How Do Primary and Early Secondary School Students Report Dealing with Positive and Negative Achievement Emotions in Class? A Mixed-Methods Approach

Abstract

Students experience both positive and negative emotions during lessons. However, little research has been conducted into students’ perspectives on how they deal with these emotions. For this reason, the current study aimed to investigate how students perceive, deal with, and interact in emotional classroom situations, using a mixed-methods approach. In Study 1, 28 primary school children in the fourth grade drew pictures of achievement-emotion situations and were subsequently interviewed. In a three-step analysis process, students’ drawings and interviews were analysed using open coding and content analysis. In Study 2, 635 students from 38 primary and secondary schools in northern Germany reported (via the DECCS questionnaire) their appraisals, achievement emotions, emotion regulation behaviour, and desired teacher support regarding two case vignettes: one involving a learning situation and the other a performance situation. The results of Study 1 reveal that peers play a crucial role in how students deal with positive emotions in class. The structural equation model of Study 2 shows that almost all pathways between emotion regulation behaviour (maladaptive as well as adaptive) and students’ desired teacher behaviour in the form of praise and support are significant. The results of both studies indicate that social support (i.e., peer interactions and teacher support) is crucial in how students deal with both positive and negative achievement emotions in class.

1. Introduction

A recent OECD report stated that “in an increasingly fast-changing and diverse world, the role of social and emotional skills is becoming more important” [1] (p. 8). Several authors even assume that socio-emotional skills are becoming as relevant as cognitive skills for occupational outcomes [1,2]. For this reason, those involved in teaching and learning have become increasingly interested in the socio-emotional skills of students, and this was even more intensified by the challenges brought on by the COVID-19 pandemic [3,4,5,6]. Although the social and emotional skills of students are currently excessively investigated, the focus has been on personality-related traits [1]. However, how students perceive, deal with, and interact in concrete situations in which they experience positive or negative achievement emotions has not been investigated in the field of social and emotional skills until now. But, this seems to be a promising approach because students are assumably better able to empathise with concrete and realistic contexts, i.e., typical, everyday-occurring, concrete achievement-emotion situations [7,8]. In particular, primary and early secondary students are more likely to understand and answer questions that are less abstract compared to those posed in traditional trait surveys [7,8].

In any case, studies that address challenging classroom situations are rare and relate mostly to the field of higher education or the perspective of teachers, rather than to primary or early secondary school students [9,10,11,12,13,14]. There are some studies on challenging classroom situations that are qualitative in approach and give hints on, for example, patterns of interaction in concrete classroom situations [15,16,17,18]. But, the findings from these studies have seldom been tested regarding their generalisability and/or they do not include student perspectives.

It was thus the aim of the current study to investigate how primary and early secondary students perceive, deal with, and interact in emotionally challenging achievement-emotion situations. This study aimed to utilise two new—and, until now, only rarely used—methods of data acquisition in order to gain comprehensive and deep insight into the child’s perspective on achievement-emotion situations. In the first study, we sought a deeper understanding of the students’ life world by focussing on primary school students’ drawings and interviews. In the second study, we explored the extent to which the state-like subscales in the DECCS questionnaire on students dealing with emotionally challenging classroom situations [8] are related to each other, including students’ situation-specific appraisals, students’ negative achievement emotions, students’ emotion regulation behaviour, and their desired teacher support.

1.1. Students’ Achievement Emotions

Over the last two decades, research has produced a large number of studies that refer to antecedents of the Control–Value Theory [19,20,21]. The CVT assumes that the subjective value that a student assigns to a classroom situation—as well as subjectively assessed levels of control—are decisive in appraising a situation, thus determining which discrete achievement emotion is experienced, that is, an emotion that occurs in learning and performance contexts [19,20,21]. Although CVT refers to the interpretation of situations in learning and performance contexts, most studies in this research field investigate primarily (non-situation-specific) traits of appraisals, achievement emotions, and their antecedents [22,23,24,25,26,27]. These studies have revealed the importance of achievement emotions and appraisals in many different ways, for example, that negative emotions are associated with poorer performance and vice versa and positive emotions are positively related to achievement, mediated through motivation and self-regulated learning [24,28,29,30].

There is evidence that achievement emotions are experienced domain-specifically [31]. This is probably due mainly to similar settings in different domains since Raccanello et al. (2013) [32] showed that it is primarily the setting of the teaching situation (i.e., whether it is a test situation, a homework situation, or a class discussion) that determines which emotions are experienced in the classroom. Moreover, classroom settings are related to significantly more joy, pride, relief, and relaxation—as well as less hopelessness and anger—than homework or oral/written exam settings [32]. However, a distinction between different settings—or between learning and performance emotions (or activity and outcome emotions; [24])—should not only be included in studies but should also be the focus of attention more directly.

Furthermore, the transition period from fourth to fifth grade, when students in Germany usually move from primary to secondary school, is already known to be an emotionally challenging time for students. This time is a critical life event that incurs new normative demands, including types of organisation, social interactions, different instructions, less autonomy, etc. [33]. Long-term studies show that there seems to be an initial transition effect (probably brief) that is characterised by students’ experience of intense positive emotions [22,34,35,36]. Whether there is a tendency for negative emotions to increase or decrease immediately surrounding the transition is still unclear, as research results and indications for different transition phases are not consistent [37,38,39]. What seems to be clear, however, is that as the fifth grade progresses, negative emotions increase and positive emotions decrease [37]. This unfavourable trend has also been observed for the earlier primary school years, beginning in second grade and continuing into the lower secondary grades through ninth grade [22,32,35,36,40,41], with the most significant decrease in positive emotions occurring in the first two years after transition [22]. Moreover, previous findings show that there are gender differences in students’ achievement emotions [24,37,42,43]. For example, girls tend to report higher levels of stronger negative achievement emotions in mathematics [37]. Taken together, these findings indicate an integration of grade level as well as gender, especially around transition and the first years of secondary school, into studies on achievement emotions, at least as covariates. Longitudinal studies that integrate the transition from primary to secondary school are rare and mostly provide information on only a few emotions, such as joy and boredom (e.g., [28]). Moreover, studies in this field often focus only on either primary or secondary school students, but rarely include the time before as well as after the transition (one example of including students before and after transition is the study of [36]).

1.2. Student’s Dealing with Emotions in Class

For an investigation of how students perceive, deal with, and interact in achievement-emotion situations in class, it is equally relevant—in addition to their inner psychological appraisals and emotions—to consider the behavioural level displayed in such situations [8]. For this purpose, the logic of the CVT and the DECCS (Students Dealing with Emotionally Challenging Classroom Situations [8]) provides a comprehensive and suitable theoretical basis: when students—as stated in the CVT [19]—interpret (appraise) a challenging classroom situation and experience an emotion, this process is—as stated in the DECCS [8,18]—reflected in their subsequent emotion regulation behaviour, which, in turn, is accompanied by a certain desire for support that they receive from their teachers (see Figure 1). This means that students’ internal emotion regulation behaviour—referring to the inhibition, enhancement, maintaining and modulation of emotional arousal and the initiation, increase, or decrease of the experience of emotions [44,45,46]—determines the extent of students’ desire for external emotion regulation in terms of emotional or instructional support from their teachers, according to the DECCS [8]. In turn, teachers react according to their own interpretation of students’ behaviour [47].

The fact that students’ emotion regulation (or the regulation of emotionally driven behaviour, for example, self-regulation, behavioural inhibition, etc. [45]) is closely related to teacher behaviour has already been shown in several studies [25,49,50]. However, studies in this research field tend to assume that teacher behaviour has an influence on students’ (emotion regulation) behaviour; for example, Wisniewski et al. (2020) showed that teacher feedback has a significant effect on students’ behavioural outcomes. However, rates of verbal and nonverbal praise appear to be low for students exhibiting more disruptive behaviour [51,52].

Furthermore, Nurmi’s (2012) meta-analysis shows that teachers report more conflicts and less closeness in their relationships with students who exhibit a high level of external or internal problem behaviour [53]. Notably, teacher behaviour is usually seen more as a vehicle to promote appropriate student behaviour and less as a mutual interaction process in which teachers also react to students’ behaviour [54]. But, this is certainly too short-sighted, as students’ desires do not automatically match teachers’ beliefs of what constitutes appropriate support or praise [55,56]. For this reason, it is important to take into account what students particularly wish for in regard to their teachers’ behaviour. Nevertheless, there remain only a few research studies that focus on this topic.

1.3. New Approaches to the Child’s Perspective on Their Own Emotions and Interactions

In recent years, research on students’ emotions and the links of these emotions has concentrated primarily on questionnaire analyses focussing on students’ traits (e.g., general tendencies of appraisals or general tendencies of certain emotion regulation strategies). But, it is not expedient to rely solely on traits, as this does not ensure that all children at different stages of development are able to understand and answer the commonly formulated questions. This raises the question of what is being measured and whether the data actually reflect what they are supposed to measure. For this reason, it is particularly important to use different approaches in research studies in order to gain as comprehensive an understanding as possible of children’s perceptions of their own lives. One possibility for obtaining a deeper understanding of the child’s perspective may be to use questionnaires, not as trait questionnaires, but as vignettes. This approach appears to be promising as students are presumably more adept at empathising with tangible and authentic scenarios, such as typical, everyday instances of emotions in class [7,8]. However, despite this advantage, there are still very few studies (e.g., [8]) and instruments in the field of students’ emotions and interactions in classrooms that utilise this method of data collection. But, so far, case vignettes have been primarily used in investigations on emotions or emotion regulation with adults such as university students or teachers [57,58,59,60].

Besides situation descriptions, another way to gain better access to the child’s perspective is to analyse children’s drawings and conduct explanatory interviews. These methods have an advantage over the state-like vignette questionnaires in that the questions can be designed in a more open format and, thus, completely new hypotheses can be generated. Children’s drawings have attracted interest in educational science and educational psychology in recent years because they not only show the child’s development (intellectually and physically) but also provide deep insight into the child’s world [61]. In particular, drawings can identify emotions that are being communicated non-verbally [62]. Driessnack’s (2005) meta-analysis shows that children’s drawings are a relatively robust interview strategy, with a large overall effect size [63]. In studies focussing on the emotions of primary school-aged children that included drawing in the research design, children were asked to make drawings directly after hearing a description of a scenario that may trigger certain emotions [64] or report the emotions they felt about the scenario [65,66].

Using children’s drawings and interviews to portray their emotions in the context of the primary and early secondary school experience appears to be an underrepresented area of research. An example is provided by Maxwell (2015), who asked students (n = 72) aged 9 to 10 years to draw themselves in a situation at school where they felt happy and in one where they felt unhappy, and the data were analysed using content analysis [67]. Drawings of a happy situation were categorised as peers (47), outside (46), social-informal (41), active (37), passive (35), formal learning (29), alone (25), inside (20), and teacher (5). Drawings of an unfortunate situation were assigned to the categories passive (52), peers (49), learning-formal (43), inside (37), outside (26), social-informal (24), alone (22), active (20), and teacher (17). Within the subcategories of learning settings and social settings of the unhappy drawings, constraints (24) and peer conflicts (13) stood out. Peers also played an important (but partly shocking) role in emotions in the context of bullying, as Romera and colleagues (2015) demonstrated [68]. The authors asked 276 children between the ages of 6 and 11 to draw a picture of bullying; emotions formed a category in the subsequent data analysis. The results revealed that 53% of aggressors were associated with different emotions such as joy, anger, disgust, and sadness, so it can be deduced that students’ emotions and their relations and interactions with their peers are closely linked to each other.

2. Aims and Hypotheses

It is particularly important to gain a deeper understanding of the child’s world of experience in order to be able to optimise the education and professionalisation of (future) teachers. This includes drawing attention to and shedding light on those classroom situations in which students experience a variety of emotions [8]. Although there are numerous findings from traditional research studies on the emotional traits of students [24,28,69], state-like perceptions of achievement-emotion situations have so far received little attention and, if so, usually in pure questionnaire studies or approaches using children’s drawings only [8,67,68,70]. Using children’s drawings with explanatory interviews, in combination with a quantitative questionnaire, seems to be a suitable approach for gaining a deeper understanding of students’ perceptions of achievement-emotion situations.

Numerous studies already exist on the different components of students dealing with achievement-emotion situations, namely, students’ appraisals, emotions, and emotion regulation and teacher behaviour [25,49,71,72,73,74]. The main focus is usually on the effect of teacher behaviour on students’ emotional experiences [75], but the extent to which students’ emotion regulation expressions affect the support they desire from teachers has not yet been studied. It has been shown that students’ appraisals can act as a mediator between social factors and positive/negative emotions [75]. But whether emotions can mediate the relationship between appraisals and emotion regulation is still a question that has not yet been conclusively answered.

The present mixed-methods study pursued the following goals and research questions (RQs):

Study 1

−

To investigate how primary school students perceive and describe their interactions in achievement-emotion classroom situations regarding joy and pride (RQ1).

−

To explore the role of peer interactions in positive achievement-emotion classroom situations (RQ2).

We anticipated that triggers, relevant factors, descriptions, and definitions of pride and joy in learning; appraisals; and the emotion regulation behaviour of children could be extracted from their drawings and interviews, consistent with the logic and subscales of the DECCS ([8], see Figure 1) (Expectation RQ1). We assumed that students are able to provide deeper insight into their own emotional experiences than teachers are able to surmise (as the DECCS questionnaire was constructed only on the basis of teachers’ perspectives on students dealing with emotionally challenging situations [8]). Based on prior results [67,68], it was assumed that peers play an essential role in joy-in-learning situations (Expectation RQ2).

Study 2

−

To explore how primary and early secondary students deal with emotionally challenging classroom situations, targeting the interplay of students’ situation-specific appraisals, students’ negative achievement emotions, students’ emotion regulation behaviour, and their desired forms of teacher support (RQ3), considering grade-level and gender-specific differences as covariates in how students deal with emotionally challenging classroom situations.

Following the CVT [21] and the DECCS [8] as well as previous studies with high school students [76,77,78], it was assumed that students’ situation-specific appraisals are linked to which emotion is experienced, and the nature of the emotion is, in turn, associated with how students report regulating the emotion (see Figure 1). Thus, a mediation of the connection between appraisal and emotion regulation (via the achievement emotion) was integrated into our model. Accordingly, we assumed that appraisals of emotionally challenging situations are directly associated with subsequent emotion regulation behaviour (Hypothesis 3a), indirectly mediated through the discrete emotion that students report feeling in such situations (Hypothesis 3b). Since teachers cannot respond to behaviour until students exhibit emotion regulation behaviour, as stated exclusively in the DECCS [8], students’ reported desired teacher support was modelled following students’ emotion regulation behaviour (Hypothesis 3c).

3. Study 1

3.1. Methods

3.1.1. Participants and Procedures

Within the framework of a methodological–triangulative, mixed-methods research design, two different types of data on the experience of the emotions of pride and joy in achievement-emotion situations were collected. In Study 1, 28 primary school children in the fourth grade drew pictures of achievement-emotion situations and were subsequently interviewed. In a three-step analysis process, the students’ drawings were evaluated inductively using open coding [79]; then, the interview data were evaluated deductively using structured content analysis [80,81].

Theoretical sampling was used to select an appropriate sample size and sample composition for the qualitative approach. In autumn 2019, a total of 28 primary school children (10 boys, 18 girls; aged 8–11 years) from a total of three fourth-grade classes (including one inclusion class) from one school in northern Germany participated. The majority of the children at this school are from households with a solid educational background. In advance, declarations of consent were obtained from the Senator for Education, the school principal, the class teachers, the students themselves, and their parents. Socio-demographic data were collected according to students’ age, gender, and class level. The data were kept strictly anonymous, thus preventing any inferences to be drawn about the children or their parents. Due to ethical concerns (such as not upsetting the children and avoiding any trauma), the data collection related exclusively to the positive classroom emotions of joy and pride; we used the data collection methods of drawings and interviews conducted during art classes in the after-school care environment.

3.1.2. Data Collection: Children’s Drawings and Interviews

With the help of a stimulus, the students were encouraged to draw a situation in which they felt pride or joy in learning. The students were free to choose their own achievement-emotion situation, regardless of subject. There was no time limit for the preparation of the drawings, so the students had enough time to develop their ideas and experiences and present these in their drawings. A guiding principle in the conception of the stimuli was that the drawings should refer explicitly to classroom contexts and not to the school context in general (such as breaks). The stimuli were developed on the basis of scientific qualitative data collection criteria [82] and were as follows:

“Draw a situation from your lessons in which you had joy in learning on this school day or week.” “Draw a situation from your lessons in which you felt proud on this school day or week.”

After the drawings had been made, the students were interviewed about their drawings in order to explicate the teaching situations as well as the sequence of situations in the drawings and avoid any potential misinterpretation. The interviews took place with students individually, in a room prepared for this purpose.

The interview guide was written in child-friendly language so that the students could understand the content of the questions clearly. The interview guide was designed in line with Helfferich (2011) and contained four leading questions, a checklist, seven concrete follow-up questions, and eight maintenance questions (a total of 19 questions). All questions were—according to adequate qualitative interview research criteria—designed as open questions and closed questions were avoided. The interviews were recorded using a recording device and then converted into written form according to common scientific transcription rules [83].

3.1.3. Data Analysis Procedures

The units of analysis were defined as follows: a coding unit was a word, a context unit was a whole interview with a student, and an evaluation unit was the totality of all 28 interviews with students. First, the data from the children’s drawings and interviews were coded according to their content (open coding). Then, both types of data were transferred to MAXQDA 2024. The situations with pride and joy in learning were then inductively reconstructed [79]. This process provided an overview of all the classroom situations that had been drawn.

Subsequently, the data were analysed using one of Mayring’s (2015) content analysis procedures, namely, deductive structuring content analysis. The main aspects of the DECCS [8] were used as the basis for structuring the interviews with the children (a priori coding): triggers, interaction factors, achievement emotions, students’ appraisals, students’ emotion regulation behaviour, teachers’ emotion regulation behaviour, and consequences. ‘Adaptations’ was used as a further category to identify feedback loops (the insight-enhancing value of this analysis step was primarily abductive).

The extracted material (from the interviews) was coded according to the DECCS categories and colour-coded to assign either the emotion “pride” or the emotion “joy”. Then, the categories were systematised, assigned to the main and sub-categories of the DECCS, bundled, and transferred to a category tree.

The data were run through several times using this method to ensure intracoder reliability. In case of a deviation of more than 5% of the categories, the material was run through again to obtain an intracoder reliability index of approximately 1.0. Furthermore, the intercoder reliability was determined according to Feng (2015) and calculated for two coders according to 10% of the interview material, as adjusted Cohen’s Kappa κ_n = 0.91 [84].

3.2. Results

3.2.1. RQ1: Interactions in Achievement-Emotion Situations (Students’ Drawings and Interviews)

The results included the main categories: triggers, interaction factors, students’ appraisal, students’ achievement emotions, students’ emotion regulation behaviour, and teachers’ emotion regulation behaviour.

The triggers in the learning environment turned out to be process-related and goal-oriented triggers (see

Table 1. Results of the qualitative content analysis: triggers of pride and joy from the students’ drawings and interviews.

Main Category	Category	Sub-Category	Structure	Codes
Triggers	Learning environment	Process-related	Learning activity with teacher	− Exercise − Exercise for performance − Explanation space by the teacher
			Usage media	− Movie
			Creative activity	− Painting a book about world religions − Tinkering with an aquarium − Drawing a wolf
			Class discussion	− Seat circle − Signing up
			Games	− Hanging from bars and putting balls into bags − New maths game − Take-it-Easy game with teacher and friend − Puzzle with one-time tasks
			Objects	− Spelling book − Maths Star magazine − Picture − Booklet
		Goal-orientated	Public recognition in class	− Teacher hands over badge
			Learning activity	− Calculation tasks
			Creative activity	− Painting together with a friend
			Games	− Maths puzzle
	Performance-related situation (performance approach)	Performance-associated	Allocation of marks (getting a test back)	− German test − Maths test − Awarded a “king” in work (everything is correct) − Exercises came back
			Presentation	− In German language, together with friend
			Victory in class representative election	− Teacher stands in front and announces results

for details of all the triggers). Performance-related situations were exclusively performance-associated, which means that they were associated only with performance but were not directly related to the performance situation (e.g., getting a test back).

Concentration capacity, social factors, the physical state of the student, the student’s temperament, competence relation, subject relation, and duration/time were explored as relevant factors (see

Table 2. Results of the qualitative content analysis: relevant factors.

Main Category	Category	Sub-Category	Structure	Codes
Relevant factors	Concentration capacity	General lack of concentration		− Often not concentrating − Usually looks out the window
	Social factors	Peer interaction	Other students in class	− Classmates also present − Several classmates watching − Interaction with classmates through turning around − Finds it “cool” that classmates trust them − Six students involved − School mates did not know what was being talked about − Classmates also signed up, but not so many students
			Partner work or partner interaction	− With best friend − Key function friend: without her, she would not have dared − Friend is present − Friend is happy for her
		Individual work		− Student enjoyed working on their own − Creates more work at home
	Physical state			− Stomach ache because too little eaten (anxious before test), felt hungry
	Temperament of the child	Extroversion		− Not signing up often (except for maths)
	Competence-related	Competence level	High competence level	− Competence level 9
	Subject-related	Maths		− Often proud in maths
	Duration/time			− Drawing a wolf for the first time − Duration: one whole week

). It could be seen from the children’s drawings and interviews that social factors seem to play a crucial role regarding achievement-emotion situations with pride and joy in learning. Students emphasised the importance of their peers in these situations (see category “social factors” in Table 2).

From the content analysis of the interview data—in which the students explained the situations that they drew—it also became apparent that they evaluated each situation and appraised/assessed whether coping resources were available, or some form of coping mechanism was possible. The relevant sub-categories for the evaluation of a situation were the student’s interest, the type of task, self-efficacy (self-assessment of own abilities), effort, and importance of the task (see

Table 3. Results of the qualitative content analysis: students’ appraisals.

Main Category	Category	Sub-Category	Structure	Codes
Students’ appraisals	Evaluation of the situation	Interest	Interest in subject	− Science education − Arts at the same time as religion − Likes music lessons very much − Likes maths − Loves painting and the arts
			Interest in theme/activity	− Pleased to perform something − Wanted to be able to paint a world for a long time, now finally learned how to do it − Interest in theme − Interest in handicrafts − Really felt like talking about it − Interest in wolves
		Type of task	Severity	− Relatively simple task − Intermediate task − “so that you could do it” − “it was easy, but also not very difficult”
			Actuality of the game	− New game
		Self-efficacy (self-assessment of own abilities)		− Thinks they are good at it
		Effort	Great effort	− Worked a long time on it − Third attempt − Procured extra large pens − Great effort − A lot of discipline − Learned a lot from it − Much effort put into learning − It was a lot of work
			No effort	− Not much effort
		Importance of task		− Important test
	Coping resources available?/Coping possible?			− Finally accomplished − Because they managed it − Finally finished − Has achieved a lot

). In the interviews, the students emphasised how much effort they made in these drawn situations and that the effort was significantly related to their positive achievement emotions: “I was quite proud of it, because I think I made quite an effort, and yes, also that I had the pens for it, because I got pens especially for it.”

According to the achievement emotions in the drawn and described situations, joy was described by the children in the interviews using adjectives such as “happy”, “nice”, “very good”, “good”, “cool”, or “fun”. Pride was mentioned as “proud”, “good”, “happy”, and “such a feeling”.

Students’ emotion regulation behaviour turned out to be both adaptive and maladaptive. Maladaptive behaviour was demonstrated when a child looked at the other children, even though they were supposed to be working on a task individually. Adaptive behaviour was demonstrated by a child working independently on the tasks and seeking help from the teacher when necessary.

The teachers’ emotion regulation behaviour was systematised in the form of demonstrating emotional support, appreciation, and sanctioning. Appreciation included a teacher constantly walking through the room to give help, turn to the students, or explain something to them. Only praise was mentioned by the students’ as sanctioning, but not rebuke or negative feedback. Examples of praise included a teacher drawing a smiley face on a child’s worksheet and the comment, “Then we got our notebooks back and he [the teacher] praised me very much”. Emotional support from teachers was shown in both Figure 2 (a classroom situation with joy) and Figure 3 (a classroom situation with pride).

3.2.2. RQ2: The Role of Peer Interactions in Achievement-Emotion Situations (Children’s Drawings and Interviews)

Both the children’s drawings and the interviews showed that peers played an essential role in achievement-emotion situations, even though peers had different functions in different settings (see Table 2). In some cases, it was only mentioned that classmates were present, but, sometimes, classmates were decisive for the development of a discrete emotion (e.g., when classmates watched a child being acknowledged by the teacher) or when the other students directly acknowledged a child (e.g., when the class representative was elected). Working in pairs also played a relevant role; joy or pride was felt particularly when a neighbouring child was also happy (see Figure 4 and Figure 5). On the other hand, some students mentioned in the interviews that they were especially proud in competitive situations (when their classmates achieved, but they achieved more).

4. Study 2

Study 1 revealed that students in the fourth grade were already able to name various emotional situations and describe their own experiences. However, they reflected less on their own behaviour and their desired teacher behaviour, so it was not possible to derive information about the complex interplay of students’ appraisals, emotions, behaviour, and interactions with their teachers. At the same time, it became apparent in Study 1 that concrete situation descriptions seemed to be appropriate and corresponded to the children’s cognitive and emotional levels of development. For these two reasons, a vignette questionnaire with two concrete classroom scenarios was used in Study 2, which surveyed their emotion regulation behaviour in addition to their individual worlds of experience (i.e., triggers, emotions, appraisals) so that we could investigate the complex interplay of these variables.

4.1. Methods

4.1.1. Sample and Procedures

A cross-sectional questionnaire research design was used in Study 2 to survey students in the 4th to the 7th grades about how they reported dealing with emotionally challenging classroom situations (their appraisals, negative emotions, emotion regulation, and the behaviour they would wish for from their teachers). Between November 2022 and January 2023, 635 students (M_age = 10.80, SD = 1.32) from 4th to 7th grade in 38 primary and secondary schools were asked via a paper-and-pencil questionnaire about how they dealt with emotionally challenging classroom situations. The reported genders of the students were 51.30% girls, 47.9% boys, and 0.6% reported diverse gender. The transition phases were 44.10% in the immediate transition phase, i.e., just before or just after the transition from 4th (33.1%) to 5th (11.0%) grade, and 55.8% in the further transition phase, in the 6th (31.5%) and 7th (24.3%) grades. Students attended different school types: 33.2% attended a primary school, 23.2% a lower or middle track school (so-called “Realschule”, “Hauptschule”, and “Oberschule”), and 43.6% a higher track school (so-called “Gymnasium”). The majority (88.5%) reported Germany as their place of birth, and 10.7% reported being born outside Germany. Approval to conduct the study was obtained from the federal state education authority of Lower Saxony, Germany, and the ethics committee of the Carl von Ossietzky University of Oldenburg. The recruitment of schools and data collection took place with the help of student teachers in the context of a course at the University of Oldenburg as part of the DECCS II project. Schools were not randomly selected but were selected based on their willingness to participate in the study. The student teachers contacted various schools in Lower Saxony, Germany. At the schools, data were collected only from primary and secondary school students who had parental consent. Participation was voluntary and anonymous.

4.1.2. Measures

In order to investigate the relationships between students’ reported situation-specific appraisals, achievement emotions, emotion regulation behaviour, and desired teacher support in emotionally challenging situations, we used the DECCS questionnaire [8]. The questionnaire started with two case vignettes, one regarding a typical, emotionally challenging classroom learning situation and one depicting a typical performance situation in mathematics. After working through the vignettes, students rated their situation-specific appraisal, achievement emotions, subsequent emotion regulation behaviour, and desired teacher behaviour for each case vignette on a 5-point Likert scale. The DECCS questionnaire can be adapted to different domains; we applied it to the subject of mathematics.

Appraisals. The performance appraisal subscale consisted of three items: situation-specific beliefs about self-concept, overstress, and attributions of competence (e.g., “You feel this way because…you think you can’t do it anyway.”; α = 0.70). The learning appraisal subscale consisted of four items including situation-specific beliefs about self-concept, overstress, and attributions of competence (e.g., “You feel this way because…it overwhelms you.”; α = 0.76). High values indicated an unfavourable appraisal.

Achievement emotions. Students rated their achievement emotions based on eight emotions provided. As a significantly stronger correlation between negative emotions and teacher support has been found for primary and middle school students [85], we used only the negative emotions with the highest mean values in both the performance and the learning situation: anxiety, disappointment, anger, frustration, and shame. The mean values of these negative achievement emotions were built upon and used in the present analyses (e.g., “You are/feel…angry”; performance emotion: α = 0.73; learning emotion: α = 0.78).

Emotion regulation behaviour. According to the DECCS [8], both situations were considered together in specific emotion regulation behaviour. Two subscales were used to measure students’ adaptive emotion regulation behaviour: overall grit (3 items, e.g., “Then…you absolutely want to do well on this test anyway.”; α = 0.70) and overall problem solving (2 items, e.g., “Then…you skip the tasks and solve the others first.”; α = 0.72). Two subscales were also used to measure students’ maladaptive emotion regulation behaviour: overall avoidance (4 items, e.g., “Then…… you don’t want to write this test anyway.”; α = 0.75) and overall rebelliousness (4 items, e.g., “Then……you tear up the paper on which the test is written.”; α = 0.56). (The alpha value was relatively low. However, from the validation paper by Schlesier et al. (2023), it appeared that the items had a very good fit with α = 0.85, so we included this scale in our analyses.) Each emotion regulation subscale referred to both of the possible emotionally challenging classroom situations, i.e., the learning and the performance situations [8].

Desired teacher support. What students desired from their teacher in emotionally challenging situations (“What would you want the teacher to do at this moment?”) was measured using two subscales of the DECCS, including both situations: overall teacher support (4 items, e.g., “That she comes to me and explains the test tasks.”; α = 0.78) and overall teacher praise (2 items, e.g., “That she tells me I’m doing great.”; α = 0.86).

Covariates. Previous studies have shown that girls and boys—as well as students in both lower and higher grades—report different levels of trait appraisals and emotions [24,28,36,37]. They also tend to use different emotion regulation behaviours and are confronted with different teacher behaviours [36,49,71,85]. Accordingly, gender and grade level were included as covariates in this study.

4.1.3. Data Analysis Procedures

Descriptive statistics. Initially, descriptive statistics and bivariate correlations between all variables of interest were conducted using Mplus Version 8.8 [86].

Structural equation modelling (SEM). SEM was conducted using Mplus Version 8.8 [86,87] with MLR estimators, a method that is recommended for small and medium-sized samples and non-normally distributed data [88,89,90,91]. To test for indirect mediation effects, a confidence interval was formed around the indirect effect estimates so that the bias due to the non-normally distributed sampling distribution of indirect effects was reduced [92,93]. Model fit was estimated using four primary fit indices, as recommended by Hu and Bentler (1999): Chi-Square Test of Model Fit (χ²), Root Mean Square Error of Approximation (RMSEA; ≤0.06), Comparative Fit Index (CFI; ≥0.95), and Standardized Root Mean Square Residuals (SRMR; ≤0.05) [94].

The Little’s Missing Completely at Random (MCAR) test validated the complete randomness of the missing data. Consequently, to address missing data in all analyses, full information maximum likelihood (FIML) with robust standard errors (MLR) was utilised, as indicated by the non-significant result of the test (χ² (154) = 180.25, p > 0.05).

The percentage of missings was 1.4–5% for the items of the DECCS. The “type is complex” feature of Mplus was employed for all analytical processes in order to take into account the multilevel structure of the data (i.e., students nested in schools). Because of cluster sampling, this feature computes standard errors of parameter estimates that account for the stratification and non-independence of observations [86,95].

The structural equation model was constructed according to the CVT [21] and the interaction pattern from the DECCS questionnaire [8]. This meant that this SEM corresponded not only to the long and widely established evident CVT [19,20,48] but also to the logic of the DECCS questionnaire. For example, in the DECCS questionnaire, children were asked how they would feel in a situation that was presented to them. Subsequently, they were asked what they would do (when they reported feeling an emotion). Then, they were asked what they would want the teacher to do.

4.2. Results

4.2.1. Descriptive Statistics

Table 4. Means, standard deviations, intercorrelations and reliabilities of DECCS subscales and covariates.

	M	SD	Skewness	Kurtosis	α	ω	2	3	4	5	6	7	8	9	10	11	12
Students’ appraisal
1. Appr Perf	2.62	1.12	0.34	0.57	0.70	0.70	0.62 ***	0.56 ***	0.44 ***	0.12 *	0.06	0.27 ***	0.11 **	0.18 ***	0.04	0.14 **	0.28 ***
2. Appr Learn	2.16	1.02	0.80	0.02	0.76	0.76	–	0.42 ***	0.56 ***	0.25 ***	0.05	0.36 ***	0.21 ***	0.18 ***	0.08	0.15 **	0.23 ***
Students’ emotions
3. Emotions Perform	2.49	0.88	0.33	0.38	0.73	0.73			0.51 ***	0.00	0.15 **	0.15 **	0.06	0.24 ***	0.07	0.15 **	0.17 ***
4. Emotions Learning	2.14	0.89	0.90	0.50	0.78	0.78			–	0.10 *	0.22 ***	0.22 ***	0.13 **	0.12 **	0.11 *	0.05	0.11 ***
Students’ adaptive behaviour
5. Grit	4.53	0.47	1.96	5.04	0.70	0.73				–	0.12 **	0.28 ***	0.29 ***	0.09	0.14	0.11 *	0.07
6. Problem solving	3.76	1.57	0.75	0.62	0.72	0.82					–	0.09	0.03	0.13 ***	0.06	0.05	0.12 **
Students’ maladaptive behaviour
7. Avoidance	1.57	0.49	1.44	1.82	0.75	0.65						–	0.42 ***	0.18 ***	0.12 **	0.02	0.06
8. Rebelliousness	1.16	0.18	3.58	15.73	0.56	0.61							–	0.02	0.04	0.04	0.04
Students’ desired teacher behaviour
9. Teacher support	3.52	1.29	0.40	0.67	0.78	0.79								–	0.35 ***	0.03	0.14 ***
10. Teacher praise	3.34	1.82	0.27	1.18	0.86	0.86									–	0.31 ***	0.02
Covariates
11. Grade level	5.47	1.40	0.09	1.50	–	–										–	0.04
12. Gender	–	–	–	–	–	–											–

Notes. All measures are standardised. Appr Perf = Appraisal Performance Situation, Appr Learn = Appraisal Learning Situation, Emotions Perform = Emotions During Performance Appraisal; Emotions Learning = Emotions During Learning Appraisal; Teacher Support = Students’ Desired Teacher Support, Teacher Praise = Students’ Desired Teacher Praise; gender = 1 (girls), 2 (boys); grades = 4th, 5th, 6th, or 7th grade; α = Reliability Cronbach’s Alpha * p < 0.05, ** p < 0.01, *** p < 0.001, N = 631.

presents the descriptive statistics and bivariate correlations in the data (see Appendix A 

Table A1. Means, standard deviations, and item–total correlations of learning and performance emotions.

Emotion Variable	M	SD	rit
Performance Emotions	2.49	0.88
Anxiety	2.58	1.43	0.72 ***
Disappointment	2.56	1.33	0.69 ***
Anger	2.83	1.38	0.69 ***
Frustration	2.54	1.35	0.72 ***
Shame	1.83	1.16	0.67 ***
Learning Emotions	2.14	0.89
Anxiety	1.63	1.10	0.72 ***
Disappointment	2.30	1.31	0.76 ***
Anger	2.51	1.34	0.71 ***
Frustration	2.10	1.31	0.75 ***
Shame	2.05	1.27	0.72 ***

Note. r_it = item–total correlation; *** p ≤ 0.001.

for means, standard deviations, and item–total correlations of learning and performance emotions).

4.2.2. SEM

The indices of the calculated SEM revealed a good fit for the model (χ² (10) = 55.31, p < 0.001, CFI = 0.96, RMSEA = 0.08 (0.06−0.11), SRMR = 0.04). Figure 6 displays the results of the SEM with covariates; Figure A1 in the appendix shows the SEM without covariates (gender and grade levels).

Direct effects. Figure 6 shows all significant direct paths of the model. In detail, appraisals of the learning situation were negatively associated with overall grit (B = −0.18, β = −0.27, SE = 0.04, p < 0.001) and positively associated with overall avoidance (B = 0.21, β = 0.31, SE = 0.05, p < 0.001) and overall rebelliousness (B = 0.09, β = 0.21, SE = 0.07, p < 0.001) as well as learning emotions (B = 0.49, β = 0.53, SE = 0.04, p < 0.001). In contrast, appraisals of the performance situation were positively connected to overall avoidance (B = 0.07, β = 0.10, SE = 0.05, p < 0.05) and performance emotions (B = 0.46, β = 0.52, SE = 0.04, p < 0.001). From both the learning and the performance emotions regarding the four variables of students’ behaviour, only one direct path was found to be significant: from performance emotions to overall grit (B = 0.09, β = 0.12, SE = 0.04, p < 0.01). Furthermore, overall grit was related to both overall desired support (B = 0.21, β = 0.13, SE = 0.04, p < 0.01) and overall praise (B = 0.25, β = 0.12, SE = 0.04, p < 0.01) from the teacher. Overall problem solving was associated with both overall desired support (B = 0.11, β = 0.12, SE = 0.04, p < 0.001) and overall desired praise (B = 0.09, β = 0.09, SE = 0.03, p < 0.05) from the teacher and overall avoidance was linked to both overall desired support (B = 0.41, β = 0.25, SE = 0.04, p < 0.001) and overall desired praise (B = 0.39, β = 0.20, SE = 0.05, p < 0.001) from the teacher. In turn, overall rebelliousness was negatively connected only to overall desired praise from the teacher (B = −0.29, β = −0.09, SE = 0.04, p < 0.05).

Some gender differences were detected. Girls tended to report higher emotions in the performance appraisal (B = −0.58, β = −0.28, SE = 0.04, p < 0.001) and the learning appraisal (B = −0.45, β = −0.22, SE = 0.05, p < 0.001). Girls also used overall problem solving (B = −0.32, β = −0.12, SE = 0.04, p < 0.01) more frequently, whereas boys tended to show overall rebelliousness more frequently (B = 0.07, β = 0.09, SE = 0.04, p < 0.05). Furthermore, girls tended to desire more overall support from their teachers than boys (B = −0.26, β = −0.11, SE = 0.03, p < 0.001).

Finally, some grade-level differences were detected. Students in higher grades reported higher levels of unfavourable appraisals in performance situations (B = 0.11, β = 0.13, SE = 0.05, p < 0.001) and learning situations (B = 0.12, β = 0.14, SE = 0.05, p < 0.01), as well as more negative achievement emotions in performance situations (B = 0.11, β = 0.08, SE = 0.03, p < 0.05). In turn, students in lower grades tended to report more overall grit (B = −0.05, β = −0.08, SE = 0.04, p < 0.05), avoidance (B = −0.05, β = −0.08, SE = 0.04, p < 0.05), and rebelliousness (B = −0.05, β = −0.09, SE = 0.04, p < 0.05) as emotion regulation behaviours. They also desired overall praise from the teacher more frequently (B = −0.34, β = −0.30, SE = 0.05, p < 0.001).

Indirect effects. Only one out of eight possible indirect effects could be identified as significant. Emotions of the performance situation fully mediated the association between the appraisal of the performance situation and overall grit (B = 0.04, β = 0.06, SE = 0.02, 95% CIs [0.02, 0.11]).

Covariances. The following covariances were found to be significant: between the learning and performance appraisals (r = 0.58, p < 0.001); overall grit and overall problem solving (r = 0.15, p < 0.001); overall grit and overall avoidance (r = −0.21, p < 0.001); overall grit and overall rebelliousness (r = −0.24, p < 0.001); overall problem solving and overall avoidance (r = −0.12, p < 0.01); overall avoidance and overall rebelliousness (r = 0.36, p < 0.001); emotions of the learning and performance appraisals (r = 0.34, p < 0.001); and overall desired support and overall desired praise from the teacher (r = 0.34, p < 0.001). The covariances between overall problem solving and overall rebelliousness (r = −0.03, p > 0.05) were not significant.

5. General Discussion

The present study used a mixed-methods approach with the aim of deepening our understanding of how students perceive, deal with, and interact in achievement-emotion classroom situations at the end of primary school (fourth grade) and around the transition to secondary school (fourth to seventh grades). In other words, the first study sought to discern insights into the intricate dynamics encompassing students’ appraisals, emotions, behaviour, and interactions with their teachers. The findings from Study 1 indicated that fourth-grade students possess the capacity to identify emotional situations and articulate their personal experiences, albeit with less reflection on their own behaviour and the preferred behaviour of their teachers. In particular, Study 1 underscored the suitability of employing specific situational descriptions that are aligned with children’s cognitive and emotional developmental stages. Hence, in Study 2, we implemented a vignette questionnaire with two concrete classroom scenarios in order to probe both emotion regulation behaviours and individual emotional experiences encompassing different triggers, negative emotions, and appraisals. This methodological approach enabled a more nuanced exploration of the multifaceted interplay among these variables.

5.1. Theoretical Discussion and Practical Implications

In Study 1, as anticipated, the students drew and mentioned triggers, relevant factors, and their perceptions of pride and joy in classroom situations, as well as appraisals and emotion regulation behaviour, as part of the DECCS (Expectation RQ1). It was noticeable that the drawings and interviews showed mainly how students perceived themselves in situations where they experienced joy and pride, i.e., many different triggers, appraisals, and descriptions of emotions could be extracted from the drawings and interviews, although they showed less reflection on their own behaviour or teachers’ behaviour. Although some students were indeed able to identify their own behaviour, they described it only rarely. Accordingly, further studies would need to be conducted to determine whether or not children at that age are mature enough in terms of developmental psychology in order to reflect on their own behaviour or whether this finding was due to the method used.

Consistent with the findings of Maxwell (2015) and Romera et al. (2015) [67,68], we also found that peers play an essential role in joyful situations (Expectation RQ2). Classmates can be decisive in the development of a discrete positive emotion (e.g., when classmates watch a child being acknowledged by the teacher) or when peers directly acknowledge a child (e.g., by electing a class representative). Therefore, further studies on positive emotions in the classroom should also take into account the essential role played by peers.

Study 2 showed (in line with Hypothesis 3a, CVT, and the DECCS) that the appraisal of an emotionally challenging situation was directly associated with the reported subsequent emotion regulation behaviour for the vignette of the learning situation. In detail, an overstressed appraisal regarding the learning situation was negatively related to overall grit (but not to overall problem solving) and positively related to overall avoidance and overall rebelliousness. In contrast, we could not confirm these correlations for the performance appraisal. There was a significant correlation only between appraisal and avoidance behaviour. This indicates that students seem to differentiate very strongly between the two triggering situations, which once again underlines the importance of the context [31,32]. This may stem from the fact that students are not evaluated in learning situations, leading to different value appraisals compared to performance situations. However, value appraisals were not included in the present study, and we can furthermore not derive causal explanations from our findings. Future research should address these aspects.

With respect to the performance situation, we were able to identify—in agreement with Hypothesis 3b—a complete mediation, insofar as the negative emotions that students felt, along with their appraisal of the performance situation, disrupted the association between learning appraisal and overall grit as an emotion regulation behaviour. Specifically, when students perceived negative learning emotions, they were at the same time unable to apply overall grit as an adaptive emotion regulation behaviour when they reported an overstressed appraisal. However, beyond that—and thus in contrast to Hypothesis 3b—we could not identify any further correlational mediation effects; this was due mainly to the fact that almost all paths between emotions and regulation behaviour were not significant, except for the path between performance emotions and overall grit. This may seem surprising at first glance but is consistent with previous findings that show that during exam preparations, the emotion anxiety in particular is associated with the use of different appraisals as well as contrary adaptive and maladaptive emotion regulation strategies [76]. That is, fewer coping potential appraisals and more importance appraisals lead to high values of anxiety, which, in turn, are associated with more social support, more problem-focussing, less distancing strategies, and more drug use [76].

Hypothesis 3c was almost entirely confirmed, as all associations between emotion regulation behaviour and students’ desired teacher behaviour in the form of praise and support were significant (except for the pathway from overall rebelliousness to overall teacher support). This implies that, regardless of whether students choose adaptive or maladaptive strategies, they desire instructional support and praise from the teacher when experiencing challenging situations, which is in line with the DECCS logic. Only maladaptive behaviour (in the form of overall rebelliousness) was associated with the desire not to receive praise from the teacher. This may also be due to the fact that rebelliousness implies a kind of “attack” on the teacher, and is in line with previous research findings that rates of verbal and nonverbal praise appear to be low for students who exhibit disruptive behaviour [51,52]. In the current form of the CVT (e.g., [48]), there is also a connection between student behaviour as an outcome and the achievement setting (e.g., teacher clarity, difficulty, enthusiasm). However, within the logic of CVT, this is primarily considered as a situational trigger for the genesis of achievement emotions and is less directly connected to the outcome. When considering our results alongside both the CVT and the DECCS, it becomes evident that teacher–student interactions in emotionally challenging classroom situations should be viewed primarily as a circular process (as presented in [8] (p. 150) and slightly indicated in [48] (p. 147)).

5.2. Strengths, Limitations, and Future Directions

The present study demonstrates several strengths. In particular, we applied an innovative data collection method (children’s drawings), as well as a new instrument (DECCS), that enabled the investigation of both students’ perspectives of and the interplay between situation-specific appraisals, achievement emotions, emotion regulation behaviour, and desired teacher support. This method allowed us to show that the interplay works differently depending on the situation and the emotions that are felt, which, in turn, underlines the role of the context and allows theoretical as well as practical inferences to be drawn. The drawings and interviews (Study 1) and the case vignettes (Study 2) had the advantage that students could more easily put themselves in a specific situation and thus provide more concrete statements to describe their own perceptions and actions [8]. In addition, our study provides important findings regarding an age group (fourth to seventh grade) that has rarely been studied with respect to this interplay.

The focus on students’ perspectives has once again clearly emphasised the importance of their social interactions with peers on the one hand and teachers providing support on the other. At the same time, this study should be seen as a first attempt to intensively explore children’s lifeworld beyond studies that employ purely trait questionnaires. In particular, further studies with a longitudinal design are essential in order to map the development of how children deal with emotional classroom situations as they grow older. Furthermore, younger students should also be surveyed, i.e., pre-school age and the beginning of primary school, in order to find out when social interactions in class (with teachers and peers) begin to be important for them.

It would also be interesting to investigate whether peers play a key role in situations with negative emotion. We were not able to find this out in Study 1 as it focused on positive emotions only for ethical reasons. Although Study 2 recorded negative emotions, it did not link these with peer interactions. It is therefore crucial to expand the DECCS instrument to include a subscale on peer interaction in order to obtain insights into the role of peer interactions for students in emotionally challenging situations.

Although the data from Study 2 were cross-sectional, the use of case vignettes and the design of the DECCS questionnaire instrument allowed the temporal sequence to be mapped appropriately. Moreover, we used only self-reported data as the perceptions of the students, their experiences, and their interpretation of the classroom situations were the focus of this study. Nevertheless, reports from teachers and parents could be a good addition to such studies in future research.

Although the DECCS survey instrument has been proven to be reliable, it shows weaknesses in reliability for one subscale (rebelliousness). This means that the results regarding rebelliousness should be interpreted with caution. Nevertheless, rebelliousness is a construct that is difficult to measure and may be confounded by the effects of social desirability; furthermore, many children may not use rebelliousness as a maladaptive emotion regulation strategy in the vignette situations described in the instrument.

6. Conclusions

In these two studies, we investigated students’ perceptions and descriptions of positive achievement-emotion situations via children’s drawings and interviews. We also researched the interplay of students’ reported situation-specific appraisals, achievement emotions, emotion regulation behaviour, and responses that they desired from their teachers via a vignette questionnaire and a structural equation modelling approach.

Overall, new findings were revealed, namely, that students seem to deal better with challenging learning situations, in that relations between appraisals and emotion regulation behaviour are rather favourable, whereas they do not always seem to have the appropriate emotion regulation strategies at hand in order to deal with performance situations. For this reason, more support and promotion seem to be necessary for students in dealing with performance contexts in challenging classroom situations. Moreover, the role of peers appears to be crucial in the experience of positive emotions. Therefore, besides teacher–student interactions, student–student interactions should be promoted in order to nurture more positive emotional school experiences.