Instructional research deems video studies to be an adequate approach to describe and analyze learning and interaction processes in classrooms. Since TIMS video study in 1995, a recording system with two fixed stationed camera perspectives in classroom (overview, dynamic teacher) is well established in empirical classroom research (Seidel, Dalehefte & Meyer, 2003). With this recording system, global ratings of teaching quality or quality of teachers' interaction with students is possible. However, due to educational reforms (e.g. implementation of an inclusive education system), the composition of the learning group has changed. It implicates an increased use of individualized and cooperative learning. Therefore, it can be assumed that a recording system with two camera perspectives stretched its limits, because there is a risk of not being able to perceive crucial moments of classroom events like discussions in student-centered teaching methods (Dumont, 2019). Hence, a further development of the usual recording system is necessary. One possibility is offered by student-centered recording systems. This approach provides an overview of all activities in the classroom and records both teacher’s as well as individual students’ behavior by using several cameras – each centered on a pair of students combined with an overview perspective. In addition to several advantages of this approach, possible limitations are discussed in the literature. Researchers frequently discuss a potential “negative effect”, which might compromise the validity of the data (Praetorius, McIntyre & Klassen, 2017). This influence of video cameras or observers on natural classroom behavior is known as reactivity or observer effect (Masling & Stern, 1969; Praetorius et al., 2017). Based on existing scientific literature, it can be expected that the presence of cameras has negative effects on teachers or students by distracting or making them nervous (Lotz, 2016) or by forcing the learners and teachers to show more positive behavior than usual compared to a non-videotaped classroom situation (Hiebert et al., 2003). Unfortunately, previous video studies predominantly focus on a teacher’s reactivity effect (Hiebert et al., 2003; Liang, 2015; Praetorius et al., 2017). So far, we know little about reactivity effects of students in video studies. Additionally, previous findings on the reactivity of students are often based on retrospective interviews with teachers about deviations of student behavior during a videotaped lesson from their behavior during a normal lesson. There is only one study known to the authors in which the students rated their one perceived emotions, behavior and cognitive involvement in the videotaped classroom, which were compared to a control group without any video recording (Praetorius et al., 2017). Because of this, the present study focuses on this desideratum and investigates whether and how students show reactivity to a student-centered recording system and how this develops over the course of a teaching week. Reactivity is understood as the attention paid to an unknown stimulus – the recording equipment, the recording team and the project participants. It manifests in the fact that students direct their gaze at the recording system or interact with it. The study addresses the following research questions:

1. To what extent is there an observable reactivity effect of students to a student-centered recording system in individual lessons or over the instructional time of a school week, as measured by the percentage expression of reactive behaviors in students? 2. What percentage of total off-task behavior is due to observable reactivity? 3. How does the percentage of reactivity to the student-centered recording system (relative to the percentage of total off-task behavior during the entire class) change over the course of a week? 4. Are there differences in reactivity due to gender, achievement level and/or special educational needs of the students?

The study is based on observations of mathematics lessons in a third primary school class (N = 24) which were videotaped for one week using a student-centered recording system. The mathematic lessons on Tuesday (45 Minutes) was the first videotaped lesson in this project. The mathematic lesson on Wednesday was the fourth (80 Minutes) and the videotaped lesson on Friday (70 Minutes) was the seventh out of eight recorded lessons. The lessons were videotaped from 17 different camera perspectives. In addition to a dynamic teacher's camera, two GoPro cameras were installed as an overview perspective next to the Smartboard; further, one or two GoPro cameras were installed at each group table. Additionally, the teacher was equipped with a clip-on microphone and the group tables were equipped with table microphones. The topic of these sessions where "weights" which was introduced in the first lesson and continued in the two following lessons. Methodically, the three lessons varied; the students worked on the topic both individually as well as in groups. The sample consists of 24 students and 1 female teacher. Twelve students are male; eight students have special educational needs; eleven students are low achiever duo to the teachers’ appraisal. In order to objectively code the reactivity of the students with regard to the recording system, the behavior was analyzed via the students’ gaze to the recording equipment, the recording team and the project participants (Praetorius et al., 2017) and via students’ talking about and with the recording system. To capture the reactivity of the students a low-inferent German coding scheme – The Munich Attention Inventory (MAI) (Helmke, 1988) – was adapted for the purpose of this study. The coding system distinguishes between on-task and off-task behavior. The following adjustments were made to the original MAI category system: Since the focus of the analyses was on the students' reactivity regarding the recording system – interpreted as distraction from classroom events – the off-task-related behavior basically captured by the MAI was further subdivided into its qualitative gradations (general distraction following MAI and reactivity). The subcategories of the reactivity category were inductively developed from the video data following the gradations of general off-task behavior according to MAI (reactivity passive and reactivity active). In one second intervals, two independent raters coded the behavior of all students in every active learning phase with a good rater agreement (Cohen´s Kappa κ =.73).

Research question 1: The descriptive analysis over all three lessons showed that the recording system causes little reactivity (Tuesday: passive reactivity M=4.19%; SD=2.83%, active reactivity M= 0.34%; SD=0.52%; Wednesday: passive reactivity M=2.02%; SD=1.42%; active reactivity, M=1.28%; SD=1.00%; Friday passive reactivity M=1.80%; SD=1.32%, active reactivity M=1.10%; SD=1.46%). Research question 2: In proportion to the general off task behavior students showed higher passive reactivity than active reactivity. The relative average of passive reactivity decreased in the course of the week. The relative average of active reactivity increased by 4.7% on Wednesday, but decreased again to an average value of M=3.89% on Friday. Research question 3: The results of a ANOVA with repeated measurements showed a main effect for passive reactivity over the three measurement points (Greenhouse-Geisser corrected test; F(1.30,29.97)=21.622, p<.00, η2=.49, n=24). All measurement points varied significantly. For active reactivity there was a main effect, too (F(2,46)=6.86, p<.05, η2=.23, n=24). Active reactivity on Tuesday varied significantly from increased active reactivity on Wednesday but not from active reactivity on Friday. Research question 4. There are no significant group effects. The present study confirms previous findings that video-based instructional research causes reactivity – even if this effect is small, which can be seen as explicitly positive considering the significantly increased presence of cameras in the classroom. The results also indicate that during the first videotaped lesson students show rather unnatural behavior compared to a regular math class. Over the course of the videotaped math lessons, the attentional focus (passive reactivity) decreases significantly, but the interaction with and about the recording system (active reactivity) increases significantly. The results suggest that students showed a more natural behavior throughout the videotaped week than in the first videotaped math lesson. Based on the results, acclimatization period of at least one day is recommended for student-centered video studies.

Dumont, H. (2019). Neuer Schlauch für alten Wein? Eine konzeptuelle Betrachtung von individueller Förderung im Unterricht. [New Hose for Old Wine? A Conceptual Consideration of Individual Learning in Classroom.] Zeitschrift für Erziehungswissenschaft, 22(2), 249–277. https://doi.org/10.1007/s11618-018-0840-0 Helmke, A. (1988). Das Münchener Aufmerksamkeitsinventar (MAI). Manual für die Beobachtung des Aufmerksamkeitsverhaltens von Grundschülern während des Unterrichts. [The Munich Attention Inventory (MAI). Manual for the observation of the attention behavior of elementary school students during class.] Paper 6/1988. München. Hiebert, J., Gallimore, R., Garnier, H., Givvin, K. B., Hollingsworth, H., Jacobs, J. et al. (2003). Teaching Mathematics in Seven Countries. Results From the TIMSS 1999 Video Study (National Center for Education Statistics, Hrsg.). Washington. Liang, J. (2015). Live video classroom observation: an effective approach to reducing reactivity in collecting observational information for teacher professional development. Journal of Education for Teaching, 41(3), 235–253. https://doi.org/10.1080/02607476.2015.1045314 Lotz, M. (2016). Kognitive Aktivierung im Leseunterricht der Grundschule. [Cognitive Activation in Primary Reading Class]. Dissertation. Wiesbaden: Springer Fachmedien Wiesbaden. https://doi.org/10.1007/978-3-658-10436-8 Masling, J. & Stern, G. (1969). Effect of the Observer in the Classroom. Journal of Educational Psychology, 60(5), 351–354. Praetorius, A.‑K., McIntyre, N. A. & Klassen, R. M. (2017). Reactivity effects in video-based classroom research: an investigation using teacher and student questionnaires as well as teacher eye-tracking. Zeitschrift für Erziehungswissenschaft, 20(1), 49–74. https://doi.org/10.1007/s11618-017-0729-3 Seidel, T., Dalehefte, I. M. & Meyer, L. (2003). Aufzeichnen von Physikunterricht. [Recordings of physics lessons]. In T. Seidel, M. Prenzel, R. Duit & M. Lehrke (Hrsg.), Technischer Bericht zur Videostudie "Lehr-Lern-Prozesse im Physikunterricht" (IPN-Materialien, S. 47–98). Kiel: IPN.