Aarhus University Seal

Feedback from ChatGPT on written group products

Case description

The students were asked to write popular science texts about the historical background for Great Prayer Day for different media platforms and target groups. They were asked to use chatbots to get suggestions for the content and structure of the text.

Motivation

This was during the time of the Great Prayer Day controversy, and the topic was therefore a relevant case study for the course.

ChatGPT was hot news at the time, and I therefore decided to use it in my teaching, so that the students could learn how to use ChatGPT as a writing tool and become aware of its limitations.

          

      

Execution

THE PROCESS

Student preperation

The students worked in groups with texts they had read as preparation. The texts were on the topic of written communication of history-related content. They also read a short academic text about the story behind Great Prayer Day on Danmarkshistorien.dk.

Individual analysis

(15 min)

The students were allocated a specific media (e.g. the Danish children's newspaper Børneavisen, Twitter, Facebook, the Danish Broadcasting Corporation’s online media site, an article for a tabloid, a theatre sketch) and were asked to prepare a short analysis of the relevant media's target group and typical content/style of communication (if the target group was very broad, they were allowed to narrow it down).

Drafts in groups

(15 min)

The students were asked to write a first draft popular science text about the historical background for Great Prayer Day for their specific media.

Feedback from ChatGPT

(15 min)

After this, the students asked ChatGPT for different approaches to the content and structure of their text. For example:

  • What is one way to write an article about Great Prayer Day aimed at children aged 10?
  • What is one way to structure an article about Great Prayer Day?
  • How do you make a good headline about Great Prayer Day?
  • What are the genre conventions for an article for a tabloid media?

Reflection in groups

(30 min)

Then each group was asked to write a short text in which they reflected on the usefulness and limitations of ChatGPT in this type of work.

Group presentations

(5x6 min)

The activity ended with each group doing a brief presentation before the rest of the class.

RESOURCES FOR STUDENTS

SUPPORT FOR STUDENTS

I showed them an article that I had written for the Washington Post and explained how I used ChatCPT to overcome writer’s block during my writing.

Learning objectives

  • Translate history knowledge for popular media
  • Translate history knowledge for various target groups

Experiences

Outcome:

  • The students found that ChatGPT was poor at describing historical events in an academically appropriate way, but that it was good at giving concrete writing advice for specific genres.
  • They also found that ChatGPT needs good prompts to give good answers.

Advice for teachers:

  • Test ChatGPT carefully before the teaching session.
  • It is important to have a reflective, academic dialogue with the students about the use of ChatGPT in academic communication of history.

Activities

    Examples of practice

      Teaching plans

      Under development


      Basic information

      The teacher Helle Strandgaard Jensen
      Faculty and Department/Institute ARTS, School of Culture and Society
      Discipline History
      Study level Master’s degree, second semester
      Course/subject Cultural-historical dissemination and governance
      Number of students (approx.) 17
      Extent Activity
      Teaching Format Classroom 
      Implementation Spring 2023

      Contact

      Please contact the editors at AU Educate if you have any questions about the content of the platform or if you need consultation on your teaching from one of the many skilled professionals at the Centre for Educational Development