Educational Technology in 2030: AI Driven Chatbots as Teacher’s Assistants

In the year 2030, higher education institutions will continue to grow in both faculty and student populations.  The increased hiring of part-time faculty over full-time in those institutions will force them into a position where they will have to rely on technological intervention to provide their students with engaging and valuable learning experiences.  Chatbots, software applications that allow for simulated conversations with users on the internet, will fill in the gaps where educators are unavailable to communicate and will record interactions with students to develop learning profiles.

It has been the trend in the past ten years to hire part-time faculty in greater numbers than full-time and there is little reason to believe the hiring patterns will change in the next ten.  There are a number of reasons that support the decision to hire part-time faculty.  Wyles (1998) observed the many benefits part-time faculty bring to their educational institutions, including reduced cost, flexibility in course delivery, and a continued connection to industry knowledge and practices (pp. 89, 92).  In addition, there appears to be no distinction between the quality of education provided by part-time faculty when compared to full-time.  Rogers (2015) asserted that students were not negatively impacted by the teaching practices of part-time faculty and that there appeared to be little support for the argument to increase full-time hiring to increase learning outcomes for students (p. 682).  These benefits are unlikely to shift within the next ten years.  On the contrary, observing the hiring practices of one Canadian practical college, as represented in Figure 1, the trend towards hiring part-time faculty over full-time is increasing, showing a greater divide between the two designations.  Since part-time faculty are limited in the number of course hours they are permitted to teach, the volume of educators employed at educational institutions is likely to increase by the year 2030.  At the same time, the number of students for which each educator is responsible, is also increasing.  Eicher et al. (2018) noted during the implementation of an Artificial Intelligence (AI) as a teacher’s assistant, that the growing number of students in their online course had grown to a size where interactions with students by humans was becoming unmanageable (p. 88).  As a result, it is likely that an increasingly part-time faculty will require additional technical assistance in the future.

Figure 1

Fanshawe College Faculty Distribution

Note. This figure represents the categories of faculty employed at Fanshawe College between 2012 and 2019.  Part-time and Partial Load are both considered part-time faculty and are combined as Total Part-time.  Full-time and Sessional are both considered full-time faculty are combined as Total Full-Time. a) Data was acquired from the OPSEU Local 110 October College Staffing Survey available at http://www.opseu110.ca/for-stewards/college-staffing-survey-data/. b) Data for the 2013 academic year was unavailable and is not represented here.

In order to keep the cost of human resources to a minimum, educational institutions will increasingly rely on technological interventions to support educators.  Chatbots powered by AI are a likely candidate to act as a teacher’s assistant for faculty, allowing them to focus their time effectively.  To begin with, there could be concerns that interacting with a machine rather than a human teacher could reduce student engagement.  Crutzen et al. (2011) indicated that students would react positively to communicating with a chatbot (p. 519).  If students enjoy their interactions with a chatbot, they will be more likely to make use of it when they need questions answered when the teacher is not available.  Smutny and Schreiberova (2020) demonstrated that chatbots are instantly available and can provide students with assistance while communicating with them in a conversational manner (p. 2).  In this way, students need not wait for a response from an unavailable educator and can continue to work on their studies on their own time, increasing the accessibility of their education.  Additionally, the use of chatbots could increase course participation by alleviating the anxiety of some students.  Burke (2019) suggested that some students avoid asking teachers questions due to a fear of being ridiculed or judged and that the use of a system that provided them with an opportunity to engage in class anonymously increased their participation.  At the same time, Crutzen et al. (2011) showed that adolescents perceived chatbots as both faster and more anonymous than information lines and search engines (p. 518).  These two facts combined; an environment of perceived anonymity, and an increase in participation under that condition; suggests that students would be more engaged in a course with the usage of chatbots as a teacher’s assistant.

Another positive outcome of the use of chatbots as teachers’ assistants would be the ability to gather data on student interactions to be used for developing personalized learning.  By the year 2030, it is very likely that students will be accessing their virtual learning environment on a number of devices such as their computer, mobile device, and even wearable technology such as a smart watch.  Those devices would not only be used to gather information on students’ interactions with the chatbot, but physiological data would also be recorded to be used to determine their emotional state.  Knox (2020) argued that wearable technology could be used to capture physiological data including facial expressions, neurological responses, and heart rate amongst others to determine an individual’s feelings to identify at risk students, which could then be used to modify their behaviour (pp. 35, 39).  The information gathered could be used to inform the chatbot’s interactions with the student in order to provide subtle suggestions that would gradually shift their behaviour in a direction which would be more conducive to their academic success.  This will only be the beginning as AI becomes more efficient with data usage and can make analyses with less information.  Sucholutsky and Schonlau (2020) have provided evidence that AI can be used to infer accurate conclusions based on an extremely limited amount of data.  As the efficiency of machine learning accelerates, it will be possible for AI powered chatbots to come to conclusions about students’ performance and emotional states without having to be exposed to all of the data currently required.  By 2030, it is feasible to believe that this technology will have advanced sufficiently to have a significant impact in education.

In conclusion, it is likely that educational institutions will continue the trend of hiring part-time faculty over full-time, significantly increasing the size of their teaching staff. At the same time, the student populations will also increase, and educators will be responsible for increasingly large class sizes.  As a result, those institutions will be forced into a position, as a cost saving measure, to employ AI powered chatbots to maintain, or even improve, the student learning experience.

References

Burke, L. (2019). Behind the Back Channel: Can Giving Students Anonymity Help Them Engage In Class? Inside Higher Ed. https://www.insidehighered.com/digital-learning/article/2019/12/06/students-may-benefit-anonymous-back-channel-communications

Crutzen, R., Peters, G. J. Y., Portugal, S. D., Fisser, E. M., & Grolleman, J. J. (2011). An artificially intelligent chat agent that answers adolescents’ questions related to sex, drugs, and alcohol: An exploratory study. Journal of Adolescent Health, 48(5), 514–519. https://doi.org/10.1016/j.jadohealth.2010.09.002

Eicher, B., Polepeddi, L., & Goel, A. (2018). Jill Watson Doesn’t Care if You’re Pregnant: Grounding AI Ethics in Empirical Studies. AIES 2018 – Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society, 88–94. https://doi.org/10.1145/3278721.3278760

Knox, J., Williamson, B., & Bayne, S. (2020). Machine behaviourism: future visions of ‘learnification’ and ‘datafication’ across humans and digital technologies. Learning, Media and Technology, 45(1), 31–45. https://doi.org/10.1080/17439884.2019.1623251

Rogers, G. S. (2015). Part-time faculty and community college student success. Community College Journal of Research and Practice, 39(7), 673–684. https://doi.org/10.1080/10668926.2014.908426

Smutny, P., & Schreiberova, P. (2020). Chatbots for learning: A review of educational chatbots for the Facebook Messenger. Computers and Education, 151(February), 1-11. https://doi.org/10.1016/j.compedu.2020.103862

Sucholutsky, I., & Schonlau, M. (2020). ’Less Than One’-Shot Learning: Learning N Classes From M<N Samples. http://arxiv.org/abs/2009.08449

Wyles, B. A. (1998). Adjunct faculty in the community college: realities and challenges. New Directions for Higher Education, 104, 89-93. https://doi.org/10.1002/he.10409

Leave a Reply

Your email address will not be published. Required fields are marked *