In cases of cyberbullying, missing persons cases, kidnappings and others, the analysis of linguistic samples (messages in social networks, recordings, ransom notes, etc.) with the aim of constructing a sociolinguistic profile of the author or speaker can be a great hint for investigators. The description of the sociolinguistic profile of the possible author of the criminal language samples is one of the tasks included in the branch of applied linguistics known as forensic linguistics. This discipline at the intersection of linguistics and the law places sociolinguistic knowledge at the service of the delivery of justice, especially in the case of sociolinguistic profiles. In this framework, the research presented in this communication is a step forward to discovering linguistic variables that can help forensic linguists to discriminate Catalan-speaking teens according to their age and sex. Its aim is to quantify the progression in the use of specific linguistic variables found in verbal productions made by the same speakers in the same communicative situation in different years. In this way, it is expected to contribute to the theoretical knowledge required for the construction of profiles based on linguistic samples of real cases and the development of forensic linguistics in our country. In addition, since the participants in the study are high school students, the conclusions of this research may also be relevant for the field of Catalan teaching and learning of Catalan in adolescence.