A Meta-Analytic Evaluation of Naglieri Nonverbal Ability Test: Exploring Its Validity Evidence and Effectiveness in Equitably Identifying Gifted Students

The Naglieri Nonverbal Ability Test (NNAT) was developed to more equitably identify students of color, as it advertises itself as a culture-fair measure. In this meta-analytic evaluation, we aimed to investigate (a) the generalizability of validity evidence of NNAT by checking its construct and criterion validity with other measures (Part I) and (b) whether NNAT truly meets its goal to identify more culturally diverse students (Part II). After reviewing 1,714 studies, a total of 29 studies met our criteria (59 effect sizes from 22 studies for Part I and 7 effect sizes from 7 studies for Part II). In Part I, we investigated empirical evidence of validity of NNAT in relationship with different types of measures (overall effect size of r was .44); The results revealed that the correlation between NNAT and the achievement test results was 0.68, followed by the intelligence measures similar to NNAT (e.g., Cognitive Abilities Test, Otis–Lennon School Ability Tests; r = .31) and other alternative measures often used to identify gifted students (e.g., teacher-rating scale; r = .20). The moderator analysis results showed high correlations between NNAT and other measures when Naglieri is an author of the study. In Part II, although NNAT identified more students of color compared with other nonverbal tests (overall effect size of risk ratio was 0.42), findings revealed that students of color remain underrepresented in gifted programs and services.

Differences in Using the Cognitive Abilities Test (CogAT) 7 Nonverbal Battery Versus the Naglieri Nonverbal Ability Test (NNAT) 2 to Identify the Gifted/Talented

The two most commonly used nonverbal tests for gifted identification, the Naglieri Nonverbal Ability Test (NNAT) and the Cognitive Abilities Test (CogAT) nonverbal battery, have not been compared in their newer versions to explore the effects of their use on the identification of underserved populations. Additionally, the effects of the use of various norming groups and cutoff scores on both instruments’ identification abilities has not been compared. This study compared 15,733 CogAT7 nonverbal battery scores and 14,421 NNAT2 scores of kindergartners between 2013 and 2015 from one large urban school district to explore the differences between how each test relates to major demographic variables and examine the effects on who is selected for participation in gifted programming based on which instrument, which norming group, and which cutoff scores are used. Both instruments were less likely to identify students from demographic groups that have been traditionally underrepresented than students from traditionally overrepresented demographic groups, but identification varied based on the type of norming plan used and which instrument was taken. Suggestions are made as to the best instrument for use with various demographic groups and norming plans.

Measuring Goal Setting in School-Aged Children: Studying the Effects of Demographic Variables in Regression-Based Norms

The aim of the study was to establish demographically representative norms for tasks measuring goal setting, and more specifically planning and reasoning in children. Three tasks were administered to n = 195 Ukrainian children aged 5.10 to 14.5 years old: the Spatial Working Memory (SWM), the Stockings of Cambridge (SOC) test, and the Naglieri Nonverbal Ability Test (NNAT). Main outcome per test was accuracy: i.e., the total number correct for the SOC and NNAT, and the total amount of incorrect responses for the SWM. Correlations among accuracy measures varied from − 0.51 to 0.60, indicating these tasks measure related but at the same time unique constructs. Higher age was associated with more accurate test performances on all outcome measures. On the NNAT, we found a curvilinear association between age and accuracy, indicating that younger children’s NNAT accuracy scores increased more with age compared with older children. We found a cubic age effect on accuracy for the SWM and SOC: i.e., test scores were relatively stable at younger and older ages, with a curvilinear increase in test scores in the other age groups. Demographically corrected norms were calculated and presented per test. These indicated that sex was not associated with accuracy scores on any of the tests. Last, a higher level of parental education (LPE) was associated with higher accuracy scores, but only on the NNAT. We conclude that demographic variables in norm analyses enhance insight in the scores and allow for application in clinical settings and research.