Limits...
Improving IQ measurement in intellectual disabilities using true deviation from population norms.

Sansone SM, Schneider A, Bickel E, Berry-Kravis E, Prescott C, Hessl D - J Neurodev Disord (2014)

Bottom Line: We compared the distributional characteristics and Q-Q plots from the standardized scores with the deviation z-scores.Use of the deviation z- score method rectifies this problem, and accounts for significant additional variance in criterion validation measures, above and beyond the usual IQ scores.However assessment of cognitive abilities is substantially improved by measuring true deviation in performance from standardization sample norms.

View Article: PubMed Central - HTML - PubMed

Affiliation: Medical Investigation of Neurodevelopmental Disorders (MIND) Institute, University of California at Davis Medical Center, 2825 50th Street, Sacramento, CA 95817, USA.

ABSTRACT

Background: Intellectual disability (ID) is characterized by global cognitive deficits, yet the very IQ tests used to assess ID have limited range and precision in this population, especially for more impaired individuals.

Methods: We describe the development and validation of a method of raw z-score transformation (based on general population norms) that ameliorates floor effects and improves the precision of IQ measurement in ID using the Stanford Binet 5 (SB5) in fragile X syndrome (FXS; n = 106), the leading inherited cause of ID, and in individuals with idiopathic autism spectrum disorder (ASD; n = 205). We compared the distributional characteristics and Q-Q plots from the standardized scores with the deviation z-scores. Additionally, we examined the relationship between both scoring methods and multiple criterion measures.

Results: We found evidence that substantial and meaningful variation in cognitive ability on standardized IQ tests among individuals with ID is lost when converting raw scores to standardized scaled, index and IQ scores. Use of the deviation z- score method rectifies this problem, and accounts for significant additional variance in criterion validation measures, above and beyond the usual IQ scores. Additionally, individual and group-level cognitive strengths and weaknesses are recovered using deviation scores.

Conclusion: Traditional methods for generating IQ scores in lower functioning individuals with ID are inaccurate and inadequate, leading to erroneously flat profiles. However assessment of cognitive abilities is substantially improved by measuring true deviation in performance from standardization sample norms. This work has important implications for standardized test development, clinical assessment, and research for which IQ is an important measure of interest in individuals with neurodevelopmental disorders and other forms of cognitive impairment.

No MeSH data available.


Related in: MedlinePlus

Subtest raw scores (open triangles), standardized scores (closed circles), and deviation (z) scores (closed squares) for two case examples. FR, fluid reasoning; KN, knowledge; NV, non-verbal; QR, quantitative reasoning; V, verbal; VS, visual spatial processing; WM, working memory.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4613563&req=5

Figure 3: Subtest raw scores (open triangles), standardized scores (closed circles), and deviation (z) scores (closed squares) for two case examples. FR, fluid reasoning; KN, knowledge; NV, non-verbal; QR, quantitative reasoning; V, verbal; VS, visual spatial processing; WM, working memory.

Mentions: Figure 3 provides two case examples of how the practical application of this scoring method may improve a clinician’s understanding of a patient’s cognitive strengths and weaknesses compared to the traditional standardized scoring method. ‘Alex’ is a nine-year-old male diagnosed with autism. His cognitive assessment yielded a FSIQ of 40, EXIQ of 10, and VABC score of 61. As can be seen in panel (A), Alex received a standard score of 1 across all subtests producing the horizontal line with no variability and possibly leading to the conclusion that he is equally affected across all domains on the SB5. In contrast, we see that his performance on nine of the ten subtests are actually more than four standard deviations below the mean and he shows a larger deficit on many of the verbal subtests relative to their non-verbal equivalents, especially in visual spatial processing.‘Jake’ is a 19-year-old male diagnosed with FXS, who received a FSIQ of 40, EXIQ of 10, and a VABC score of 32. While Jake and Alex had identical FSIQ, EXIQ, and standardized subtest scores, their deviation score profiles are substantially different. In panel (B) (Figure 3), despite a flat profile of standardized scores, deviation subtest scores reveal that non-verbal fluid reasoning and quantitative reasoning are areas of relative weakness while he performs better on non-verbal knowledge and verbal visual spatial processing.


Improving IQ measurement in intellectual disabilities using true deviation from population norms.

Sansone SM, Schneider A, Bickel E, Berry-Kravis E, Prescott C, Hessl D - J Neurodev Disord (2014)

Subtest raw scores (open triangles), standardized scores (closed circles), and deviation (z) scores (closed squares) for two case examples. FR, fluid reasoning; KN, knowledge; NV, non-verbal; QR, quantitative reasoning; V, verbal; VS, visual spatial processing; WM, working memory.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4613563&req=5

Figure 3: Subtest raw scores (open triangles), standardized scores (closed circles), and deviation (z) scores (closed squares) for two case examples. FR, fluid reasoning; KN, knowledge; NV, non-verbal; QR, quantitative reasoning; V, verbal; VS, visual spatial processing; WM, working memory.
Mentions: Figure 3 provides two case examples of how the practical application of this scoring method may improve a clinician’s understanding of a patient’s cognitive strengths and weaknesses compared to the traditional standardized scoring method. ‘Alex’ is a nine-year-old male diagnosed with autism. His cognitive assessment yielded a FSIQ of 40, EXIQ of 10, and VABC score of 61. As can be seen in panel (A), Alex received a standard score of 1 across all subtests producing the horizontal line with no variability and possibly leading to the conclusion that he is equally affected across all domains on the SB5. In contrast, we see that his performance on nine of the ten subtests are actually more than four standard deviations below the mean and he shows a larger deficit on many of the verbal subtests relative to their non-verbal equivalents, especially in visual spatial processing.‘Jake’ is a 19-year-old male diagnosed with FXS, who received a FSIQ of 40, EXIQ of 10, and a VABC score of 32. While Jake and Alex had identical FSIQ, EXIQ, and standardized subtest scores, their deviation score profiles are substantially different. In panel (B) (Figure 3), despite a flat profile of standardized scores, deviation subtest scores reveal that non-verbal fluid reasoning and quantitative reasoning are areas of relative weakness while he performs better on non-verbal knowledge and verbal visual spatial processing.

Bottom Line: We compared the distributional characteristics and Q-Q plots from the standardized scores with the deviation z-scores.Use of the deviation z- score method rectifies this problem, and accounts for significant additional variance in criterion validation measures, above and beyond the usual IQ scores.However assessment of cognitive abilities is substantially improved by measuring true deviation in performance from standardization sample norms.

View Article: PubMed Central - HTML - PubMed

Affiliation: Medical Investigation of Neurodevelopmental Disorders (MIND) Institute, University of California at Davis Medical Center, 2825 50th Street, Sacramento, CA 95817, USA.

ABSTRACT

Background: Intellectual disability (ID) is characterized by global cognitive deficits, yet the very IQ tests used to assess ID have limited range and precision in this population, especially for more impaired individuals.

Methods: We describe the development and validation of a method of raw z-score transformation (based on general population norms) that ameliorates floor effects and improves the precision of IQ measurement in ID using the Stanford Binet 5 (SB5) in fragile X syndrome (FXS; n = 106), the leading inherited cause of ID, and in individuals with idiopathic autism spectrum disorder (ASD; n = 205). We compared the distributional characteristics and Q-Q plots from the standardized scores with the deviation z-scores. Additionally, we examined the relationship between both scoring methods and multiple criterion measures.

Results: We found evidence that substantial and meaningful variation in cognitive ability on standardized IQ tests among individuals with ID is lost when converting raw scores to standardized scaled, index and IQ scores. Use of the deviation z- score method rectifies this problem, and accounts for significant additional variance in criterion validation measures, above and beyond the usual IQ scores. Additionally, individual and group-level cognitive strengths and weaknesses are recovered using deviation scores.

Conclusion: Traditional methods for generating IQ scores in lower functioning individuals with ID are inaccurate and inadequate, leading to erroneously flat profiles. However assessment of cognitive abilities is substantially improved by measuring true deviation in performance from standardization sample norms. This work has important implications for standardized test development, clinical assessment, and research for which IQ is an important measure of interest in individuals with neurodevelopmental disorders and other forms of cognitive impairment.

No MeSH data available.


Related in: MedlinePlus