Abstract:

Information security is a fundamental component in many everyday situations, such as credit card transactions, the management of large volumes of numerical data, or the operation of digital infrastructures connected to the internet. In this context, detecting potential irregularities in data sets that could affect the proper functioning of these systems has become an essential skill for future professionals working in multiple scientific and technological fields.

Despite the presence of probability and statistics content in university curricula, we have identified that simplified conceptions of randomness still persist among higher education students. As a consequence, these intuitive notions hinder the correct interpretation of real-world data sets whose statistical structures often diverge students’ expectations. This leads to difficulties in distinguishing between authentic data and data that has been manipulated or artificially generated.

This paper describes a learning activity designed to strengthen the critical analysis of data in contexts related to information security through the use of statistical reasoning. The main objective is to promote a deeper understanding of statistical patterns in different real-life situations and to show how such knowledge can be used to assess the reliability of information. The proposal was implemented with master’s level university students. The learning activity is primarily based on gamification and is presented via a competition divided into three phases.

In the initial session, students reflect on the reliability of data across various real-world contexts and are introduced to several simple statistical tools that help determine whether a data set is real or artificial. To achieve this, two example binary sequences are analyzed through brief observation and comparison procedures. This allows students to discuss how certain patterns may appear in real data and how others may suggest irregularities.

In the second phase, students are organized into heterogeneous groups, that compete between them. Each group must imagine a plausible scenario in which data is generated—for example, a company responsible for monitoring the energy consumption of different buildings or analyzing environmental sensor records. Based on this context, students create three data sets, ensuring that at least one represents plausible data generated by a real process, while another contains fabricated or manipulated data.

Finally, in the third phase, groups exchange their data sets with other teams without revealing which ones are authentic. Each group must analyze the received datasets to determine which one appears to be the most reliable. Points are awarded to teams that correctly identify the real data or that manage to create data sets that are particularly difficult to distinguish. The first team to reach the predetermined score wins the competition.

The results of the experience indicate that linking the learning of statistical concepts to problems related to data integrity encourages greater student engagement and helps overcome common misconceptions about randomness. Moreover, this approach contributes to the development of a critical attitude toward the interpretation of information, a skill that is particularly relevant in scientific and technological contexts where data reliability is crucial.

Keywords:

Data integrity, digital security, higher education, randomness, statistical reasoning.