Objectives: Premorbid estimation of cognitive function is essential for the interpretation of the presence and severity of actual cognitive impairment. The most reliable method in Spanish speaking countries is with the Word Accentuation Test (WAT) / Test de Acentuación de Palabras (TAP). This is used to predict intelligence test scores by linear regression. Results: In an Ecuadorian sample the TAP was found to have good internal consistency and test-retest reliability. The correlation of the TAP with WAIS-IV full IQ scores was high (r= .827), allowing the development of a regression equation to estimate IQ scores from TAP performance. Furthermore, a sample of dementia patients was found to perform normally on the WAT compared to a matched control group. This suggests that WAT performance holds in the presence of neurological illness with associated cognitive impairment. Conclusions: The WAT has good psychometric properties and can be used to rapidly estimate actual intelligence test scores in healthy participants. It can also estimate premorbid intelligence scores in patients with neurological or psychiatric illnesses, allowing a clearer interpretation of the severity of impairment. This simple assessment could be used in various research and clinical contexts.