An independent examination of the equivalence of the standard and digital administration formats of the Wechsler Intelligence Scale for Children-5th Edition
Autor: | Kacey Gilbert, John H. Kranzler, Nicholas Benson |
---|---|
Rok vydání: | 2020 |
Předmět: |
Psychometrics
Large effect size 05 social sciences Wechsler Scales 050301 education Wechsler Adult Intelligence Scale Standard score behavioral disciplines and activities Education Test (assessment) Statistics Developmental and Educational Psychology Main effect Humans 0501 psychology and cognitive sciences Psychology Child 0503 education Equivalence (measure theory) 050104 developmental & child psychology Wechsler Intelligence Scale for Children Coding (social sciences) |
Zdroj: | Journal of school psychology. 85 |
ISSN: | 1873-3506 |
Popis: | A new administration format for the Wechsler Intelligence Scale for Children-Fifth Edition (WISC-V; Wechsler, 2014) was introduced in 2016 on Q-interactive, Pearson's digital platform for test administration and scoring. The current study examined the measurement unit equivalence of the WISC-V standard and digital administration formats using counter-balanced administration of the 10 primary subtests to measure intellectual ability. The results indicated that correlations (r) between standard scores on subtests and composites administered in each format were generally moderate, with mean rs of 0.64 for subtests and 0.71 for composites after correction for attenuation, with the lowest rs for processing speed. Split-plot ANOVAs were conducted to examine within-subjects main effects for administration format and order and their interaction. The results of these analyses revealed significant main effects for format for the Full Scale IQ and Processing Speed composite scores, with small to medium effect sizes (ds > 0.40). These format effects largely stemmed from the non-equivalence of the Coding subtest, which is used to derive both composites. For Coding, the main effect for format was statistically significant, with a large effect size (d = 0.69). Statistically significant administration order by format interaction effects were also observed for a number of composites and subtests, with medium to large effect sizes (ƞps2 > 0.20). In each case, higher mean scores were observed when the WISC-V was administered first in digital format. Implications of these results for research and practice are discussed. |
Databáze: | OpenAIRE |
Externí odkaz: |