Study objective: To explore how Student's t test, which assumes normal
errors, is affected by the nonnormal distribution of Glasgow Coma Sca
le (GCS) scores.Methods: A sample of 145,295 GCS scores from the Penns
ylvania Trauma Systems Foundation was assumed to represent the true GC
S distribution. Four Monte Carlo simulations, each based on 40,000 rep
lications, were conducted to approximate the distribution of t values
arising from samples drawn from the GCS distribution, using sample siz
es of 10, 30, 60, and 100 per group. Results: The histograms of t valu
es derived from the GCS distribution resembled the corresponding Stude
nt t distributions for sample sizes of 30 or more per group. The t tes
t was conservative in samples of size 10, conservative to acceptable i
n samples of sizes 30 and 60, and acceptable in samples of size 100. I
n no case was the t test liberal. Conclusion: The t test can be applie
d to GCS data with a type I error rate that never exceeds the investig
ator's chosen significance level. However, a sample size exceeding 30
per group is required for the error rate to adequately approximate the
chosen significance level.