Our objective in this study was to examine children with and without listening difficulty (LiD) longitudinally to ask whether LiD was developmentally transient or persistent. Like many other developmental problems, we hypothesized that LiD persists into adolescence. Behavioral and physiological data were collected from children initially aged 6-13 years at baseline and two and four years later. Of 169 enrolled participants, 147, 100, and 31 children completed required testing in study visits at baseline, 2 years, and 4 years later, respectively. All children had clinically normal audiograms at all visits. Listening skills scores from a caregiver report scale (ECLiPS), and auditory and cognitive skills were significantly poorer in the LiD group than in a typically developing (TD) group throughout the study period. In both groups, age-adjusted listening and auditory skills increased over time. Using the longitudinal data, a parsimonious prediction model for ECLIPS scores (pooled across groups) was created. The final model included maternal education, spatial listening skills, and cognitive performance, which explained 54.8% of the variance in the ECLiPS score. Children with LiD but normal audiograms have, relative to TD children, persistent listening difficulties, challenges in competing speech tasks, and poor cognitive performance through adolescence. The degree of LiD can be independently predicted by maternal education, cognitive performance, and spatial listening skills.