Sample size and the study of F waves

Ulnar nerve F waves were studied in 23 healthy volunteers and 27 diabetic patients. Latencies and chronodispersion were analyzed in each group for different sample sizes. Significant differences were not detected with the different sample sizes for mean latencies, with samples above 16 stimuli or 10 waves for minimum and maximum latencies and above 20 stimuli or 16 waves for chronodispersion. These findings suggest that these limits may be adequate for group comparison. However, for the analysis of individual patients, the evidence suggests that larger samples are required for the determination of the minimum and maximum latencies and chronodispersion.