In January, February and March of 1989 an epidemic of rubeola occurred on the campus of Texas Tech University. A vaccination programme was initiated as soon as the epidemic was confirmed. Extensive case histories of all confirmed cases were collected by the Lubbock City Health Department and given an exhaustive statistical analysis by a group from the Department of Mathematics at Texas Tech University. The data and statistical analysis were used to formulate stochastic and deterministic models of the measles epidemic based on the standard SEIR model. The analysis and the simulations indicate that in order to prevent a measles outbreak on a university campus a high rate of immunity may be required (> 98 per cent). The assumptions in the models raise some interesting questions regarding social contacts which require further investigation.