Preimpreso
Permanent URI for this collectionhttps://hdl.handle.net/11285/636050
Artículos científicos o técnicos preliminares, sujeto a revisión de pares, pero tiene la intención de ser publicado en una publicación periódica, de manera independiente o como un capítulo de libro de naturaleza académica, tal como resultados preliminares de investigación publicados en cuadernillo separado.
Browse
Search Results
- Gender gap in perceived achievement of complex thinking in engineering students: A challenge for STEM inclusion(2023-10-25) Vázquez Parra, José Carlos; Suárez Brito, Paloma; López Caudana, Edgar OmarThis paper is part of the thematic axis Bridging the diversity gap in STEM, specifically in the themes Diversity and inclusion and Gender studies in STEM. Its objective is to share the results of a study conducted on engineering students in their last semester of training, regarding their perception of achievement of the competence of complex thinking. The intention is to have an insight into how students perceive their level of competence and cognitive skills, in order to assess whether there are significant differences based on gender. For this study, a representative convenience sample of graduating candidates from a university institution in western Mexico was taken. Methodologically, descriptive analyses were made considering a validated instrument. In conclusion, it is identified that the level of achievement perception of engineering and science students is high (4.29), being critical thinking the one that yielded the best perception and sys- temic thinking the lowest. In terms of gender, it is identified that men are those who yielded the best perception in all cases, being the students of Digital Trans- formation Engineering the best self-evaluated.
- Research plan on the effects of interventions on dropout predictions for higher education institutions(2023) Talamás Carvajal, Juan Andrés; Tecnologico de Monterrey; https://ror.org/03ayjn504; Polytechnic Institute of BragançaOne of the main challenges that Higher Education Institutions face currently is dropout/ student retention. In most cases, identifying this group of stu-dents is no easy task, and doing so on time is even harder. This challenge re-quires both speed and accuracy, which makes it a prime candidate for the use of machine learning models and predictions. We are currently developing a series of models capable of early identification of students at risk of dropping out, with one key difference from classic approaches: we want to not only find out who these students are, but how we can best help them avoid that prediction. By developing methodologies capable of identifying and measur-ing the effects of a series of interventions (academic guidance courses, extra-curricular encouragement, diminished course load, etc.), we intend to devel-op a system capable of providing counterfactuals (what the student needs to change or do to reverse a prediction) based on the causal effects of the previ-ously mentioned interventions. In this manner, we would not only identify groups of students at risk of dropping out, but would be doing so on time, and with a viable and specific strategy for each individual to improve.
- A stacking ensemble machine learning method for early identification of students at risk of dropout(Springer, 2023-03-07) Talamás Carvajal, Juan Andrés; Ceballos Cancino, Héctor Gibrán; Tecnologico de MonterreyEarly dropout of students is one of the bigger problems that universities face currently. Several machine learning techniques have been used for detecting students at risk of dropout. By using sociodemographic data and qualifications of the previous level, the accuracy of these predictive models is good enough for implementing retention programs. In addition, by using grades of the first semesters, the accuracy of these models increases. Nevertheless, the classification errors produced by these models cause undetected students to be discarded from the retention programs, whereas students with no actual risk consume additional resources. In order to provide more accurate models, we propose the use of a stacking ensemble technique to obtain an improved combined dropout model, while using relatively few variables. The model results show values on the expected ranges for an early dropout model, but with considerably fewer features and historical information, and we show that deploying the models would be cost-efficient for the institution if applied towards an intervention program.