retardation with an adult population. The ABS-RC:2 was devel-oped for use with individuals up to 79 years of age, but the normativesample is not representative of the different levels of mental retardation(mild, moderate, and severe), and norms are not available for adults withtypical functioning (Harrison, 1998). Reschly et al. also concluded that“the ABS-RC:2 may not fit the psychometric criteria used in determining adiagnosis of mental retardation according to AAMR requirements” (p. 169).The Battelle Developmental Inventory is a developmental scale appro-priate only for children from birth to age 8, and thus inappropriate foruse with adults.
Area Review 7
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
Vineland Adaptive Behavior Scale. The Vineland Adaptive BehaviorScale (VABS) (Sparrow, Balla, & Cicchetti, 1984) was selected for re-view here for several reasons. First, in addition to its identification byReshley, Myers and Hartel (2002), the DSM-IV-TR lists the VABS as apotential instrument for the assessment of adaptive behavior to deter-mine mental retardation. Second, the VABS has a twenty-year historyof use by professionals in the determination of mental retardation.Third, the professional literature includes research and descriptive arti-cles in which the VABS is administered in the diagnosis of adult mentalretardation in legal contexts (e.g., Devault, 1987; Baroff, 1990; Baroff,1991; Beail, 2003). Fourth, the authors have observed the use of theVABS as a measure of adaptive behavior with adults when determiningmental retardation in death penalty cases. However, as will be describedsubsequently, the authors do not recommend the use of the VABS in thedetermination of mental retardation in defendants 19 years of age orolder.
Three versions of the VABS are available: a survey form, an ex-panded form, and a classroom form. The survey form (VAB-S) is rec-ommended by AAMR (2002) for diagnostic purposes and is the focusof this review. The VABS-S was standardized on 3,000 individualsages birth to 18 years, 11 months. The selection of the standardizationsample was based on the 1980 U. S. Census. The VABS-S manual re-ports that it can be used with a child of any age and with “low function-ing adults.” The VABS manual indicates that the normative tables forages 18-8-0 through 18-11-30 should be used for all individuals whosechronological age is 19 or above. The administration time of the VABS-Sis one hour or less. When administering the VABS-S, the interviewerasks general, open-ended questions of an informant and subjectively es-timates a rating based on the informant’s responses concerning whetheror not the individual does perform rather than can perform the behaviorin question. The rating options include: 1 (no, never), 2 (sometimes,partially), 3 (yes, usually), N (no opportunity to perform behavior), orDK (informant does not know). The VABS-S contains 297 items and isorganized around five adaptive domains, including: (a) Communica-tion; (b) Daily Living Scales; (c) Socialization; (d) Motor Skills; and (e) Mala-daptive Behavior. When all domains are administered, an Adaptive Be-havior Composite may be obtained for individuals who are older thansix years and not physically handicapped by combining all of the do-mains except the Maladaptive Behavior Domain and the Motor SkillsDomain. The results of the ratings are converted to standard scores, per-centiles, adaptive level, and age-equivalent scores.
8 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
According to the VABS-S manual, split-half reliability coefficientsfor the Adaptive Behavior Composite and the Domains range from .70to .98. Test-retest reliability coefficients range from .77 to .92. Inter-rater reliability coefficients range from .62 to .78.
Support for construct validity is described in terms of developmentalprogression, factor analysis of Domains and Sub-domains, and profilesof scores for supplementary groups. Mean raw scores are shown to in-crease with chronological age from 2-0 to 8-11, but not for the oldestage group in the normative sample (16-0 to 18-11). Supplementarynorm groups with visually-impairment, hearing-impairment, emotionaldisturbance, and mental retardation all scored “well below” the nationalstandardization sample with the mental retardation group “near the bot-tom.” No IQ data is reported for either the residential or the non-residen-tial mental retardation sample. Criterion-related validity data consist ofcorrelations between the VABS-S Composite Scores and the originalVineland Social Maturity Scale (.55) and the Adaptive Behavior Inven-tory for Children (.58). Correlations between the VABS-S DomainScores and the AAMD Adaptive Behavior Scale raw scores ranged be-tween .40 and .70. Correlations between the VABS-S Domain Scoresand several measures of intelligence ranged from .12 to .49, supportingthe notion that the VABS-S measures different areas of functioning thanintelligence tests. No other independent criterion-related or predictivevalidity data were reported.
Sattler (1989) found the VABS to be potentially useful but expressedconcern with the norming procedures and requiring information un-likely to be available from informants. The AAMR (2002) noted thatthe VABS Domains are transferable to the re-conceptualized adaptivebehavior factor structure of conceptual, social and practical skills.Reschly, Myers, and Hartel (2002) report the “data from reliability andvalidity studies of the survey form are very impressive, especially inlight of the flexible conversational procedures used for obtaining infor-mation” (p. 168); however, Salvia and Ysseldyke (1998) found severaldeficiencies with the VABS: (a) 79% of the split-half reliability coeffi-cients are .90 or less; (b) 50% of the test-retest coefficients are .90 orless; and (c) inter-rater reliability for two of the four domains are lessthan .87.
Beail (2003) pointed out that individuals age 19 and older are com-pared to a group of approximately 22 individuals 18-8-0 to 18-11-30.This means when using the VABS with adult populations, the adults arecompared to a group of approximately 22 individuals who were in thelatter part of their 18th year of life in the early 1980s. Beail concluded
Area Review 9
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
that the VABS “has some major psychometric deficiencies when usedwith adults.” He considered the reliability data to fall between satisfac-tory and high, but offered the opinion that the disadvantages of theVABS now outweighed the advantages. The disadvantages include thelack of adequate norms for adults, the lack of comparability of the sup-plemental norms to similar groups today, and the general outdated na-ture of the items in the daily living skills area. The VABS is currentlyundergoing a revision and restandardization with individuals from birthto age 70-plus (AAMR, 2002).
Adaptive Behavior Assessment System-II. The Adaptive Behavior As-sessment System-II (ABAS-II) (Harrison & Oakland, 2003) is a revisionof the original ABAS (Harrison & Oakland, 2000). The ABAS-II re-tains all the features of the ABAS. New features are: (a) rating forms,normative data, and validity coefficients for children ages birth to 5years, and (b) the inclusion of an assessment of the AAMR (2002) threedomains of adaptive behavior. Five forms are available: three forms forchildren of different age groups (0 to 5 years, 2 to 5 years, and 5 to 21years) that are rated by either primary caregivers or teachers and twoversions for adults (16 to 89 years) consisting of the adult form-rated byothers and the adult form-self report. The focus of this review will be thetwo versions of the adult form. The rated-by-others version and theself-report version of the ABAS-II were standardized on 920 and 990respondents, respectively, with overlap in the two samples. The selec-tion of the standardization sample was based on the 2000 U.S. Census.The administration time is approximately 20 minutes if the respondenthas adequate reading skills. Respondents must be very knowledgeableabout the adaptive behavior of the individual being evaluated. Gener-ally, respondents should have: (a) contact almost every day with the in-dividual; (b) contacts of several hours each contact; (c) contact duringthe past 1 to 2 months, and (d) opportunities to observe the variety of be-haviors measured by the ABAS-II. In both the rated-by-others and theself-report versions, the respondent completes the rating scale by eitherreading it himself or herself or, if reading skills are less than 5th gradelevel, the items may be read aloud to the respondent. The stimulus andresponse formats are the same in the self-report as in the rated-by-othersversion. The rating form asks the respondent to rate how often the in-dividual displays the behavior independently when appropriate orneeded. The rating options include: 0 (is not able), 1 (never or almostnever), 2 (sometimes when needed), and 3 (always or almost always).The respondent is also asked to check a box if the rating was a guess.The ABAS-II contains 239 items comprising the 10 adaptive behavior
10 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
skill areas recommended by the 1992 AAMR definition and the DSM-IVdiagnostic criteria. Each of the adaptive behavior skill areas contain 20to 27 items per skill area. The specific skill area scores are combinedyielding composite scores for the General Adaptive Composite (GAC)and for the three AAMR 2002 adaptive behavior domains: conceptual,social, and practical. Skill area raw scores are transformed into scaledscores, and the GAC and domain scores are converted to standardscores and percentiles. An assessment of response validity is obtainedby counting the number of guessed items. If the respondent guesses onfour or more items in any skill area, the manual recommends interview-ing the respondent to evaluate whether or not the respondent can serveas a reliable rater.
The technical psychometric data are reported in the ABAS-II manualfor adults for both rated-by-others and self-report versions. Internalconsistency coefficients for the rated-by-others version range from .86to .99. Test-retest reliability coefficients for the rated-by-others versionrange from .91 to .96. Inter-rater reliability coefficients for multiple rat-ings by others range from .76 to .93. Internal consistency coefficientsfor the self-report version range from .82 to .99. Test-retest reliabilitycoefficients for the self-report version range from .91 to .99. Inter-raterreliability coefficients between the rated-by-others version and the self-rated version range from .84 to .95. No reliability data were reported forthe sample of mentally retarded adults for either rated-by-others or theself-report versions.
Content validity coefficients for the intercorrelations of skill areasand the Adaptive Domains (Conceptual, Social, Practical) for the rated-by-others version range from .69 to .84. Content validity coefficientsfor the intercorrelations of skill areas with the GAC for the rated-by-others version range from .74 to .90. Content validity coefficients forthe intercorrelations of skill areas and the Adaptive Domains (Concep-tual, Social, Practical) for the self-report version range from .66 to .82.Content validity coefficients for the intercorrelations of skill areas withthe GAC for the self-report version range from .69 to .85. Construct va-lidity in the form of factor analytic data indicate that the ABAS-II pro-vides a good fit for a one-factor, global factor of adaptive functioning,and a close fit for the three-factor model proposed by the AAMR(2002). Criterion-related validity data consists of correlations betweenthe ABAS- II GAC Scores and the VABS Composite Scores and rangefrom .75 to .84. Correlations between the VABS Domains and ABAS-IIDomains range from .58 to .82. Correlation between the ABAS-II GACScores and the SIB-R Broad Independent Score is .59. None of the
Area Review 11
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
above criterion-related validity coefficients used either adult version ofthe ABAS- II. The ABAS-II GAC and Domain Scores correlate withthe WAIS-III Full Scale IQ at a range from .24 to .67. ABAS-II clinicalvalidity data revealed significant differences between the mean GACScores for adults with mental retardation and a matched control groupon the rating-by-others version. No other independent criterion-relatedor predictive validity data were reported for either the rated-by-other orthe self-report adult versions of the ABAS-II.
AAMR (2002) gave a positive rating for the psychometric propertiesof the ABAS but stated that the ABAS is “too new to have beencritiqued by others” (p. 90). Similarly, Reschly, Myers, and Hartel(2002) found the ABAS to be psychometrically sound and to have“good potential for assessing adaptive behavior for diagnostic purposes”(p. 170). Burns (2005) reviewed the ABAS-II positively, commenting onits sound theoretical and empirical basis, large and representative normgroup, and impressive reliability and validity. The review judged theABAS-II to be technically superior and recommended it for use in eligibil-ity and entitlement decisions. However, we see the lack of peer-reviewedresearch on the ABAS-II as well as the absence of criterion-related va-lidity for the adult version as deficiencies.
Scales of Independent Behavior-Revised (SIB-R).The original Scalesof Independent Behavior (Bruininks, Woodcock, Weatherman, & Hill,1984) was revised and published as the SIB-R in 1996 (Bruininks,Woodcock, Weatherman, & Hill, 1996). Two forms of administrationare available: an interview procedure and a checklist procedure. In theinterview procedure, a respondent familiar with the target person beingevaluated is interviewed about the target person’s adaptive behavior.When the checklist rating procedure is used, the respondent completesthe rating independently by answering the items directly in the responsebooklet. The SIB-R also allows the target person to provide the ratingsby self-report. The SIB-R was standardized on 2,182 individuals rang-ing in age from 3 months to 90 years. The selection of the standardiza-tion sample was based on the 1990 U.S. Census. The task of the rater isconfounded in that each item includes three rating criteria, each contain-ing two considerations: (1) how well the target person (does or coulddo) the task; (2) how often the target person does (or could do) the task;and, (3) the degree of independence with which the target person does (orcould do) the task. The SIB-R contains 259 items that are organized into14 subscales containing 16 to 20 items each that are further organized intofour clusters containing 2 to 5 subscales each. The clusters are: (1) MotorSkills; (2) Social Interaction and Communication; (3) Personal Living;
12 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
and (4) Community Living. A total score called Broad Independencecan be obtained. A Problem Behavior Scale also is included. Raw scoresare transformed into several types of relative standing scores includingstandard scores, percentiles, age-equivalent scores, and adaptive behav-ior skill level.
Split-half reliability coefficients for cluster scores obtained for adultsrange from .91 to .98. No test-retest reliability coefficients were re-ported for individuals over the age of 13. Split-half reliability coeffi-cients range from .55 to .92 on subscales for adults. No inter-rater relia-bility coefficients were reported for individuals over the age of 21.
Construct validity reported for the SIB-R includes its correlationswith the original SIB with correlations ranging from .50 to .99. Corre-lation coefficients between SIB-R scores and chronological age iscurvilinear and developmentally predictable. SIB-R scores for develop-mentally disabled groups, including adults with mild retardation, are re-ported but the differences were not statistically significant between thegroups with disabilities and matched samples of individuals withoutdisabilities. For adults, correlation coefficients for SIB-R Subscales andSIB-R Cluster Scores range from �.02 to .92. Criterion-related coef-ficients for the SIB-R Clusters and the Woodcock Johnson-RevisedBroad Cognitive Ability scores range from .01 to .45, revealing thepredictable lack of strong relationship between adaptive behaviorand intelligence. No other independent criterion-related or predic-tive validity data were reported for the SIB-R with adults.
The AAMR (2002), Reschly, Myers, and Hartel (2002), and Mac-Cow (2001) consistently reviewed the SIB-R as useful but not outstand-ing. The rating system is complicated and the hand-scoring is tedious.The Problem Behavior Scale appears positively received in these re-views. Zlomke (2001) found the SIB-R to be well-developed and psy-chometrically sound. Nevertheless, we find the lack of test-retest andinter-rater reliability data for adults remains problematic. Also, no sepa-rate psychometric data of any type were reported for the self-report ad-ministration format.
Conclusions Regarding Suitable Adaptive Behavior Measures forAdults. From our reviews, the ABAS-II and SIB-R instruments weredesigned and normed for adult populations and have basically adequatepsychometric properties, and therefore seem appropriate for obtainingadaptive behavior scores for adults when administered according to thestandardization procedures for informant responding. The psychometricreliability of the ABAS-II for adults is superior to the SIB-R, but crite-rion-related validity data for both instruments remain somewhat deficient.
Area Review 13
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
We do not recommend using either instrument’s self-report versionsince no empirical data were reported in the manuals to validateself-reporting with individuals functioning in the borderline to mildlymentally retarded ranges. In addition, we do not support the use of in-struments under conditions that violate the standardized administrationprocedures such as using an inappropriate informant.
THE ASSESSMENT OF ADAPTIVE BEHAVIORIN DEATH PENALTY CASES
Defining and measuring adaptive behavior in adults with IQ scoresfalling in the borderline to mildly mentally retarded ranges are difficult(Reschly, Myers, & Hartel, 2002). Additional problems emerge whenadaptive behavior is assessed in a death penalty case with the objective ofdiagnosing or ruling out mental retardation. This section reviews impor-tant issues in the assessment of adaptive behavior in death penalty cases.
Reviewing Records. The most useful information in Atkins caseslikely comes from documents generated prior to the capital crime inquestion including records from schools, employers, medical and men-tal health providers, and government agencies. Valid intelligence test-ing, adaptive behavior assessment, and other evaluation data reportedprior to the defendant’s 18th birthday that provides or fails to provide adiagnosis of mental retardation is invaluable (Ellis, 2003). That said, theimportance of locating and reviewing archival records cannot be over-stated even though school and agency records may be difficult to obtain.Counsel in death penalty cases should secure and serve subpoenas forarchived records to both the central school administration and the spe-cial education department in every school system attended by the defen-dant. An in-depth record review can lead to vital information regardingadaptive behavior strengths and weaknesses in previous life activities.If the defendant has a prior criminal record this also should be ob-tained and analyzed. Information obtained from records review can becombined with current information from family members, friends, co-workers, and others to strengthen clinical judgment and expert opinion(Brodsky & Galloway, 2003).
Selection of Rating Scales. The AAMR (2002) and Reschly, Myers,and Hartel (2002) provide guidelines for the selection of adaptive be-havior assessment instruments. These guidelines include considerationof the following: (a) determine the purpose of assessment (e.g., diagnosisor program planning), (b) assure technical adequacy (e.g., appropriate nor-
14 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
mative data and psychometric qualities), and (c) match characteristicsof the individual being assessed with the normative sample (e.g., age,level of functioning, setting, language).
Administration of Rating Scales. The Ethical Principles of Psycholo-gists and Code of Conduct (APA, 2002) requires that the administrationof all psychological instruments–which would include adaptive behav-ior rating scales–follow the exact procedures used in the standardizationof the instrument (APA, 2002.). For example, if the standardized ad-ministration dictates that a single informant familiar with the behaviorof the target individual complete the adaptive behavior rating scale, itwould be inappropriate to use multiple informants, record reviews, and/or self-report to obtain ratings and calculate scores for one adaptive be-havior instrument. The use of multiple sources is not the issue but ratherthe inappropriate administration of the instrument. In fact, adminis-tering independent rating scales to multiple informants may provideconvergent validity. In other words, several non-biased informants pro-viding similar adaptive behavior ratings strengthens the validity of theassessment findings. On the other hand, obtaining greatly divergentadaptive behavior ratings from multiple informants requires the evalua-tor to investigate the array of opportunities when informants observedthe defendant and/or to consider the possibility of biased ratings.
Forensic evaluations sometimes require a retrospective judgment ofa defendant’s mental status at an earlier time period in his or her life.Such evaluations involve the use of multiple sources of information andthe clinical judgment of the evaluator. No standardized methodologyfor the retrospective reconstruction of mental states exists; therefore,retrospective ratings of adaptive behavior should be carefully devel-oped with a basis of supportable fact (Simon & Shuman, 2002). The as-sessment of adaptive behavior is retrospective when informants areasked to recall the defendant’s typical adaptive behavior functioning atan earlier time in the defendant’s life. In Atkins cases, adaptive behaviorratings may refer to the developmental period, the time of the crime, orthe time of post-crime assessment. In Atkins appellate cases, it is not un-usual for an inmate to have been on death row for eight to ten years.Even at trial stage, the defendant may well have been incarcerated formore than a year before an evaluation is conducted. No adaptive behav-ior rating scale was normed with such a long-term retrospective admin-istration. In other words, no standardization procedure used in normingany adaptive behavior rating scale requires raters to think back and remem-ber how the target person behaved during the developmental period or attime years earlier. When such retrospective ratings are obtained, scores
Area Review 15
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
are not valid, and information gained in this manner should be used clin-ically and cautiously. Certainly the forensic evaluator should be veryclear in describing the referent time period used by the informant. Whennonbiased and credible informants are available, ratings for all relevanttime frames may be obtained and used clinically within limits to formu-late expert opinions. However, the use of scores from retrospective ad-ministrations of the instruments is strongly discouraged. Certainly,great care should be taken to describe the limitations of such data.
Selection of Informants. All adaptive behavior-rating scales use third-party informants as the primary source of data. Informants are typicallyinterviewed regarding their observations and knowledge of the individ-ual’s adaptive behavior. Thus, the accuracy of the results is only as goodas the accuracy of the ratings from the informant. An informant asked toparticipate in the adaptive behavior rating of an individual suspected ofhaving mental retardation is expected to be well-acquainted with thetypical behavior of the person over an extended time period in multiplesettings and be unbiased in terms of the outcome of the adaptive behav-ior rating assessment (Reschly, Myers, & Hartel, 2002). These require-ments can be difficult to meet even in the evaluation of a child, but whenassessing adults for death penalty eligibility, the selection of informantswill be more problematic. In the evaluation of children, parents andteachers are the most often selected informants and biases do not go tolife or death considerations. In a death penalty eligibility assessment,the potential bias of the defendant’s family members must be consid-ered. Correctional officers are sometimes selected as informal infor-mants, yet they may be biased regarding the death penalty, inhibited byinstitutional policies or peer pressure, and/or poorly informed. Formerteachers or employers may be appropriate informants, but gaining theircooperation often can be difficult, and their potential bias also should beconsidered.
Presentation of Results. The types of scores reported must illustratethe clearest, most accurate means for describing results. Reschly, Myers,and Hartel (2002) recommend the use of standard scores when reportingadaptive behavior results. Standard scores allow score comparisonswithin and across tests and are well suited for discussion and descrip-tion when numerous scores are presented. Percentiles also are helpfulfor describing performance of examinees due to their single purpose ofillustrating where an individual scores in relation to similar individualsin the normative sample. Using age-equivalent scores to report adaptivebehavior performance is highly discouraged. The psychological and edu-cational communities have criticized the use of age equivalent scores
16 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
(Salvia & Yssekdyke, 1998; Anastasi & Urbina, 1997). Although it maybe tempting to use age-equivalent scores due to their seemingly easy-to-understand value; in reality, age-equivalent scores are statistical esti-mations that are frequently misleading by promoting typologicalthinking and encouraging inaccurate imagery. For example, when a30-year-old death row inmate is reported as having adaptive behaviorsequivalent to a child four to eight years of age, judges and juries maytend to think of a stereotypical young child and the behaviors attributedto such a child, when in fact the defendant may appear and behave sig-nificantly different than the imagined child. In such cases, reportingage-equivalent scores can erode the credibility of the expert witness. Itis the duty of any testifying clinical psychologist to assist judges and ju-ries to understand scores that most appropriately report findings (e.g.,standard scores, percentile scores) and to avoid using scores that are lesspsychometrically sound and potentially misleading (AERA, APA, NCME,1999).
OTHER SOURCES OF ADAPTIVE BEHAVIOR ASSESSMENTS
Self-Report. Whether used with an adaptive behavior rating scale orin a clinical interview, self-reports in forensic assessments are repletewith difficulties, especially considering the potential consequences in-herent in a criminal proceeding. To accept a criminal defendant’s self-report without skepticism is forensically naïve and constitutes a deviationfrom acceptable practice that may result in unreliable data (Rogers,1997). The criminal defendant may be blatantly malingering or feigningsevere adaptive behavior deficits to avoid the death penalty. On theother hand, the defendant may be inaccurately describing adaptive be-havior to avoid being labeled ignorant or mentally retarded. No researchdata were found concerning self-reported adaptive behavior assessmentof individuals involved in death penalty cases. No psychometric datafor individuals with mental retardation were reported on the self-reportversions of available adaptive behavior rating scales. Therefore, self-re-port adaptive behavior information should be cautiously considered andcorroborated when possible.
Other Psychometric Instruments. Normed instruments are availablethat directly measure adaptive abilities and knowledge, such as theStreet Survival Skills Questionnaire (SSSQ) (Giller, Dial, & Chan, 1986;Janniro, Sapp, & Kohler, 1994; Linkenhoker & McCarron, 1993). TheSSSQ is a normed-referenced instrument requiring the individual being
Area Review 17
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
assessed to choose between four visual stimuli in response to a verbalquestion related to various adaptive abilities such as telling time, using atelephone book, and using money. The SSSQ is less than comprehen-sive with limited norms, but the psychometric data is impressive, and ithas been recommended as an adjunctive assessment instrument of valueas one component of an adaptive behavior assessment (Haring, 1992).The Independent Living Scales (ILS) (Loeb, 1996) are designed to as-sess activities of daily living and functional competence in older adults,but Cohen (2001) suggested that a flexible clinical use of the ILS ismore appropriate than a rigid psychometric interpretation. Also, themental retardation normative sample is limited. The ComprehensiveTest of Adaptive Behavior-Revised (Adams, 1999) uses a combinationof third-party ratings and direct assessment of adaptive skills, but someof the content and tasks of this instrument do not appear to be appropri-ate for use in Atkins evaluations. Direct assessment of adaptive behaviorabilities need not be limited to norm-referenced instruments but mayalso include criterion-referenced instruments (Spreat, 1999). Examplesof such instruments include The Brigance Inventory of Essential Skills(Brigance, 1981), The Brigance Life Skills Inventory (Brigance, 1994),and The Brigance Employability Skills Inventory (Brigance, 1995).Certain neuropsychological tests that have been studied in terms of their“ecological validity” or their relationship to the daily activities of livingalso may be useful in assessing adaptive abilities (Long & Sbordone,1996). Certainly, the use of relevant norm-referenced psychologicaltests such as academic achievement is appropriate for the assessment ofsome adaptive skills.
Use of Facts of the Instant Offense. Controversy surrounds the use ofthe defendants behavior just prior to, during, and after the crime in anadaptive behavior analysis. For example, in Penry, the defendant con-fessed to having ridden his bicycle around the victim’s house before themurder to see if anyone else was present in the house or observing. This“planning ahead” was presented as evidence of intact adaptive behav-ior. Evidence such as this is typically presented through prosecutionwitnesses on direct examination or as hypothetical questions to defensewitnesses. Defense attorneys may object, citing that the prejudicial ef-fect of using the facts of the crime outweighs the probative value, but inmost jurisdictions the evidence likely will be admitted as relevant. Ofcourse, the facts of an offense might indicate a poorly planned and exe-cuted crime or one in which an intelligent co-defendant led a gullibledefendant.
18 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
Assessment Context. In Atkins cases, the defendant’s behavior whileincarcerated may be evaluated in terms of its adaptive quality. Duringan appeal case, the prosecution may focus on the inmate’s adaptation tolife in prison and specifically, on death row. While this may have intu-itive appeal, evidence such as this runs counter to a basic assumption ofproper adaptive behavior assessment-evaluation should occur in thecontext of the individual’s age peers in a typical community environ-ment. Consider the assessment conducted in special education for achild being served in a highly structured residential or educational pro-gram designed for youngsters with disabilities. The child’s adaptationto that environment should not be taken as evidence of his ability toadapt in a less structured, more typical environment. Similarly, an in-mate’s adaptation to the extremely restricted and structured environ-ment found on death row should not be taken as valid evidence of hisability to adapt to the free world. Life on death row allows limited expe-riences and few opportunities. On the other hand, forensic evaluatorsshould consider adaptation to prison incarceration as one factor of manyin their assessment of adaptive behavior.
CURRENT PROFESSIONAL PRACTICES:ADAPTIVE BEHAVIOR ASSESSMENT
IN DEATH PENALTY CASES
The importance of current practice studies and the development ofbest-practice guidelines cannot be over estimated. The literature relatedto current assessment practices by psychologists regarding issues indeath penalty cases is extremely limited. Information regarding the as-sessment practices in the measurement of adaptive behavior in Atkinscases is especially scant. Salekin (2004) and Brodsky and Galloway(2003) represent initial attempts at addressing this critical issue.
Salekin (2004) surveyed more than three hundred psychologists withpast involvement in death penalty cases to ascertain the methods cur-rently used to assess adaptive behavior. Approximately 80% of thosesurveyed believed that the available adaptive behavior instruments wereappropriate for use in correctional settings, that psychometric measuresof adaptive functioning are always necessary, and that detailed historiesof the defendant are necessary. Approximately 50% believed that usingjail or prison personnel as raters was beneficial. The sample was evenlydivided regarding the advantage of having more raters rather than less.
Area Review 19
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
Regarding the selection of raters, clinicians who specialize in mental re-tardation reported using family and friends as raters; forensic evaluatorswere less likely to use family and friends as raters. Regardless of spe-cialization, the sample was equally split regarding the appropriateness ofusing previous employers, high school teachers, or correctional officers asraters. The majority of the sample reported that using middle school and el-ementary teachers, caseworkers, probation officers, or police officers asraters was inappropriate. According to this survey, clinical judgment ap-peared to be extremely important when diagnosing mental retardation indeath penalty cases. Salekin concluded that no consensus or standard ofpractice exists for conducting adaptive behavior evaluations in death pen-alty defendants.
Brodsky and Galloway (2003) analyzed the ethical issues regardingAtkins evaluations and specifically addressed the assessment of adap-tive functioning. They emphasized the inadequacies of existing adap-tive behavior measurements and the lack of consensus in usage acrossmedical or correctional institutions. While stating that there is no simpleresolution, they offered the general recommendation that the best pro-fessional choice is to base adaptive functioning evaluations in deathpenalty cases on “a clinical synthesis of both pre-incarceration func-tioning and current functioning” (p. 7).
A consideration of evidentiary issues is necessary as the purpose ofadaptive behavior assessment in death penalty cases is to provide evi-dence to the trier of fact concerning the presence or absence of mentalretardation. In Daubert v. Dow (1993), the United States Supreme Courtheld that scientific expert testimony is admissible only if it is both relevantand reliable. In states where Daubert is the precedent, the trial court mustdecide if an expert’s proposed testimony is: (a) based on reasoning andmethodology that is scientifically valid and (b) whether that reasoning andmethodology properly can be applied to the facts. In states that haveadopted Daubert, evidentiary rules may prohibit the admissibility of as-sessment results that are not based on scientifically valid methods or proce-dures. States that use the Frye standard for the admissibility of evidencerequire that the scientific techniques used are commonly accepted in theexpert’s professional field (Frye v. U.S., 1923). Given the variability in theselection and administration of adaptive behavior instruments and the lackof consensus regarding the assessment of adaptive behavior among profes-sionals in death penalty cases, one has to wonder what the courts’ responsewill be in light of evidentiary rules from Frye and Daubert.
20 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
RECOMMENDATIONS FOR FUTURE RESEARCH
Considering the problems discussed related to the use of adaptive be-havior rating scales, how then should mental retardation and eligibilityfor the death penalty be determined? Limiting the determination ofmental retardation to a specific IQ score is unacceptable due to multiplefactors such as measurement error and cultural bias issues. Future re-search is vital in addressing many of the issues related to mental retarda-tion and the legal system. The scientist-practitioner model in psy-chology dictates that practitioners base their methods on peer reviewed,empirical research. Few would disagree that many research gaps existin areas related to the assessment of adaptive behavior and mental retar-dation in death penalty cases
Conduct Reliability and Validity Studies on the Adult Versions of Ex-isting Adaptive Behavior Rating Scales. The adult versions of both theABAS-II and the SIB-R are somewhat psychometrically deficient.Psychometric investigations of criterion-related validity are needed onthe adult version of the ABAS-II, and the SIB-R lacks inter-rater reli-ability data on its adult version. Empirical investigation of the technicalpsychometric characteristics of both these instruments is an importantstep toward increased accuracy in adaptive behavior assessment.
Establish Norms on a Correctional Population. Professional stan-dards regarding the use of psychological and educational tests recom-mend the collection of new normative evidence if a test is to be used in away for which it has not been validated (AERA, APA, NCME, 1999).Currently, no norms are available for the retrospective assessment of in-mate’s adaptive behavior functioning. Although the collection of suchnorms represents an ambitious endeavor, psychometric standardizationprojects are regularly completed on a large scale, especially when fundedby a research grant or commercial test publisher. Using specific criteria,an expert panel might select a sample from inmates deemed mentally re-tarded based on established prison classification procedures (e.g., in-mates placed in units specifically for mentally retarded individuals).The sample of inmates with mental retardation could then be comparedto a sample of inmates without mental retardation drawn from the gen-eral prison population. Appropriate third-party informants could retro-spectively rate adaptive behavior from both groups using either anexisting adaptive behavior rating scale or a new instrument. Norms re-sulting from this process could then be applied to inmates involved inAtkins evaluations. In addition, adaptive behavior assessment strategiescould be developed in the context of the prison environment. Such in-
Area Review 21
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
struments might also prove useful in prison classification and manage-ment.
Develop Systematic Clinical Guidelines. The subjectivity of clinicalassessment can be significantly reduced by the use of explicit system-atic guidelines based on a comprehensive review of the scientific litera-ture and input from experienced clinical and forensic evaluators. Thedevelopment of an adaptive behavior functioning assessment schemefashioned after instruments such as the HCR-20 Violence Risk Assess-ment Scheme (Webster, Douglas, Eaves, & Hart, 1997) holds promise.The HCR-20 rating items for risk assessment were based on the peer-re-viewed, empirical literature concerning factors related to violence. TheHCR-20 produces no specific risk score but rather the ratings form thebasis for an empirically-based clinical opinion regarding the likelihoodof future violence. Adaptive behavior assessment factors might includehistorical information verified by third-party informants via review ofarchival document, an assessment of current adaptive abilities, and arating of the types of environmental support needed by the defendant.
Investigate the Validity of Direct Assessment and Self-Report Mea-sures of Adaptive Behavior. Since locating appropriate informants foradaptive behavior rating scales is difficult in death penalty cases, researchaimed at other methodologies appears necessary. The direct psycho-metric testing of defendant’s adaptive behavior capacities and the utiliza-tion of the self-report versions of adaptive behavior rating scales such asthe SIB-R and the ABAS-II represent alternative methods for assessingadaptive behavior constructs; however, the reliability and validity ofthese instruments must be investigated. The results of adaptive behaviorrating scales by appropriate informants should be correlated with boththe direct assessment of adaptive abilities and with self-report ratings.Such research could form the empirical basis for the use of alternativeadaptive behavior assessment strategies that do not necessarily requirean informant.
Conduct Jury Research on Effect of Lay Testimony. Expert testimonyis limited by the rules of evidence to only that testimony that falls out-side what a trier of fact would be expected to understand without suchtestimony. Perhaps the adaptive ability as required to function inde-pendently in the real world is something that a judge or jury could eval-uate based on the testimony of a variety of lay witnesses who knew thedefendant and could describe the defendant’s behavior across time, set-tings, and circumstances. Lay-witness testimony on adaptive behaviormay be a viable alternative or supplement to expert testimony. Researchusing mock juries and simulated case information could provide valu-
22 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
able information for use in Atkins cases where strategic decisions mustbe made by attorneys regarding the presentation of credible evidence ina case.
Conduct an In-Depth Analysis Involving Death Penalty Cases andMental Retardation. A comprehensive investigation of death penaltycases involving mental retardation and the death penalty was far beyondthe purview of this article; however, such a review has potential for pro-viding information on the range of adaptive behavior assessment prac-tices. Reviewing the evidence presented in past Atkins cases acrossjurisdictions would be useful in determining best practices for forensicevaluators.
RECOMMENDATIONS FOR FORENSIC EXAMINERS
While future research is imperative, forensic practitioners must meetthe challenges related to the Atkins decision now. The remainder of thisarticle will focus on recommendations for forensic evaluations involv-ing the assessment of adaptive behavior and mental retardation in deathpenalty cases.
Know Relevant Legal Standards. All forensic evaluators and expertwitnesses should be knowledgeable about the relevant statutes and pre-cedent-setting case law in the states where he or she testifies. Recom-mended legal standards and procedures relevant to Atkins cases havebeen promulgated (Ellis, 2003); but states differ in their definitions ofmental retardation and the legal processes involved in Atkins cases. Ad-ministrative rules and regulations or statutory laws exist in most statesthat define mental retardation for treatment or entitlement purposes, butmany state legislatures have passed statutes defining mental retardationand specifying examiner qualifications for Atkins cases. Several stateshave adopted definitions of mental retardation that parallel closely thoseof the 1983, 1992, or 2002 AAMR or the DSM-IV-TR but other stateshave adopted a different definition. For example, the definition adoptedin Utah for Atkins cases focuses the adaptive behavior criteria on defi-ciencies that exist primarily in the areas of reasoning and/or impulsecontrol. Individual states also have either legislative statutes or rulingcase law governing important legal processes involved in the determi-nation of mental retardation in Atkins cases. For example, states may al-low a judge or a jury to decide the issue of mental retardation either in aseparate pre-trial hearing or during the punishment phase of a bifurcatedtrial. States also differ on the admissibility of the facts of the criminal
Area Review 23
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
conduct during the proceedings on the issue of mental retardation. Theprudent forensic evaluator always will be knowledgeable about the rele-vant statutes and precedent-setting case law in the jurisdiction where beor she will testify.
Use Standardized Adaptive Behavior Rating Scales When Appropri-ate. When the circumstances surrounding a defendant’s need for adap-tive behavior assessment meet the prerequisite psychometric demands,administering standardized adaptive behavior instruments to credible,nonbiased, honest informants is recommended. Under such circum-stances, the informants should be objective individuals who meet orclosely approximate the qualifications described in the Adaptive Be-havior Assessment System-II, which we previously presented but feelcompelled to repeat here. Informants should have: “(a) frequent contactwith the client (e.g., almost every day), (b) contacts of long duration(e.g., several hours for each contact, (c) recent contact (e.g., during thepast one-to-two months), and (d) opportunities to observe the variety ofskills measured . . .” (Harrison & Oakland, 2003, p. 15). Finally, we be-lieve the use of the self-report versions of adaptive behavior ratingscales lacks empirical support at this time.
Include Direct Testing of Adaptive Abilities. Although the traditionalconcept underlying the assessment of adaptive functioning has been touse third-party informants to rate the target individuals adaptive behav-ior in multiple real-life settings, perhaps Atkins evaluations demand adifferent approach. Assessing adaptive skills directly is a departure fromthe tradition in the assessment of mental retardation, but it has value asan adjunctive assessment strategy. Traditional tests measuring aca-demic abilities are often used in mental retardation evaluations, but sev-eral measures reviewed in this article directly assess adaptive behaviorknowledge and the ability to engage in activities of daily living. Neuro-psychological tests measure other functional limitations important inAtkins evaluations such as abstract thinking and logical reasoning. As inany forensic assessment, additional methods including effort and moti-vation testing must be employed to identify defendants attempting tomalinger on psychometric testing.
Information Integration and Clinical Judgment. The following rec-ommendations expand Brodsky and Galloway’s (2003) advice that the“immediate best professional choice appears to base adaptive function-ing evaluation of capital case and death row inmates on a clinical syn-thesis of both preincarceration functioning and current function” (p. 7).The most appropriate assessment of adaptive behavior functioning in-volves an integration of all available information including:
24 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
1. If standardization and informant standards are met, use the scoreson the ABAS-II and/or the SIB-R as one component of adaptivebehavior assessment.
2. Complete a review and analysis of collateral interview informa-tion, written statements, and affidavits from the least biased per-sons who are knowledgeable about the defendant.
3. Conduct an exhaustive search for and review of all archival recordsmade contemporaneously with pre-crime functioning (includingschool, medical, military, and employment records) focusing onadaptive behavior whether or not others thought the defendantwas mentally retarded, and the defendant’s ability to understandand process information, follow commands, learn from experi-ence, comply with rules, reason logically, and control impulses.
4. Design and implement a detailed adaptive behavior analysis of thedefendant’s likely actions, thoughts, and feelings surrounding thecriminal conduct using both the facts of the crime and the defen-dant’s account while focusing on the presence or absence of fore-thought, planning, complex execution of purpose, perceptions ofthe actions of others, and his or her relationship with any co-de-fendants.
5. Conduct an evaluation of the defendant’s adjustment to presentand past incarcerations, whether it has been a few months in thecounty jail or years on death row, focusing on any network of peersupport, maintenance of contact with the free world, quality ofpersonal hygiene and health maintenance, involvement in treat-ment or work programs (if available), use of free time, and disci-plinary actions.
6. Administer psychometric tests of the defendant’s present abilitiesand adaptive capacities including functional academic testing, rel-evant neuropsychological instruments, as well as effort and moti-vation validity measures.
7. Conduct an extended clinical interview comprised primarily ofopen ended questions and focus on the defendant’s response toquestioning, receptive and expressive language, judgment andplanning, decision-making ability, interpersonal skills, acceptanceof responsibility, self-esteem, social roles and status, gullibility,naiveté, perception of rules and laws, reaction to authority figures,avoidance of victimization, activities of daily living, perception ofoccupational skills/potential, and how “street smart” and “jointsavvy” the defendant appears to be. Modified inquiries may be
Area Review 25
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
generated from adaptive behavior assessment instrument itemsfor clinical use in assessing adaptive strengths and weaknesses.
Forensic evaluators should thoroughly and thoughtfully integrate theabove information in a fashion consistent with the relevant legal stan-dard regarding the adaptive behavior components of a diagnosis ofmental retardation and in a manner that is sensible and helpful to thejudge and/or jury. Forensic evaluators should present the results of theirinvestigation of adaptive behavior and testify affirmatively, not defen-sively, that the methodology of integrating all available informationinto a clinical judgment is consistent with the highest ethical standardsof their profession and current recommended best practices (AAMR,2002; Brodsky & Galloway, 2002; Reschley, Myers, & Hartel, 2002).Doing so will set the proper stage for the trier of fact to assign the appro-priate weight to the expert testimony.
REFERENCES
Adams, G. L. (1999). Comprehensive Test of Adaptive Behavior–Revised. Seattle, WA:Educational Achievement Systems.
American Association on Mental Retardation. (2002). Mental retardation: Definitionclassification, and systems of support (10th ed.). Washington, DC: Author.
American Education Research Association, American Psychological Association, & Na-tional Council on Measurement in Education. (1999). Standards for educational andpsychological testing. Washington, DC: American Educational Research Association.
American Psychological Association. (2002). Ethical principles of psychologist andcoed of conduct. [Electronic version]. Retrieved May 18, 2005 from http://www.apa.org/ethics/code2002.html
American Psychiatric Association. (1952). Diagnostic and statistical manual: Mentaldisorders. Washington, DC: Author.
American Psychiatric Association. (1980). Diagnostic and statistical manual: Mentaldisorders (3rd ed.). Washington, DC: Author.
American Psychiatric Association. (1987). Diagnostic and statistical manual: Mentaldisorders (3rd ed., rev.). Washington, DC: Author.
American Psychiatric Association. (1994). Diagnostic and statistical manual: Mentaldisorders (4th ed.). Washington, DC: Author.
American Psychiatric Association. (2000). Diagnostic and statistical manual: Mentaldisorders (4th ed., text rev.). Washington, DC: Author.
Anastasi, A., & Urbina, S. (1997). Psychological Testing (7th ed.). Upper SaddleRiver, NJ.
APA Ethics Code (2002). http://www.apa.org/ethics/code2002.htmlAtkins v. Virginia, 122 S. Ct. 2242 (2002).
26 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
Baroff, G. S. (1990). Establishing mental retardation in capital defendants. AmericanJournal of Forensic Psychology, 8, 2, 35-45.
Baroff, G. W. (1991). Establishing mental retardation in capital cases: A potential mat-ter of life and death. Mental Retardation, 29, 343-349.
Beail, N. (2003). Utility of the Vineland Adaptive Behavior Scales in diagnosis and re-search with adults who have mental retardation. Mental Retardation, 41, 286-289.
Brigance, A. H. (1981). Inventory of Essential Skills. North Billerica, MA: CurriculumAssociates.
Brigance, A. H. (1994). Life Skills Inventory. North Billerica, MA: Curriculum Associ-ates.
Brigance, A. H. (1995). Employability Skills Inventory. North Billerica, MA: Curricu-lum Associates.
Brodsky, S. L., & Galloway, V. A. (2003). Ethical and professional demands for foren-sic mental health professionals in the post-Atkins era. Ethics & Behavior, 13(1) 3-9.
Bruininks, R. H., Woodcock, R. W., Weatherman, R. F., & Hill, B. K. (1984). Scales onIndependent Behavior. Itasca, IL: Riverside.
Bruininks, R. H., Woodcock, R. W., Weatherman, R. F., & Hill, B. K. (1996). Scales onIndependent Behavior–Revised. Itasca, IL: Riverside.
Burns, M. K. (2005). Test review of the Adaptive Behavior Assessment System– Sec-ond Edition. From B. S. Plake, J. C. Impara, & R. A. Spies (Eds.), The sixteenth men-tal measurements yearbook [Electronic version]. Retrieved May 18, 2005 from theBuros Institute’s Test Reviews Online Website: http://www.unl.edu/buros
Carey, K. T. (1998). Test review of the AAMR Adaptive Behavior Scale–Residentialand Community, Second Edition. From J. C. Impara & B. S. Blake (Eds.), The thir-teenth mental measurements yearbook [Electronic version]. Retrieved May 18, 2005from the Buros Institute’s Test Reviews Online Website: http://www.unl.edu/buros
Cohen, L. (2001). Test review of the Independent Living Scales. From B. S. Plake &J. C. Impara (Eds.), The fourteenth mental measurements yearbook [Electronic ver-sion]. Retrieved 5/18/05, from the Buros Institute’s Test Reviews Online Website:http://www.unl.edu/buros
Committee on Ethical Guidelines for Forensic Psychologists. (1991). Specialty guide-lines for forensic psychologists. Law and Human Behavior, 15, 655-665.
Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993).DeVault, S. (1987). Assessment of adaptive behavior in helping to determine compe-
tencies. American Journal of Forensic Psychology, V, 3, 45-53.Ellis, J. W. (2003). Mental retardation and the death penalty: A guide to state legisla-
tive issues. Mental and Physical Disabilities Law Reports, Jan.-Feb. 27 (1), 11-24.Frye v. United States. 293 F. 1013 (DC Cir. 1923).Giller, V. L., Dial, J. G., & Chan, F. (1986). The Street Survival Skills Questionnaire: A
correlational study. American Journal of Mental Deficiency, 91, 67-71.Haring, T. (1992). Test review of the Street Survival Skills Test. From J. J. Kramer & J.
C. Conoley (Eds.), The eleventh mental measurements yearbook [Electronic version].Retrieved 4/25/05, from the Buros Institute’s Test Reviews Online Website: http://www.unl.edu/buros
Harrison, P. L. (1987). Research with adaptive behavior scales. Journal of Special Edu-cation, 21, 37-68.
Area Review 27
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
Harrison, P. L. (1998). Test review of the AAMR Adaptive Behavior Scale-Residentialand Community (2nd ed.). From J. C. Impara & B. S. Plake (Eds.), The thirteenthmental measurements yearbook [Electronic version]. Retrieved May 18, 2005, from theBuros Institute’s Test Reviews Online Website: http://www.unl.edu/buros
Harrison, P. L., & Oakland, T. (2000). ABAS: Adaptive Behavior Assessment System. SanAntonio, TX: Psychological Corp.
Harrison, P. L., & Oakland, T. (2003). ABAS-II: Adaptive Behavior Assessment System–II. San Antonio, TX: Psychological Corp.
Heal, L. W., & Tasse, M. J. (1999). The culturally individualized assessment of adap-tive behavior: An accommodation to the 1992 AAMR definition, classification, andsystems of support. In R. L. Schalock (Ed.), Adaptive behavior and its measurement:Implications for the field of mental retardation (185-205). Washington, DC: AmericanAssociation on Mental Retardation.
Janniro, F., Sapp, G. L., & Kohler, M. P. (1994). Validating the Street Survival SkillsQuestionnaire. Psychological Reports, 74, 191-194.
Kelly, T. (1927). Interpretation of educational measurements. Yonkers, NY: World Book.Linkenhoker, D., & McCarron, L. (1993). The Street Survival Skills Questionnaire. Dal-
las, TX: McCarron-Dial Systems.Loeb, P. A. (1996). Independent Living Scales. San Antonio, TX: The Psychological Cor-
poration.Long, C., & Sborbone, R. (1996). Ecological validity of neuropsychological testing. San-
ford, FL: DC Press.MacCow, G. (2001). Test review of the Scales of Independent Behavior-Revised.
From B. S. Plake & J. C. Impara (Eds.), The fourteenth mental measurements year-book [Electronic version]. Retrieved May 18, 2005 from the Buros Institute’s Test Re-views Online Website: http://www.unl.edu/buros
MacMillan, D. L., Gresham, F. M., & Siperstein, G. N. (1993). Conceptual and psycho-metric concerns about the 1992 AAMR definition of mental retardation. AmericanJournal on Mental Retardation, 98, 325-335.
MacMillan, D. L., Gresham, F. M., & Siperstein, G. N. (1995). Heightened concernsover the 1992 AAMR definition: Advocacy versus precision. American Journal onMental Retardation, 100, 87-97.
McGrew, K. S., & Bruininks, R. H. (1989). The factor structure of adaptive behavior.School Psychology Review, 18, 64-81.
Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York:McGraw-Hill.
Penry v. Lynaugh, 492 U.S. 302 (1989).Reschly, D. J., Myers, T. G., & Hartel, C. R. (Eds.). (2002). Mental retardation: Determin-
ing eligibility for social security benefits. Washington, DC: National Academy Press.Rogers, R. (Ed.). (1997). Clinical assessment of malingering and deception (2nd ed.).
New York: Guilford Press.Salekin, K. L. (2004). Forensic evaluation, mental retardation, and methodology: Is there
a consensus? Paper presented at the annual conference of the American Psychology-Law Society, Scottsdale, AZ: March.
Salvia, J., & Ysseldyke, J. E. (1998). Assessment (7th ed.). Boston, MA: HoughtonMifflin.
28 JOURNAL OF FORENSIC PSYCHOLOGY PRACTICE
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4
Sattler, J. M. (1989). Test review of the Vineland Adaptive Behavior Scales. From J. C.Conoley & J. J. Kramer (Eds.), The tenth mental measurements yearbook [Electronicversion]. Retrieved May 5, 2005 from the Buros Institute’s Test Reviews OnlineWebsite: http://www.unl.edu/buros
Schalock, R.L. (1999a). The merging of adaptive behavior and intelligence: Implica-tions for the field d of mental retardation. In R. L. Schalock (Ed.), Adaptive behaviorand its measurement: Implications for the field of mental retardation (43-59). Wash-ington DC: American Association on Mental Retardation.
Schalock, R. L. (1999b). The future of the construct of adaptive behavior. In R. L.Schalock (Ed.), Adaptive behavior and its measurement: Implications for the field ofmental retardation (209-222). Washington DC: American Association on Mental Re-tardation.
Simon, R. I., & Shuman, D. W. (Eds.) (2002). Retrospective assessment of mental states inlitigation: Predicting the past. Washington, DC: American Psychiatric Publishing, Inc.
Sparrow, S. S., Balla, D. A., & Cicchetti, D. V. (1984). Vineland Adaptive BehaviorScales. Circle Pines, MN: American Guidance Service.
Spreat, S. (1999). Psychometric standards for adaptive behavior assessment. In R. L.Schalock (Ed.), Adaptive behavior and its measurement: Implications for the field ofmental retardation (103-117). Washington, DC: American Association on Mental Re-tardation.
Webster, C. D., Douglas, K. S., Eaves, D., & Hart, S. D. (1997). HCR-20: Assessing riskfor violence (version 2). Lutz, FL: Psychological Assessment Resources, Inc.
Zlomke, A. C. (2001). Test review of the Scales of Independent Behavior-Revised.From, B. S., Blake, J. C., & J. C. Impara (Eds.), The fourteenth mental measurementsyearbook [Electronic version]. Retrieved May 18, 2005 from the Buros Institute’s TestReviews Online Website: http://www.unl.edu/buros
doi:10.1300/J158v06n03_01
Case Report 29
Dow
nloa
ded
by [
Uni
vers
ity o
f A
lber
ta]
at 1
6:24
15
Oct
ober
201
4