In this case study, Alan Chadwick, Head of Classics at City of London Freemen’s School, shares how technology has helped with retrieval practice in his teaching of Latin.
I wanted to improve both engagement and achievement in Latin. Pupils who were studying Latin in Years 7 and 8 were neither sufficiently enthused by the subject nor, in my opinion, were they making enough progress. I also wanted to reduce the burden of assessment as marking was becoming increasingly onerous. I wanted to make the presentation of the subject more appealing, promote higher attainment and improve the efficiency of assessment.
I decided to use Moodle as a platform for my course. Moodle is an open source VLE and is extremely easy to customise through the vast array of plugins available. I had some previous experience of using Moodle and the highly effective online assessment tool in its Quiz module. I felt that Moodle could help deploy all the finer points of AfL, providing formative feedback to students in response to their answers.
First, I had to set up Moodle on a web server. I set up my version of Moodle (which I called CyberCaesar) on a powerful server so that it could cope with the many responses and requests that each quiz would make. Setting up Moodle isn’t difficult but you do need to co-opt the assistance of the ICT administration team to install the program on a school server.
Once Moodle was installed, I started to populate the course with instances of the quiz module to practise certain aspects of the language. The question types found in the Moodle Quiz module provide an enormous amount of variety towards assessment. The ubiquitous multiple-choice type question is among the easiest to construct. The short answer question type provides enormous scope for retrieval practice. A question can be set that demands recall of the precise terminology required for mastery of the subject. These short answer questions can be made more flexible by the use of wildcards.
Quizzes can be adapted in many different ways. The order of questions can be randomised in a test to deter collaboration. Questions can be randomly selected from a bank of possible items to ensure that no student ever takes the same exercise twice. Quizzes can have a time limit. They can be made available only at certain times and access can be limited to certain groups or unlocked by completing other activities. Moodle quizzes can be tailored to the needs of every group of students or teachers, and the variation enables students to grasp the deeper concepts of a subject rather than assessing their surface structure. The quality of feedback has been improved considerably through the addition of the RegExp plugin to the Moodle Quiz module. This helps the quiz engine scan ahead for any common or recurrent elements in an answer. The plugin can accept variations of a correct answer that can take spelling errors into account, detect the absence of punctuation or incorporate text patterns into an answer.
Once a student has completed a quiz, both their responses and grades are available for analysis. Students can review their own work and the teacher can review the responses of everyone. This analysis is an extremely helpful way of identifying misconceptions or misunderstandings of the subject matter. This analysis empowers teachers to quickly spot any areas that require further elucidation or reinforcement. The Moodle quiz is an extremely powerful tool for assessment for learning. It made the entire assessment process far more efficient, as students didn’t have to hand their books into the teacher, wait for them to be marked and then attempt to clarify misunderstandings days after having originally completed the exercise. The immediate responsive feedback meant that students could gain a firmer grasp of the material through constant retrieval practice.
I have composed quizzes to enable students to recall vocabulary items, manipulate features of grammar and to translate simple Latin sentences. Most of these quizzes consist of short answer questions.. Unless I was creating a test to assess retention of material, all of the questions in the quiz were deployed in adaptive mode. This option enables students to make multiple attempts at each question and also to receive immediate feedback to an incorrect response. Initially, I tried to anticipate the kinds of mistakes that students might make. But as the course evolved, I was able to analyse student attempts and incorporate common errors into the feedback.
The quiz exercises have been designed in such a way that material is spaced out over the course of several exercises and then repeated in a later activity to further reinforce learning. A student begins by learning new vocabulary. Knowledge of new lexical items is reinforced by practising its forms in both the target language and in English. Later exercises require students to use these terms in longer sentences. Other features of the subject are then explored before students return to using these terms again, interleaving their learning of the language with that of other aspects of the subject.
The Moodle Quiz module is, in my opinion, the most effective digital tool for retrieval practice in education. Students can constantly check their understanding of material whenever they have the opportunity. We know how important it is for students to test their mastery of material. The Moodle Quiz module does this in a highly effective way. By utilising some of the other options in the administration of the quiz (such as access times), the Moodle Quiz also becomes a means of practising spaced learning. Finally, by the careful crafting and composition of assessment material, the Moodle Quiz module can also be constructed to enhance interleaving.
The use of CyberCaesar has been extremely successful within the school. Recruitment for GCSE has been much higher than before and results at GCSE have improved considerably. In the letter grade GCSE, 97% of students achieved A* or A grades, with 84% achieving A*. In the new system, 85% of students achieved grades 8 or 9 at GCSE and 69% achieved grade 9. In the harder language paper, the average mark of the cohort was 90%. Students continuing their studies of Latin to A Level have also had a much fuller grasp of the material than I have encountered previously in my career. The use of digital technology really improved pupils’ learning and understanding of the subject. Progress was made at an astonishing rate. As they got closer to their examinations, more work had to be completed beyond the digital platform, but the effects of using CyberCaesar were clear from their comprehension of difficult material.
The tool mentioned by Alan is:
Note: The author of this case study runs CyberCaesar, a chargeable online resource.
Whilst Alan references the use of Moodle quizzes, consider what might enable you to achieve similar in your own context if you don’t have access to the same technology.
If you choose to focus on this case study as part of this week’s learning, you can share any initial reflections and questions with the course community in the comments space below.
- How might these approaches be applied in your own context to solve a challenge you’ve identified?
When you are ready click the ‘Mark as complete’ button below and then select ‘Retrieval practice with Socrative’ to see the next case study. Just keep clicking until you arrive at a case study you’ve chosen to focus on this week.
© Chartered College of Teaching