Probit regression is an alternative log-linear approach to handle the dependent variable category analysis. Probit regression that can be used in experimental data, so you want to be applied to the survey data on the age of the first menstrual coming in teenage women. Each woman has a different age of menarche (first menstruation) and cannot be ascertained. The goal to be achieved to answer the problem is to find out the probit regression equation obtained from the age data of menarche in adolescent women so that the value of changes in variables is not free to changes in each unit of independent variables such as linear models. To find out the probability of menarche at a certain age. To find out the age of the coming menarche the most experienced by teenage women. Starting from the problems we have explained before, several conclusions can be drawn including the probit regression equation obtained for the age of menarche in teenage women, namely: $y = -11,8189 + 0,907823 X$. The model is good enough because the parameter value Significant/ meaningful regression is not equal to 0 (zero). The probability of menarche at the age of 9.2 years has the greatest value among all the ages of teenage women, equal to 0,999728. The age of menarche comes at most at the age of 17,5 years.