System of Empirically Based Assessment

Achenbach System of Empirically Based Assessment

 

Preview of the Achenbach System of Empirically Based Assessment by ROSEMARY FLANAGAN, Assistant Professor/Director, Masters Program in School Psychology, Adelphi University, Garden City, NY: DESCRIPTION. The Achenbach System of Empirically Based Assessment (ASEBA) available in two versions for children (ages 1.5-5 and 6-18), is a multiple-rater system used to assess the behavior and personality of youth. Compared to the previous editions (Achenbach, 1991; Achenbach, 1992), the beginning age for the preschool version has been extended downward, and the beginning age for the school-age form has increased. Parents complete the Child Behavior Checklist (CBCL/1 1/2-5, CBCL/6-18), those working in school or care settings complete the Teacher Report Form (TRF/6-18) or the Caregiver-Teacher Report (C-TRF), and youngsters aged 11-18 complete the Youth Self-Report (YSR). The version for school age youth contains 113 items and the version for preschoolers contains 100 items. Subscales for the YSR and CBCL/6-18 are subsumed under groupings called Competence, Syndrome, and DSM-Oriented, the latter of which is based on the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV; American Psychiatric Association, 1994). The TRF/6-18 yields 20 scales categorized as Adaptive, Syndrome, and DSM-Oriented. The CBCL/1 1/2-5 and C-TRF yields 8 Syndrome and 6 DSM-Oriented scales. The Language Development Survey (LDS; Rescorla, 1989) is part of the CBCL/1 1/2-5; it is composed of questions about the child’s birth history, ear infections, and speech problems within the family. Parents are also asked to report the child’s best multiword phrases and indicate whether their child knows words commonly known by preschoolers. Although not a substitute for a more established speech-language screening, it is helpful given the comorbidity of delays in language development with psychopathology. Practical applications for these forms are in an array of settings serving children that include schools, mental health settings, medical settings, forensic settings, and child and family service settings. The manual provides case studies to illustrate the application and interpretation of the data. Summary manuals are available to guide professionals who are consumers of the data (but not necessarily direct test users) in several settings as to the nature of the scales and their usage.

The authors recommend using the computer-scoring program, but scoring can also be accomplished by hand, by laboriously transferring item ratings to a profile sheet that groups the items by scales; clerical errors seem likely. Cross-informant comparisons can be made readily with the computer-scoring program only. DEVELOPMENT. The revision of the scales includes an updating of the norms and refinement of the scales; the procedures used to accomplish this were thorough. Competence scales were derived by comparing the responses of referred versus nonreferred youth, with items having been retained from the CBCL/4-18, TRF, and YSR; some refinement in scoring has been incorporated. Extensive procedures were followed to scientifically obtain a nationally representative nonreferred/nontreated sample. The norming samples contain 1,753, 1,057, and 2,319 individuals for the CBCL/6-18, YSR, and TRF, respectively. The sample for the TRF initially derived from the norming sample for the CBCL/6-18 was small. Given that the 1989 sample did not score differently from the 2001 sample, the data from the two cohorts were combined to have a larger normative sample. Normalized T-scores were assigned to the raw scores for each gender at two age levels: 6-11 and 12-18 years. Particular attention was given to areas of the score distribution that were skewed, as it was thought desirable to make these more sensitive to differences in functioning.

The Syndrome scales were developed to produce information about patterns of problems. In addition to the sample used to norm the Competence scales, additional children who received inpatient and outpatient treatment were included. This resulted in sample sizes for the CBCL/6-18, TRF, and YSR of 4,994, 4,437, and 2,551, respectively. Sample sizes for the CBCL/1 1/2-5 and the C-TRF are 1,728 and 1,113 youth, respectively. Some items from the 1991 versions of the CBCL, TRF, and YSR were eliminated because these were endorsed by considerably less than 5% of the respondents; the net result was that six items were replaced in the CBCL/6-18 and YSR; three items were replaced in the TRF. Similarly, two items were replaced in the CBCL/1 1/2-5 and C-TRF. Principal components analysis yielded an eight-factor solution, for the CBCL/6-18 and the YSR, which also proved to demonstrate the best fit upon confirmatory factor analysis. A seven-factor solution was realized for the TRF. A subsequent factor analysis grouped the Syndromes according to Total Problems, Internalizing Problems, and Externalizing Problems. The names of two Syndrome scales were changed from the 1991 versions; Withdrawn is now Withdrawn/Depressed, and Delinquent Behavior is Rule Breaking Behavior. Factor analysis of the TRF scales indicates that the Attention Problems Scale is composed of items that may be categorized as Inattentive and Hyperactive-Impulsive, parallel to DSM-IV (American Psychiatric Association, 1994). Normalizing the data at points on the distributions that were skewed was done to derive T-scores. Segments of scales that correspond to those individuals making an appropriate adjustment were truncated. Thus, the statistical treatment focuses on the portions of the scales that potentially yield the greatest amount of diagnostic and classification data. To limit false negatives, the borderline clinical range was extended downward, now beginning at T = 65, which is consistent with other commonly used rating forms. T-scores of 70 and above are clearly in the clinical range. The Syndrome scales for the CBCL/1 1/2-5 were similarly developed, using a norming sample of 700 individuals. Norms for the LDS are based on the mean length of utterance and vocabulary development by 5-month intervals. The DSM-Oriented scales for the CBCL/1 1/2-5 and the CBCL/6-18 were developed by psychologists and psychiatrists indicating the degree of consistency of each item with nine DSM categories. Only items that were rated as very consistent with a particular diagnosis were retained.

TECHNICAL. Psychometric properties are generally strong. For the CBCL/6-18, internal consistency reliability (coefficient alpha) ranged from .55-.90 for Competence and Adaptive scales, from .71-.97 for the Syndrome scales, and from .67-.94 for the DSM-Oriented scales. Mean stability for the CBCL/6-18 at 12 months is .65, .51 for the YSR at 7 months, and .65 for the TRF at 2 months. Mean test-retest reliability ranged from .88-.90, .79-.88, and .85-.90 for 8- or 16-day intervals for CBCL/6-18, YSR, and TRF, respectively. The CBCL/1 1/2-5 similarly demonstrates strong internal consistency, with coefficient alpha ranging from .66-.96 for the Syndrome scales, and .from .63-.93 for the DSM-Oriented scales. Test-retest reliability for an 8-day interval ranged from .68-.92 for the Syndrome scales and from .57-.87 for the DSM-Oriented scales. The mean stability for the CBCL/1 1/2-5 at a 12-month interval is .61, and is .59 for the C-TRF at a 3-month interval. Mean cross-informant agreement for the CBCL/1 1/2-5 and the C-TRF are .61 and .65, respectively. For the Competence and Adaptive scales of the CBCL/6-18 and the TRF, the mean cross-informant agreement values are .69 and .49, respectively. For the Syndrome scales, these values are .76 and .60 for the CBCL/6-18 and the TRF, respectively. For the DSM-Oriented scales, the mean cross-informant agreement for the CBCL/6-18 is .73, for the TRF it is .58. These values are substantial. Validity evidence is extensive, with analysis of the scores on scales as well as items that document the successfulness of the scale in youth scoring differently, based on referral status. For some scales and items at shorter item intervals, the test-retest score relationship may be attenuated. Evidence substantiating content validity is based on prior research with the scales. Items that failed to differentiate between referred and nonreferred children were excluded from the scales. Evidence of criterion-related validity of the CBCL/6-18, YSR, and TRF is based on multiple regression analyses and indicates that 2-33% of the variance on individual scales is accounted for by referral status. Additional evidence is based on classification accuracy by referral status using discriminant analysis procedures (79-85%). Information is in the manual that will assist practitioners interpreting the data for youth who are not clearly in the clinical range, but may be exhibiting behavior or affect of concern; this is of considerable importance to school-based practitioners. Construct validity was evaluated on the basis of correlations with similar instruments, in particular the BASC (Reynolds & Kamphaus, 1992), the Conners’ Rating Scales-Revised (Conners, 1997), and the DSM-IV Checklist (Hudziak, 1998). Correlations with the Conners’ Rating Scales and the DSM-IV Checklist are moderate; correlations with the BASC are more substantial.

Similarly, the content validity for the CBCL/1 1/2-5 was examined based on prior research with the scales. Content validity of the LDS was evaluated by repeated correlational studies using other measures of language development. Evidence of criterion-related validity is based on multiple regression analyses of the CBCL/1 1/2-5 and C-TRF that yielded percentages of explained variance accounted for by referral status ranging from 2-25% for the individual scales. Moreover, classification accuracy according to referral status was documented using discriminant analysis at 84.2%. Criterion validity of the LDS is demonstrated through a series of studies that report correlations with cognitive and language measures; these correlations range from .56-.87, with most values exceeding .70. Evidence of construct validity of the CBCL/1 1/2-5 and C-TRF is based on correlations with measures not common to clinical practice. The correlations for series of studies range from .46-.72. Construct validity evidence for scores from the LDS includes correlations that predict language scores at age 13, ranging from .38-.55. COMMENTARY. Compared to the previous editions, the new ASEBA is improved and refined. The manuals are clear, providing technical information in a format understandable to practitioners. Extensive covariance analyses are reported in the manuals, substantiating that each item retained on the scales effectively differentiates youth based on referral status. The computer-scoring program is more effective. Important to practitioners, in particular, is that the borderline clinical range was made broader to limit the false negatives and aid in the identification of youth who might need attention. The Competence scales were made more sensitive, having expanded the possible score range. These features bring the ASEBA more in line with its main competitor, the BASC. A companion semistructured clinical interview and a direct observation form are available. The LDS of the ASEBA is an important feature that does not appear on the BASC. Nevertheless, the BASC continues to be easier to score and interpretation is less complicated. The CBCL/1 1/2-5 and CBCL/6-18 scales are substantially correlated across age levels. Thus, the manner in which the components of these assessment systems are related differs. The BASC forms may be easier for respondents to complete, as the items are in one format. One reason to use the ASEBA over the BASC for researchers is to take advantage of the lengthier research history as compared to the BASC. SUMMARY. The new ASEBA (CBCL/1 1/2-5, C-TRF, CBCL/6-18, TRF, YSR) is composed of multiple-respondent rating forms and companion scales that may be used in any combination to rate the behavior and affect of youth aged 1 1/2-5, or 6-18. Procedures to develop the scales and examine their psychometric properties are exemplary. Although the psychometric properties are stronger for the school-aged versions than for the preschool versions, this may reflect the inherent variability of preschoolers. The scales are supported by a solid research base and are technically sound, both from test development and psychometric perspectives. Applicability in various settings serving children is apparent. Practitioners and researchers alike should expect the new versions to be useful as were their predecessors. REVIEWER’S REFERENCES Achenbach, T. M. (1991). Child Behavior Checklist for Ages 4-18. Burlington, VT: University of Vermont Department of Psychiatry. Achenbach, T. M. (1992). Manual for the Child Behavior Checklist/2-3 and 1992 profile. Burlington, VT: University of Vermont Department of Psychiatry. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed). Washington, DC: Author. Conners, C. K. (1990). Manual for the Conners’ Rating Scales. North Tonawanda, NY: Multi-Health Systems. Hudziak, J. J. (1998). DSM-IV Checklist for Childhood Disorders. Burlington, VT: University of Vermont, Research Center for Children, Youth. Rescorla, L. (1989). The Language Development Survey: A screening tool for delayed language in toddlers. Journal of Speech and Hearing Disorders, 54, 587-599. Reynolds, C. R., & Kamphaus, R. W. (1992). The Behavior Assessment System for Children. Circle Pines, MN: AGS Publishing. Salvia, J., & Ysseldyke, J. E. (2000). Assessment (8th ed.). Boston: Houghton-Mifflin.

 

Review of the Achenbach System of Empirically Based Assessment by T. STEUART WATSON, Professor and Chair of Educational Psychology, Miami University, Oxford, OH: GENERAL DESCRIPTION. The Achenbach System of Empirically Based Assessment (ASEBA) is a revision of the popular Achenbach scales of 1991 and 1992. The ASEBA is a set of integrated instruments designed to assess children’s problems and competencies and includes the Child Behavior Checklist for ages 1 1/2-5 and 6-18 (CBCL), the Teacher Report Form (TRF) for Ages 6-18, the Youth Self Report (YSR) for Ages 11-18, and the Semistructured Clinical Interview for Children and Adolescents (SCICA). The ASEBA is designed for assessment, intervention planning, and outcome evaluation in a number of settings and is one of the most widely researched and used behavioral rating scales. The 21st Century edition of the CBCL for ages 6-18 is a revision of the 1991 version for children ages 4-18 (Parent Report Form; PRF) and ages 5-18 (TRF). The CBCL 1 1/2-5 is a revision of the CBCL/2-3. Although there is a high degree of consistency between the previous and current versions of the CBCL, some items have been removed or reworded in order to more accurately discriminate between referred and nonreferred children. There is also a Language Development Survey (LDS) that is part of the CBCL 1 1/2-5. It is intended for parents of all children under the age of 3 and children over the age of 3 who are suspected of having language delays. After answering some basic questions on the front page, parents are instructed to circle any of 310 words that their child spontaneously emits. Two percentile rank scores are provided-one based on the child’s age and number of words circled and another based on the average length of phrases emitted by the child. Scores below the 15th and 20th percentiles, respectively, are considered delayed. The LDS is a valuable addition to the ASEBA and allows the clinician to assess the impact of language on other problems noted by parents and teachers. The parent and teacher versions of the CBCL can be completed in about 15-20 minutes. The parent version may be completed by a parent, caregiver, or anyone else who has experience with the child in a residential setting. The teacher version may be completed by any of the children’s teachers or school personnel who are familiar with the child. The first two pages ask for demographic information, children’s competencies at home or school, and either activities or academic information. As with previous editions, this information is more descriptive of incompetence than competence and is not particularly useful for intervention planning or diagnostic purposes. The remaining two pages contain 120 items that are rated: 0 = not true (as far as you know); 1 = somewhat or sometimes true; or 2 = very true or often true. Although the authors contend that a wider gradient of ratings would add little in the way of increased discriminative power, some respondents may find it difficult to accurately rate behaviors that have differing rates of occurrence that meet the descriptors of the numerical scale. For instance, Item 57 “Physically attacks people” may require only 1 or 2 instances in the past 6 months to warrant a 2 whereas Item 109 “Whining” may require several incidents per day over 6 months to warrant a rating of 2.

 

There are separate scoring profiles for boys and girls ages 6 to 11 and 12 to 18. Raw scores, T scores, and percentiles for Total Competence, three Competence scales (Activities, Social, and School), eight Syndrome scales, six DSM-Oriented scales, Internalizing Problems, Externalizing Problems, and Total Problems are provided. The Syndrome scales include Anxious/Depressed, Withdrawn/Depressed, Somatic Complaints, Social Problems, Thought Problems, Attention Problems, Rule-Breaking Behavior, and Aggressive Behavior. The DSM-Oriented scales include Affective Problems, Anxiety Problems, Somatic Problems, Attention Deficit/Hyperactivity Problems, Oppositional Defiant Problems, and Conduct Problems. The Internalizing scale is the sum of scores for the Anxious/Depressed, Withdrawn/Depressed, and Somatic Complaints syndromes. The Externalizing scale is the sum of scores from the Rule-Breaking Behavior and Aggressive Behavior syndromes. The Total Problem score is the sum of scores on all 120 items. The Syndrome and DSM-Oriented scales are similar across the PRF, TRF, and YSR. A number of different scores are available to assist in the interpretation of the CBCL. T scores and percentile rank scores allow for comparisons to the normative sample to determine if a child’s competencies and problems differ from what is considered typical of a child that age and gender. On the Activities, Social, and School scales, T scores of 31 to 35 are in the borderline range and T scores below 31 fall in the clinical range. On Total Competence, T scores of 37 to 40 are borderline range, whereas T scores below 37 are considered to fall in the clinical range. T scores of 65-69 on the Syndrome and DSM-Oriented scales are considered borderline whereas T scores above 69 are in the clinical range. T scores of 60 to 63 on the Total Problems, Internalizing, and Externalizing scales are borderline and T scores above 63 fall within the clinical range. The scoring profile for the CBCL/6-18 was normed on a sample of 1,753 children ages 6 to 18 who had not been referred for professional help for behavioral or emotional problems within the preceding 12 months. All 48 contiguous states were represented in the sample and were stratified by socioeconomic status, ethnicity, region, and urban-suburban-rural residence. Separate norms are provided for boys and girls ages 6 to 11 and 12 to 18.

 

The YSR is for children ages 11 to 18 and is very similar to the 1991 edition. Four pages in length, it requires fifth grade reading skills to accurately complete the form, which takes about 20 minutes. For children with reading difficulties, items may be read aloud and respondents indicate the score orally. It is similar in format to the CBCL, thus there is a high degree of consistency among the items. As with the CBCL, most users will probably find the first two pages of information to be useless. The real meat of the instrument is the two pages with problem items. Each item is rated for how true it is for them in the past 6 months: 0 = not true (as far as you know); 1 = somewhat or sometimes true; 2 = very true or often true. Scores and scales on the YSR are identical to those on the CBCL. Rule-Breaking Behavior is on this edition, which replaces the Delinquent Behavior Scale from the 1991 version. The YSR scoring profile was normed on a sample of 1,057 children ages 11 to 18 from the contiguous 48 states and was stratified by ethnicity, geographic region, and SES. Separate norms are provided for boys and girls. The T scores and percentile ranks allow the clinician to compare a child’s score with children of the same age and gender to determine if they are exhibiting more problems than is typical for someone of similar age and gender. Although responses on the CBCL, YSR, and TRF may be hand scored, which takes experienced examiners about 15 minutes per instrument, computerized scoring is much simpler, is far less likely to result in errors than the hand-scoring option, and yields several different types of reports. The computer scoring is in a Windows format called the Assessment Data Manager (ADM). One type of report generated by the ADM is cross-informant comparisons. This printout allows comparisons for up to eight informants for the 93 items that are similar across the YSR, TRF, and PRF Syndrome scales and the 45 items that are similar on the DSM-Oriented scales. Other comparisons include the T scores for the eight Syndrome scales, the six DSM-Oriented scales, and the three Problems scales (Internalizing, Externalizing, and Total). This is a particularly useful feature, especially when there are multiple informants and the clinician is attempting to identify patterns within an individual. An additional feature based on cross-informant data is Q correlations that indicate whether the agreement between pairs of informants is average, above average, or below average. Having agreement information readily available is particularly helpful in clinical situations by providing areas for probing regarding disagreements.

 

A second type of report compiled by the ADM is a narrative report. This report summarizes results for the Competence and Problems scales. A noted improvement over earlier versions of the Achenbach is the recognition of critical items from the PRF, TRF, and YSR that are important for further assessment and intervention. The narrative report also lists the scores for each of these items. SEMISTRUCTURED CLINICAL INTERVIEW FOR CHILDREN AND ADOLESCENTS (SCICA). The SCICA is a standardized interview for children ages 6 to 18 and includes interview questions, tasks, and standardized rating forms for scoring observations and self-reported problems. Only experienced clinicians should use the SCICA, which takes about 60-90 minutes. Nine areas are covered by the SCICA and include Activities, Friends, Family Relations, Fantasies, Self-Perception, Parent/Teacher Reported Problems, Achievement tests (optional), for ages 6-11, a screen for fine and gross motor problems (optional), and for ages 12-18, somatic complaints, alcohol, drugs, and legal trouble. Children aged 6-11 are also asked to make a drawing of their family doing something, which can provide information for further questioning. The protocol is extremely user-friendly as it contains instructions, open-ended questions, and interviewing tasks. There is also ample space to record and make notations regarding the child’s behavior during the interview. For children ages 6 to 11, interviewers may administer other tests in order to gather a more complete picture of the child’s functioning. Suggestions for additional testing include brief forms of standardized achievement tests, writing samples, and assessing gross motor functioning. For children ages 12 to 18, the SCICA includes more structured questions to assess somatic complaints, alcohol and drug use, and legal difficulties. After completing the SCICA, the interviewer scores both the SCICA Observation and Self-Report Forms. The Observation Form contains 96 problem items, many of which are similar to the problem items from the YSR, TRF, and PRF. The Self-Report Form contains 114 items for ages 6 to 18 and several additional items for ages 13 to 18. As with the observation form, many of the items on the SCICA self-report are similar to those on the CBCL/6-18 PRF and TRF. All items from the SCICA observation and self-report forms are scored on a 4-point scale: 0 = no occurrence; 1 = very slight or ambiguous occurrence; 2 = definite occurrence with mild to moderate intensity and less than 3 minutes duration; 3 = definite occurrence with severe intensity or 3 or more minutes duration. For the additional self-report items for ages 12 to 18, a 4-point scale is used with varying time lines for reports of somatic complaints, alcohol and drug use, and legal trouble.

 

The current version of the SCICA provides separate scores for children ages 6 to 11 and 12 to 18. This represents an improvement over the 1994 version, which only provided scores for children ages 6 to 12. Raw scores, T scores, and percentile ranks are provided for Total Observation, Total Self-Report, Externalizing Problems, Internalizing Problems, the eight Syndrome scales, and the six DSM scales. Of the eight Syndrome scales, five are derived based on interview observations (Anxious, Withdrawn/Depressed, Language/Motor Problems, Attention Problems, and Self-Control Problems) and three are derived from the child’s self-report during the interview (Anxious/Depressed, Aggressive/Rule-Breaking, and Somatic Complaints). The Somatic Complaints scale is scored only for children ages 12-18. For children ages 12 to 18, items on the Aggressive/Rule-Breaking syndrome are scored on separate Aggressive and Rule-Breaking scales. DEVELOPMENT. The Syndrome scales for the PRF were derived by analyzing data from 4,994 children referred for mental health services whose parents completed the CBCL. The Syndrome scales for the TRF were developed by statistically analyzing data from TRFs completed by teachers from 4,437 children referred for mental health or special education services. The DSM-Oriented scales for the PRF, TRF, and YSR were derived from the ratings of psychiatrists and psychologists who rated items from the three versions of the CBCL. The items were rated as being not consistent, somewhat consistent, or very consistent with the diagnostic criteria of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association, 1994). To be included on the DSM scale, items had to be rated as very consistent with diagnostic criteria by at least 14 of the 22 raters. High scores on the DSM-Oriented scales can assist clinicians in determining whether a DSM-IV diagnosis is appropriate for a particular child. As is true with any assessment situation, clinicians should not rely solely on the results of the CBCL to make a diagnosis. The Syndrome scales of the SCICA were developed by statistically analyzing interviewers’ ratings of 381 children ages 6 to 11 and 305 children ages 12 to 18. All of the children in the samples had been referred for either mental health or special education services. Items endorsed for fewer than 5% of the samples in each age group were excluded from further analysis.

 

TECHNICAL. The ASEBA is a well-researched instrument with exceptionally strong reliability and validity data. The scales from the parent version of the CBCL/1 1/2-5 with the lowest test-retest reliabilities are Anxious/Depressed and ADHD Problems. Those from the Caregiver-Teacher Report Form (C-TRF) were the Anxiety Problems and Anxious/Depressed. This is not surprising given the difficult nature of identifying these types of behaviors in young children. In fact, one may reasonably question the clinical validity of attempting to identify ADHD behaviors in such a young sample. Internal consistency coefficients were acceptable with the possible exceptions of Anxiety Problems and Somatic Problems from the DSM-Oriented scales. Cross-informant correlations ranged from extremely low to extremely high. As might be expected, correlations between the YSR and TRF were uniformly low whereas mother and father correlations on the PRF were the highest. Given the purposes of the ASEBA, perhaps the most important psychometric aspect is criterion-related validity. In this case, do the scores on the instrument(s) differentiate referred from nonreferred children? Without going into undue detail, it is accurate to say that all 120 items from the PRF, YSR, and TRF significantly discriminated at p<.01. A number of other computations are provided in the manual that demonstrate criterion-related and other types of validity of scores from the ASEBA. Overall, the ASEBA is a psychometrically sound instrument with weaknesses on some of the scales, particularly at the younger age ranges and with some of the more ubiquitous scales (e.g., Anxiety, Thought Problems). MANUALS. A number of manuals and guides are included as part of the ASEBA package: (a) Manual for the ASEBA Preschool Forms & Profiles including the CBCL for ages 1 1/2-5, Language Development Survey, and Caregiver-Teacher Report Form; (b) Manual for the ASEBA School-Age Forms & Profiles including the CBCL for Ages 6-18, Teacher’s Report Form, and Youth Self-Report Form; (c) Mental Health Practitioner’s Guide for the ASEBA; (d) School-Based Practitioner’s Guide for the ASEBA; (e) Child and Family Service Worker’s Guide for the ASEBA; and (f) Medical Practitioner’s Guide for the ASEBA. The first two are comprehensive in scope and contain far more information than the average practitioner requires to use the scales. The guides are clearly directed at practitioners who will find them more useful than the technical manuals. Although there are four separate guides, the distinction between them is minor but is an excellent marketing strategy to reach a wider audience of users. Although it may simply be a matter of personal preference, this reviewer would find it helpful to have a “Users Manual” for all the instruments and a “Technical Manual” for all the instruments. Dividing the manuals in such a manner would make it easier for both practitioners and researchers to easily access desired information without having to wade through two information-dense manuals. SUMMARY. Overall, the ASEBA is a well-researched, empirically derived battery of instruments that allows a clinician to assess a wide range of behaviors across a variety of settings and informants. There are some minor drawbacks%2