Author(s):Antonia Scholkmann, Jens Siemon (presenting)

Conference:ECER 2017

Network:02. Vocational Education and Training (VETNET)


Session Information

02 SES 06 A, How Teachers and Work Shape Learning

Paper Session



Chair:Lorenz Lassnigg


Supporting the student in the open naturalistic classroom: What the teacher does. Evidence from a video analysis study in German VET-education

Teachers supporting students during open classes have become one of the key issues of educational practice – in the VET-classroom, and beyond. Giving adequate support of students is considered pedagogically relevant both in the context of individualized student-centred learning (e.g. Weimer, 2002) and with respect to implementation of competence-oriented teaching methods (Bruijn & Leeman, 2011). In German VET, this has been formalised in the concept “Handlungsorientierung” (e.g. Gudjons, 2001;  engl: activity-oriented teaching).

Despite the importance that educational theorists put into individualized teacher support, the body of empirical evidence about which behaviour on the teacher’s side might be actually adequate, effective and thus supportive is scarce. This is especially the case for research in naturalistic classroom settings, i.e. without experimental or quasi-experimental interventions. Because only scarce evidence exists with respect to those questions, and none of this evidence was addressed to the specifics of VET-education, the present study follows a descriptive-explorative approach, aiming to give a clearer picture of teacher supportive behaviours in the naturalistic VET-classroom.

The lack of evidence for teacher supportive behaviour in naturalistic classroom situations might be at least partly due to specific methodological challenges. To validly and reliably asses naturalistic educational situations they need to be researched via behaviour observation methods, which are difficult to realize, time consuming and difficult to operationalize (e.g. Pellegrini, 2004). Multi-modal video-and audio-analysis here provides a couple of advantages, amongst others it provides insights on more than one perspective and allows for repeated analysis of a recorded situation under more than one perspective (Siemon, Boom, & Scholkmann, 2015).

Theoretically, research around the construct ‘scaffolding’ provides a sound framework to analyse teacher supportive behaviour, and thus was used as a baseline and starting point for the present study, Scaffolding has been originally described as verbal interventions that facilitate students’ cognitive learning processes by giving hints and prompts that help the student to reach his/her zone of proximal development (Vygotsky, 1976; Wood, Bruner, & Ross, 1976). Subsequently, scaffolding has been further distinguished into different means that can be applied (Tharp & Gallimore, 1988; van de Pol, Volman, & Beishuizen, 2010), and further defined by the attributes of ‘contingency‘, ‘fading‘ und ‘transfer of responsibility‘ (van de Pol et al., 2010: 275). However, studies in which scaffolding has been reconstructed empirically (e.g. Belland, Glasewsik, & Richardson, 2008; Hong, Wei, Guanghua, & Wanxia, 2011; van de Pol & Elbers, 2013; van de Pol, Volman, & Beishuizen, 2012) normally lack the perspective on the larger context of all interactions, thus indicating that ‘scaffolding’ in  total might be too narrow a concept to describe the wealth of teacher supportive behaviours in the naturalistic classroom.

At this point the present study sets on by asking the question:

  • Which types of supportive behaviours can be inferred from observation of teacher-student interactions in the naturalistic VET-classroom?

Additionally, we were interested in complementing aspects of the support process described in previous literature, namely:

  • Who initiated the supportive behaviour sequence  - teacher or student?
  • How did the teacher make use of diagnostic strategies to assess the student’s need for support and direction?
  • Who profited from the teacher support - a single student, a group of students or a whole class?


In order to assess the above-mentioned research question video study approach was applied. For that, open and individualized teaching was recorded in vocational education classes across a wide variety of subjects and teachers. Recording was done with the Multimodal Video- and Audioanalysis-setup (Siemon u. a., 2015), which uses up to six cameras and individualized audio-recording for teachers and students. The resulting dataset at the moment comprises a total of 19 teachers, 466 students and over 28 hours of recorded material. The dataset is still enlarged at the moment, with an envisioned total of 25-30 teachers in total.
The material was analysed with the video-analysis software Interact (e.g. Mangold, 2006), using two separated codings. In a first step, supportive events were identified in an event-sampling procedure. For that, a coding scheme was applied that focused on deductively inferred complementing aspects of supportive interactions, such as the initiator of the supportive interaction, the focus the supportive interaction had (task-oriented or organisational), the use of diagnostic strategies by the teacher, the negotiation of a shared understanding of the problem at hand and the recipients of the supportive behaviour. In this step all interactions were coded according to the prevalence of these aspects.
In a second step, a second coding was applied to categorize the types of supportive behaviours teachers showed. This coding was developed inductively from parts of the available material, resulting in six distinct categories. Those are:
• Explaining instructional decisions (EID)
• Directing students’ learning activities (DLA)
• Hints, instructions and explanations with respect to learning (HIE)
• Positive feedback (PFB)
• Diagnosing (DGN)
• Cooperative dialogue at eye level (CDE)
In an event-within-event procedure, supportive events were further categorized by allocating each sequence of interaction to one of these categories.
Both codings required low-inference decisions. In order to ensure the quality of the codings, parts of the material for both codings were reliability-checked between at least two (partly three or four) independent raters, using Kappa (κ, Cohen, 1960) and the Holsti-procedure (rH, Holsti, 1969). Resulting coefficients for the first coding lie between κ = .717 (use of diagnostic strategies) and κ = .914 (negotiation of shared understanding); for the second coding they lie between rH = . 75 (explaining instructional decisions) and rH = .87 (diagnosing). With that both codings were considered sufficiently reliable to be used to analyse the complete material.

Expected Outcomes

Preliminary results over the dataset as it presented by January 2017 showed a total of 632 identifiable supportive events and 3030 events-within-events distinguishing types of supportive behaviours. The amount of both real open teaching phases and supportive events varies widely across classes and teachers, with open phases ranging from 0,22% to 98% of total class-time and supportive events from 9 to 49. However, there is no strong correlation between relative open teaching time and number of supportive events, thus substantiating variety of supportive patterns applied by various teachers.
In contrast to theoretical assumptions about supportive interactions inferred from experimental or quasi-experimental procedures, data from our study showed that only approximately two-thirds of supportive interactions were initiated by the teachers, whereas one–third was initiated by a single student or a group of students. Almost half of the supportive events in the naturalistic classroom setting were concerned with organisational aspects of the learning process, i.e. they did not directly centre on the actual content of the task at hand. Moreover, in more than 75% of the cases the teacher didn’t use diagnostic strategies to assess the students need for support, and only in about 2% of the cases did teachers negotiate a shared understanding with the students.
With respect to the types of supportive behaviours, the data also showed a strong tendency towards the use of instructional activities (category HIE) and explanations of instructional decisions (category EID). The category cooperative dialogue at eye level (CDE) is the one most seldom used, thus indicating that teachers in German VET-classroom are still not familiar with supportive behaviour patterns treating students more equally and cooperatively.
In total, the findings illustrate the divergence between patterns of supportive behaviour reconstructed from naturalistic classroom observation opposed to textbook assumptions. Further implications need to be inferred and discussed.


Belland, B. R., Glasewsik, K. D., & Richardson, J. C. (2008). A scaffolding framework to support the construction of evidence-based arguments among middle school students. Education Tech Research Dev, (56), 401–422.
Bruijn, E. de, & Leeman, Y. (2011). Authentic and self-directed learning in vocational education: Challenges to vocational educators. Teaching and Teacher Education, 27(4), 694 – 702.
Cohen, J. (1960). A Coefficient of Agreement for Nominal Scales. Educational and Psychological Measurement, 20(1), 37–46.
Gudjons, H. (2001). Handlungsorientiert lehren und lernen: Schüleraktivierung - Selbsttätigkeit - Projektarbeit. Bad Heilbrunn: Klinkhardt.
Holsti, O. R. (1969). Content analysis for the social sciences and humanities. Addison-Wesley Pub. Co.
Hong, L., Wei, Y., Guanghua, W., & Wanxia, C. (2011). Scaffolding in Teacher-Student Interaction: A Case Study in Two Oral English Classes in China. Chinese Journal of Applied Linguistics, 34(3).
Mangold, P. (2006). Getting better results in less time. When using audio/video recordings in research applications make sense. In H. Voss (Hrsg.), Abstracts of European Society of Family Relations (Bd. 3rd InternationalCongress, S. 73). Darmstadt, Germany.
Pellegrini, A. D. (2004). Observing children in their natural worlds: a methodological primer (2nd ed). Mahwah, N.J: Lawrence Erlbaum Associates.
Siemon, J., Boom, K.-D., & Scholkmann, A. (2015). Multimodale Video-und Audioauswertungen (MuVA). Manuskript präsentiert auf der 3. Frankfurter Tagung zu Videoanalysen in der Unterrichts- und Bildungsforschung, Frankfurt am Main.
Tharp, R. G., & Gallimore, R. (1988). Rousing minds to life: teaching, learning, and schooling in social context. Cambridge [Cambridgeshire] ; New York: Cambridge University Press.
van de Pol, J., & Elbers, E. (2013). Scaffolding student learning: A micro-analysis of teacher–student interaction. Learning, Culture and Social Interaction, 2(1), 32–41.
van de Pol, J., Volman, M., & Beishuizen, J. (2010). Scaffolding in Teacher–Student Interaction: A Decade of Research. Educational Psychology Review, 22(3), 271–296.
van de Pol, J., Volman, M., & Beishuizen, J. (2012). Promoting teacher scaffolding in small-group work: A contingency perspective. Teaching and Teacher Education, 28(2), 193–205.
Vygotsky, Lev. (1976). Interaction between learning and development. In Mind and Society (S. 79–91). Cambridge, MA: Harvard University Press.
Weimer, M. (2002). Learner-centered teaching: five key changes to practice (1st ed). San Francisco: Jossey-Bass.
Wood, D., Bruner, J. S., & Ross, G. (1976). The role of tutoring in problem solving. Journal of Child Psychology and Psychiatry, 17, 89–100.

Author Information

Antonia Scholkmann
Universität Hamburg
Faculty of Education
Jens Siemon (presenting)
Universität Hamburg