| Janet Sheehan-Holt
Northern Illinois University
Factors affecting adult literacy: A model-testing approach for investigating associations among variables in the NALS
FINAL REPORT:
The purpose ofthis study was to demonstrate how multilevel analyses can be used with data such as the NALS. The results of ordinary least squares (OLS) regression and hierarchical linear modeling (HLM) analyses were compared for modeling predictors of literacy from the National Adult Literacy Survey. Our results indicate that (a) contextual factors, such as mean income ofthe neighborhood, are important to take into account with the NALS, particularly when studying race effects, (b) contextual effects estimates and their standard errors differed between HLM and OLS, and (c) contextual effects studies of adult literacy using OLS produced a different model of the predictors of adult literacy than did HLM. Statistical justification is given for the discrepant results between the two methods and HLM is recommended as the appropriate statistical tool for studying predictors of literacy using the NALS.
Back to Funded Research Grants Page |