Measuring biology teachers’ professional vision: Development and validation of a video-based assessment tool

Abstract A key challenge for teachers is the evaluation of teaching and learning situations in the classroom. Video-based instruments are considered as effective tools for measuring teachers’ skills for evaluating classroom situations. This article reports the development and validation of a video-based assessment tool for measuring biology teachers’ professional vision. Analyses of interviews using think-aloud protocols with in-service biology teachers showed that they perceived the staged videos in the video-based assessment tool as authentic. Furthermore, they identified biology-specific dimensions of instructional quality in the videos, and used aspects of professional vision while reasoning about the identified situations. Thus, results indicate the suitability of the video-based assessment tool for capturing professional vision with the potential to be used in further interventions within teacher education programs.

ABOUT THE AUTHORS Maria Kramer is a doctoral candidate at the Biology Education, LMU Munich, Germany. Her main research focuses on biology teacher education and professional competence of biology teachers. As a member of the research team COSIMA, she investigates the facilitation of diagnostic competences in simulation-based learning environments in higher education.
Christian Förtsch is a post-doctoral researcher at the Biology Education, LMU Munich, Germany. He is interested in research on instructional quality of biology lessons as well as on biology teachers' professional competence, including professional knowledge, professional vision, diagnostic competences, and digital skills.
Birgit J. Neuhaus is Professor in Biology Education at LMU Munich, Germany. Her research interests are mainly connected to biology teachers' professional competence and instructional quality of biology lessons.

PUBLIC INTEREST STATEMENT
A crucial ability of a science teacher is that he or she is able to identify subject-specific problems in the classroom quickly and to make appropriate decisions that result in a good teaching practice, and thus, to improve students' achievement. The ability to notice such problems and interpret them appropriately is called professional vision. Professional vision is considered an important key characteristic of teachers, and thus, should be promoted already in university education. To assess the effectiveness of supportive programs, subject-specific measuring instruments are needed. The use of video-based tools is considered to be particularly promising, since videos can be used to measure professional vision in a situated context of real-world demands. This article reports the development and validation of a video-based learning environment that in future might fill the gap between theory and practice within university science teacher education.

Introduction
Effective teaching can be seen as knowing about instructional quality and its implementation in the classroom. To evaluate effective teaching, systematic observation of teaching components is considered as crucial part of teachers' professional competence. Systematic observation of teaching components includes observation and assessment of teachers' behavior, which results from knowledge teachers draw on in the moment of action (Alonzo et al., 2019;Biggs & Tang, 2007;Sherin, 2002). The concept of professional vision followed this approach of investigation, for example, by using videos. Professional vision can be seen as mediating situation-specific skill between cognitive dispositions and classroom performance. Teachers' professional knowledge, as part of cognitive dispositions, influences their professional vision that in turn affects their performance in the classroom and, therefore, instructional quality of their teaching (Blömeke et al., 2015;Goodwin, 1994;Meschede et al., 2017).
It is assumed that teachers' professional competence develops bi-directionally, because their knowledge and their teaching performance interact with each other while planning, teaching, and reflecting upon their instruction (Blömeke & Kaiser, 2017;Santagata & Yeh, 2016). Because of the bi-directionality of the competence development, biology teachers' application of knowledge during their assessment of classroom situations brings an opportunity to support them in developing their knowledge. This approach is also used within the Refined Consensus Model, a framework, which illustrates the circular shaping process of teachers' subject-specific knowledge during planning, teaching, and reflecting (Alonzo et al., 2019). Since the assessment of classroom situations depends on the subject-specific professional knowledge of teachers, their pedagogical content knowledge as well as their subject-specific professional vision must be taken into account (Steffensky et al., 2015).
Consequently, science teachers should have various opportunities to apply their subject-specific knowledge to teaching, gain experiences, reflect on their teaching, and develop their routines of teaching (Kane et al., 2013;Santagata & Yeh, 2016). In order to provide classroom situations, in which subject-specific knowledge can be applied to assess science teaching, we developed the video-based assessment tool DiKoBi Assess (German acronym for diagnostic competences of biology teachers in biology classrooms). Here, a biology teacher's instruction on a specific biological content, and thus, subject-specific instructional quality in a real-life teaching situation can be assessed by other biology teachers who act as observers. The video-based assessment tool offers opportunities for biology teachers to put their knowledge into practice. This article reports the validation of the content of the videos used in DiKoBi Assess with regard to their authenticity and the recognizability of the integrated challenges (part A) as well as the validation of the assigned tasks with regard to their suitability for capturing aspects of professional vision (part B).

Effective teaching
The behavior of a teacher in the classroom is considered to be decisive for teaching effectiveness. Research on teaching effectiveness has described generic as well as subject-specific features of instructional quality, which influence student achievement (e.g., Brunner et al., 2006;Hattie, 2009;Kyriakides et al., 2013;Seidel & Shavelson, 2007;Wüsten, 2010). In German-speaking countries, generic features have been summarized into three basic dimensions of instructional quality. Similar approaches can also be found in English-speaking countries, such as the CLASS-S framework (Pianta et al., 2012). The three basic dimensions are classroom management, supportive climate, and cognitive activation (Baumert et al., 2010;Praetorius et al., 2018). Classroom management describes how a teacher structures, organizes, and monitors teaching and learning situations, and how he or she deals with student disruptions and establishes rules and routines to increase time on task and students' interest in the subject (Kunter et al., 2007;Praetorius et al., 2018;Seidel & Shavelson, 2007). The basic dimension supportive climate contains features such as a positive relation between teachers and students as well as constructive feedback that benefits the learning atmosphere in the classroom (Praetorius et al., 2018). Cognitive activation refers to the facilitation of a deeper understanding of the content and the consideration of prior knowledge and ideas of students. Since cognitive activation cannot be observed directly, it is often indirectly described by the teachers' instruction. Important subdimensions of cognitively activating instruction are the level of students' cognitive activities, and conceptual instruction. In order to analyze classroom situations for evidence of these subdimensions of cognitive activation, elements such as exploring and activating prior knowledge, using challenging tasks and questions, or supporting knowledge linking have been used Lipowsky et al., 2009;Praetorius et al., 2018).
With regard to interrelations between the three basic dimensions,  underlined the establishment of good classroom management and supportive climate as prerequisites for implementing cognitively activating strategies. Furthermore, cognitive activation is the only basic dimension, in which subject-specificity is considered Schlesinger & Jentsch, 2016). Dorfner, Förtsch, Boone et al. (2019) showed the suitability of the basic dimensions for content-independent descriptions of instructional quality, but they also emphasized the need of features that are closer to the specificity of teaching a particular subject, especially science, of which they discussed several subject-specific features of instructional quality. Table 1 shows specific features of instructional quality that can extend the basic dimensions for science education. These features, exemplified for biology instruction, were found to be empirically effective and can be implemented in the course of a lesson (Dorfner, Förtsch, Spangler et al., 2019).
As part of teachers' professional competence, teachers need to know about the subject-specific instructional quality features and transfer them to content-specific situations of teaching and learning. For effective biology teaching, teachers have to pay attention to biology-specific features of instruction and implement them in their classroom practice (Biggs & Tang, 2007;Steffensky & Neuhaus, 2018). A crucial part of biology teachers' expertise is their ability to identify effective biology teaching as well as to reason about the identified situations in an evidence-based and concept-oriented manner (Kane et al., 2013;Santagata & Yeh, 2016). Promoting professional competence of teachers is therefore an important element in their development of effective teaching practice. Blömeke et al. (2015) modelled professional competence on a continuum between dispositions and performance, mediated by situation-specific skills. The model can be applied to varying situations such as evaluating effective teaching. Cognitive dispositions such as professional knowledge and affective-motivational dispositions such as beliefs influence the application of situationspecific skills within instruction, and thus, foster student outcomes (Baumert & Kunter, 2013;Blömeke & Kaiser, 2017). Teachers need perceptual skills to identify characteristics of classroom situations; they have to analyze and interpret them, and finally make a decision on using appropriate teaching strategies. In teacher education, the ability to notice and reason about relevant classroom situations is discussed in terms of the concept of professional vision (Goodwin, 1994;Van Es & Sherin, 2002). Professional vision requires conceptual knowledge of effective teaching as well as the ability to transfer this knowledge in form of interpretations to the situation observed (Borko, 2004;Van Es & Sherin, 2008). Thus, Stürmer and Seidel (2015) considered professional vision as a situation-specific skill that combines knowledge and practice.

Role of professional knowledge
Existing models have already described the structure of teachers' professional knowledge. Based on the work of Shulman (1986), professional knowledge of teachers can be divided into general pedagogical-psychological knowledge (PK), content knowledge (CK), and pedagogical content knowledge (PCK) (cf. Baumert & Kunter, 2013;Blömeke et al., 2014). PK includes interdisciplinary It describes cognitively demanding instruction that fosters students' deep understanding of the content. Cognitive activities can be assessed with regard to a single task or the whole teaching process and its structure. Indicators are the use of cognitive conflicts, focus questions, the activation of prior knowledge and conceptions (without targeting a specific answer), and the use of challenging and complex tasks to foster high-level thinking (Förtsch et al., 2016Lipowsky et al., 2009;Nawani et al., 2017;Praetorius et al., 2018).
Lesson starts with a cognitively stimulating component: comparison of a knight's armor with human skin.
Conceptual understanding Instruction should include high knowledge linking to foster students' knowledge structure. By teaching interrelated facts and concepts instead of isolated facts, students are cognitively activated and thus the development of their knowledge structure can be fostered. Knowledge linking also plays a central role, when students' prior knowledge and ideas are activated and linked with already known concepts, as well as when knowledge is applied and transferred to multiple contexts (Förtsch et al., 2020;Lipowsky et al., 2009;Steffensky & Neuhaus, 2018;Wadouh et al., 2014).
At the end of the lesson, the teacher refers back to the initial comparison of a knight's armor with human skin. Students have to apply their acquired knowledge on this context.

Creation of situational interest
Situational interest can be understood as a psychological state that involves conditions like focused attention and affective involvement. Subject-specific characteristics like referring to everyday examples or experiments can enhance intrinsic motivation and situational interest Schiefele, 2009;Todt, 1977).
The comparison of a knight's armor with human skin functions as a catch that creates focused attention and affective involvement.

Dealing with students' ideas and errors
An appropriate way of dealing with students' errors means to detect technical errors and use these errors as an opportunity to their learning. Examples of a formative handling are: listing errors as assumptions on the board and discuss them later in class, encourage students to explain their ideas and solution methods, or simply correct the errors right away. The aim is to change students' misconceptions (Herppich et al., 2013;Rach et al., 2013;Spychiger, 2008).
The teacher uses a misconception of a student (e.g., the skin notices when you are touched) as assumption that is noted at the board. After working on the topic, students can discuss the misconception and correct it.
(Continued) knowledge about several teaching methods, learning strategies, and classroom management. CK is described as knowledge about subject-specific content and its conceptual understanding; and PCK is knowledge that teachers need to make this content accessible to their students (Fischer et al., 2012;Shulman, 1986;Voss & Kunter, 2013). Even though different conceptualizations of PCK exist, they all take into account knowledge about students' misconceptions, knowledge about subject-specific structures of biology instruction, and corresponding teaching strategies (Depaepe

Subject-specific dimension Description of subjectspecific features and empirical evidence
Biology-specific example on the topic skin Use of technical language Using minimally necessary topicrelevant terms in biology instruction with additional use of biological core ideas can positively affect students' learning of conceptual knowledge. The overall number of technical terms should be reduced. Explaining and understanding new concepts in everyday terms prior to using scientific language supports students' conceptual and linguistic understanding (Brown & Ryoo, 2008;.
The teacher explains the structure and function of sweat glands without mentioning all technical terms for the different types of glands that exist.

Use of experiments
To solve problems, scientific inquiry methods, such as experiments, can be used. Following the concept of scientific inquiry, process variables can be described. These variables have to be implemented during experimentation to foster scientific reasoning. Hence, formulating scientific research questions and hypotheses, planning experiments, analyzing data, and drawing conclusions are seen as necessary, cognitively activating process variables when using experiments in instruction (Dorfner, Förtsch, Germ et al., 2018;Mayer, 2007;Simon, 1992;Tesch & Duit, 2004).
Steps of scientific inquiry are embedded in the biology lesson. For example, to understand the specific sensibility of different skin areas, first, the students hypothesize the initial phenomenon, second, they plan and conduct an appropriate test, and finally they interpret their results.
Use of models An elaborate use of models (e.g., by using models as tools for scientific reasoning) can foster students' learning. It is assumed that using models for explaining learning content rather than just for illustration fosters students' achievement and scientific reasoning skills. When using models on an elaborate level, students need to think about the content more deeply, which is considered more cognitively activating. A critical reflection of the model as representation of something is also part of the modelling process Oh & Oh, 2011).
Three different models showing the profile of different areas of the skin (fingertip, upper arm, back) are used as a method of scientific inquiry. Additionally, the models are critically reflected and compared with the original (that is our skin).
Note. References to associated concepts as well as research findings are added after each description of subject-specific features.
In terms of biology teachers' professional competence, the subject-specific knowledge dimension PCK is considered to be a determining factor for effective biology teaching (Förtsch et al., 2016). To measure science teachers' PCK, researchers used methods such as classroom observations or semi-structured interviews, they collected and analyzed lesson plans, students' work samples, teachers' written reflections, as well as researcher's field notes. Others developed tools included Content Representation (CoRe) and Pedagogical and Professional Experiences Repertoires (PaP-eRs) for documenting and portraying PCK; and some studies also used a combination of these methods in order to form a genuine representation of the complex construct of PCK (cf. Barendsen & Henze, 2019;Lee & Luft, 2008;Lehane & Bertram, 2016;Loughran et al., 2004;Park & Oliver, 2008). Over the past years, researchers discussed different aspects of PCK that may form this complex construct. For example, Alonzo and Kim (2016) emphasized the differentiation between declarative PCK (knowing that) and more dynamic forms of PCK "that cover teachers' activities during a lesson, for example, if a teacher is able to react appropriately to students' questions and mistakes" (Schmelzing et al., as cited in Alonzo & Kim, 2016, p. 1260. The acquisition of PCK and its transfer to effective teaching performance is described as a circular shaping process, in which knowledge and practical experience influence each other (Blömeke & Kaiser, 2017;Santagata & Yeh, 2016). The Refined Consensus Model (RCM) (Carlson & Daehler, 2019) is an attempt to represent this process in a modeling way. The RCM distinguishes between three different forms of PCK that develop and interact during teachers' professionalization: collective PCK (cPCK), personal PCK (pPCK), and enacted PCK (ePCK). The general canonical knowledge someone needs to know to teach biology represents cPCK. This form is held by multiple educators of biology education, and collected and reported in the literature on biology education (Carlson & Daehler, 2019). In contrast, pPCK and ePCK are more individual forms. In this vein, pPCK describes the personal knowledge resulting from the teaching and learning experiences of an individual biology teacher and his/her reflection on such experiences. pPCK interacts largely with ePCK, which is considered to be the core of teachers' professional competence that only occurs in the moment of acting in biology classrooms, and therefore, has a rather tacit nature (Alonzo et al., 2019;Sorge et al., 2019). Through reflection in and on action, the generated ePCK can be transformed into pPCK afterwards (Alonzo et al., 2019;Schön, 1987). By contributing to pPCK and sharing it with a group of biology teachers, aspects of pPCK can also be transferred into cPCK and thus, can complement the canonical knowledge of the domain.
Considering ePCK as a dynamic form including skills and knowledge to plan (ePCK plan ), teach (ePCK teach ), and reflect (ePCK reflect ) on instruction and student outcomes within the moment of acting, the proximity to Blömeke and Kaiser (2017) conception of situation-specific skills that teachers use for planning, teaching, and reflecting might come to mind. The crucial difference between these two conceptions of teacher knowledge is the explicit consideration of tacit knowledge within the RCM (Alonzo et al., 2019), whereas Blömeke and Kaiser (2017) did not mention tacit knowledge in their definition of situation-specific skills. The emphasis of the tacit nature of ePCK also raises the question of how to measure this form of PCK valid and reliable. Due to the definition of ePCK as tacit knowledge and the contextualized nature thereof, researchers demand other approaches of inferring or approximating ePCK. One way to approximate ePCK might be seen in the conception of situation-specific skills. A situation-specific skill that researchers used for investigating reflection on instruction is professional vision.

Professional vision as central part of teachers' competence
The transfer of professional knowledge to high-quality instruction in real-life performance depends on the execution of situation-specific skills to perceive events relevant to learning, to interpret them, and decide on pursuing strategies. Transferring the RCM to the competence as a continuum model, the execution of situation-specific skills can be considered as activation and generation of ePCK during teachers' planning of instruction, teaching during instruction, or reflecting on instruction (Carlson & Daehler, 2019).
Based on the situation-specific skill professional vision, which is described as "socially organized ways of seeing and understanding events that are answerable to the distinctive interests of a particular social group" (Goodwin, 1994, p. 606), noticing and reasoning are described as two components for professionally observing and interpreting classroom situations (Borko, 2004;Seidel & Stürmer, 2014;Van Es & Sherin, 2002). Whereas these components are characterized by explicit elements, which are measurable, there are also indications that they can contain tacit elements, which have an impact on teachers' noticing and reasoning . Therefore, if we understand observing and interpreting classroom situations as a way of reflecting on instruction, the situation-specific skill professional vision might be considered as approximation to ePCK reflect , but not as exact representation. Seidel and Stürmer (2014) described noticing as paying attention to events that are relevant for students learning, and reasoning as knowledge-based interpretation of classroom situations including the three aspects: description, explanation, and prediction. Description means referring to "relevant aspects of noticed teaching situations and learning components without making further judgements" (Seidel & Stürmer, 2014, p. 745); explanation means to use someone's knowledge to reason about the noticed aspects by linking them to concepts and theories; and prediction means to derive consequences from "observed events in terms of student learning" (Seidel & Stürmer, 2014, p. 746). In other words, explaining and predicting helps teachers to come to pedagogical decisions. These decisions can also occur as suggestions of alternative responses to student answers or alternative teaching strategies that can make teaching more effective (Santagata & Yeh, 2016;Van Es & Sherin, 2002). As the aspects of professional vision are perceived to develop over time, teacher education has to support teachers in gaining them as part of their professional development. An increasingly used method for supporting teachers' professional vision is video viewing (Gaudin & Chaliès, 2015).

Video use in teacher education
Professional competence can be measured with different instruments (e.g., Baumert et al., 2010;Blömeke et al., 2014). When measuring professional vision as part of teachers' professional competence, a video-based approach is promising. Video-based instruments are considered as appropriate to foster the transfer of knowledge about effective teaching and learning principles to practice, and thus, they are considered as a potentially effective tool for professional development (Roth et al., 2017;Seidel & Stürmer, 2014;Shavelson, 2012;Stürmer et al., 2013;Treisch, 2018).
In teacher education, the number of recently designed video-based instruments for assessing professional vision or similar constructs with general or domain-specific perspectives is growing (e.g., Gold & Holodynski, 2017;Kersting et al., 2010;König & Kramer, 2016;Meschede et al., 2017;Michalsky, 2014;Seidel et al., 2010;Wiens et al., 2020). These video-based tools have in common that skills such as professional vision are captured using task prompts. By completing the tasks given in the instrument, teachers' ability to apply knowledge about teaching and learning within videotaped classroom situations can be measured (Seidel & Stürmer, 2014).
Besides video-based assessment tools, real-life classroom videos are used in video clubs to observe and interpret videotaped classroom situations as part of pre-service teachers' higher education. Videos work as item prompts to evoke professional knowledge that can be applied for reasoning about instructional quality, and thus, can improve both professional knowledge and reasoning skills (Behling et al., 2019;Roth et al., 2011;Santagata & Guarino, 2011;Star & Strickland, 2008;Van Es & Sherin, 2008). Referring to the RCM, watching classroom situations and reflecting upon them might raise teachers' ePCK and pPCK, and thus gives opportunity to increase teachers' situation-specific skills as well (Alonzo et al., 2019;Meschede et al., 2017). Additionally, researchers also emphasized the importance of video in raising students' awareness of practices and actions that facilitates transformation (Roth et al., 2011;Siry & Martin, 2014).
Besides the use of videos that recorded real-life classroom situations, scripted (staged) classroom scenarios become more and more relevant, as they can be tailored to specific purposes and observation foci, and thus, can provide very condensed situations. Research underlined that staged videos are considered as an appropriate method to measure teachers' expertise to analyse complex classroom situations. Similar to original real-life performance, staged scenarios can activate teachers' professional knowledge and teaching strategies by the school context shown in the videos (Gaudin & Chaliès, 2015;Hoth et al., 2018;Kersting, 2008). In order to enable teachers to transfer acquired knowledge and skills to real-world situations, the shown classroom practice must be perceived as authentic. "Authentic" means that the situation shown in the video relates to real-life teaching contexts, requires skills of the discipline used similarly by more experienced teachers, and allows immersion in the situation (Barab et al., 2000;Belland et al., 2013;Messick, 1994b;Seidel et al., 2010). Therefore, a sufficient level of presence is also of considerable importance (Schubert et al., 2001). Overall, previous research indicates that staged videos of classroom situations are generally experienced as authentic (Piwowar et al., 2018).
So far, different components for achieving authenticity have been suggested. Components are: selecting key topic areas from national curricula to develop and review video situations that elicit rich discussion, ensuring an explicit focus of each video, and including mistakes to provide a true experience of exemplary teaching situations (Bliss & Reynolds, 2004;Kersting et al., 2010). To strengthen ecological validity, experts of the discipline (e.g., more experienced teachers) should be included in the development and review process of the practice representation and the instructional contexts, since experts are expected to know the cognitive processes and skills required for effective science teaching (Codreanu et al., 2020;Piwowar et al., 2018). Additionally, researchers emphasized that both the teacher and the students should be visible in the videos to enhance authenticity, and that the video situations are suggested to be relatively self-contained and short (Bliss & Reynolds, 2004;Kersting et al., 2010). Furthermore, authenticity can be increased by providing additional material that would be available in real-life situations as well. On the other hand, additional subject-specific information and material, subtitles or hints for task solution may cause distraction. Therefore, researchers emphasized to carefully select type and number of additional material but also the camera perspective to avoid heavy extraneous cognitive load that might decrease authenticity (Alonzo & Kim, 2016;Codreanu et al., 2020;Gold & Holodynski, 2017;Müller, 2016).

Research questions and objectives
Current research regarding evaluation of effective science teaching showed that subject-specific knowledge and professional vision are crucial aspects for instructional quality and teaching effectiveness (e.g., Kyriakides et al., 2013;Seidel & Shavelson, 2007;Steffensky et al., 2015). There is evidence that more practical experience that activates teachers' ePCK positively affects the development of teachers' professional competences, and thus, the development of their cognitive dispositions and cognitive skills as well (Stahnke et al., 2016). However, for better understanding of the interaction between ePCK generated by practical experience, and the other PCK-forms, research needs to define ways of measuring knowledge facets and forms validly. Practical approaches addressing subject-specificity of teachers' professional competence might be useful, but need to be tested thoroughly (Van Es & Sherin, 2017;Gaudin & Chaliès, 2015).
Considering the subject-specificity of teaching effectiveness, researchers need valid instruments to systematically measure the structure and the development of professional vision with respect to biology-specific, situated contexts of real-world demands (Blömeke et al., 2015). As far as we know, subject-and content-specific instruments for investigating biology instruction are still missing, but they are important to provide a necessary subject-specific lens for examining the development of biology teachers' professional vision.
The aim of the study reported in this article is the development of a valid instrument to measure professional vision, and thus, approximate teachers' ePCK. We assume that the development of skills for assessing classroom situations is highly subject specific and necessary in teacher education as it prepares biology teachers to teach their subject effectively. To provide a tool, which represents real-world teacher practice for biology instruction, we developed the video-based assessment tool DiKoBi Assess. In our further studies, we want to use the biology-specific instrument to stimulate teachers' reflection on teaching performances as part of their ePCK reflect (Carlson & Daehler, 2019). In the first step, we tested the validity of the instrument with inservice teachers. Following a construct-centered approach (Messick, 1994b), our validation process was split into two parts, in which Part A referred to the validation of the content of the videos we used in DiKoBi Assess, and Part B to the validation of the tasks we assigned to trigger aspects of teachers' professional vision. The following research questions were guidelines for our study.

Methods
For measuring professional vision of biology teachers, we developed the video-based assessment tool DiKoBi Assess, and conducted interviews using think-aloud protocols with in-service teachers.

Development of the video-based assessment tool DiKoBi Assess
To develop a video-based assessment tool to investigate pre-service teachers' professional vision, we recorded a set of staged video cases depicting biology-specific classroom situations, which were then embedded in an online-survey platform (Questback GmbH, 2018).
We scripted three consecutive lessons of the same class on the topic skin. Each lesson deals with a certain subtopic within the main topic skin and consists of six videos showing six different classroom situations (see Figure 1). In case participants use the assessment tool DiKoBi Assess more often, they do not have to watch the videos of the same lesson twice but can work on the classroom situations of one of the other lessons. Instruction of the scripted lessons followed standards from the Bavarian curriculum on the topic senses and sensory organs (State Institute of School Quality and Educational Research Munich, 2018). This topic can be taught at a similar level in both primary and early stages of secondary schools.
For each lesson, challenging aspects are shown in six classroom situations, each of which focuses on a different biology-specific dimension of instructional quality (in accordance with the dimensions listed in Table 1). The order of the situations follows the course of a lesson. The classroom situations include empirically proven, subject-specific dimensions of instructional quality as described in the theoretical section (see Section 2.1). In each classroom situation, relevant features of the focused dimension are missing. We consider these missing features as challenges that participants have to identify when watching the videos (see Table 2). In order to assess the aspects of professional vision by using DiKoBi Assess, the tasks Describe, Explain, and Alternative Strategy were designed to optimally reveal the respective constructs (Messick, 1994a). Additional information-about sequence scheduling, the lesson plan, and the observation situation-was embedded before working on the video cases to minimize cognitive load during task solution (cf. Müller, 2016).
Before videotaping the staged video cases, scripts were prepared. Initially, we conducted a literature research about the topic skin. Additionally, the basic structure of the lessons in our scripts is based on lessons from teaching concepts of Förtsch et al. (2018) and Heidenfelder (2016) as well as on a lesson planning model for biology instruction (Dorfner, Förtsch, Spangler et al., 2019). We videotaped the staged video cases in school afternoon workshops at a secondary school in Bavaria. The students participated voluntarily in the workshop. Before starting the workshop, we obtained students' and their parents' written consent of participation in the workshop. The 90 minutes lasting workshop took place weekly. For videotaping, we used a special science classroom. This classroom was equipped with cameras for videotaping lessons, which gave us the possibility for videotaping our staged videos (see Behling et al., 2019). Each workshop day began with handing out the scripts to the students and determining the speaking students in our video case. The teacher and all speaking students used microphones for improving the audio-recording quality. After text learning and practicing, video cases were recorded, in which both the teacher and the students were visible (cf. Bliss & Reynolds, 2004). The selected perspective of the video situations was comparable to the perspective of an observing person sitting in the back of the classroom and monitoring activities in the classroom. The average length of the relatively selfcontained video cases was 2.56 min (SD = 2.08; Min =.67; Max = 5.23).

Sample size
We contacted German secondary schools to recruit participating teachers, which completed university education and their teacher trainee program. Once the theoretical and practical training had been completed, teachers were expected to have the necessary knowledge and skills to plan, design, and reflect on lessons according to scientific standards. Therefore, we assumed that, while watching the videos in DiKoBi Assess, these in-service teachers must be able to identify the challenging classroom situations we had scripted in the video cases (cf. Piwowar et al., 2018). After the recruitment process, five teachers agreed to participate in the study. This qualitative approach follows other studies in which constructs such as PCK could be validly recorded with eleven or less participants (e.g., Alonzo & Kim, 2016;Bertram & Loughran, 2012;Sorge et al., 2019).
For validation, we interviewed the teachers using think-aloud protocols (see Table 3). The interviews were videotaped. All teachers taught the subject biology in German secondary schools (with  Table 1). For each classroom situation, participants answer three tasks representing the aspects of professional vision. The concept is broken down for the lesson on protective functions of the skin as example.
Grade 5 to 12). The interviewed teachers had an average age of 40.4 years (SD = 9.2) and an average teaching experience of 9.4 years (SD = 6.9) after their teacher trainee program.

Description of the validation process
As DiKoBi Assess is intended to measure aspects of teachers' professional vision when they are noticing and reasoning about biology instruction, we follow a construct-centered approach (Messick, 1994b). We examined the validity of the content of the videos (RQ1, RQ2) (hereinafter referred to as "Part A"), and that of the assigned tasks (RQ3) (hereinafter referred to as "Part B") by conducting interviews using think-aloud protocols. Using think-aloud protocols in interviews is considered as an appropriate method in the development of instruments and items, as they provide insight into the thoughts and views of teachers (Hill et al., 2007).

Classroom situation Challenging aspects
Level of students' cognitive activities and creation of situational interest (1) Activation of prior knowledge and ideas Videotaped interviews using think-aloud protocols within this study were conducted with the participating teachers in April 2018. Lasting for an average of 120 minutes, each interview began with an introduction and an exercise for training to use the method thinking aloud (see Figure 2). Next, the five participating teachers watched a video showing a classroom situation of another topic of biology instruction and thought aloud how they would assess the classroom situation. After this training, they were informed about the aim of the study, the implementation of the interviews, and the tasks to be executed. Furthermore, the topic of the given classroom situation was linked to the curriculum using a sequence schedule. Then, Parts A and B of the validation using think-aloud protocols started as the main study. Note. Teachers' names were changed to pseudonyms.

Figure 2. Steps of the validation shown as a flow diagram for analysis of Part A. First, based
on empirical studies subjectspecific features of instructional quality were described, and a category system was derived. Subsequently, the staged videos were recorded. Then interviews using thinkaloud protocols with in-service teachers have been conducted. After the transcription of the responses to these protocols, the category system was used for coding and analyzing the transcripts.
To illustrate how teachers thought aloud, and how we analyzed Parts A and B of the interviews, the following sections give an overview about the category systems used for coding, examples of the coding process, and a description of the empirical analysis of the data for each part.

Part A: validation of content of videos
First, in-service teachers watched the staged videos showing the six classroom situations. When they identified challenging aspects about teaching and learning, they stopped the video and talked about their thoughts aloud. Each time when teachers stopped the video, the time was noted on the log sheet by the interviewer.

Assessing authenticity.
To assess the extent to which each staged classroom situation was perceived as authentic, teachers rated three follow-up questions about the authenticity of the watched classroom situations with five-point Likert-scale response options (see Table 4; Schubert et al., 2001;Seidel et al., 2010). When teachers rated the items low, they had to justify their answer.
4.3.1.2. Category system. The category system for coding Part A of the interview using think-aloud protocols was derived from research on biology-specific instructional quality features, and included challenging aspects of biology instruction, which can be found in the recorded classroom situations (see Figure 2). The categories of the category system are identical with the challenges listed in Table 2. In addition, we added the categories further PCK-aspects and not relevant. The category further PCK-aspects was added to each classroom situation for coding in case teachers mentioned other challenges of the specific situation.
The categories of the classroom situations used for coding are therefore as follows: • (1) Activation of prior knowledge and ideas: ○ level of students' cognitive activities,

Item 2
The teaching situation seemed to be just like a real classroom situation.

Item 3
The experience of the teaching situation in the video was similar to the experience of a real classroom situation.
Note. In-service teachers rated the three listed items with Likert-scale response from 1 (disagree) to 5 (agree) for each classroom situation.
• (6) Summary/transfer of knowledge: ○ referring back to the beginning of the lesson ○ conceptual understanding ○ further PCK-aspects • not relevant 4.3.1.3. Coding procedure. We analyzed the transcripts through qualitative content analysis (Mayring, 2014). Thus, we first divided the transcripts into coding units. For validation of the content of the videos, one coding unit of the transcript corresponded to one content-specific analytic proposition (N = 397 coding units). As we wanted to analyze if the teachers could identify challenging classroom situations in the staged videos, we defined each identified challenging aspect as one coding unit. After dividing the transcripts into coding units, we analyzed the material using a structuring analysis technique (Mayring, 2014). All structuring was done using MAXQDA 2018 (VERBI Software, 2017). For analysis of the transcripts, we coded each coding unit of the transcripts with the category system. In case teachers identified challenges other than the ones we staged, we discussed the actual visibility of the perceived challenge in the classroom situation, and added appropriate challenging aspects as examples of the category further PCK-aspects in the coding manual. The steps of the validation process are graphically shown for Part A in Figure 2.
The following excerpt of the transcript of Teacher Paul's interview illustrates how the teacher thought aloud when he worked on the first staged video clip, as well as the coded categories from the category system (see Table 5).

Empirical analysis.
After we had coded all identified challenging aspects by means of the category system, we checked each teacher's thoughts and each classroom situation to find out whether or not our staged challenging classroom situations were identified. Eventually, we summed up the numbers of identification. For later evaluation of the data, it was important that each interviewed teacher referred to at least one challenging aspect of each of the classroom situations. The staged challenges thus served as indicators of the classroom situations, which focused on different biology-specific dimensions of instructional quality, whereby the identification of one indicator was sufficient for a positive coding of the challenging classroom situation.

Part B: validation of assigned tasks
After watching the six staged videos of the first lesson, the teachers worked on the first classroom situation (Activation of prior knowledge and ideas) in DiKoBi Assess. Therein, teachers worked on the tasks in the assessment tool while simultaneously thinking aloud (Aitken & Mardegan, 2000;Ericsson, 2006). As the three tasks Describe, Explain, and Alternative Strategy are the same for every classroom situation, we focused on the first classroom situation for the construct validation as an example.
4.3.2.1. Category system. The category system for validating the assigned tasks was derived from the theory of professional vision, and included a description of the aspects of professional vision and coding examples (Seidel & Stürmer, 2014;Van Es & Sherin, 2002; see Table 6). First, the teachers watched thevideo of the first classroom situation Activation of prior knowledge and ideas and identified challenging aspects by making notes about their identified aspects in open text fields on the computer (Task Describe). We assumed Task Describe to measure description as an aspect of professional vision. Second, participants reasoned about their described challenges by linking them to scientific theories and concepts (Task Explain). Additionally, they estimated their confidence about their reflections on the classroom situation by adjusting a 10-point-slider on a questionnaire scale (between "completely unconfident" to "very confident"). We assumed Task Explain to measure explanation as an aspect of professional vision. Third, the participants proposed an alternative teaching strategy, and justified their selected strategy regarding the question why it would improve the classroom situation (Task Alternative Strategy). Again, for each classroom situation they had to estimate their confidence

Explain
[2] there is nothing that functions as a catch.

Explanation
Answer is a rationale including a theoretical reference, which is made using PCK/PK theories, concepts, and technical terms.
[3] Therefore the students are not taken to the topic of the lesson.

Prediction as a learning consequence
Effects or consequences of the observed events for the learning/learning development of the students are derived.

Alternative Strategy
[4] During a productive stage of the lesson one could compare skin and its functions to a jacket and its functions, for example.

Prediction as an alternative strategy
Suggestions of alternative responses to student answers or alternative teaching strategies that can make teaching more effective.
Note. Statements translated from German by bilingual speaker.
about their described and explained teaching alternatives. For Task Alternative Strategy, we assumed predicting as giving an alternative teaching strategy as an aspect of professional vision. Table 6 shows excerpts from Part B of the transcripts as well as the coded categories, which are the aspects of professional vision. Additionally, the description about how the aspects of professional vision are defined and operationalized in the coding manual is given. For Task Explain, two subcategories were defined to subdivide the coding rubrics more finely, and to consider explanation and prediction as aspects of professional vision as described by Seidel and Stürmer (2014). The subcategories could either contain rationales that referred to a specific concept or contained a conceptual component as a technical term (explanation), or statements that touched on conceivable consequences of the observed events (prediction as a learning consequence). However, intended answers to task Explain should focus on the linking between observation and appropriate theories that were coded with the professional vision aspect explanation.

4.3.2.2.
Coding procedure. Part B of each transcript was cut into coding units on the statement level (N = 433 coding units). By doing so, we were able to check whether the statements reflect the assumed aspect of professional vision. For analysis of the transcripts, we coded each single statement with the category system of professional vision.

Empirical analysis.
After coding all statements, the codings were aggregated per task. Thus, we summed up the codings of each task separately and set the sum of each coded category in relation to the total number of statements for the corresponding task.

Interrater reliability
For reliability and validity assurance, two independent raters analyzed Parts A and B of one transcript with the category systems. The raters were research assistants, who first familiarized themselves with the different category systems and the connected concepts. Afterwards they were trained in coding by taking another transcript for practicing. The training included several coding and discussion loops within a period of about one month. For assessing interrater reliability, 124 coding units were analyzed for Part A, and 265 coding units for Part B. Calculation of Cohens Kappa showed very good agreement: κ = .86 for Part A; κ = .95 for Part B; κ = .93 in total (Wirtz & Caspar, 2002).

Results
We summarize the results here regarding the research questions of Parts A and B. Part A refers to the validation of the content of the videos, and shows the results of teachers' self-reports on authenticity (RQ 1), and their identified challenges of the classroom situations (RQ 2). Part B refers to the validation of the assigned tasks, and to the results of teachers' use of the aspects of professional vision (RQ 3).

Part A: validation of content of videos
To assess authenticity of the simulated classroom situations, the teachers answered follow-up questions after watching each simulated video. Table 7 shows the results of the teachers' selfreports on authenticity of the classroom situations. All of the authenticity ratings show high values. As reported by the teachers, none of the classroom situations seemed to be artificial or unnatural. Table 8 shows how often the challenging classroom situations were identified by the teachers. For a deeper look into the teachers' identifications, we also counted each identified challenging aspect that served as indicator of the challenging classroom situations.
The teachers identified almost every challenging classroom situation, though with varying frequency. A lack of problem identification occurred in the second classroom situation that focused on teachers' dealing with students' ideas and errors. Here, we embedded three different challenging aspects in the situation. The first identifiable challenging aspect referred to the observation that the teacher does not use students' answers and errors' in a formative and supportive way, but ignores them mostly. Furthermore, the teacher in the video does not notice wrong statements given by two students. We listed the wrong statements as another two indicators (student's error 1 and student's error 2). An example of Teacher Ben's statement to identify this challenge for which he was given a positive coding is: "Well, er, of course, this [teachers' behavior in the video] is problematic, because the student gave a completely different answer. And the teacher, well, she has her own idea of what should be written down". Ben referred to the teacher's inappropriate handling of the answer given by the student, because the teacher in the video did not deal with the original answer of the student. Apart from Ben, none of the other teachers mentioned this behavior of the teacher in the video clip.
Even if the other numbers of identification were acceptable, a deeper look at the individual challenges revealed another lack of identification within classroom situation (6). One problem that should be identified related to the teacher's missing reference back to the beginning of the lesson. Teacher Clara said that "the teacher should go back to her mind map [from the beginning]. I think it did not help the students to memorize the results of the lesson, the way she did it". However, none of the other teachers mentioned this challenge.
Additionally, the teachers identified further biology-specific challenges in the classroom situations other than the scripted ones. Further challenges they identified mostly referred to different aspects of cognitive activation, complexity of instruction, dealing with students' answers, and scientific working methods. Apart from that, they mentioned general pedagogical issues such as the use of media, social arrangements, general teaching methods, time management, classroom management, achievement of learning goals, and the general structuring of the lesson. Note. The grey colored area shows the relevant number of identification of each classroom situation representing the subject-specific dimensions. In-service teachers had to refer to at least one challenge of the corresponding dimension. For a deeper look, the indicators of the classroom situations are listed separately as challenges that could be identified. For each challenge, the number of in-service teachers who named the challenge is given. For analysis, the identification on classroom-situation level is sufficient.

Part B: validation of assigned tasks
To investigate whether the tasks in DiKoBi Assess measure the assumed aspects of the construct of professional vision, we assigned to each of the tasks (Describe, Explain, and Alternative Strategy) aspects of professional vision, which occurred while teachers used DiKoBi Assess. For analysis, 219 relevant coding units were identified. When working on Task Describe, teachers mostly used description (57 % of the coding units) and alternative strategies (37 % of the coding units).
Teachers' answers regarding Task Explain could mostly be assigned to explanation (88 % of the coding units), whereas those regarding Task Alternative Strategy could be assigned to prediction as alternative strategy (100 % of the coding units) (see Figure 3).

Discussion
Regarding our objectives, we wanted to develop an instrument for measuring aspects of professional vision of biology teachers to approximate their ePCK reflect . Therefore, we developed DiKoBi Assess that represents real-world biology classroom situations. Through the implementation of instructional tasks that guided in-service teachers' professional vision, combined with the request to think-aloud, we believe that our approach enabled us to infer teachers' ePCK reflect . The participating teachers had to directly articulate their thoughts on particular classroom situations and on the performance of the teacher in the video, and thus, had to draw on a specific subset of knowledge and skills while doing these assessments (cf. Alonzo et al., 2019). Therefore, we claim that using professional vision as a measure to approximate forms of PCK opens promising possibilities.
This study addressed three research questions, which are discussed below.

Part A: validation of content of videos
When validating the content of the videos, we wanted to find out whether the interviewed in-service teachers assessed the simulated classroom situations as authentic (RQ1), and whether the teachers were able to identify the staged challenging classroom situations (RQ2). The results of this study revealed that the staged videos of the six classroom situations were perceived as authentic representations of practice, thus supporting present results from research on videos capturing the authenticity of classroom practice (Codreanu et al., 2020;Piwowar et al., 2018). Besides the inclusion of in-service teachers in the validation process to strengthen ecological validity, we made several efforts to achieve authenticity of the staged classroom situations. For example, we have based our conceptual design on the national curriculum and on previous research results on instructional quality in order to define selfcontained challenging aspects for each classroom situation, thus ensuring an explicit focus (cf. Bliss & Reynolds, 2004;Piwowar et al., 2018). The use of video situations in which features of instructional quality were not perfectly implemented may have additionally contributed to an authentic perception of the classroom situations. Other possible authenticity enhancing components are the chosen camera perspective showing both the teacher and students (thus imitating the perspective of a person sitting in on classes), the relatively short length of the videos not exceeding 5.23 min, and providing contextual information (cf. Bliss & Reynolds, 2004;Gold & Holodynski, 2017;Müller, 2016). However, our present results do not provide any information about the level of presence or the degree of immersion, which can also be regarded as indicators of authenticity, and thus, need to be evaluated in subsequent studies (Codreanu et al., 2020;Schubert et al., 2001).
Regarding the second research question, the results of content validation showed that almost all of the challenging classroom situations were identified sufficiently. Individual challenging aspects served as indicators of the classroom situations so that the participants who were using the assessment tool did not have to perceive two challenges at the same time. Thus, a positive coding could already be reached by referring to one of the challenging aspects. We justify this decision on coding by two reasons. First, we took into account the difficulty of relating conceptual professional knowledge to specific practical assessment situations in classrooms. Our decision followed the idea of the decomposition of practice into aspects of classroom situations that support teachers in their professional development (Grossman et al., 2009). Second, since we wanted to measure aspects of professional vision with our instrument, we focused on teachers' abilities to describe, explain, and suggest alternative strategies, but we did not require them to name all indicators of a classroom situation.
Nevertheless, the teachers' varying abilities to identify the challenging aspects also depends on their knowledge, beliefs, and prior experiences that might change the way they noticed complex situations (Borko, 2004;Van Es & Sherin, 2008). Although we did not systematically evaluate the responses of the participating teachers in this study, we would like to emphasize that we believe that specific experiences led to teachers' more differentiated abilities to notice and reason in the classroom. From our observations, this was particularly the case for teachers who, in addition to their own teaching practice, were also engaged in the education of future teachers (e.g., as seminar instructors in teacher training). At the same time, due to the small number of participants in this study, such observations cannot be generalized.
However, the result of teachers' very poor identification of challenging aspects within the second and the last classroom situations called for instructional changes to improve the instrument as a whole. The second classroom situation referred to teachers' dealing with students' errors. For successful problem identification, the teachers had to focus on student activities to find out students' errors first. This was an essential condition to further judge the situation. In previous studies, researchers emphasized that in-service teachers in particular focus on teachers' behavior in videos rather than on the activities of their students (Colestock & Sherin, 2009;Gaudin & Chaliès, 2015;Santagata, 2009;Van Es & Sherin, 2008). Our results confirmed this observation. Instead of focusing on the students' answers, the participating teachers in our study focused on teachers' behavior. For example, the teachers mentioned a low level of students' cognitive activities, the complexity of the given instructions, and missing methods for visualization. All of these examples of criticism result from teachers' actions in the classroom situation. Hence, we decided to adjust the instructional task to reduce construct-irrelevant variance by adding an observation focus on the handling of students' responses (Messick, 1994a). Another in-service teacher's response to an additional think-aloud protocol showed that the scripted challenge could be identified clearly after task modification. Furthermore, one of the challenges included in the last classroom situation, in which the acting teacher in the video did not refer back to the start of the lesson, was also not identified satisfactorily by the participating biology teachers. We assume that this aspect was not obvious to them to criticize, because there was neither a specific part during the start of the lesson like a focus question which needs to be answered, nor a problem which needs to be solved, nor a context to which the teacher in the video can refer back. Thus, we inserted another video clip before the sixth classroom situation. The alternative teaching strategy illustrates how to introduce the topic skin and its functions by using an analogy of a knight's armor. During the lesson, students have to elaborate essential outcomes regarding the topic. After summarizing the main points, the teacher should refer back to the analogy used at the beginning of the lesson. Thus, the acquired knowledge of the topic skin can be transferred to teaching other content areas so that content linking, and therefore knowledge linking, can be supported (Förtsch et al., 2020;Scott et al., 2011). When the teacher refers back to a focus question, a problem or a context, students have to use their gained knowledge to explain the underlying relations. In doing so, unconnected knowledge elements are activated and connected. This way of science teaching requires the consideration of the conceptual change theory that is a fundamental theoretical perspective of how science learning takes place (Nachreiner et al., 2015;Özdemir & Clark, 2007).
Besides our scripted challenges, the participating biology teachers noticed further biologyspecific problems that they mostly referred to as different aspects of cognitive activation, complexity of instruction, dealing with students' answers, scientific working methods, and general pedagogical issues. When using videos of classroom situations, it is not possible to show a perfect teaching situation as there are too many influencing variables (Seidel & Stürmer, 2014). Among all the occurring events of teaching and learning, teachers have to notice those events with particular importance for students' science learning within our video-based assessment tool. This ability to notice events that are relevant to students' science learning is the professional competence of teachers' subject-specific vision (Seidel & Stürmer, 2014;Steffensky et al., 2015).

Part B: validation of assigned tasks
In response to RQ3, we examined whether the tasks measured the aspects of the conceptualizations of teachers' professional vision. Overall, teachers' answers to the tasks fitted our assumptions for most parts. The intended aspects describing, explaining, and predicting were measured during reasoning.
Regarding Task Describe, we asked teachers for a brief description of challenging aspects to measure their professional vision ability of describing. The results showed that the interviewed teachers used both description and prediction as alternative strategy for answering this task. It was not our expectation that teachers would use alternative strategies to answer Task Describe. However, as in-service teachers possess more knowledge about classroom practice than preservice teachers, the strategy to reason by explaining alternative teaching performance might be part of this expertise (Jong et al., 2012;Palmer et al., 2005). Additionally, knowledge can occur as encapsulated knowledge (Boshuizen & Schmidt, 1992;Schmidt & Rikers, 2007) that can be described as "existing declarative-conceptual knowledge […] enriched with experience from cases" (Heitzmann, 2014, p. 28). Thus, this experience may be used for evaluating effective teaching. However, a close look at the teachers' answers also showed that the description of the challenge was indirectly present in their answers. For example, when a teacher said that it would be better to choose an introduction that is close to everyday life, the reverse conclusion is that the teacher in the simulated video did not use an introduction that was close to everyday life. Therefore, we decided to expand our coding manual and include examples of indirect descriptions as alternative answers of Task Describe. However, it would also be necessary to analyze whether the same answering behavior occurred among pre-service teachers, or whether the observed behavior in our study was based on the practical experience of the in-service teachers.
Teachers' answers to Task Explain were mainly coded as explanation. This result is in line with our expectation of measuring the professional vision aspect explanation. For positive codings, participating teachers had to reason about the observed classroom specifics by connecting them with professional terms and concepts (Seidel & Stürmer, 2014). However, the rationales could theoretically also refer to consequences for students and their learning (prediction as learning consequence), even if this was not explicitly required in the task definition. Since this type of a rationale was rare and not obviously detectable due to the similarity of the given rationales, we decided to combine the subcategories into just one coding category for future coding. Regarding teachers' answers of Task Alternative Strategy, we measured the situation-specific skill prediction. All answers could be categorized as giving alternative teaching strategies that benefit the acquisition of professional decision making (Kersting et al., 2010;Van Es & Sherin, 2002). In light of a construct-centered approach (Messick, 1994b), we can summarize that our assigned tasks are suitable for measuring the aspects of biology-specific professional vision.

Conclusions and outlook
Our validation study showed that DiKoBi Assess can be used as a valid test instrument for measuring professional vision of biology teachers. It extends the basic dimensions for science education that currently used instruments can measure when investigating teachers' professional development in a subject-specific environment. To complete the validation process, data of preservice teachers and in-service teachers are currently being compared. Initial analyses indicated that in-service teachers may achieve better scores in the area of professional knowledge based on their teaching experience.
In addition to the validation of the instrument, the interviews using think-aloud protocols showed that DiKoBi Assess triggered both teachers' reflection on their perceptions of good or bad practice and their reflection on whether their perceptions were based on a theoretical basis or only on their teaching experience. A glimpse of this conclusion can be seen in the answer of teacher Ben, who talked about his reflection: "[you start thinking about your judgment], whether it is just a gut feeling or whether there is more behind it". For reflection on teaching strategies, empirical evidence and theories, teachers need to draw on their ePCK reflect (Alonzo et al., 2019). Thus, reflection fosters teachers' professional development, and thus, increases their professional competence (Blomberg et al., 2013;Gaudin & Chaliès, 2015;Santagata & Yeh, 2014). Learning becomes even more effective when video clips are used with guidance from facilitators, who help learners to identify and make sense of relevant content in video clips. This is a promising approach in actual teacher education programs (Blomberg et al., 2013;Van Es et al., 2014). However, the knowledge teachers likely apply in the moment of classroom instruction is considered to be important for high-quality instruction (Kersting et al., 2012). Thus, using DiKoBi Assess as a constructive alignment (Biggs & Tang, 2007) of biology classroom situations activates teachers' subject-specific knowledge in the context of biology education and supports the establishment of learning opportunities for students.
In future studies, DiKoBi should not only be used for measuring, but also for fostering professional vision of biology teachers (DiKoBi Learn). Hence, the suitability for fostering aspects of professional vision needs to be examined systematically.