Test of Problem Solving 3E-Normative Update (TOPS-3E:NU)
Authors: Linda Bowers, MA, CCC-SLP / Rosemary Huisingh, MA, CCC-SLP / Carolyn LoGiudice, MA, CCC-SLP
- Critical Thinking
- Individual Administration
- Ages 6-0 through 12-11
- Qualification Level B
The TOPS-3E:NU assesses a school-age child’s ability to integrate semantic and linguistic knowledge with reasoning ability by way of picture stimuli and verbal responses.
The TOPS-3E:NU focuses on students’ linguistic ability to think and reason. Language competence is the verbal indicator of how a student’s language skills affect his ability to think, reason, problem solve, infer, classify, associate, predict, determine causes, sequence, and understand directions. The test focuses on a broad range of language-based thinking skills, including clarifying, analyzing, generating solutions, evaluating, and showing affective thinking.
While other tests may assess students’ thinking skills by tapping mathematical, spatial, or nonverbal potential, the TOPS-3E:NU measures discreet skills that form the foundation of language-based thinking, reasoning, and problem-solving ability.
Although the skills tested on the TOPS-3E:NU are necessary for developing social competence, it is not primarily a test of pragmatic or social language skills. Rather, it should be part of a battery of tests and observations used to assess pragmatic competence.
- Characteristics of the normative sample were stratified by age relative to region, gender, ethnicity, and socioeconomic factors, and other critical variables are the same as those reported for the school-age population reported in the ProQuest Statistical Abstract of the United States, 2016 (ProQuest, 2016).
- The Total Score was renamed the Problem Solving Index and calculated as a standard score with a mean of 100 and a standard deviation of 15.
- Each item on the test was evaluated using both conventional item analysis to choose “good” items and differential item analyses to find and eliminate potentially biased items.
- The index score was thoroughly examined for floor and ceiling effects.
- The test was subjected to diagnostic accuracy analyses, particularly rigorous techniques involving the computation of the receiver operating characteristic/area under the curve (ROC/AUC) statistic.
- The Examiner’s Manual was reorganized and rewritten to provide more detailed information on the administration, interpretation, and statistical characteristics of the test.
Administration and Scoring
The TOPS-3E:NU has three components: an Examiner’s Manual, Examiner Record Booklets, and a Picture Book. The Examiner’s Manual includes a comprehensive discussion of the test’s theoretical and research-based foundation, item development, standardization, administration and scoring procedures, norm tables, and guidelines for using and interpreting the test’s results. The Examiner Record Booklet provides space to record responses and transform the raw score to an age equivalent, percentile rank, and the Problem Solving Index. The test kit also includes a Picture Book, which includes the picture stimuli for the test items.
Reliability and validity studies were conducted with individuals who have typical language ability and individuals who had been previously diagnosed with a language impairment or received other special education services. The average coefficient alpha is .82 for the Problem Solving Index. Studies were conducted to examine the ability of the test to differentiate students who receive special education services or have language impairments from those who do not. The results demonstrate that a Problem Solving Index cutoff score of 90 resulted in a sensitivity of .75, a specificity of .85, and a ROC/AUC of .74 for differentiating students who receive special education services; and a cutoff score of 92 resulted in a sensitivity of .69, a specificity of .89, and a ROC/AUC of .73 for differentiating students who have a language impairment. Validity of the test composites was demonstrated by correlations to the Universal Nonverbal Intelligence Test–Group Abilities Test (UNIT-GAT; Bracken & McCallum, in development). The coefficient for the Analogic Reasoning subtest was .73, and the coefficient for the Quantitative subtest was .89, both very large.