The Education Sector Revolution: The Automation of Education

The education sector is about to undergo a revolution in which automation of instruction delivery using Intelligent Tutoring Systems (ITSs) will vastly improve accessibility to learning at a fraction of what that education costs today. This will be achieved while obtaining better student outcomes, and a more individualized learning experience as compared to traditional learning using human teachers in classroom settings. In this paper, we will present a review of the current state of ITSs along with the characteristics of a new more innovative ITS. The potential and consequences of this new learning paradigm will also be explored. Automation and Revolutions During the past 200 years a number of revolutions have taken place in the three main sectors of the world economy: the agricultural, industrial, and service sectors. The main traits of these revolutions have been a push towards automation that has caused a quantum leap in productivity, and a corresponding vast reduction in the need for human labor. The agricultural revolution for example resulted in an increase in cereal equivalent production from 1,000 kg per worker per year to 500,000 kg per worker per year using motorized and input intensive farming. The industrial revolution started with the introduction of the steam engine and is ongoing today with advances such as automated factories and industrial robots. Several sub-sectors of the service sector are also undergoing their own revolutions. For example the financial sector’s use of money counting machines, Automated Teller Machines (ATMs) and internet banking are all aspects of the automation revolution that this sub-sector is undergoing. Most sectors of the economy are moving at various rates towards more automation with the goal of reducing costs and increasing efficiency, quality, and reliability. Revolutions generally happen in 3 stages; the stage before the revolution, the stage during the revolution and the stage after the revolution. The stages can be discerned by plotting the employment data as the percentage of total workers employed in a given sector versus time (Figure 1). In the pre revolution stage, the percent of workers employed in the sector under investigation is stable and the curve is flat. During the revolution, the employment curve slopes downwards, as fewer workers are needed to perform the same functions. Finally post revolution, after the new technology matures in terms of development and implementation, the downwards slope of the percentage of workers employed in the sector decreases as new advances in automation technology bring evolutionary rather than revolutionary improvements in productivity. Looking at the percentage of the United States workforce that is employed in the agricultural sector (Figure 1), we can see that the agricultural revolution entered its post revolution stage in the US around the 1980s. The data for the manufacturing sector shows a downwards trend in the percentage of workers employed in this sector that starts around the 1960s. This is due to a number of causes including higher automation, a shift towards a service economy, and outsourcing of manufacturing jobs out of the US. The employment data for school and higher education teachers, on the other hand, shows a flat curve around 3% of the total workforce as would be expected in the pre revolution stage. P ge 23188.2 Figure 1. United States employment data. There are three main requirements for automation in any field to be viable: the availability of a technological solution that can replace human labor, the need for a large number of similar items, and a lower cost for automation as compared to manual methods at a similar or superior quality. Today, computer based Intelligent Tutoring Systems (ITSs) provide the required technological solution for instruction delivery automation at a fraction of the cost of traditional education. With more than 70 million students enrolled in primary, secondary and higher education in the US in 2010, and many times that number worldwide, the cost of developing and delivering high quality ITSs can be spread out over a very large population of education services customers. Existing Automated Online Learning Systems Online courses where 80% or more of the content is delivered over the internet can be classified in terms of the degree of automation, which is inversely proportional to the degree of involvement of a human teacher in the education process. Instructor delivered online courses are 0 to 29% automated, and are similar to a traditional course except for the fact that most of the instructor’s activities including lecturing, answering questions and giving assignments and exams happen over the internet instead of face to face in a classroom. Partially automated online courses have 30 to 79% of their content automated, while fully automated courses with more than 80% automation can be delivered with little or no involvement from a human teacher. Fully automated online courses currently fall into two main categories: Massive Open Online Courses (MOOCs); such as EdX, coursera, and Udacity, and Intelligent Tutoring Systems (ITSs); such as ALEKS, Carnegie Leaning’s Cognitive Tutor, AutoTutor, and ANDES. Table 1 provides a contrast between the characteristics of MOOCs and ITSs. MOOCs provide a venue for course designers to seamlessly develop and publish their courses. In ITSs, on the other hand, course development and course publishing are decoupled. Some ITSs have tools to facilitate the authoring of course content, but the publishing of these courses is usually handled by separate Learning Management Systems (LMSs). MOOC websites provide course credit for their enrolled students. ITSs, since they are customarily used as learning aids for an online or P ge 23188.3 traditional course, simply report student performance to their LMSs but do not generally provide course credit. Table 1. Comparison between the characteristics of MOOCs and ITSs.