clinician can compare the ratings of each informant to the responses of other parents or teachers with children of the same age and gender. Similarly, an adolescent’s self-report ratings can be compared to the reports of other youths of the same age and gender. The results are shown as T scores that reflect the degree to which the child’s functioning deviates from other youths. Problem scores two or more standard deviations above the mean (≥70) or adaptive skill scores two or more standard deviations below the mean (≤30) indicate clinically significant difficulties that may merit treatment (Kamphaus & Dever, 2018).
Sara’s mother and homeroom teacher completed the BASC-3 to assess Sara’s functioning at home and school, respectively (Table 4.3). Her mother’s ratings suggested that Sara is experiencing significant anxiety and somatic complaints at home compared to other girls her age. In contrast, her teacher reported few problems at school. These data indicate that Sara might manifest anxiety in terms of physical problems, like headaches and stomachaches, and that she might be worried about her family.
Specific Symptom Inventories
Clinicians can also administer other tests to assess specific disorders. For example, the Autism Spectrum Rating Scales (ASRS; Goldstein & Naglieri, 2013) are widely used to screen children suspected of autism. Clinicians administer the scales to parents and teachers who rate DSM-5 symptoms of the disorder. The ASRS also assess the child’s communication and socialization skills; tendency to engage in rigid, repetitive, or stereotyped behaviors; sensitivity to sensory stimuli (e.g., certain textures or sounds); and capacity for self-regulation. The ASRS are norm-referenced; scores allow clinicians to compare the child to other youths of approximately the same age as well as to children previously diagnosed with autism.
Table 4.3
Note: Compared to other girls her age, Sara shows clinically significant internalizing problems at home characterized by anxiety and somatic (i.e., physical health) problems. Her teacher reported no significant problems at school.
*Problem scores ≥ 70 and adaptive skill scores ≤ 30 suggest clinically significant problems.
The Conners 3 (Conners, 2015) is a behavior rating scale used to screen children for ADHD and disruptive behavior disorders. The test assesses DSM-5 symptoms of ADHD and can be administered to parents, teachers, and older children to provide multi-informant data regarding the child’s functioning at home and school. The test also assesses other potential problems such as oppositional behavior toward adults, learning difficulties, and peer rejection. The test yields T scores that allow clinicians to compare children to youths of the same age and gender.
The Revised Children’s Anxiety and Depression Scale (Weiss & Chorpita, 2011) might be administered to girls like Sara who show internalizing problems. This self-report questionnaire assesses five DSM-5 anxiety disorders as well as symptoms of depression. The scale yields T scores, which allow clinicians to compare a child’s ratings to other children of the same age and gender. Sara reported significant problems with separation anxiety compared to other girls her age (Figure 4.4). She experiences intense anxiety or panic when she must leave her parents and worries about them when she is away from them for extended periods of time.
Figure 4.4 ■ Sara’s Scores on the Revised Children’s Anxiety and Depression Scale
Note: Clinicians administer specific symptom inventories, like this one, to assess particular psychological problems. Sara reported significant (T ≥ 70) fears of separation compared to other girls her age.
Altogether, data from the diagnostic interview, observations, and norm-referenced tests indicate that Sara’s somatic symptoms and school refusal are caused by underlying anxiety about separating from her parents. Sara’s symptoms developed shortly after her father’s stroke. Because of her mother’s busy work schedule, Sara cared for her father over the summer as he recovered. As the academic year approached, Sara became preoccupied by thoughts that he might experience another stroke if she left him to attend school. Her anxiety about her father and fears of separation increased until she began to develop physical symptoms. By allowing Sara to stay home from school, her mother inadvertently reinforced these symptoms, which maintained Sara’s school refusal over time.
Review
The MMPI-A-RF is a broad, self-report measure of adolescents’ social–emotional functioning. It yields scores on three composites (i.e., emotions, behaviors, and thoughts) and nine clinical scales.
The BASC-3 can be completed by parents, teachers, or older children and adolescents to obtain an overall estimate of behavior problems and adaptive functioning.
Many tests of personality and social–emotional functioning yield T scores with a mean of 50 and standard deviation of 10.
What Makes a Good Psychological Test?
There are many kinds of psychological tests, but not all tests are created equal. The accuracy of a clinician’s diagnosis and recommendations for treatment depends on the quality of the tests he selects and the manner in which he uses them. In this section, we will discuss the three most important features of evidence-based testing: (1) standardization, (2) reliability, and (3) validity.
Standardization
Most tests used in clinical settings follow some sort of standardization—that is, they are administered, scored, and interpreted in the same way to all examinees. For example, all 7-year-old children who take the WISC–V are administered the same test items. Items are presented in the same way to all children according to specific rules described in the test manual. These rules include where participants must sit, how instructions must be presented, how much time is allowed, and what sort of help (if any) examiners can provide. Children’s answers are scored in the same way, using specific guidelines presented in the manual (Wechsler et al., 2014).
Standardized test administration and scoring allows clinicians to compare one child’s test scores with the performance of his or her peers. Two children who obtain the same number of correct test items on an intelligence test are believed to have comparable levels of cognitive functioning only if they were administered the test in a standardized fashion. If one child was given extra time, additional help, or greater encouragement by the examiner, comparisons would be inappropriate.
Most standardized tests, like the WISC–V, are norm-referenced. Norm-referenced tests allow clinicians to quantify the degree to which a specific child is similar to other youths of the same age, grade, and/or gender. These tests are called norm-referenced because the child is compared to a normative sample of children, a large group of youths whose demographics reflect a larger population, such as all children in the United States or children with ADHD. Examples of norm-referenced tests include intelligence tests, personality tests, and behavior rating scales (Achenbach, 2015).
Children’s scores on norm-referenced tests are compared to the performance of other children, in order to make these scores more meaningful. Imagine that a 9-year-old girl correctly answers 45 questions on the WISC–V. A clinician would record her “raw score” as 45. However, a raw score of 45 does not allow the clinician to determine whether the girl is intellectually gifted, average, or delayed. To interpret her raw score, the clinician needs to compare her raw score to children in the normative sample, that is, other children who have already completed the WISC–V. If the mean raw score for 9-year-olds in the normative sample was 45 and the girl’s raw score was 45, the clinician might conclude that the girl’s cognitive functioning is within the average range. However, if the mean raw score