Multilevel modeling of cognitive ability in highly functioning adults
The goal of this research was to study differences in cognitive performance on verbal and quantitative measures among subjects of different ages. Data was gathered on subjects ranging in age from 16 to 80 years of age from birth-cohorts from 1927 to 1990. In addition to year of birth, personal characteristics of gender, race/ethnicity and undergraduate area of study were obtained. Multilevel models were built that predict cognitive performance as a function of age, cohort and other non-independent personal characteristics . Verbal performance rises as the age of the test-taker rises; quantitative performance declines as the age of the test-taker rises. After controlling for the race/ethnicity and gender of the test-taker, there are both age and cohort effects for verbal and quantitative models. On the verbal measure, the cohort effect favors those test-takers born at an earlier time. There is an interaction between age and cohort on the quantitative measure. This data is secondary analysis and the records are from those test-takers who choose to take a consequential assessment. When the multilevel models are produced independently for those test-takers ages 20-39 and those ages 40-64, different results are seen between the two age groups. There is little difference in performance for 20-39 year olds on the verbal measure other than a positive effect for age at time of test. For the test-taker aged 40-64, there is a positive effect due to age, a positive cohort effect and a negative interaction effect between age and science study. Comparing the 20-39 year olds with the 40-64 year olds on the quantitative measure, the decline in performance for the older group is one-fourth the rate of decline in the younger group. For the quantitative measure, after controlling for age, there is a positive cohort effect for both age groups. ^
Psychology, Developmental|Psychology, Psychometrics|Psychology, Cognitive
Trapani, Catherine Schuler, "Multilevel modeling of cognitive ability in highly functioning adults" (2013). ETD Collection for Fordham University. AAI3564971.