Individual Variability in Reading Processes: The Compound Remote Associates Test
Even among skilled readers, there is a considerable degree of variability in specific sub-skills and processing proficiency. A number of behavioral and physiological tests have been designed and implemented in an attempt to uncover the foundations of these differences. The Compound Remote Associates Test (CRAT) was designed in 2003 (Bowden & Jung-Beeman, 2003a; Bowden & Jung-Beeman, 2003b), examining participants' ability to identify and comprehend compound words in the English language. That study was designed to examine problem-solving aspects related to insight and conscious analysis. The purpose of the current study is to determine if a modified version of this test can be used as a measure of variability in reading profiles. The empirical data was collected at Wesleyan University during the fall of 2015. Participants (N= 70) completed four separate assessments of reading sub-skills; a modified CRAT, the vocabulary and reading comprehension portions of the Nelson- Denny Reading Test (NDRT), and a spelling dictation task. Results revealed that 20.9% of the variability in CRAT performance could be accounted for by the other measures which were used. Multiple regression analyses revealed that spelling was a significant predictor of CRAT performance and vocabulary a moderate predictor when performance on the other measures was accounted for. Additionally, a marginal effect of solution type emerged whereby Type 2 solutions were most readily completed. Together, results indicate that the modified CRAT is indeed a useful tool for determining individual differences in reading, and may be indicative of variability in lexical expertise and morphological awareness.