The Primary English Classroom Corpus

The Primary English Classroom Corpus (PECC) is a collection of 30 transcripts from different primary classrooms in Northern Germany where English is taught as the first foreign language. The transcripts are based on video-recordings of the entire lessons, which were afterwards transferred to a text to make them accessible to students, teachers, and researchers. Each transcript renders the utterances of the teachers and their students in a sequential order as they occurred during the lesson. In addition, short descriptions of nonverbal actions complement the portrayal of the interactive processes, so readers are provided with a more comprehensive view of what goes on during the lesson.

The transcription was done with the help of the program Transana (Woods & Fassnacht 2005), using standard orthography and applying basic transcription conventions that are widespread in the field of Conversation Analysis (cf. Atkinson & Heritage 1984: ix-xvi; Selting et al. 2009). The recordings of the ‘zone-of-interaction’ camera were the main source for the transcripts, while a second camera was used for additional audio checking and to capture any background actions of the students (see Figure 1 below). The results are reader-friendly texts in which the interaction is guided and controlled by the teacher. Since the purpose of using these transcripts is not exclusively discourse or conversation analytical, but also pedagogical (e.g., by studying teaching techniques and classroom activities), fine-grained CA notations in which every single pause or prosodic features of each turn are traced, have not been applied (cf. transcripts in Walsh 2011; but see also Seedhouse 2004; Schwab 2009; for multimodal transcripts of primary classrooms see Cowan 2014). During the transcription process an attempt was made to balance two main criteria: expenditure (time) and precision (representation). A compromise was found in terms of readability and accessibility of the data, although it is clear that the richness of classroom interactions can never be fully represented adequately in a text (cf. Cowan 2014).

The PECC has a word count of 86,346 spoken words distributed over 30 transcripts. It only contains spoken English contextualized in a specific classroom setting. The speakers are teachers who act as language instructors and pupils who are learners of English as a foreign language (EFL). The recordings took place over two years between 2013 and 2015. Four primary schools, six primary teachers, and eight classes participated in this project. The majority of recordings were made in grades 3 and 4 since these are the two main years of English language instruction in German primary schools. In addition, six recordings show English lessons in Year 1 and 2, which were documented in a primary school that has an official permission by the Ministry of Education to start with English in Year 1.

Volume 1 of the PECC contains the first 15 transcripts from Year 3 and 4. It counts 46,481 words, with an average of 3,099 words per transcript.

The following table presents on overview of the number of recordings made in each grade, including their length.

Table 1: Overview of recordings in the PECC (Vol. 1 & 2)

Grade Number of recordings Length (min:sec)
1 3 119:32
2 3 137:10
3 14 657:08
4 10 393:25
1-4 30 1307:15

As Table 1 shows, the corpus reflects a cross-section of primary school foreign language teaching and learning in different grades. The video material consists of over 21 hours of lesson recordings. Most of the lessons are 45 minutes long. Only one school had 60-minute lessons once a week (5 recordings). Two schools are located in an urban area, the other two in more rural areas in Northern Germany. As far as the number of students in each class is concerned, figures differ, wich usually a bit more than 20 students per class.

At the time of recording, all teachers were the current English teachers of the recorded classes. Some of them were also form or homeroom teacher (Klassenlehrer/in), and they had all taught English in Grade 3 and 4 before. Four of the six teachers have a university degree in teaching English as a foreign language and have spent some time abroad in an English-speaking country. Thus, they are formally qualified to teach English in primary school. The other two, one who taught a first-year and one a third-year English class, have no university degree which certifies them to teach English at this level. Instead, they took part in a further education program to receive the necessary qualification for teaching the subject English at this level.

This sample reflects a typical situation in German primary schools. A lack of fully qualified teachers points to one of the downsides of TEFL still evident in primary schools (Wilden & Porsch 2017: 12). Not every primary school can draw upon teachers who are academically qualified to teach a foreign language. In fact, this seems to be an issue in Europe as a whole, as Enever (2011: 27, [HL]) concludes in the ELLiE study:

"[T]he contemporary picture [of teacher qualifications] across Europe may vary substantially, indicating a continuing need for investment in primary FL teacher education if an adequate teacher supply with appropriate expertise is to be available in the foreseeable future."

A similar conclusion has been drawn by TEFL researchers in Germany, who have repeatedly pointed out the continued lack of qualified primary school English teachers as well as the importance of having competent and certified teachers be responsible for EFL instruction at the primary school level (e.g., BIG-Kreis 2015: 71; Elsner 2017: 114f.; Wilden & Porsch 2017: 20). The formal training of English teachers and their employment in every primary school across the country is a necessary precondition for successful early foreign language teaching.

Data Collection

Guiding principles in the data collection were authenticity, comprehensiveness and accessibility. The transcripts portray authentic interactions in primary school EFL classrooms. Lesson topics and learning outcomes were neither imposed by the researcher, nor coordinated with the teacher in advance. The avoidance of external influence was important to get as close to what a 'real' lesson is like. Permitting that everyone involved had given their formal written consent, and teachers had agreed to a recording on a specific day, videos were made without further organizational turmoil. Despite these decisions, any open recording somehow affects the behavior of the participants involved.

Moreover, the level of authenticity was enhanced by having teachers as opposed to university students or pre-service teachers conduct the lessons. Two thirds of the teachers in this project are formally qualified to teach English as a foreign language in the primary school. At the same time, the PECC also includes lessons taught by general teachers who had to take over an English class since there were no other foreign language teachers available at hat school. What often happens if there are no (or not enough) qualified EFL teachers at a school, colleagues who are interested in the subject English and have had prior experience or contact with the target language have to help out. Ministries usually demand that these teachers undergo a training program to receive a Basic qualification before they take over an English class.

Capturing classroom discourse in a comprehensive way meant having to record entire lessons from beginning to end. The recordings were made with two cameras, one in the front and the other in the back of the classroom (cf. Figure 1). The front camera was a mobile device which focused on the interaction between teacher and pupil(s), while the back camera was set on a tripod in the corner to capture an overview of the classroom. One person was present during the lesson to operate both cameras. An additional external microphone attached to the overview camera was used to record a large part of the talk. Figure 1 shows how the two cameras were positioned in the classroom. The seating plan is just one example of how pupils can sit in a primary school classroom:

These two camera positions provided a good basis to record classroom activities with a focus on the ongoing interaction, yet still observe other marginal events in the classroom. Different video studies have shown that using two cameras is a practical set-up for recording classrooms (cf. DESI study, Helmke et al. 2007: 38; IPN-video study, Seidel et al. 2003: 51ff.). This set-up preserves the normality of a lesson, but still widens the angle, making it possible to better observe different classroom activities. In addition to the recording, the teaching material used in each lesson was collected to have access to the exercises the pupils worked on during individual or group work phases (e.g., worksheets, textbook tasks).

Corpus data become accessible if researchers, teachers and students can use it without restriction, and if no special training is needed to work with the transcripts. The transcript corpus is available to anyone studying language teaching, classroom interaction and teaching principles in primary school foreign language teaching. Transcripts are readable without much knowledge of conversation analysis. Standard transcription conventions are followed in order to provide additional information on the talk and actions in the classroom.

Volume 1 of the PECC can be downloaded on this Website (see Transcripts).



Atkinson, John M. & Heritage, John (eds.) (1984). Structures of Social Actions: Studies in Conversation Analysis. Cambridge: Cambridge University Press.

Cowan, Kate (2014). Multimodal transcription of video: examining interaction in Early Years classroom. Classroom Discourse 5 (1), 6-21. doi:

Enever, Janet (ed.) (2011). ELLiE: Early Language Learning in Europe. United Kingdom: British Council. Online: [01.07.17]

Schwab, Götz (2009). Gesprächsanalyse und Fremdsprachenunterricht. Landau: Verlag Empirische Pädagogik.

Seedhouse, Paul (2004). The Interactional Organization of the Language Classroom: A Conversation Analysis Perspective. Malden, MA: Blackwell.

Selting, Margret et al. (2009). Gesprächsanalytisches Transkriptionssystem 2 (GAT2). Gesprächsforschung - Online-Zeitschrift zur verbalen Interaktion 10, 353-402.

Walsh, Steve (2011). Exploring Classroom Discourse: Language in Action. New York: Routledge.

Woods, David & Fassnacht, Chris (2005). Transana 2.05. Madison, WI: Spurgeon Woods LLC. Accessed Mai 6, 2006.