This paper presents the zero-inflated generalised Poisson distribution, which is useful when there is a large presence of zeros in the sample. After presenting the model, we develop a specific program based on Mathematica, overcoming some limitations of alternative approaches such as STATA or EViews, which do not include the zero-inflated Poisson distribution among its routines. The advantages of the model used and the proposed program are illustrated with a real example that is very appropriate to its features, namely an analysis of the factors influencing university students’ attendance at tutoring sessions. This example is particularly suitable to show the usefulness of the methodology presented because it includes a large number of zeros, reflecting the many occasions on which the students do not attend these sessions. The students’ place of residence, their attendance at lectures and the application of continual assessment are variables that seem to account for attendance at tutoring sessions.