It has been noted that clinical populations complaining of excessive daytime sleepiness (EDS) frequently have disrupted or fragmented nocturnal sleep. The relation between sleep fragmentation and daytime sleepiness has not been systematically studied. This study was designed to use correlational techniques evaluating the relation between these variables in patients complaining of EDS, patients complaining of insomnia, and asymptomatic controls. The four groups studied included patients complaining of EDS with sleep apnea (n = 15) or with periodic leg movements (n = 15), patients complaining of insomnia (n = 15), and healthy volunteers with no sleep complaint (n = 10). One night of polysomnography followed by a Multiple Sleep Latency Test was obtained for each subject. Each recording was evaluated using standard criteria and also by a four-level arousal scoring system. Across all subjects, the total number of arousals correlated significantly with sleepiness index (r = 0.48, p less than 0.001). Closer analysis of the data shows that, depending upon the sleep complaint, different types of arousals are predictive of degree of daytime sleepiness. It is concluded that the number and type of nocturnal arousals play an important role in subsequent daytime sleepiness.