As part of a larger project on the effects of benzodiazepine and caffeine on daytime sleepiness, performance and mood, this study examined the relationship among the Multiple Sleep Latency Test, lapses during a tapping task, a Visual Analog Scale, and the Stanford Sleepiness scale. Subjects were 80 male, adult nonsmokers aged 20.3±2.7 years. The Multiple Sleep Latency Test, Stanford Sleepiness Scale, and the Visual Analog Scale were obtained at two-hour intervals beginning at 0700 h and ending at 1700 h. The tapping task (lapses) was administered each day at 0600 h, 1000 h, and 1400 h. A lapse was a 3-s or greater pause between taps. Correlations between the Multiple Sleep Latency Test and subjective (Visual Analog Scale and the Stanford Sleepiness Scale) measures were significant at 0600 h, but became nonsignificant as the day progressed. Correlations between lapses and the two subjective measures were generally nonsignificant. The two objective measures were significantly correlated in the total group but not in all treatment groups. The subjective measures were significantly correlated in the total sample and in each treatment group. This study reaffirms the importance of time of day when measuring sleepiness, and suggests that subjective and objective measures may measure different aspects of sleepiness.