1 Introduction

Self-regulated learning (SRL) is essential for overcoming learning challenges and a prerequisite for successful lifelong learning (OECD, 2019; Richardson et al., 2012). Self-regulated learners are masters of their learning processes as they purposefully plan, monitor, and regulate their learning to acquire new knowledge and skills (Zimmerman, 2002). SRL is a complex and effortful process that requires acquiring, coordinating, and consolidating metacognitive knowledge and various strategies (Pressley et al., 1987). Research has shown that students struggle to successfully self-regulate their learning, especially those with low metacognitive knowledge about strategies (Askell-Williams et al., 2012; Karlen, 2016). Thus, the assessment and promotion of SRL is an essential objective of education today.

The diagnostic competences of teachers play a central role in improving the quality of teaching and students’ learning in general (Chernikova et al., 2020; Heitzmann et al., 2019; Loibl et al., 2020) and can specifically influence the quality of teachers’ SRL promotion and students’ SRL development (Karlen et al., 2020). Based on their judgments about students’ SRL, teachers might adjust their instructional strategies to adaptively support the individual learning processes of students (Corno, 2008). The diagnostic competences of teachers include their assessment practices (activities such as the use of assessment strategies and methods of data collection), processes (diagnostic thinking such as processing and interpreting information and decision-making), and products (judgment accuracy, e.g. Chernikova et al., 2020; Loibl et al., 2020). The judgment accuracy of teachers is shaped by situational or contextual (e.g. domain, the available assessment tools) as well as personal characteristics (e.g. job experiences) and professional competences (e.g. knowledge, motivation, Südkamp et al., 2012). The diagnostic competences of teachers in SRL encompass a comprehensive set of skills, knowledge, beliefs, and motivation that enable teachers to assess various aspects of students’ SRL effectively. These diagnostic competences enable teachers to identify students’ strengths, weaknesses, and areas for improvement in SRL, collect and analyse data from multiple sources, interpret the findings, and provide timely and targeted feedback.

So far, research on SRL has examined how teachers promote SRL and explained variations in teachers’ SRL promotion based on teachers’ professional competences in SRL, but hardly any evidence is available on teachers’ assessment of SRL (e.g. Callan & Shim, 2019; Dignath & Sprenger, 2020; Spruce & Bol, 2015). Gaining further insight into how accurately teachers judge their students’ SRL and how their knowledge and assessment activities relate to their judgment accuracy is an important step forward in this research field. The current study contributes to these objectives by examining in-service teachers’ assessment activities and the accuracy of their judgments of their students’ SRL skills.

2 Theoretical framework

2.1 Self‑regulated learning

SRL refers to the process by which individuals take an active and intentional approach to learning. In SRL, individuals are considered proactive, goal-directed, strategic, and reflective in their learning endeavors. They plan, monitor, and regulate their cognition, motivation, emotions, and behavior to acquire knowledge and develop skills (Pintrich, 2000; Zimmerman, 2002). Self-regulated learners believe that SRL skills can be developed and that the deliberate use of strategies is essential to achieving high performance. Learners who understand the connection between their performance and SRL are more likely to persist in using SRL strategies than learners who do not recognize the significance of SRL (Hertel & Karlen, 2021). In the literature, various strategies (e.g. metacognitive, cognitive, motivational, emotional, and management strategies) and strategy classifications are presented (e.g., Boekaerts, 1996; Hattie & Donoghue, 2016; Pintrich, 2000). However, to engage effectively in SRL, learners need to acquire, coordinate, and apply different strategies to meet the challenges specific to the tasks they are working on and the situations they encounter (Pintrich, 2000; Pressley et al., 1987; Zimmerman, 2002).

Metacognitive knowledge helps self-regulated learners to use and coordinate strategies effectively (Boekaerts, 1996; Veenman et al., 2006). There are many definitions and conceptualisations of metacognitive knowledge (e.g. Efklides, 2011; Flavell, 1979; Pintrich, 2002; Veenman et al., 2006). According to Flavell (1979), metacognitive knowledge is verbalisable and consciously accessible knowledge. Metacognitive knowledge includes understanding the characteristics of different tasks and strategies, knowing how and when to use strategies to solve specific tasks and determining their effectiveness compared to alternative strategies (Pintrich, 2002). Metacognitive knowledge about strategies and tasks is interconnected, as the selection of strategies should be aligned with the specific demands of the task (Efklides, 2011). Moreover, metacognitive knowledge enables learners to recognize that specific strategies are not limited to one particular task but can be adapted and applied to different learning situations (Pressley et al., 1987). Various studies showed that students’ metacognitive knowledge is an essential prerequisite for the deliberate use of strategies and is positively related to achievement (Karlen, 2016; Richardson et al., 2012; Wolters, 2003). Metacognitive knowledge is often a stronger predictor of student achievement than strategy use (Maag Merki et al., 2013).

Self-regulated learners need metacognitive knowledge about higher-order skills to monitor and regulate their cognitive activities during the learning process. This includes metacognitive knowledge of various metacognitive strategies that enable students to plan, monitor, and adjust their learning effectively (Veenman et al., 2006). Monitoring strategies, such as tracking one’s learning progress and identifying areas for improvement, coupled with self-evaluation or reflection strategies, such as evaluating the learning process, play a vital role in helping students stay on track and enhance their learning process (Zimmerman, 2002).

Self-regulated learners possess metacognitive knowledge concerning management strategies that facilitate the creation of an optimal learning environment. These management strategies are intended to assist and support strategic learners in cultivating and sustaining an effective state of learning, which includes aspects such as concentration management, help-seeking, and effort regulation. Due to their role in supporting learners, these strategies are often referred to as support strategies (Dansereau, 1985). Management strategies helping to regulate the learning environment involve setting up the workspace and adjusting the learning situation to support and enhance the learning process (e.g. well-organised, quiet, availability of necessary resources and materials, and free from distraction) (Pintrich, 2000). These strategies create an environment conducive to effective learning that facilitates better engagement, productivity, and overall learning outcomes.

To successfully maintain concentration and effort in the face of distractions, self-regulated learners possess metacognitive knowledge about self-control strategies (Zimmerman, 2000). Self-control regulates one’s thoughts, feelings, and actions and suppresses impulses to attain a higher goal. Self-control becomes especially evident when long-term valuable goals conflict with more satisfying short-term goals (Duckworth et al., 2014). Self-control strategies are relevant to the actual implementation of learning in the performance phase, as they help to protect learning intentions in the presence of competing action tendencies or obstacles (Zimmerman, 2000). While learning, strategic learners might use self-control strategies (e.g. attention focus, cognitive change, self-instruction, and situation modification) to keep themselves cognitively engaged and motivated to finish the task (Duckworth et al., 2016).

To successfully regulate their motivation, strategic learners possess metacognitive knowledge about motivational regulation strategies. These strategies can help increase or maintain motivation at different stages of the learning process and, therefore, support academic achievement (Schwinger & Stiensmeier-Pelster, 2012). Motivational regulation strategies encompass thoughts, actions, or behaviours learners use to influence their choice, effort, or persistence while learning (Wolters, 2003). Students can use motivational regulation strategies such as mastery self-talk, enhancing personal significance, or self-consequating in situations where they risk losing motivation. These strategies can boost their level of effort and persistence, enabling them to maintain or increase their motivation towards achieving their goal (Boekaerts, 1996; Wolters, 2003).

2.2 Teachers’ professional competences in teaching self-regulated learning

Models of teacher competences offer valuable frameworks for integrating the key factors that influence teaching practice. These models of teacher competences delineate a variety of professional competences that assist teachers in navigating the specific demands of the instructional context and in facilitating efficient decision-making during classroom instruction (Blömeke et al., 2015). Researchers in the field of SRL have argued that for teachers to succeed in supporting students to become self-regulated learners, they must undergo essential dual professional processes. Thus, an important development in research on professional competences of teachers in SRL has been the distinction between teachers’ self-regulation of their own learning and the teaching of SRL (Kramarski & Heaysman, 2021). Hence, models or frameworks about teachers’ professional competence in SRL include their knowledge, beliefs, and motivation as self-regulated learners and as agents of SRL (Karlen et al., 2020).

As self-regulated learners, teachers’ actions and considerations involve proactive processes to monitor, regulate, and control their learning process (Pressley et al., 1987). With their knowledge and experience as self-regulated learners, teachers are assumed to better recognize and cope with the needs, obstacles, and difficulties students face in SRL (Dembo, 2001; Karlen et al., 2023). As self-regulated learners, teachers follow their own SRL trajectories and have different experiences in SRL that shape their self-concept about SRL. Teachers’ self-concept about their own SRL is an important competence aspect as self-regulated learners (Karlen et al., 2020). Self-concept beliefs are relatively stable, multidimensional, and hierarchical cognitive representations of one’s perceived level of abilities (Bong & Skaalvik, 2003). Efklides (2011) describes a self-concept as a mental representation of one’s own SRL abilities that interacts with the learners’ SRL skills. Karlen et al. (2020) showed that the self-concept of teachers as self-regulated learners is not only related to their competences as learners but also positively correlated with their competences as agents (e.g. with self-efficacy to promote SRL). However, empirical results on the relationship between teachers’ self-concept about SRL, assessment activities, and judgment accuracy are missing.

The actions and considerations of teachers as agents of SRL are directed to promote their students’ SRL (Kramarski & Heaysman, 2021). Teachers as agents of SRL have professional knowledge about SRL, beliefs consistent with SRL (e.g. growth mindsets about SRL), and motivational orientation (e.g. self-efficacy) regarding SRL that form a foundation for teachers to master specific professional situations and to guide them toward a powerful SRL instruction (Karlen et al., 2020; Kramarski & Heaysman, 2021).

When examining how teachers conceptualize SRL to assess students’ SRL, it is relevant to consider their professional knowledge about SRL. The professional knowledge of teachers can be subdivided into different domains as proposed by Shulman (1987). Although SRL is ideally taught in conjunction with subject content, the teaching of SRL is an educational objective independent of subject content and is therefore content in its own right. Thus, SRL-specific knowledge can be divided into knowledge about the construct of SRL (SRL is the content or subject here), referred to as content knowledge (CK-SRL), and knowledge about the promotion of SRL (in the sense of teaching it), referred to as pedagogical content knowledge (PCK-SRL). PCK-SRL encompasses knowledge of various direct and indirect methods to promote SRL. In this study, our focus is on teachers’ CK-SRL, which encompasses their theoretical knowledge and understanding of the fundamental concepts of SRL. This includes their familiarity with terminology, theoretical models, strategies, and components of SRL. Researchers reported a wide range of teachers’ levels of CK-SRL, including teachers who hold misconceptions or possess only fragmented knowledge about strategies and metacognition and teachers who know various strategies (Callan & Shim, 2019; Dignath & Sprenger, 2020).

2.3 Diagnostic competences as a facet of teachers’ professional competences

The diagnostic competences of teachers are essential for successfully assessing students’ SRL, which in turn is the basis of adaptive and individualised teaching of SRL (Corno, 2008). In general, diagnostic competences is an umbrella term used to refer to teachers’ skills needed for the entire process of assessing students’ knowledge and skills. These competences include the ability to identify students’ misconceptions as well as their strengths and areas of proficiency (Südkamp et al., 2012). Diagnosing is a goal-oriented procedure of collecting and integrating specific information about students’ learning, tasks, and instructional situations (Heitzmann et al., 2019; Kramer et al., 2021). While diagnosing, teachers assess the current state of their students’ knowledge and skills in relation to predefined learning objectives and tasks. The aim is to identify students' misconceptions, difficulties and strengths so that teachers can tailor their instructional support to meet these needs and align with learning goals (Chernikova et al., 2020). Teacher diagnostic competences regarding SRL encompass a diverse set of skills, knowledge, beliefs, and motivational orientations that enable teachers to effectively assess students’ SRL strengths, weaknesses, and areas for improvement. These diagnostic competences involve gathering, analyzing, and interpreting data from multiple sources concerning SRL. Moreover, teachers with diagnostic competences in SRL can provide timely and targeted feedback, design appropriate instructional interventions, and make informed decisions regarding instructional strategies and adaptations based on their comprehensive assessment of students’ SRL (Karlen et al., 2020).

2.4 Teachers’ diagnostic knowledge and assessment practices regarding self-regulated learning

Teachers’ diagnostic knowledge about SRL and their assessment practices are evident in their selection and use of appropriate assessment instruments. Diagnostic knowledge includes understanding how learners acquire knowledge and being familiar with assessment instruments suitable for collecting the necessary information (Klug et al., 2016). Thus, teachers’ diagnostic knowledge about SRL might be closely linked to their CK-SRL, as CK is a vital prerequisite for recognizing students’ specific knowledge, skills, or behaviors (Heitzmann et al., 2019).

Assessment instruments play a crucial role in generating accurate information about students by capturing relevant diagnostic information or cues that provide valuable insights into their learning progress, strengths, and areas for improvement (Heitzmann et al., 2019; Südkamp et al., 2012). Cues refer to information about tasks, students, or contextual aspects that teachers can use in the assessment process to make a judgment (Loibl et al., 2020). For instance, teachers could examine students’ learning traces, such as note-taking or highlighting keywords, while they read a text to evaluate the quality of the strategies employed. Therefore, teachers need to possess knowledge of assessment practices, including instruments and cues, to assess their students’ SRL. Furthermore, they should be capable of critically evaluating these practices to determine which ones to employ in specific situations (Kramer et al., 2021; Loibl et al., 2020).

Teachers’ assessment practices comprise activities such as different assessment approaches (e.g. formative and summative techniques) and methods (use of instruments and cues) of data collection (Kramer et al., 2021). To assess students’ SRL, teachers can apply several assessment instruments to collect data. In SRL research, instruments to assess SRL can be categorised into offline or online instruments (Michalsky, 2017; Winne & Perry, 2000). Online methods (e.g. think-aloud, observation while learning, log traces) assess SRL during students’ learning. In contrast, offline methods (e.g. questionnaires, interviews, learning diaries) measure SRL independently from students actual learning process or immediately after learning. The limited number of empirical studies conducted on this topic have revealed that certain teachers do not assess their students’ SRL in their classrooms and they encounter challenges in identifying cues that can be utilized to identify their students’ SRL (Callan & Shim, 2019; Dignath & Sprenger, 2020; Michalsky, 2017). According to Michalsky (2017), a majority of teachers (63.3%) only apply offline SRL aptitude assessments (e.g. self-report questionnaires, oral interviews), and 28.2% of the teachers also reported incorporating offline SRL event assessments (e.g. learning diaries). Only a minority of teachers (8.5%) reported incorporating online SRL event assessments in their classrooms. Limited digital tool availability in classrooms may explain the lack of teachers who assess students’ SRL with online SRL event assessments. In another study, teachers reported mainly using a learning diary (55.6%,), followed by interviews (reflective talks, 34%), and unsystematic (13.2%) and systematic observations (11.7%). Further, Callan and Shim (2019) asked teachers what cues of deficient SRL they use. Most teachers reported that they interpret off-task behaviour, underachievement, and disengagement as cues of deficient SRL (Callan & Shim, 2019). However, these cues are not necessarily diagnostic of SRL, as achievement or disengagement can have many other causes than poor SRL. Similarly, Dignath and Sprenger (2020) concluded that teachers mainly reported practices that are not necessarily diagnostic of SRL, such as observation of off-task behaviour or students’ learning results (e.g. poor work completion). The three studies suggest variability in teacher assessment practices. However, further research is required to better understand how teachers assess their students’ SRL.

2.5 Teachers’ judgment accuracy of students’ self-regulated learning

The quality of teacher judgments is commonly evaluated based on the accuracy of their assessments. Accuracy refers to the extent to which teacher judgments align with the actual condition of the diagnosed learners. There are several measures to express judgment accuracy. A common measure, known as person-related judgment accuracy, refers to the degree to which teacher judgments correspond to actual levels of student characteristics or task difficulty (Urhahne & Wijnia, 2021). This accuracy is typically measured by examining the correspondence between teacher judgments and the student’s actual level of characteristics (most often indicated by a test score). To compare the results between studies, teachers' judgment can be quantified through the correlation between teacher-predicted scores and the students' actual scores (Südkamp et al., 2012; Urhahne & Wijnia, 2021). The premise is that a higher correlation between teacher judgment and actual student characteristics indicates a higher level of accuracy in the judgment. However, the question arises regarding determining which judgments are considered accurate or inaccurate. One approach is to utilize established statistical benchmarks to define the degree of alignment between teacher judgments and actual student characteristics (Cohen, 1988): Correlations between 10 and 30 are described as low, between .30 and .50 as moderate, and between .50 and .90 as high or very high. From a theoretical perspective, teachers' accuracy could be considered high when their instructional actions effectively result in adaptive learning support, facilitating the individual development of learners (Corno, 2008). However, determining the specific level of accuracy required for effective student support remains an open question. Further research and discussions are needed to establish guidelines or benchmarks for determining the desired level of accuracy that leads to meaningful student development.

Only a few researchers examined how accurately teachers judge students’ SRL skills. In an early study, Carr and Kurtz (1991) reported that primary school teachers were moderately accurate in judging students’ metacognitive knowledge (r = .40). However, teachers ranked high achievers compared to low achievers as having higher SRL skills beyond actual group differences. Several studies have shown that teachers judge students’ motivation and emotions (e.g. self-concept, learning motivation, anxiety) with low to moderate accuracy (Carr & Kurtz, 1991; Givvin et al., 2001; Spinath, 2005). In contrast, Lee and Reeve (2012) found that high school teachers judgments of students motivation did not correlate with students’ self-reported motivation. However, the authors found higher levels of judgment accuracy when teachers had to judge behavioural outcomes such as student engagement (r = .29–.40). Comparable values have been reported by Lohbeck et al. (2015) for learning effort (r = .29). Stang and Urhahne (2016) investigated mathematics teachers’ judgment accuracy. Whereas teachers judged student achievement with moderate accuracy (r = .43), they had more difficulties assessing students’ learning behaviour (r = .30; e.g. ability to organise learning, students’ independence) and concentration (r = .24). Very few studies assessed students’ SRL strategies. Friedrich et al. (2013) revealed that mathematics teachers’ judgment accuracy of students’ SRL strategies was low (r = .13–.29). Comparable to Stang and Urhahne’s (2016) results, those judgments were of lower accuracy than for mathematics-specific skills, highlighting that it might be more difficult for teachers to judge SRL than subject-specific skills.

Based on the previous studies, it is evident that teachers encounter greater challenges in making accurate judgments about students’ SRL compared to their judgments about students’ achievement. Overall, the accuracy of teacher judgments for SRL can be described as relatively low. Possible reasons for this may be that teachers lack experience in assessing students’ SRL, that there is a lack of own experience regarding SRL, and that they may not be specifically trained to assess their students’ SRL (Karlen et al., 2020; Kramarski & Heaysman, 2021). Moreover, Urhahne and Wijnia (2021) argued that it might be more difficult for teachers to accurately assess non-cognitive characteristics because they can rely on less concrete or reliable information than, for example, for academic achievement or classroom engagement. Non-cognitive characteristics (e.g. motivation) sometimes cannot be assessed directly and must thus be interpreted indirectly through different information or cues (Lee & Reeve, 2012).

2.6 Predictors of teachers’ judgment accuracy

Studies revealed variability in the accuracy of teacher judgments based on teacher characteristics, such as teaching experience. Teaching experience is often operationalised as a teacher’s job experience and is a well-studied teacher characteristic (Urhahne & Wijnia, 2021). Whereas some studies found a positive association between job experience and judgment accuracy (e.g., Ready & Wright, 2011), other studies, however, suggest that job experience is not or only weakly associated with teachers’ judgment accuracy (e.g. Praetorius et al., 2011; Zhu & Urhahne, 2015).

Although it has been emphasised that different teachers’ professional competence aspects influence their assessment activities and judgment accuracy, studies on the correlates of teachers’ professional competence aspects (e.g. knowledge, motivation) and their judgment accuracy are scarce and show varied patterns (Ohle et al., 2015; Südkamp et al., 2012; Urhahne & Wijnia, 2021). Since there are only a few studies on SRL, we also include studies about teachers’ judgment accuracy about other students’ characteristics (e.g. achievement) in the literature review.

Research with teachers has stressed the role of professional knowledge in teachers’ assessment procedures, especially judgment accuracy. Kramer et al. (2021) reported a positive relationship between pre-service biology teachers’ PCK and their assessment activities and judgment accuracy. However, CK was not significantly correlated to teachers’ assessment activities and judgment accuracy. In contrast, Kron et al. (2022) found a positive effect of pre-service teachers’ CK on their judgment accuracy. Moreover, Radkowitsch et al. (2023) found analogous to Kramer et al. (2021), that only PCK and not CK was related to judgment accuracy. Klug et al. (2016) showed that teachers’ diagnostics knowledge correlated with their judgment of student learning behaviour. Finally, Michalsky (2017) also reported that only teacher knowledge about SRL assessments correlated with their assessment activities. Unfortunately, the accuracy of the teachers’ judgments has not been examined. Altogether, there is evidence that specific professional knowledge aspects influence teachers’ judgment accuracy.

When examining teachers’ motivational dispositions and their impact on judgment accuracy, empirical evidence is scarce. Few studies have been conducted on this topic, and the results have been mixed, providing limited conclusive findings. Klug et al. (2016) found that pre- and in-service teachers’ motivation was positively related to their judgment accuracy of student learning behaviour. However, Praetorius et al. (2017) found that teacher self-efficacy did not affect their judgment accuracy of students’ motivation. In a more recent study, Radkowitsch et al. (2023) found that only teachers’ success expectancy but not their task value and self-efficacy was related to their judgment accuracy. However, teachers’ motivation seems related to their engagement in the assessment activities. For example, Ohle et al. (2015) reported that teachers’ self-efficacy positively affected their time spent with assessment activities.

3 The present study

Teachers play a critical role in fostering students’ SRL development. By being knowledgeable about their students’ SRL needs and strengths, teachers can adaptively promote SRL. To accomplish this, teachers require diagnostic competences,  which are part of their professional competences in SRL. However, there is a notable lack of research investigating whether teachers accurately assess their students’ SRL and which aspects of teachers’ professional competences in SRL influence the accuracy of their judgments. In this study, we aim to address three research questions:

  1. 1.

    What assessment activities do teachers use to assess their students’ SRL?

The assessment activities of teachers regarding students’ SRL have received limited research attention, making this research area exploratory. However, incorporating diverse assessment activities and utilising different methods and perspectives offer potential advantages. This comprehensive approach holds the potential to provide a more holistic and thorough understanding of students’ SRL skills. Consequently, it is desirable for teachers to apply several assessment activities to effectively assess their students’ SRL.

  1. 2.

    To what extent do teachers’ judgments of students’ SRL skills correspond to students’ actual SRL skills?

Previous studies suggest that teachers find it more difficult to accurately judge students’ SRL than student achievement or domain-related skills (e.g. Friedrich et al., 2013) and that teachers often achieved low accuracy (correlation lower 0.30) in their judgments about students (e.g. Givvin et al., 2001; Lee & Reeve, 2012; Stang & Urhahne, 2016). Thus, we expected a relatively low agreement between teachers’ judgment of students’ SRL skills and students’ actual test scores for SRL skills.

  1. 3.

    To what extent do teacher characteristics (job experience, experience in assessing SRL) and professional competences (self-concept, knowledge, and assessment activities) influence their accuracy in judging students’ SRL?

Studies examining the correlates of various aspects of teachers’ professional competences and judgment accuracy are limited and the reported patterns are inconsistent (Südkamp et al., 2012; Urhahne & Wijnia, 2021). Particularly, there is a lack of research on the significance of teachers’ competence in SRL for their judgment of students’ SRL skills. Given the absence of clear hypotheses derived from previous research findings, an exploratory approach was adopted to investigate this question.

4 Methods

4.1 Procedure and participants

The study involved emailing several public schools in Switzerland, specifically the school principals. Four lower secondary schools representing urban and rural areas voluntarily agreed to participate in the study. The recruitment of teachers was carried out via email invitations sent by their respective principals. Participation was voluntary for teachers and their classes. This study was approved by the Ethics Committee of the Univerity of Applied Sciences and Arts Northwestern Switzerland. Before participation, informed consent was obtained from parents or guardians of the students.

Forty-one lower secondary school teachers and their classes participated in this study. The participants taught in all three grades of the lower secondary school level. In Switzerland, three years of lower secondary schooling follow six years of primary school. Notably, 41.5% of the teachers were female and 58.5% were male. The teachers covered the entire age range, with five teachers being 20–29 years old and one 60 years or older (43.9% were 40–49 years). The work experience of the teachers ranged from 2 to 35 years (M = 15.0; SD = 8.2). Regarding experience of assessing students’ SRL, 9.8% of the teachers indicated having no experience, 61.0% had little experience, and 29.3% had some experience.

Teachers had to rate the SRL skills of the first five students on their class list. Thus, this study includes a sample of five students per teacher. However, not all rated students filled out the questionnaire due to illness or missing parental consent. The final student sample consisted of N = 173 students (49.7% female) who were M = 14.42 years old (SD = 0.97).

4.2 Measures

All participating students completed an online survey during two class lessons. The student questionnaire was divided into two parts. In the first part, students were required to complete metacognitive knowledge tests that assessed their understanding of various SRL skills. The second part of the questionnaire included closed questions on demographic information and additional inquiries related to their SRL.

Teachers were required to complete their online questionnaire before accessing the students’ questionnaire. The teacher questionnaire consisted of three parts. The first part included closed questions concerning teachers’ demographic information. In the second part, to minimize potential response bias, teachers were initially asked to report their knowledge about SRL and their SRL assessment activities before being presented with further scales related to SRL. The final part of the questionnaire focused on teachers’ judgment accuracy of students’ SRL.

4.2.1 Student questionnaire

The students’ questionnaire includes various knowledge tests on different components of SRL. The SRL components were selected based on current research on the importance of different skills in SRL for academic success (e.g., Richardson et al., 2012). Further, to ensure content validity, high practicability, and a great benefit for practice, we asked a separate group of teachers which dimension of SRL they felt was most relevant to school learning. This procedure led to the identification of eight SRL components.

For this study, students’ metacognitive knowledge of four different SRL components is included: monitoring and reflection strategies, strategies for management of the learning environment, self-control strategies, and motivational regulation strategies. In a preliminary study, the four SRL components showed expected positive correlations with students’ GPA, home educational resources, cultural possessions at home, and books at home, confirming the validity of the scales (Bäuerlein et al., 2022).

4.2.1.1 Metacognitive knowledge about monitoring and reflection

The scenario test for assessing students’ knowledge about monitoring and reflection strategies was developed by drawing upon existing scenario-based procedures commonly used to assess metacognitive knowledge about strategies (Händel et al., 2013; Maag Merki et al., 2013). In each of the four scenarios presented (e.g. the learner gets stuck solving a difficult task), students were given a list of six strategies that could potentially be employed to address the specific problem. For each scenario, students were required to analyze the goals and characteristics of the particular learning problem. They then had to activate their knowledge regarding the characteristics of the listed strategies and assess the extent to which each strategy would be useful in solving the given task. Students rated each strategy on a scale from 1 (not useful at all) to 6 (very useful). The strategies (items) were scored via pair comparisons (strategy X is more or less useful than strategy Y), referencing experts’ judgments of the listed strategies since scenario-based tests are evaluated based on experts’ judgments. Several experts from the specific field also rated the usefulness of all listed strategies, offering qualitative standards and a clear benchmark to evaluate students’ answers. Thus, based on theoretical assumptions validated by experts (at least 75% agreement about the relation of two strategies), pairs of strategies were built.

Sum scores were calculated based on pair comparisons of the ratings of different strategies. The students’ pair comparisons were then compared with the expert ratings. One point was awarded for each correct answer and zero points for every false answer. The value of the score varied between 0 (0% correspondence with the experts and low knowledge) and 1 (100% correspondence with the experts and high knowledge). The monitoring and regulation scale comprised 27 pair comparisons and showed good internal consistency (α = 0.83).

4.2.1.2 Metacognitive knowledge about management of the learning environment

A new scale consisting of seven tasks was developed to assess students’ knowledge about managing the learning environment. Two tasks presented students with four pictures depicting different workplaces and they had to select the most conducive to learning. The remaining five tasks described scenarios of a young learner facing difficulties studying at home and students were required to choose one option out of three that provided the best supported learning in each scenario. For each correctly answered task, students were awarded one point. The scale contained seven tasks with a moderate internal consistency (α = 0.60).

4.2.1.3 Metacognitive knowledge about self-control

Four situations were described to assess students’ metacognitive knowledge about self-control strategies. In each situation, a learner was depicted as studying but faced a distraction, such as receiving a cell phone call. Students were then presented with five possible strategies and asked to select the best option for managing the distraction in each situation. The presented control strategies differed in usefulness based on Duckworth et al. (2016) theoretical and empirical findings. Students received one point for each correct answer. The scale had a moderate internal consistency (α = 0.64).

4.2.1.4 Metacognitive knowledge about regulation of motivation

The development of the scenario-based test to assess students’ knowledge about the regulation of motivation was constructed per existing and frequently validated and tested scenario-based procedures for assessing metacognitive knowledge about strategies (e.g. Händel et al., 2013; Maag Merki et al., 2013). The metacognitive knowledge of students about the regulation of motivation was assessed using four scenarios that describe situations where a learner experienced motivational difficulty. Several strategies were listed in each situation, differing in their usefulness in influencing a learner’s motivation (Wolters, 2003). Several experts from the specific field of motivation rated the usefulness of all listed strategies, offering qualitative standards and a clear benchmark to evaluate students’ answers. The students evaluated the strategies for usefulness in mastering the scenario-specific motivational challenges on a 6-point Likert scale ranging from 1 (not at all useful) to 6 (very useful). The students had to reach the same assessment as the experts to achieve a high knowledge score. For each matching pair comparison, the students received one point. The value of the score varied between 0 (0% correspondence with the experts and low knowledge) and 1 (100% correspondence with the experts and high knowledge). Scores were calculated based on 22 pair comparisons. The scale had a good internal consistency value (α = 0.88).

4.2.2 Teacher questionnaire

4.2.2.1 Demographics

Teachers’ job experience (“How many years of professional experience do you have?”, 1 = 1 year to 45 = 45 years) and experience in assessing SRL (“Do you already have experience in diagnosing self-regulated learning?”, 1 = I have no experience at all to 4 = I have much experience) were assessed each with one question.

4.2.2.2 Self-concepts about SRL

Teachers’ self-concept of their own SRL was assessed with a scale developed by Karlen et al. (2020) consisting of three items (e.g. “I am good at self-regulating my learning.”). Answers were provided on a 4-point scale from 1 (does not apply) to 4 (applies). The scale’s internal consistency was good (α = 0.89).

4.2.2.3 Content knowledge about SRL (CK-SRL)

We used one open-ended question to assess teacher CK-SRL. Teachers were asked to provide a typed response to an open-ended question regarding their definition of SRL: “Please describe in your own words what you understand by self-regulated learning.” The answers of the teachers were qualitatively analysed. In the first step, we applied a structured content analysis (Mayring, 2015) to create the coding framework for teachers knowledge about SRL, which was developed both inductively by analysing the data mentioned by teachers and deductively from a priori categories (area of regulation and process phases of SRL) derived from the available research literature (Pintrich, 2000; Zimmerman, 2000). In the second step, teachers’ answers were analyzed using evaluative content analysis (Kuckartz & Rädiker, 2022). This procedure makes it possible to draw conclusions from the written answers and classify the teachers’ answers into four levels of knowledge. Answers of teachers were assigned to level one (= one point) if they only had misconceptions about SRL in their answers, to level two (= two points) if they mentioned correct aspects of SRL in combination with incorrect ones, to level three (= three points) if they mentioned more than one correct aspect of SRL but no misconceptions, and to level four (= four points) if they mentioned (almost) all phases and areas of SRL. Higher scores represent teachers possessing a higher level of CK-SRL. Two coders conducted all coding consensually (Schmidt, 2013).

4.2.2.4 Diagnostic knowledge about SRL

To assess teachers’ diagnostic knowledge about SRL, they had to complete a learning interview task developed by Dignath et al. (2021). The task was to read excerpts of a teacher’s interview with a student about his or her learning. For 15 different student statements, teachers had to decide whether the student indicated SRL. For each correct decision, teachers received 1 point. For each wrong decision, teachers received zero points. The teacher could reach 0 to 15 points. Higher values represented a higher level of diagnostic knowledge about SRL.

4.2.2.5 Self-reported assessment activities

Analogous to Michalsky (2017), teachers were asked to provide a typed response to an open-ended question regarding their self-reported assessment activities to identify their students’ SRL: “How do you identify what SRL skills your students possess? Describe how you assess SRL skills.” First, all responses were analysed following a structured content analysis with the deductive and inductive formation of categories for teachers’ activities when assessing SRL. The theoretical basis for deductive categories was Michalsky’s (2017) classification of different instruments for assessing SRL. This resulted in a comprehensive coding manual (see Appendix). Second, we coded the teachers’ answers based on four predefined levels, utilising evaluative content analyses (Kuckartz & Rädiker, 2022). The four levels were based on the theoretical consideration that it is important to apply various assessment methods to obtain a comprehensive picture of students’ SRL (Dörrenbächer-Ulrich et al., 2021; Michalsky, 2017). Teachers received one point if they described mainly approaches showing misconceptions about assessing SRL (e.g. they described approaches that do not allow inferences about students’ SRL competencies), two points if they described only one appropriate approach of assessing students’ SRL or several appropriate approaches combined with inappropriate approaches, three points if they mentioned at least two appropriate approaches to assess students’ SRL without mentioning any inappropriate procedures, and four points if they described more than two appropriate approaches of assessing students’ SRL using different categories (online and offline) without mentioning inappropriate methods. The higher the score or level, the greater the variety of appropriate assessment activities teachers conduct to assess their students’ SRL.

4.2.2.6 Judgment accuracy

Teachers rated the metacognitive knowledge about SRL of the first five students on their class list for four SRL skills on a 4-point scale ranging from 1 (disagree) to 4 (agree). A random selection of students per class was used because it would have been too time-consuming for teachers to judge all students. The introduction was as follows: “Please estimate what the first five students on your class know about self-regulated learning. The focus is not on the students’ actions but on their metacognitive knowledge about SRL. Indicate how well the following statements apply to these students.” Single items were used to assess each SRL skill: “This student knows how to monitor the progress during learning” (monitoring and reflection); “This student knows what a workplace should look like to work well” (regulation of the learning environment); “This student knows what to do to focus well on learning” (self-control); and “This student knows how to motivate him/herself to learn” (regulation of motivation).

4.3 Data analyses

To determine the accuracy of the teachers’ judgments, we correlated the teachers’ judgments of their students’ SRL with the students’ results in the SRL knowledge tests, controlling for data clustering (five students per teacher). We calculated one deviation score per teacher for each assessed SRL aspect to have judgment accuracy per teacher that can be correlated with other aspects of teacher professional competence. After classifying the students’ test results on a 4-point scale, we tested to what extent the classifications deviated from the teachers’ assessments on the 4-point rating scale, using the mean deviation of teachers’ assessments to students’ test scores across all five students per teacher to control for clustering. The significance of the deviation was tested using a one-sample t-test.

We calculated correlations to determine if teacher characteristics and professional competence aspects were related to their judgment accuracy. Using Spearman rank correlations, we correlated the mean deviation scores with teachers’ knowledge about SRL and their SRL assessment activities. Teachers’ work experiences, experiences with diagnosing SRL, self-concepts about SRL, and diagnostic knowledge about SRL were correlated with the mean deviation score using Pearson correlations.

5 Results

5.1 Descriptive statistics of teachers’ competences

Table 1 offers descriptive statistics for various aspects of teachers’ characteristics and professional competences, along with showing these variables' relationships (intercorrelations). Overall, the correlation analysis indicated non-significant or weak to moderate correlations between the variables. The diagnostic knowledge of teachers was positively related to their self-reported assessment activities. Further, teachers’ CK-SRL correlated positively with their diagnostic knowledge about SRL.

Table 1 Descriptive statistics and correlations for teachers’ professional competences in self-regulated learning
Full size table

5.2 Teachers’ self-reported assessment activities

Considering how teachers described their assessment activities to assess their students’ SRL (see Table 2), it was striking that they most frequently mentioned conversations or interviews about learning with students (82.5%). Second most often, but well behind, they mentioned observing students’ learning behavior (37.5%). Finally, some teachers wrote about activities not diagnostic of SRL (e.g. “tally sheet for on-time submissions”). Classifying these descriptions into online and offline assessments revealed that teachers most often assess their students’ SRL using offline assessment practices (95.0%) compared to online practices (42.5%).

Table 2 Teachers’ self-reported assessment activities of students’ self-regulated learning
Full size table

Focusing on the level of teachers’ assessment activities, we found that 20.0% of the teachers described only assessment activities that were not diagnostic of students’ SRL (Level 1). 45% of the teachers named only one appropriate SRL assessment activity or one or more assessment activities combined with at least one inappropriate assessment activity (Level 2). 15% of the teachers reported at least two appropriate assessment activities without mentioning inappropriate ones (Level 3). Finally, 20% of the teachers described at least three appropriate ways of assessing their students’ SRL, covering different categories (online-offline), without mentioning inappropriate methods (Level 4).

5.3 Teachers’ judgment accuracy of students’ self-regulated learning

The correlation of teachers’ judgments of students’ SRL with their students’ actual metacognitive knowledge test scores yielded small but statistically significant relationships for all four SRL skills (see Table 3). However, all the correlations are between 0.04 and 0.19, indicating no or low judgment accuracy.

Table 3 Student test scores, teachers judgments, and teacher judgment accuracy
Full size table

5.4 Relationship between teachers’ competences in self-regulated learning and their judgment accuracy

Table 4 shows the correlation between teachers’ work experience, competences in SRL, and judgment accuracy. Teachers’ judgment accuracy value represents the mean deviation between the teachers’ judgment of their five students’ SRL and their students’ actual metacognitive knowledge test scores. The correlations are negative because a smaller deviation indicates higher judgment accuracy. The results revealed two significant correlations: Firstly, there was a significant correlation between teachers’ assessment activities and their judgment accuracy (mean deviation) for regulating the learning environment. Secondly, teachers’ self-concept about their own SRL correlated with their judgment of students’ knowledge about self-control.

Table 4 Correlations between teachers’ work experience, competences in self-regulated learning, and judgment accuracy
Full size table

6 Discussion

Students might need support developing their SRL skills to become successful self-regulated learners. To adaptively promote SRL, teachers need to accurately assess student needs and strengths in SRL (Corno, 2008; Karlen et al., 2020). Previous research has mainly focused on investigating the accuracy of teachers’ judgments regarding student achievement (Südkamp et al., 2012). This study adopts a new perspective by examining teacher assessment activities and their judgment accuracy concerning their students’ SRL. Furthermore, the study explores the relationship between teachers’ professional competences in SRL and their judgment accuracy. The main findings and their implications for future research are discussed in the following sections.

Our first research question aimed to investigate the assessment activities employed by teachers to evaluate their students’ SRL. Consistent with previous research (Michalsky, 2017), the findings indicate that teachers primarily rely on offline assessment approaches rather than online methods when assessing their students’ SRL. In this study, teachers reported predominantly utilizing reflective talks and interviews about learning to identify strengths and weaknesses in their students’ SRL. They mentioned scheduling dedicated time slots for one-on-one discussions with students about their learning. Additionally, some teachers reported observing their students’ behaviour as part of their assessment activities for SRL. However, most teachers did not mention specific student activities they observed. Thus, it remains unclear whether SRL-relevant behaviour is being observed since only a few teachers reported having checklists to help observe and assess SRL behaviour. Furthermore, similar to findings in other studies (Dignath & Sprenger, 2020), teachers in this study were familiar with using learning diaries to assess SRL among their students. From this, it may be inferred that offline methods are perceived to be more practical in everyday teaching practices than online methods. Furthermore, it is presumable that teachers are more familiar with offline methods than online methods, which could explain their preference for offline approaches in assessing students’ SRL.

Consistent with previous research findings (Dignath & Sprenger, 2020; Michalsky, 2017), a significant number of teachers in our study (65.0%) exhibited low variability in their assessment activities and even displayed misconceptions when assessing students’ SRL. However, it is worth noting that a significant proportion of teachers (35.0%, levels 3 and 4) reported employing multiple assessment methods. Using a combination of various assessment instruments is crucial to capture the diverse range of skills encompassed within students’ SRL (Callan & Cleary, 2018; Dörrenbächer-Ulrich et al., 2021). This multi-methodical approach may enable teachers to gain deeper insights into their students’ SRL, encompassing cognitive, motivational, emotional, and behavioral aspects. However, further research is needed to understand the factors that influence teachers’ decision-making processes when selecting SRL assessment methods and the specific contexts in which these choices are made. Exploring these factors will contribute to a better understanding of how teachers can effectively assess and support students’ SRL in various educational settings. Additionally, it is crucial to identify the specific SRL skills targeted by various assessment methods. This identification will enable teachers to effectively assess and support specific aspects of students’ SRL. By aligning assessment methods with targeted SRL skills, teachers can provide tailored support and interventions that address the specific needs of students in their self-regulated learning journey.

Our second research question aimed at examining the accuracy with which teachers judge their students’ SRL. In line with previous results (Carr & Kurtz, 1991; Lee & Reeve, 2012; Spinath, 2005; Urhahne & Wijnia, 2021), we found that teachers judged their students’ metacognitive knowledge about SRL with low accuracy. A possible explanation for the observed findings could be that teachers were required to assess students’ metacognitive knowledge about SRL rather than relying solely on observable SRL behavior or self-reports regarding strategy use. Previous empirical findings have underlined that teachers possess strengths in assessing observable student characteristics (e.g. Südkamp et al., 2012). The quantity and quality of information available to teachers can facilitate or impede judgment accuracy (Südkamp et al., 2012; Urhahne & Wijnia, 2021). Thus, it is plausible that teachers may encounter challenges in identifying suitable indicators or cues that provide information about students’ metacognitive knowledge regarding SRL. The metacognitive knowledge of students may need to be made explicit to teachers for accurate judgment. Therefore, the question arises: How do teachers obtain information about students’ metacognitive knowledge about SRL? Teachers could obtain information about students’ metacognitive knowledge by employing various methods, such as questioning students during learning and asking specific questions about strategy application. Teachers may also utilize learning diaries in which students can reflect on their SRL experiences and provide insights into their understanding and application of SRL strategies. However, this will likely require specific training sessions in which teachers can be instructed.

Our third question focused on how far teachers’ characteristics and professional competence aspects might influence teachers’ judgment accuracy. As there is limited research on teacher diagnostic competences in SRL, with the literature primarily focusing on teachers’ judgment of students’ achievement, this question was investigated in an exploratory manner. Consistent with previous studies examining teachers’ judgment of students’ achievement (Praetorius et al., 2011; Zhu & Urhahne, 2015), teachers’ work experience was uncorrelated with their judgment accuracy in SRL. A possible explanation for why job experience was not associated with higher accuracy of judgment could be that in-service teachers do not usually receive feedback on their perceptions of students’ SRL and therefore have little opportunity to improve their specific accuracy of judgment about students’ learning. In addition, in the context of SRL, work experience is often not associated with higher promotion of SRL (Jud et al., 2023).

Further, teachers’ experience in assessing SRL was uncorrelated with their judgment accuracy in SRL. A possible explanation for the lack of correlation between teachers’ experience in assessing SRL and their accuracy of judgment could be that accuracy of judgment in SRL requires specific expertise and training beyond general assessment experience. In addition, assessing metacognitive knowledge about SRL may involve complexities and nuances beyond teachers’ previous experiences. The question therefore arises whether teachers with more experience in assessing SRL would more accurately assess other aspects of SRL (e.g. strategy use) that might be easier to assess with the instruments teachers use to assess students’ SRL.

The self-concept of teachers as self-regulated learners only correlated with their judgment of students’ self-control but not the other students’ SRL skills. This finding partially confirmed the findings of previous studies that did not find a relationship between teachers’ motivational aspects and their judgment accuracy (Ohle et al., 2015; Praetorius et al., 2017). It is worth noting that on average, teachers rated themselves as proficient in SRL, which raises the possibility of a bias in self-concept scores. The overestimation of their own SRL abilities could potentially contribute to the lack of correlations between self-concept and judgment accuracy observed in the study. However, it has been discussed in the literature that through their own experiences as self-regulated learners, they might better understand and recognise the development of their students’ SRL and address the needs and difficulties their students face during SRL (Dembo, 2001). Thus, further research is needed to examine how teachers’ actions and judgment in the classroom are guided by their competences as self-regulated learners. Very few studies have focused on teachers’ competences as self-regulated learners (Karlen et al., 2023). To this end, it would be interesting to include other aspects of teachers’ competence as self-regulated learners (e.g. teachers’ use of strategies and their metacognitive knowledge).

The CK-SRL of teachers was unrelated to their judgment accuracy, which aligns with studies that reported that teachers’ CK does not correlate with their assessment activities and judgment accuracy (e.g. Kramer et al., 2021; Michalsky, 2017). CK alone may not be sufficient to predict judgment accuracy in SRL assessment. Assessing SRL involves understanding and evaluating students’ metacognitive processes, strategies, and reflections, which may not solely rely on CK-SRL. In line with that argument, research showed that judgment accuracy could depend more on teachers’ PCK and familiarity with students’ misconceptions about SRL than their CK alone (e.g., Kramer et al., 2021). Furthermore, the results also raise the question of what kind of CK (e.g. about theoretical models of SRL vs. about metacognition) is needed to possibly provide an important basis for teachers’ judgment accuracy of specific SRL skills. This study assessed CK-SRL with open-ended questions. Thus, future studies could use more sophisticated measures to assess teachers’ CK-SRL. Interestingly, no significant correlation was observed between teachers’ diagnostic knowledge about SRL and their judgment accuracy. This finding raises the question of what specific knowledge and at which quality level teachers need to assess and judge students’ SRL accurately. Further investigation is needed to determine the specific components and depth of knowledge contributing to teachers’ ability to judge students’ SRL accurately. Drawing on literature from other fields, PCK could be seen as a central knowledge facet within teachers’ diagnostic competence, particularly when the diagnosis emphasises content-specific aspects of teaching (Kramer et al., 2021). While CK-SRL might not be essential for teachers’ accurate judgment about students’ SRL skills, it does exhibit a positive relationship with PCK-SRL (Karlen et al., 2020). This highlights the role of CK-SRL in delineating the extent of PCK-SRL development. However, further research is needed to understand teachers’ SRL knowledge development and possible links between different aspects of SRL knowledge and teachers’ behaviour.

We found a positive correlation between teachers’ assessment activities and their accuracy in judging students’ regulation of the learning environment. One possibility is that teachers who actively engage in various assessment activities may have a deeper understanding of the components and indicators of students’ regulation of the learning environment. By regularly assessing this aspect of students’ SRL, teachers may develop a more refined judgment and sensitivity towards recognizing the cues, behaviours, and metacognitive knowledge associated with effectively regulating the learning environment. Moreover, it is possible that students’ metacognitive knowledge regarding the regulation of the learning environment is reflected in observable behaviours within the classroom. Teachers may observe how learners actively organize their workspace, ensuring the availability and accessibility of school materials and witnessing the overall organization and structure demonstrated by students. These observable behaviors can serve as cues for teachers to infer students’ metacognitive knowledge and application of strategies related to the regulation of the learning environment.

6.1 Practical implications

The results of this study emphasise the importance of supporting teachers in assessing their students’ SRL. There is a specific need for professional development programmes for teachers to support their judgment accuracy in SRL. Professional development programmes should address how teachers can identify their students’ needs and strengths in SRL and what cues and assessment methods might be helpful. Researchers and educators could help teachers develop effective ways of integrating SRL assessments into their regular teaching. It might be necessary to offer teachers explicit support in using SRL assessment methods in their classrooms and explain why, when, and how to assess SRL. Further, teachers might need specific support in judging the various SRL skills of their students. One promising approach could be to provide teachers with simulation-based learning opportunities (Chernikova et al., 2020). This would enable teachers to practice their diagnostic competences in a very targeted way, using fictional examples. Finally, it would be desirable to develop digital diagnostic tools to support teachers in assessing their students’ SRL in classrooms. With such digital tools, teachers could compare their judgment with students’ test scores, verifying their judgments.

6.2 Limitations and future directions

Several limitations should be acknowledged in this study. Firstly, the study focused on four specific SRL components and specifically examined students’ metacognitive knowledge. While metacognitive knowledge is a significant predictor of effective strategy use and academic performance, it is important to recognise that SRL is a multidimensional construct, encompassing additional components and skills, including cognitive and emotional aspects of SRL and students’ strategy use in addition to metacognitive knowledge (Pintrich, 2000; Pressley et al., 1987; Zimmerman, 2000).

Secondly, it should be noted that not all the SRL scales used as objective measures in this study demonstrated high reliability values. This suggests that there is still room for improvement in the measurement of SRL, and there is a need to develop additional tasks and scales that can accurately assess the various components and skills of SRL. Enhancing the reliability and validity of SRL measurements will contribute to more robust research and a better understanding of students’ self-regulatory processes.

Thirdly, the non-significant correlations between teachers’ professional competences and their judgment accuracy as well as the relatively low judgment accuracy scores may be influenced by additional factors not considered in this study. Future research could explore the impact of various professional competences, including teacher beliefs, motivational orientations, and PCK-SRL, on teachers’ judgment accuracy. Additionally, it is important to acknowledge that teachers’ judgments may be susceptible to biases related to student characteristics, such as their performance or achievement levels and their behaviour within the school context (Ready & Wright, 2011). These factors may act as moderators of teachers’ judgments and should be considered when interpreting the results and implications of this study.

Fourthly, it is necessary to acknowledge certain methodological issues that should be considered. The study’s sample size might have limited the generalizability of the findings. A larger and more diverse sample would enhance the external validity of the results. The cross-sectional nature of the study design limits our ability to establish causal relationships between variables. Longitudinal or experimental designs would provide stronger evidence for understanding the dynamics and effects of teachers’ assessment activities and judgment accuracy about students’ SRL. The literature indicates that teachers’ familiarity with a task or test can affect judgment accuracy (Südkamp et al., 2012; Urhahne & Wijnia, 2021). In our study, teachers assessed students’ metacognitive knowledge about various aspects of SRL using a 4-point scale, then compared this to the students’ actual knowledge scores assessed through a test. In the literature, the terms “indirect judgments” and “uninformed judgments” are commonly used to describe this approach. Previous studies have demonstrated lower correlations for indirect or uninformed judgments than direct and informed ones (e.g. Südkamp et al., 2012; Urhahne & Wijnia, 2021). It is also worth noting that teachers rated students’ metacognitive knowledge about SRL using a single item in our study. While single items with a differentiated response format can still possess predictive power comparable to multiple-item rating (Zhu & Urhahne, 2015), it may be beneficial to consider using a format that aligns more closely with the tasks presented in the knowledge test administered to students. Lastly, it is important to consider the potential impact of sequence effects in the study design. The requirement for teachers to judge the first five students on their class list may introduce a bias due to the order in which students were presented. This sequence effect could influence the teachers’ judgments and potentially impact the results. Future research could address this issue by employing randomization techniques to minimize any potential sequence effects and ensure more robust and unbiased judgments from teachers.

Lastly, it is important to acknowledge that our study relied on teachers’ self-reported SRL assessment activities. However, it is recognized that the self-reports of the teachers may not always align with their actual practices in the classroom (e.g. Spruce & Bol, 2015). Therefore, it is recommended that future research explores teachers SRL assessment activities using additional methods, such as classroom observations and incorporating multiple data sources. This approach would provide a more comprehensive and nuanced understanding of teachers’ SRL practices and help bridge the gap between self-reported activities and actual implementation in the classroom.

6.3 Conclusion

Teachers require professional competences in SRL to support students in becoming effective self-regulated learners (Karlen et al., 2020). To teach SRL adaptively, teachers should assess their students’ needs and strengths in SRL accurately. This study showed that some teachers lack diagnostic competences in SRL. Not all teachers are familiar with different SRL assessment methods and some teachers mentioned assessment activities that are not diagnostic in assessing their students’ SRL. Furthermore, the findings showed that teachers struggle to judge students’ SRL accurately. Thus, this study highlights the importance of teacher educators in supporting teachers' ability to accurately assess their students SRL.