The context of this study centres on Higher Education in South Africa, the role that this sector plays in terms of economic development and the implications that face Institutional management in retaining students in the Higher Education system. Central to this study is the notion that student development theory can be used as a basis towards understanding the customers of Higher Education, how to better serve the customers needs and finally, retain students in the system through more effective management practices. The education sector is growing at an increasingly rapid rate as a result of strategic goals of countries and organisations such as the United Nations promoting the notion of education for all people (Altbach, Reisberg & Rumbley, 2009). The aim of the strategic goals adopted by developing countries in particular is to enhance the human capital or skills and knowledge of its people since education is a known contributor towards economic, social and political development. Higher Education in particular is known to contribute towards the human capital and economic development of a country. The Higher Education sector in South Africa for instance contributes approximately 1.5 percent to the country‟s gross domestic product (GDP), significantly higher than other industry sectors (apart from gold and agriculture) in the country (van Heerden, Bohlmann, Giesecke, Makochekanwa, & Roos, 2007). Figure 1.1 provides a context of the relevant importance of the higher education sector towards economic growth.

A reformulação do Exame nacional de ensino médio (Enem), em 2009, e a implementação da plataforma Sisu, em 2010, levaram à transição do sistema de seleção do ensino superior público brasileiro: passou-se de um modelo descentralizado para um majoritariamente centralizado. Apesar da relevância dessa política, a avaliação de seus impactos ainda é principiante. Nesse sentido, utilizando dados de ingressantes entre 2006 e 2014, a pesquisa buscou investigar os impactos dessa política na migração inter e intraestadual e na evasão dos estudantes. Utilizamos, de forma inédita, bases derivadas do cruzamento entre microdados identificados de responsabilidade do Inep, que possibilitaram uma medida mais precisa da migração estudantil. Além disso, investigamos amostras agregadas por instituições e por estados para a inclusão de controles tradicionais de atração e repulsão na literatura de migração. Encontramos que o ingresso do aluno em um programa que oferta vagas pelo SISU eleva a probabilidade de ele ser um migrante interestadual em 2,9 pontos percentuais (p.p.), mas reduz a probabilidade de ele ser um migrante intraestadual em até 3,95 p.p.. Quanto aos resultados para a evasão, verificamos que a adesão ao Sisu eleva a probabilidade de evasão no primeiro ano em 4,5 p.p.. Além disso, um aluno que ingressa em um programa que oferta vagas pelo Sisu tem uma probabilidade maior de mudar de instituição antes de completar o curso. Nesse caso, o auxílio social mostra-se relevante para manter o aluno na mesma instituição, mas não no mesmo curso. Por fim, verificamos que mulheres são mais estáveis no sentido de serem menos propensas a migrar, a evadir e a mudarem de instituições depois do ingresso. / This paper studies the policy that unified the admission system to Brazilian Higher Education. The policy consists of the reformulation of the Enem, in 2009, and the implementation of Sisu platform, in 2010. By using student level admission data from 2006 to 2014, this research seeks to investigate the impact of this policy on both the inter and intra state migration and on the dropout of students. We matched identified databases from the National Institute of Study and Research (Inep) to achieve a more accurate measure of student migration. We found that the admission of a given student in a program that offers places through SISU increases the probability of him being an interstate migrant by 2.9 percentage points (pp), but reduces the likelihood of him being a intrastate migrant by up to 3.95 pp. Concerning the results for dropout, we found that the university\'s commitment to Sisu raises the first year dropout probability by 4.5 pp. In addition, a student who joins a program that offer places via Sisu is more likely to change to another institution before finishing his course. In this case, social aid is effective in keeping the student in the same institution, but not on the same course. Finally, we found that women are more stable in the sense of being less likely to migrate, to dropout or change institutions after their admission.

A reformulação do Exame nacional de ensino médio (Enem), em 2009, e a implementação da plataforma Sisu, em 2010, levaram à transição do sistema de seleção do ensino superior público brasileiro: passou-se de um modelo descentralizado para um majoritariamente centralizado. Apesar da relevância dessa política, a avaliação de seus impactos ainda é principiante. Nesse sentido, utilizando dados de ingressantes entre 2006 e 2014, a pesquisa buscou investigar os impactos dessa política na migração inter e intraestadual e na evasão dos estudantes. Utilizamos, de forma inédita, bases derivadas do cruzamento entre microdados identificados de responsabilidade do Inep, que possibilitaram uma medida mais precisa da migração estudantil. Além disso, investigamos amostras agregadas por instituições e por estados para a inclusão de controles tradicionais de atração e repulsão na literatura de migração. Encontramos que o ingresso do aluno em um programa que oferta vagas pelo SISU eleva a probabilidade de ele ser um migrante interestadual em 2,9 pontos percentuais (p.p.), mas reduz a probabilidade de ele ser um migrante intraestadual em até 3,95 p.p.. Quanto aos resultados para a evasão, verificamos que a adesão ao Sisu eleva a probabilidade de evasão no primeiro ano em 4,5 p.p.. Além disso, um aluno que ingressa em um programa que oferta vagas pelo Sisu tem uma probabilidade maior de mudar de instituição antes de completar o curso. Nesse caso, o auxílio social mostra-se relevante para manter o aluno na mesma instituição, mas não no mesmo curso. Por fim, verificamos que mulheres são mais estáveis no sentido de serem menos propensas a migrar, a evadir e a mudarem de instituições depois do ingresso. / This paper studies the policy that unified the admission system to Brazilian Higher Education. The policy consists of the reformulation of the Enem, in 2009, and the implementation of Sisu platform, in 2010. By using student level admission data from 2006 to 2014, this research seeks to investigate the impact of this policy on both the inter and intra state migration and on the dropout of students. We matched identified databases from the National Institute of Study and Research (Inep) to achieve a more accurate measure of student migration. We found that the admission of a given student in a program that offers places through SISU increases the probability of him being an interstate migrant by 2.9 percentage points (pp), but reduces the likelihood of him being a intrastate migrant by up to 3.95 pp. Concerning the results for dropout, we found that the university\'s commitment to Sisu raises the first year dropout probability by 4.5 pp. In addition, a student who joins a program that offer places via Sisu is more likely to change to another institution before finishing his course. In this case, social aid is effective in keeping the student in the same institution, but not on the same course. Finally, we found that women are more stable in the sense of being less likely to migrate, to dropout or change institutions after their admission.

La presente investigación propone un modelo para la automatización de predicción de la deserción de estudiantes universitarios. Esta investigación surge de una problemática existente en el sector educativo peruano: la deserción estudiantil universitaria; es decir, aquellos estudiantes universitarios que abandonan sus estudios de forma parcial o definitiva. La investigación tiene por finalidad brindar una solución que contribuya a reducir la tasa de deserción universitaria, aplicando tecnologías de análisis predictivo y minería de datos, que detecte anticipadamente a estudiantes con posibilidades de abandonar sus estudios, brindando así a las instituciones educativas mayor visibilidad y oportunidades de acción ante esta problemática. Se diseñó un modelo de análisis predictivo, en base al análisis y definición de 15 variables de predicción, 3 fases y la aplicación de algoritmos de predicción, basados en la disciplina del Educational Data Minig (EDM) y soportada por la plataforma IBM SPSS Modeler. Para validar, se evaluó la aplicación de 4 algoritmos de predicción: árboles de decisión, redes bayesianas, regresión lineal y redes neuronales; en un estudio en una institución universitaria de Lima. Los resultados indican que las redes bayesianas se comportan mejor que otros algoritmos, comparados bajo las métricas de precisión, exactitud, especificidad y tasa de error. Particularmente, la precisión de las redes bayesianas alcanza un 67.10% mientras que para los árboles de decisión (el segundo mejor algoritmo) es de un 61,92% en la muestra de entrenamiento para la iteración con razón de 8:2. Además, las variables “persona deportista” (0,29%), “vivienda propia” (0,20%) y “calificaciones de preparatoria” (0,15%) son las que más contribuyen al modelo de predicción. / This research proposes a model for the automation of prediction of university student dropout. This research arises from an existing problem in the Peruvian educational sector: university student dropout; that is, those university students who partially or permanently abandon their studies. The purpose of the research is to provide a solution that contributes to reducing the university dropout rate, applying predictive analysis technologies and data mining, which detects in advance students with the possibility of dropping out of their studies, thus providing educational institutions with greater visibility and opportunities. of action before this problem. A predictive analysis model was designed, based on the analysis and definition of 15 prediction variables, 3 phases and the application of prediction algorithms, based on the Educational Data Mining (EDM) discipline and supported by the IBM SPSS Modeler platform. To validate, the application of 4 prediction algorithms was evaluated: decision trees, Bayesian networks, linear regression, and neural networks; in a study at a university institution in Lima. The results indicate that Bayesian networks perform better than other algorithms, compared under the metrics of precision, accuracy, specificity, and error rate. Particularly, the precision of Bayesian networks reaches 67.10% while for decision trees (the second-best algorithm) it is 61.92% in the training sample for the iteration with a ratio of 8: 2. In addition, the variables "sports person" (0.29%), "own home" (0.20%) and "high school grades" (0.15%) are the ones that contribute the most to the prediction model. / Tesis

La deserción universitaria es un problema latente en la región de América Latina, que origina limitaciones en el desarrollo de los individuos y en el crecimiento económico de los países. A pesar de la importancia del tema, en Perú aún no se cuenta con literatura económica de metodología sistemática y replicable que permita comprender el fenómeno en el país. Debido a esto, el presente estudio tiene por objetivo predecir la deserción de los estudiantes mediante las variables disponibles que son asociadas a la retención y al abandono estudiantil en la Educación Superior Universitaria. / University dropout is a latent problem in the Latin American region, which causes limitations in the development of individuals and in the economic growth of the countries. Despite the importance of the subject, in Peru there is still no economic literature with a systematic and replicable methodology that allows us to understand the phenomenon in the country. Due to this, the present study aims to predict student dropout through the available variables that are associated with retention and student dropout in Higher University Education. / Trabajo de investigación

Según la firma Penta Analytics, en el 2017, el 27% de los estudiantes peruanos que ingresan a una universidad privada, abandonan la carrera en el primer año de estudios. Este dato se ve respaldado con la información que el director de Grupo Educación Futuro brinda que consiste en que el 80% de los escolares que cursan su último año, no saben qué estudiar ni donde estudiar. Estos son datos que revelan el problema actual en el Perú sobre la indecisión y falta de información por parte del alumnado sobre qué y donde continuar sus estudios superiores. Por este motivo, se planteó la solución Unilink, cuya principal propuesta de valor es ofrecer información condensada, y oportuna sobre las carreras universitarias, institutos de educación superior y sus características. Además, que para los institutos de educación superior supone una oportunidad de que a través de una suscripción mensual puedan acceder a beneficios específicos que Unilink ofrece. Para la formulación e implementación, se realizó la validación del problema, solución planteada, experiencia del usuario, y de cada elemento que comprende su Business Model Canvas. Se utilizaron las herramientas de investigación como entrevistas a profundidad, focus groups, reuniones con expertos, y testeos de mercado. Como resultado, se obtuvo que este modelo de negocio, desde un análisis financiero, es viable, rentable y atractivo para la inversión, ya que, a partir del segundo año de inicio de actividades comerciales se obtienen utilidades positivas. / According to the Penta Analytics firm, in 2017, 27% of Peruvian students who enter a private university, dropped out their carriers in the first year of studies. This data is supported by the information that the director of Grupo Educación Futuro provides, which consists of the fact that 80% of the students in their final year do not know what to study or where to study. These are facts that reveal the current problem in Peru about indecision and lack of information on the part of school students about their future in higher education. According to the mentioned context, the Unilink solution was proposed, whose main value proposition is to offer condensed, suitable, and timely information on university degrees, higher education institutes and their characteristics. It also offers different benefits to the Universities that subscribe to the Unilink plan. For the formulation, we made the validation of the problem, proposed solution, UX, and of each element that comprises its Business Model Canvas was carried out. Research tools such as in-depth interviews, focus groups, meetings with experts, and market tests were used. As a result, it was obtained that this business model, from a financial analysis, is viable, profitable and attractive for investment, since, from the second year of commencement of commercial activities, positive profits are obtained. / Trabajo de investigación