Comparing how songs, chants and stories contribute to Year 3 beginner French learners’ oral language development in UK primary school lessons

It seems that primary MFL teachers frequently use songs to teach languages, with the expectation that young children are able to pick up the foreign language through songs (as explored in this Deanery Digest). In other words, songs are assumed to be not just ‘a bit of fun’ but a reliable part of the language learning toolkit. However, there is a surprising lack of trustworthy research evidence that clearly demonstrates any effect of singing on language learning outcomes. And there is even less reliable evidence that demonstrates a more beneficial effect of singing compared to other things that teachers might include in MFL lessons like stories or chants (see this Deanery Digest for more about the existing evidence base).

To help find out if songs contribute meaningfully to foreign language learning, we conducted a fair test (known as a ‘randomised controlled trial’) with young language learners in British classrooms. We hoped to provide a sound basis for drawing conclusions about whether using songs for teaching French to beginner Y3 learners produces different effects on the pupils’ oral language development in French compared to using chants or stories. We worked with two primary schools in South-West England from June 2023 to January 2024, and 96 beginner learners of French aged 7 and 8.  Their contribution to this research was vitally important to helping teachers make evidence-informed choices about what they do in their classrooms.

To run an experiment where you can draw a conclusion that one method of teaching ‘works’ differently (i.e., better or worse on average) than another, you have to make a fair comparison between the methods. This involves keeping everything in each group as similar as possible except for the method of teaching. This means that if the groups perform differently when tested at the end, you can be more confident that it was the method that made the difference, and not pre-existing differences, such as prior attainment.

The best way to ensure that groups of children are as similar as possible at the start of the experiment is to randomly allocate them to different methods of teaching. We worked with the two schools (one at a time) and randomly allocated pupils who had consented to participate in the study into one of four different methods of teaching French. Each group received 11 French lessons, for a total of 220 minutes of teaching over a three-week period. This was the equivalent of about eight weeks of normal French lessons.

The first group listened to and sang along with traditional French songs, learning two or three songs each lesson. The second group did the same thing with chants – that is they listened to and chanted along with the same songs, minus the ‘melody’ to make a rhythmic chant. The third group listened to and recited aloud a story, made up of all the words from the songs and chants to make the comparison fair, because songs often have a lot of repetition. The story was about a rabbit who went on a journey through the forest with his friend the owl to try and cheer himself up because he was worried about being eaten by the wolf: a gripping tale of friendship and adventure in 11 installments. The fourth group, known as the ‘control group’, received the same amount of time learning French, but followed a mini course based on materials that the schools were already using. They learned some numbers, colours, animals, seasons and key phrases to go with the topics. All groups had visual support on the PowerPoint in the form of the written words with accompanying illustrations to aid comprehension. All groups were taught by the same teacher (the researcher, who is a French teacher).

We wanted to find out whether the different methods produced different results in terms of how well the children repeated French sentences that were read aloud to them, i.e. what the effect on their French oral language development was. On three occasions (before the lessons, straight after the three-week programme, and again six weeks later), we asked the children to listen to and repeat some simple sentences that appeared in the songs/chants/story input (but not in the control group’s materials). We recorded what they said, transcribed it, and then listened to each of the 7668 sentences produced by the children and gave each a score from zero (no response) to five (accurate repetition).

We also measured some key things at the start like children’s English and French vocabulary knowledge, their non-verbal reasoning, rhythmic ability, language background, music experience, and age. In addition to the random allocation, this helped to reassure us that the groups were similar at the start of the experiment.

We did not find any evidence of a difference in performance on the repetition task in the three experimental groups: the songs, chants and story input all helped children learn how to listen to and repeat the French sentences to approximately the same degree on average over time.

The ‘control’ group also got better at perceiving and repeating French. They did not perform as well the other groups on the tests but that is unsurprising, since the test sentences included words that the children in the experimental groups heard and repeated during their lessons, but which the children in the control group did not hear. The purpose of the control group was to see if the overall experiment worked, not whether their teaching materials ‘worked’.

What we learn from this fair test is that children’s ability to listen to and repeat new French input is improved through songs, chants and stories similarly well. Therefore, if teachers enjoy using songs to teach French, they should continue doing so in the knowledge that songs are making a valuable contribution.

If teachers are not so keen on songs, they should not feel guilty if they don’t use them. Chants and stories (where the children read the stories aloud, not just listen to the teacher reading them) help pupils just as much. This is great news because we can now say that songs are not ‘just a bit of fun’ for primary languages lessons. Teachers probably already knew this, but research had not previously provided reliable evidence to support this view.

In conclusion, teachers should use a mixture of rich language input in the form of songs, chants or stories. Even if children do not understand every word explicitly, they enjoy them and make progress in their spoken French. Combined with explicit teaching of vocabulary and grammar at an age-appropriate level, this could be a valuable approach for beginner learners of French.