Longitudinal and Concurrent Correlates of Reading Ability: Two Meta-Analytic Approaches

McWeeny, Sean

doi:https://doi.org/10.21985/n2-dbgx-jm79

Work

Longitudinal and Concurrent Correlates of Reading Ability: Two Meta-Analytic Approaches

Public

Download PDF

Download All Files (.zip)

Learning to read is a complex process that requires integration across sensory, cognitive, and linguistic processes. Accordingly, there are numerous process that may lead to early reading difficulties. The earlier these difficulties are found, the more effective interventions can be, and the deleterious effects of falling behind in reading cannot be understated. This dissertation tested correlates of early reading ability with two meta-analytic approaches to address the considerable heterogeneity present in reading research. Study 1 investigated rapid automatized naming (RAN), a predictor of future reading across different ages, ability levels, and languages, which may be useful in literacy screening for reading disability (RD). To investigate the longitudinal relationship between preschool/kindergarten RAN and future reading, we conducted a systematic review and meta-analysis (N = 60 samples; k = 373 effect sizes; n = 10,513 participants), in which we also tested whether characteristics of the RAN tasks, reading measures, or sample demographics moderate this relationship. Our results show that kindergarten/preschool RAN is correlated with grade-school reading at r = -.38, similar in magnitude to previous concurrent meta-analyses. We also found that RAN has independent predictive ability above and beyond phonological awareness (PA), which has clear theoretical and practical impacts. This meta-analysis was the first to measure RAN’s unique effect on reading, as well as the first to test practical and theoretical moderators longitudinally. Study 2 tested another early correlate of reading ability, auditory processing. Several hypotheses exist regarding the link between RD and auditory processing impairments, but none fully account for the range of impairments reported. These impairments have been primarily summarized by qualitative reviews, but these reviews fall short in numerous key domains. To understand the full range and size of deficits in individuals with RD, we conducted a systematic review and meta-analysis (N = 63, k = 135; n = 3,545) on four auditory task domains: frequency, duration, and intensity discrimination, as well as gap detection. Our results show large impairments (g = .6 to g = .8) in each domain for individuals with RD, undermining causal hypotheses of RD from highly specific deficits. These results motivate future testing of auditory processing abilities as a correlate of reading ability, as our meta-analysis was the first to quantify deficits in duration and intensity discrimination, as well as gap detection. These studies have clear implications relating to universal screening in reading research and meta-science more broadly.

Creator