Assessment for Learning: Effective Strategies and Techniques

Assessment for learning (AfL) is a teaching approach that focuses on generating feedback for students to improve their performance. Rather than solely concentrating on results, AfL emphasises assessment processes that actively help students make progress in their learning journey. This methodology encourages learners to become more involved in the learning process, resulting in increased confidence in their understanding of the material and the expected achievement standards.

Effective learning assessments create conditions for the emergence of new knowledge and practice. This encompasses student learning and skill development and guiding instructors in refining their pedagogy and teaching methods. AfL strategies range from simple observation of class discussions and asking questions to reviewing students’ work in progress.

By prioritising assessment for learning, educators can initiate improvements in teaching and learning experiences, fostering an environment where students and instructors work together to facilitate the growth and transformation of knowledge and skills.

Read more

Meet your new virtual editorial assistant

How ChatGPT can help publishers develop educational material. Spoiler alert – it can’t write a textbook for you!

The beginning of 2023 has seen a flood of new artificial intelligence (AI) applications hitting our browsers, with ChatGPT leading the charge. AI is simultaneously embedded in applications we use daily, including search, note-taking and typing assistants. You’ve probably used these features unconsciously, as they are practical and ubiquitous. 

What is ChatGPT and how the heck does it answer all my questions?

ChatGPT is different, and the buzz around it is deserved. ChatGPT is an advanced language model developed by OpenAI designed to process and understand natural language in a way similar to how humans do (basically by chatting with it or asking it things). 

One of the key features of ChatGPT is its ability to engage in conversations and respond to natural language queries. ChatGPT can generate human-like responses to text-based prompts using neural network architecture called a transformer. But it doesn’t just make up these responses. Instead, its answers are based on a vast amount of data, including books, articles, and other sources of information, which it constantly learns and updates to provide accurate and relevant responses. The software behind ChatGPT uses natural language processing (NLP). Through NLP, computers are taught to understand and interpret human language. This involves breaking down language into its parts, such as words, phrases, and sentences, and analysing how they fit together to create meaning. 

To engage with ChatGPT, a user provides a prompt. This prompt is a text you provide to the model to give it context and help it generate a relevant response to your query. You can input your prompt into an interface that allows you to interact with the model, such as a website or an app. The quality of ChatGPT’s responses can vary depending on the complexity of the prompt and the version of the model being used.

Robots like reading and writing, thanks to NLP

Some schools and teachers have experienced an existential crisis as ChatGPT has already been shown to be able to answer test questions and draft assignment essays. Other teachers have embraced it to help draft lesson plans, write assessment questions and generate ideas for classroom activities. ChatGPT also has potential as a virtual tutoring tool to answer questions, analyse student-produced text, and suggest corrections to language and grammar.

One of the more exciting uses of the tool is to generate text, particularly marketing copy. It can quickly and easily write you a blog headline (and, in fact, a whole blog post), a blurb for a book or a snappy Tweet. It can also suggest a content outline for a topic and provide related and extension ideas. You can give ChatGPT examples of text written by an author and ask it to write in that author’s style, which it can learn to do. You can also provide it with chunks of text and prompt it to rewrite it in specific ways or generate additional text outputs. When it comes to long-form content writing, ChatGPT struggles. With a limit on the number of characters it can produce, you need to divide a more extended article into sections and feed it prompts to get it to write paragraphs at a time. This limitation, I assume, will change as the next version of ChatGPT is released. 

ChatGPT, write me a textbook?

So what does it mean for the age-old art of textbook publishing, which essentially entails putting structured bits of writing into a coherent structure? Quite a lot, but it depends on having a good editor or publisher who is handy with prompt writing

Read more