If you haven’t yet experimented with ChatGPT, I can’t tell you how mind-blowing a chatbot which uses natural language processing and deep learning algorithms to generate conversational text is to this genX’er. With easier access to artificial intelligence educators need to be asking themselves, “How should we also evolve the ways in which we check for understanding and reimagine student assessment in our schools?”
While I have utilized this technology to produce “soup starter” language for a variety of needs, I’ve never taken its output on blind faith. In my opinion it requires some amount of review to check for accuracy. Sometimes it’s just plain wrong, or the tone does not align with my written voice. With this in mind, a soft word smith dusting will do. In fairness, when I force myself to stare at that blank page and write messaging myself, carefully choosing my words to express meaning, this thought process cannot be replaced.
But I have a good dose of deliberative in my personality, and know when choosing artificial intelligence (AI) is acceptable or not. For students however, it’s easier to assume that choosing chatbot technology as a quick replacement for the time and effort assignments require is some powerful, tempting stuff.
Let’s think about why a student would choose the path of least resistance instead of doing the work. Are they bored? Is the lesson not meaningful to them? Why aren’t they running to school with the same vigor as any other activity of their choosing? If any answer to these questions acknowledge an understanding why a student would resort to trading real effort and learning for the instant gratification of AI, than we must admit that some amount of change needs to take place for summative assessment.
Also, let’s talk about how traditional testing can make us feel – the desks in rows, and dividers to keep eyes from going astray. The silent room, and worry about taking too long to finish. I am personally so averse to traditional testing I would do almost anything to avoid it; the fear of failure so heart wrenching I’d rather find another way to prove my knowledge. That students are still sitting through testing experiences like this is awful, to summarize their growth through multiple choice, fill in the blank, and matching questions is not necessarily an indicator of knowledge. If you’ve never been in a classroom when test results are returned to kids, or simply don’t remember your own experience when you were a student, then I don’t know that you can understand why anyone may want to find a way to use AI to help them do well on an exam.
While I’d be thrilled if my role as a future educator will include reimagining assessment in my classroom, I have a feeling this is mostly dictated by the school district, especially at older grade levels. If I had my druthers however, I’d make these two forms of formative assessment a priority, while using traditional testing as a means of diversification in testing models.
Performance Assessments: such as oral presentations, written essays, journals, and portfolios.
Process Assessments: such as evaluating milestones and reaching educational goals.
I’d also have students talk with me about their knowledge, and create a space where having debate and conversation is fun. I didn’t meet teachers who used this strategy until I attended university, and believe that my love for learning would’ve begun a whole lot sooner if teachers during my primary years believed in the same methodology.