British Journal of Clinical Psychology · 1993 · 383 citations · 1 references
EducationEducational AssessmentCognitive PsychologyPsycholinguisticsReading DevelopmentCognitive AgingCognitive AssessmentCognitive DeclineCognitive ImpairmentIntelligence AssessmentWord RecognitionAge-graded VocabularyAge-related Intelligence DifferencesNormative Data CollectionParallel Test FormsPremorbid Cognitive EstimationPremorbid Intelligence EstimationRecognition Memory TestsVerbal Intelligence Tests
The authors developed the Spot‑the‑Word test to estimate premorbid verbal intelligence and examined its stability across age and cognitive decline. The test presents subjects with word/non‑word pairs, requiring word identification, and was evaluated using two parallel forms and normative data from 224 participants stratified by age and socioeconomic status. Spot‑the‑Word scores strongly correlate with Mill Hill Vocabulary and NART scores, remain stable with age, and show high reliability (r ≈ 0.88 between forms) and validity, supporting its use as a premorbid intelligence estimate.
The development of a test aimed at estimating premorbid intelligence is described. The test, Spot-the-Word, involves presenting the subject with pairs of items comprising one word and one non-word, and requiring the subject to identify the word. Data show that performance correlates highly with verbal intelligence as estimated by Mill Hill Vocabulary score and by performance on the National Adult Reading Test (NART). Performance does not decline with age, in contrast to an associated test of verbal recognition memory. A second study attempted to test the effect of intellectual deterioration due to age on Spot-the-Word performance. Elderly subjects who had high vocabulary scores scored well on the Spot-the-Word regardless of whether fluid intelligence as measured by the AH4 test was well preserved, or was low, implying intellectual deterioration. A final study collected normative data on a sample of 224 subjects stratified by age and socio-economic status, with each subject performing two parallel forms of the test, A and B, together with the NART. Correlation between the two forms was .884, while correlation with NART was .831 for Form A and .859 for Form B, suggesting adequate reliability and validity. It is concluded that the test provides a potentially useful additional method of estimating premorbid intelligence.
1