Generative artificial intelligence (AI) is transforming the learning and teaching landscape, and large language models, such as ChatGPT, Gemini, or Copilot, are demonstrating their outstanding ability to hold conversations and generate relevant content. However, given their probabilistic nature, it is currently not possible to trust completely in their academic performance. A detailed comparison of the accuracy of two AI chatbots, ChatGPT-4o and Gemini 1.5 Pro is presented by measuring their ability to correctly answer multiple-choice questions of standard undergraduate physics questionnaires. A total sample of 790 questions was analyzed, covering 13 mechanics' themes, from units and measurements to gravitation. Each chatbot was given the same prompt and was asked to solve each question. Their answers were compared to the correct answers, and an accuracy indicator was derived for each chatbot. In partial agreement with similar studies, it was found that ChatGPT achieved an overall accuracy of only 58 (±7) points (on a 0–100 scale), while Gemini achieved 79 (±6), which represents a 36% improvement. An analysis of variances proved this difference to be significant. The stability of ChatGPT was also tested through different runs, yielding basically the same performance. Its average accuracy may vary up to 11 points for a given theme. On the other hand, ChatGPT running with the Wolfram Alpha plugin did not achieve a systematic improvement, and its average accuracy may vary up to 17 points for a given theme. A PhysicsGPT by pulsr.co.uk from the OpenAI gallery was also tested, and it only achieved an overall accuracy of 57. When using these AI chatbots to solve multiple-choice questions in undergraduate physics topics requiring mathematical procedures, teachers and students should take these results as a warning about their current lack of precision and potential misleadingness.

This research paper describes the factors that contribute to the academic success of university students who receive scholarships due to their outstanding performance in high school, along with a track record of social impact and leadership activities, despite their disadvantaged economic status. The data analysis encompasses information from 2014 to 2023 for a sample of 1,796 students at Tecnologico de Monterrey, with over 60% enrolled in engineering programs. This study examines the factors that positively impact the academic performance of students enrolled in the “Leaders of Tomorrow” scholarship program. The most relevant factors are participation in student group activities, international experiences, community activities, and projects related to technological development or research. Additionally, certain demographic factors, such as not being the first family member to attend university, younger age, coming from urban areas, proximity to the university campus, and female gender, also had a positive impact on academic results. It was found that previous work experiences, sports or cultural activities improve academic performance when properly balanced with the time dedicated to the academic study. Furthermore, the scope of student projects in high school, participation in leadership activities or in academic contests were not relevant to the university academic performance. Understanding the diversity of factors that contribute to academic success will enable more effective support, and the implementation of timely interventions aimed at maximizing the educational opportunities of these students.

Student dropout poses a major challenge for universities. This study seeks to identify its specific causes. An initial database of 143,326 student records was used to which data cleaning and preparation processes were applied to apply machine learning techniques. The original table showed 8.1% dropout, with information on 35 categorical and 14 numerical variables. The analysis of the categorical variables made it possible to identify segments with a greater tendency to drop out, supporting specific interventions. Through principal component analysis and clustering, profiles of students prone to dropout were identified, the results of which support our previous work where we used Random Forest as the most appropriate classifier to predict dropout. The combination of these findings will make it possible to formulate recommendations for more effective interventions, supporting universities in reducing student dropout and improving academic success.

Visuo-haptic simulators (VHS) incorporate the sense of touch into a visual simulator to enrich the learning experience of the user. Through haptic feedback, VHS allow manipulating bodies in a visual simulator and feel the forces applied to them as if they were real. This work presents the design and implementation of a 3D VHS update aimed to support the understanding of friction forces, using the VIS-HAPT methodology proposed by our group. The simulator presents a block on an incline, where the values of the block's mass, incline angle, and the static and kinetic friction coefficients can be changed. The interface also shows dynamically the forces acting on the block in a free-body diagram. The VHS was tested in the August-December 2022 term, with a sample of 309 undergraduate engineering students, distributed in experimental and control groups with 164 and 145 students, respectively. A comparison of pre and post-tests applied to both groups showed that the experimental group obtained larger learning gains than the control group to a statistical significance level ofp = 0.014. This result suggests that the addition of the sense of touch in the VHS promoted a better student understanding of the nature of friction forces. By means of a perception questionnaire, students of the experimental group highly expressed opinions about the haptic activity and that it supported their learning of friction forces. All these findings motivate authors to design additional appropriate visuo-haptic scenarios to foster students learning of real-life physical forces.

In this study, the creation of predictive models for the detection of students at risk of failure is presented. A sample of 618 student profiles , 19% of which failed a given course, were used to detect students at risk of failing. The student profiles were determined from the constructs of Self-Regulation Learning and Affective Strategies (SRLAS) and Multiple Intelligences (MI). The first part of this work, includes an Exploratory Factor Analysis of the data. The predictive phase of the study uses nine Machine Learning classification techniques (KNN, SVM, LDA, QDA, Decision Trees, Random Forest, ADA_Boosting, XGBoosting, and Bayes) to classify students that passed or failed a given course. The Log-Math and the Anxiety dimensions turned out to be relevant variables regarding the success of the student. Decision Trees and Random forests provided the best predicting power. Our results are encouraging in the sense that the methodology followed proves successful at identifying the main factors related to student failure. This may help instructors to timely identify students at risk of failure and their possible causes, to implement appropriate strategies to mitigate this undesirable outcome.

Full Paper; Innovate Practice. Challenge-based learning has been a key component of Tecnologico de Monterrey’s Tec21 disruptive educational model, which focuses on the development of competencies through working on a solution to a real-life, ad-hoc challenge. A brief description of the previous Tec20 and current Tec21 models is presented. The first part of this work presents a comparison of four Tec21 physics courses with 185 students working face-to-face and 205 students working online. The average challenge report score was 5% higher for the online courses than for the face-to-face courses. This difference was significant at α = 0.05 level through t-tests. Closer monitoring by the instructor to achieve course objectives or less rigorous course evaluation policies might have favored the online courses. The second part of this research compares the average grades of the Tec21 model vs. the Tec20 model, based on traditional Problem-Based Learning (PBL). A comparison of Tec20 PBL reports (372 students) and Tec21 Challenge reports (205 students) taking Electricity and Magnetism courses is presented. Average report grades were 3–7% higher for the Tec21 model than for the Tec20 model. The Tec21 model contains promising features, but a more rigorous evaluation approach is suggested.

After the 2017 earthquake, instruction-at-a-distance was massively adopted at Tecnologico de Monterrey Mexico City Campus (CCM). This tragic experience gave CCM staff, faculty, and students truly online expertise, which is now used due to the pandemic. The pros and cons of in-site classes versus online sessions are presented and discussed for engineering courses. A 25-item survey was designed containing categories such as human interaction, command on oneself, and academic performance, among others. A total sample of 396 engineering students was considered. Most students agree that the face-to-face scenario favors their enthusiasm for class attendance, their learning process as well as teacher and peer-to-peer communication. The lack of human contact, the lack of outdoor activities, and the difficulty in making new friends were also factors against education at a distance. Along the same line, distraction, depression, and stress incremented in the online model. However, from the academic point of view, student performance remained similar in the face-to-face model compared to the online model. The teacher's efforts in order to implement active learning activities through novel apps and simulations in the online model were remarkable.

This Innovative Practice Full Paper presents the implementation of a scenario posed to freshman engineering students taking an Introductory Physics short course related to Kepler's Laws for planetary motion. The objective is to guide them to structure a coherent explanation for this scenario. The above is part of the realization of a new educational Tec21 model that Tecnologico de Monterrey has started, where the whole curriculum is focused on challenge-based learning. The emphasis is given to the development of student competencies when real-life situations are faced, in contrast to traditional lecture-based programs. In order to evaluate the student achievement of this competence, a final exam was applied including both standard end-of-the-chapter problems and questions that required higher argumentative thinking levels. The latter was aimed to assess the reasoning followed by students when prompted to analyze different situations from an initial standard context. For a sample of 319 students, it was found that students overall got grades about 20% smaller in the argumentative questions, as compared to the knowledge questions. This can be explained in part due to the fact that most students are only beginning to develop such higher level reasoning and thinking competencies. These results suggest the need to design learning strategies to reinforce the students' reasoning processes.

Statics is a backbone course for several engineering disciplines and also a pre-requisite for dynamics and mechanics of materials. Researchers have identified a lack of understanding of statics as a significant source of difficulties in terms of both conceptual understanding, representation of free body diagrams (FBD) and problem-solving ability. Our approach to improve the learning of the concept of friction focuses on students' understanding of acting forces and specific components of such forces of a system. The presented quasi-experimental study investigates how the use of visuo-haptic simulations can improve students understanding and use of FBD. Specifically, we compared two visuo-haptic simulations; one that explicitly visually depicts FBD of multiple objects interacting with different surfaces, while the other only provides haptic feedback while students engage in the same forms of interaction. Our results suggest that using the visuo-haptic simulator with FBD leads to better learning results. These findings support the hypothesis that appropriately sequenced visuo-haptic simulators with well-designed visual cues can help students to better understand and use FBD.

This paper describes the adaptation and validation of an instrument aimed to determine self-regulation skills, learning strategies and affective strategies of engineering students from the Tecnologico de Monterrey, Mexico City Campus. A statistical validation with Cronbach's alphas and a social validation in which students indicated whether or not they agree with their results on each dimension were carried out. An online system where students answered an appropriate questionnaire to determine their student-profile was developed. The social validation shows a very good agreement among the profiles obtained with the instrument and the perception of the students. Suitable statistical techniques allow classifying samples of students in different clusters according to their profiles, where members of each cluster have similar profiles. Finally, this instrument, along with additional student academic information, allows to predict their academic performance based on statistical methods, and provides support for instructors to focus their teaching strategies and methods, in order to work and pay attention on those students whose self-regulation dimensions were relatively poor.

The assessment process in engineering courses is an arduous task. It requires a lot of effort from teachers. In Mathematics and Physics courses, there are additional issues related to the evaluation of questions with numerical and symbolic answers. In this project, an on-line dynamic assessment system called TecEval was developed. It is a web platform that can be accessed via desktop and mobile browsers. This system enables the management of mathematical equations, and it can randomize the order of the questions and the values of the coefficients in them within certain parameters. Consequently, every student will have to answer a different test every time he/she uses the platform. TecEval allows professors to generate dynamic tests for engineering courses, which are graded automatically. TecEval can create tests with 4 types of questions: i) numerical questions, where the answer is a predetermined number, ii) multiple choice questions (including multiple answers), iii) dynamic questions, where the response is based on a equation, and iv) dynamic-equation questions, where the response is a mathematical equation that may contain dynamic variables. During the evaluation process TecEval performed successfully with students and it obtained positive comments from surveys applied to professors.