Joaquin Ponce is 20 years old and was born in Guayaquil, Ecuador. In October last year, Joaquin became the first member of his family to successfully enroll in a higher education institution, after being admitted to the Software Development program at the Technological College of Guayaquil. From a young age, Joaquin has shown a passion for computers. With the help of his parents and a part-time job in the construction sector, Joaquin was able to buy a personal computer at the age of 17. Since then, his dream is to become a software developer.
However, Joaquin's experience after entering higher education has not been easy. Once his classes began, Joaquin realized that he had significant cognitive gaps, especially in mathematics. After his first month of classes, Joaquin began to lose motivation because he could not properly follow the contents of his classes. He could not do the projects demanded by his teachers and was simply not understanding or learning the contents taught in the classrooms. After 5 weeks of classes, Joaquin was seriously considering dropping out of college.
Many students enter college with significant cognitive gaps
The underlying problem is that the quality of the secondary education received by Joaquin was not good enough to ensure that he had the cognitive skills necessary to assimilate university-level academic contents. Joaquin's case is not an atypical case in Latin America. Regional results indicate that half of the students entering higher education do not continue their studies after completing their first year.
High dropout rates of first-year university students in Latin America are mostly explained by the fact that many, especially those from low- and middle-income households, are not ready to assimilate the curricular content taught in college. Moreover, leveling and remediation courses offered by universities in the region are scarce, expensive, and limited in scope.
Adaptive computer-assisted remedial programs: a potential solution
An alternative to leveling out cognitive skills of first-year university students is to use information technology to personalize each student's skills development process. Adaptive Computer Assisted Remediation Programs (ACARPs) have the potential to improve higher education students' academic readiness at scale and in a cost-effective manner, by providing personalized remedial instruction and using easily accessible technologies. ACARPs are essentially specialized software for the development of cognitive skills, such as mathematical computing or reading comprehension. ACARPSs can improve learning outcomes because of their potential to "personalize" education by providing content that is tailored to the learning needs of students, what is commonly known as "teaching at the right level". In fact, one of the main promises of ACARPs is their adaptability since teachers and students can choose the menu of curricular competencies, they need to develop to be successful in higher education. ACARPs take advantage of emerging AI and machine learning techniques to model students' cognitive processes, deliver content accordingly, and ensure, through periodic assessments, that students master curricular content appropriately.
ACARPs can also provide students with different pedagogical strategies for learning methods and, in turn, provide immediate feedback to students and teachers with quick and regular data that can be used to monitor student learning. ACARPs also offer the opportunity to accurately monitor the actual time that students are exposed to the intervention. Many programs of this type can be accessed through PCs, tablets and telephones with Internet access, which makes them even more relevant in situations of restricted access to normal classes, such as the current coronavirus (COVID-19) pandemic.
International evidence indicates that ACARPs are a promising approach to improve learning outcomes and decrease student dropout, especially among students with important curricular knowledge gaps. Several studies that assess the use of ACARPs for mathematics remediation show that college students exposed to them show significantly lower dropout, failure, and dropout rates in courses and majors that require a curricular mastery of mathematics. The available results also indicate that these interventions compare favorably (in terms of impact and cost-effectiveness) with interventions such as remedial classes, study groups, and in-person tutoring. The cost per student to access ACARPs can oscillate between 10 and 20 dollars per student per year, while the cost of tutoring and remedial classes can oscillate between 200 and 500 dollars per student per year.
ACARPs showed very promising results in a pilot conducted in Ecuador
As part of the activities of the World Bank supported Project "Reconversion of Technical and Technological Institutes in Ecuador” (PRETT for its initials in Spanish), the Secretariat of Higher Education, Science, and Technology (SENESCYT), implemented a ACARP pilot in 5 technical and technological institutes throughout the country, benefiting more than 800 first-year students enrolled in technical and technological institutes. The pilot was carried out between January and May 2020, by giving all 800 students access to licenses to use ACARPs. The first step to implement the pilot, done in close coordination with teachers and career directors, was to identify the math-related curricular areas that students should master in order to be successful in their freshman courses. After that, each student was evaluated using the platform in their domain of said curricular content. Based on this diagnosis, the platform generates a personalized learning plan for academic leveling and remediation. Students can access the platform from the institutes' laboratories, from their homes, using tablets or smartphones, or from any other place with internet access. One of the advantages of ACARPs is that its delivery does not require classroom attendance. This feature contributed to the success of the pilot even though technical institutes in Ecuador have been closed since mid-March due to the COVID-19 pandemic.
Results in Figure 1 summarize the main results of the pilot. The blue bar shows the percentage of the curricular content mastered by the students in each technical and technological institute (TTI) when they made the initial knowledge evaluation, which ranged around 20%. After using the platform for 3 months (for 90 minutes a week on average), the knowledge of the students’ mathematical curricular skills reached 61.2% (orange bar), which represents an increase in the of curricular learning of between 8 and 10% per month. This constitutes a great achievement and represents approximately the “learning” equivalent that students would acquire after a full year of schooling.
Joaquin was one of the students who benefited from this pilot program. According to the results of the initial evaluation, Joaquin only mastered 10% of the curriculum content in mathematics required for his major. After using the platform for about 20 minutes per day for 3 months, he was able to master almost half of the required curricular content, which allowed him to start participating more actively in the rest of his classes.
The results of the pilot show that, for students like Joaquin, ACARPs have the potential to improve the academic readiness of higher education students at scale and in a cost-effective manner. The SENESCYT, with funding from the PRETT, is planning to scale up the intervention in order to provide access to ACARPs for a period of 12 months to approximately 16,000 first-year students in 90 technical institutes nationwide, which will amount to an investment of approximately US$250,000. This intervention will be carefully evaluated with PRETT’s funding and with technical support from the World Bank. The results of this pilot seek to inform national and regional public policy, in order to find cost-effective policy options that support students in Latin America to be successful in their passage through higher education.