Introduction
Hagmann-von Arx, P., Grob, A., & Hagmann, R. (2016) investigated the comparability of results on five intelligence tests commonly used in German-speaking countries, administered to 206 typically developing children. The authors found high correlations between test scores, indicating that they measure a similar underlying construct, interpreted as general intelligence. However, at the individual level, scores differed significantly between tests for 12% to 38% of children.
Critical Analysis
Research by Hagmann-von Arx et al. (2016) contributes significantly to the literature on intelligence assessment, addressing the issue of interchangeability of different instruments. The use of a relatively large sample and the inclusion of widely used tests increase the clinical relevance of the findings. Additionally, the study employs robust statistical analyses, including generalizability theory, to investigate variability in scores.
However, some limitations must be considered. The sample, despite being large, is homogeneous in terms of typical development, limiting the generalization of results to clinical populations. Furthermore, the study does not explore the reasons for individual differences in scores, which can be attributed to factors such as anxiety, motivation or learning style.
Implications for Clinical Practice
The results of the study by Hagmann-von Arx et al. (2016) have important implications for clinical practice. The high correlation between test scores supports the convergent validity of the instruments, suggesting that they measure a common construct. However, the individual differences found alert to the need for caution when interpreting results, especially in high-risk contexts, such as decisions about referral to special education services.
The research highlights the importance of considering measurement error and the use of confidence intervals when interpreting IQ scores. Furthermore, it suggests that administering multiple tests can increase the accuracy of the assessment, especially in cases where decisions made based on the results have a high impact on the child’s life.
Tests Used and Comparison of Results
The study used five intelligence tests common in German-speaking countries:
- RIAS (Reynolds Intellectual Assessment Scales): Assesses verbal, non-verbal and compound intelligence. This test, together with the SON-R 6-40, presented the lowest average scores.
- SON-R 6-40 (Snijders Oomen Nonverbal Intelligence Test): Nonverbal intelligence test.
- IDS (Intelligence and Development Scales): Measures fluid intelligence and provides a developmental profile across six domains. This test, along with the WISC-IV and the CFT 20-R, presented the highest average scores.
- WISC-IV (Wechsler Intelligence Scale for Children): Assesses intellectual abilities in children and adolescents, focusing on verbal comprehension, perceptual reasoning, working memory and processing speed. More complete test.
- CFT 20-R (Culture Fair Intelligence Test): Measures non-verbal fluid intelligence.
Despite the high correlation between test scores, indicating that they measure a similar construct, the study by Hagmann-von Arx et al. (2016) found significant differences in the scores of the same child on different tests. These differences ranged from 12% to 38% depending on the test combination. The differences between test means were small, ranging from 1 to 5 IQ points, and can be attributed to the Flynn effect.
For example, when comparing the results of the RIAS and WISC-IV, 38% of children showed greater differences in scores than expected considering the tests’ margin of error. This means that a child could be classified as having average intelligence on one test (RIAS) and above average intelligence on another (WISC-IV).
Conclusions and Future Perspectives
In short, the study by Hagmann-von Arx et al. (2016) provides valuable evidence on the comparability of intelligence tests in children. The results reinforce the importance of a careful and individualized assessment, considering the specific characteristics of each child and the context of the assessment.
Future research could investigate the reasons for individual differences in scores, exploring the role of factors such as motivation, anxiety and learning style. Additionally, studies with diverse clinical samples would allow a better understanding of the comparability of tests in different populations.
Reference
Hagmann-von Arx, P., Grob, A., & Hagmann, R. (2016). Comparability of IQ Scores from Five Frequently Used Intelligence Tests in Children with Typical Development. PLoS ONE , 11 (11), e0165148.
