FirstPass is being developed by Bolton College to support students and teachers with the formative assessment of open-ended questions. The service affords teachers the opportunity to move away from traditional closed questioning techniques which have been so dominant in the field of computer mediated assessment for many decades. Whilst the project team at Bolton College is primarily focused on the development of FirstPass to support the formative assessment of open-ended questions, there are other potential use cases for the FirstPass service; especially as it develops as a modular platform. In this short article I would like to take the opportunity to explore some of these use cases.
FirstPass - discover more with open-ended questions
Student feedback surveys
Schools, colleges and universities regularly ask their students for feedback about campus services or about the courses or modules that they are undertaking. The ability to label and classify text means that free-form text responses from students can be labelled and classified on behalf of campus administration or course teams; thereby lessening the difficulty in processing and summarising student feedback. These activities which previously took many hours or days to complete would now be completed in a matter of seconds. This particular use case for natural language classification services like FirstPass would be welcomed by institutions who process module evaluations by the thousands each year. The nature of natural language classification services means that as more and more students respond to open-ended survey questions, the more accurate the natural language classification models become at labelling new responses from students.
One of the interesting scenarios that could play out goes as follows. A service delivery team at Bolton College wishes to conduct a micro survey with all registered students. The open-ended question is presented to students across different delivery channels; namely, text message, email, an Ada nudge or via the institution's proprietary messaging app that is hosted on the student home page. Free form text responses from the student body will be analysed and summarised in real-time by FirstPasss as students respond to the micro survey.
Complaints handling or online student enquiries
Natural language classification services like FirstPass may also support educational institutions with complaint handling or online student enquiries about courses. In these cases, automated workflows or agents will channel the complaint or student enquiry to designated individuals or teams on the campus according to what has been typed or said by the individual. In this use case FirstPass is a single component in a larger chain of activities or workflow that handles these complaints or enquiries.
Live transcription services
There are numerous transcription services that reliably convert speech to text in real-time. Their value lies in producing a textual record of what has been said by teachers and making it available to students following a class or lecture. However, what if the transcript could be used as training data for a suite of natural language classifiers? For example, a teacher delivers an overview of a new subject topic to her class. The transcript is labelled by FirstPass and is reviewed and quality assured by the teacher following the class. The labelled data is then placed into the existing library of classifiers within FirstPass; which becomes better placed to support students in real-time as they respond to open-ended subject topic questions from the teacher. Transcripts could be garnered from radio or television programmes, podcasts, from video channels or from a simple audio interview with a subject expert. This means that everything that is said or written about a subject topic can be used to inform, refine and improve the performance of a natural language classification service like FirstPass.
Distributing open-ended questions across different channels
At the present moment in time, teachers are able to present open-ended questions to their students in situ on the FirstPass website. Over a period of time it will be natural for teachers to present these questions to their students on any platform or modality of their choosing; such as a web-based learning platform, smartspeakers, smartphone apps, voice first apps or VR AR and mixed reality environments. At Bolton College, one can imagine our students having a two-way dialogue with Ada, their campus digital assistant. Students will be able to pose questions to Ada about a subject topic and Ada will ask open-ended questions to assess the students understanding about a subject topic. The choice of questions that are posed by these digital assistants will adapt to reflect the student's knowledge and understanding about the subject at hand.
The following video shows how FirstPass makes use of natural language classification models to support the formative assessment of open-ended questions.
This brief article begins to demonstrate how digital services like FirstPass can start to be used to perform more complex cognitive tasks on behalf of individuals and teams in a modern campus setting. One of the most compelling messages is that everything that has ever been said or written about a subject topic has the potential to be used by natural language classification services like FirstPass to support the formative assessment of open-ended questions.