Scholarlink

<p>This study was designed to examine the utility of embedded validity indicators (EVIs) within the WAIS-IV in the differential diagnosis of intellectual disability (ID). Archival data were collected through a retrospective clinical chart review from a consecutive case sequence of 88 community dwelling adults (54.5% men; 30 with ID) from the UK referred for a comprehensive evaluation to determine the presence or absence of ID (<i>M</i><sub>Age</sub> = 31.1; <i>M</i><sub>FSIQ</sub> = 63.7). Base rates of failure (BR<sub><i>Fail</i></sub>) on commonly used EVIs [age-corrected scaled scores on Digit Span, Symbol Search (SS) and Coding (CD) subtests] were very high (62.1–96.6%) within the ID subsample. Surprisingly, multivariate models were ineffective at containing BR<sub><i>Fail</i></sub>. However, a cutoff of ≥ 4 on two derivative EVIs (|CD – SS| and Digit Span minus Vocabulary) achieved low BR<sub><i>Fail</i></sub> (1.7–3.4%). None of the participants with ID scored ≥ 5 on these EVIs (i.e., zero false positive rate). Univariate EVIs within the WAIS-IV based on the <i>method of threshold</i> are inappropriate during the differential diagnosis of ID due to unacceptably high false positive rates likely driven by genuine and severe cognitive deficits. Aggregating multiple EVIs is unlikely to effectively mitigate the threat of inflated false positive rates in this population. However, EVIs based on contrast scores are robust to severe cognitive impairment, as they were designed to detect internal inconsistencies rather than “suspiciously low performance”. Therefore, they provide effective measures of performance validity in this psychometrically challenging signal detection environment.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Mission Impossible? Identifying Bona Fide Intellectual Disability Using Embedded Validity Indicators within the WAIS-IV

  • 期刊论文

  • 发表日期：

    2025年9月11日

查看全文
  • Michael P. Murphy,
  • Sarah Schneider,
  • Alastair L. Barrowcliff,
  • Michael Bayrhammer-Savel,
  • Iulia Crisan,
  • Laszlo A. Erdodi

摘要

This study was designed to examine the utility of embedded validity indicators (EVIs) within the WAIS-IV in the differential diagnosis of intellectual disability (ID). Archival data were collected through a retrospective clinical chart review from a consecutive case sequence of 88 community dwelling adults (54.5% men; 30 with ID) from the UK referred for a comprehensive evaluation to determine the presence or absence of ID (MAge = 31.1; MFSIQ = 63.7). Base rates of failure (BRFail) on commonly used EVIs [age-corrected scaled scores on Digit Span, Symbol Search (SS) and Coding (CD) subtests] were very high (62.1–96.6%) within the ID subsample. Surprisingly, multivariate models were ineffective at containing BRFail. However, a cutoff of ≥ 4 on two derivative EVIs (|CD – SS| and Digit Span minus Vocabulary) achieved low BRFail (1.7–3.4%). None of the participants with ID scored ≥ 5 on these EVIs (i.e., zero false positive rate). Univariate EVIs within the WAIS-IV based on the method of threshold are inappropriate during the differential diagnosis of ID due to unacceptably high false positive rates likely driven by genuine and severe cognitive deficits. Aggregating multiple EVIs is unlikely to effectively mitigate the threat of inflated false positive rates in this population. However, EVIs based on contrast scores are robust to severe cognitive impairment, as they were designed to detect internal inconsistencies rather than “suspiciously low performance”. Therefore, they provide effective measures of performance validity in this psychometrically challenging signal detection environment.