Skip to content
Teaching and Education Research

An international forum for rigorous research on teaching, learning and teacher education

Research articleOpen access

Dynamics of professionals’ daily negative emotions and assessments of the emotional climate among children in Finnish early childhood education and care

Show affiliations and correspondence
  1. aUniversity of Jyväskylä
  2. bJAMK University of Applied Sciences

DOI: https://doi.org/10.1016/j.tate.2026.105442

Outline

Abstract

This study examined dynamics between Finnish early childhood education and care (ECEC) professionals' ( N = 77) daily negative emotions and both harmonious and discordant climates in the child group using a mobile diary during five consecutive working days. Professionals' negative emotions and their assessments of the emotional climate were associated within the workday, but the previous day's interpretation of the emotional climate predicted next-day negative emotions only from Monday to Tuesday. Professionals' negative emotions alone showed day-to-day continuity. The findings highlight the need to improve professionals' awareness of the effects of their negative emotions on the child group. • Daily continuity was found in ECEC professionals' negative emotions. • Negative emotions and emotional climate assessments on the same day were related. • Emotional climate assessments predicted negative emotions from Monday to Tuesday. • Raising professionals' awareness of negative emotions and assessments is crucial.

1. Introduction

While teaching and educating young children in early childhood education and care (ECEC) classrooms,1 ECEC professionals are faced with diverse emotions on a daily basis due to the dynamic and interactive nature of their work (Frenzel et al., 2021; Jennings & Greenberg, 2009). It is essential to understand the emotions of ECEC professionals in the child group for various reasons. First, professionals' emotions have multiple effects on their own professional outcomes, such as influencing their behaviors (Frenzel, 2014; Frenzel et al., 2021), shaping teacher–- child relationships (e.g., Ansari et al., 2022; Jennings, 2015; Liu et al., 2025), contributing the emotional climate of the child group (Jennings, 2015; Shewark et al., 2018), and affecting their well-being at work (Denham et al., 2017; Jennings, 2015).

Second, research has demonstrated associations between professionals' and children's emotions (Denham & Bassett, 2019), including emotional contagion (Rodrigo-Ruiz, 2016), and the effects of children's behaviors (Denham & Bassett, 2019; Denham et al., 2017) on professionals' emotions, attributions and occupational well-being (Panthi et al., 2025). Third, ECEC professionals' emotions also influence children's outcomes, as professionals' function as models for children in emotion expression and regulation (e.g., Denham et al., 2012, 2017). Moreover, professionals' emotions can have direct transmission effects on children's emotions, cognitions, and behaviors (Frenzel et al., 2021), with implications for children's social-emotional development (Arace et al., 2021; Denham et al., 2017).

Despite growing interest in ECEC professionals' emotions and their associations with child and child group outcomes, previous research has not revealed how these emotions fluctuate or how they may be associated with professionals' assessments and decisions in their daily work related to relationships and climate among children. To address these gaps, intensive assessment strategies are essential for advancing research in this area. Accordingly, our study examines the continuity of Finnish ECEC professionals' (N = 77) day-to-day negative emotions and their assessments of the emotional climate among children, as well as the associations between professionals’ negative emotions and their assessments of the emotional climate using mobile diary data.

Studying ECEC professionals' negative emotions are particularly relevant due to their intense, attention-grabbing nature (e.g., Tugade et al., 2016), as well as their comprehensive impact on professionals' attention direction, perception of children's behavior, their own behavior within the child group, interpersonal dynamics with children, * Corresponding author. 1 The present article uses the term “child group” instead of “classroom", as in the Finnish ECEC context, the concept of “child group” is commonly employed. and both their own and the children's well-being (e.g., Frenzel et al., 2021). Furthermore, ECEC professionals have a multifaceted role in shaping peer social experiences and the emotional climate within the child group through the quality of their relationships with individual children and through daily pedagogical practices related to peer interactions (Chen et al., 2020; Wang et al., 2019).

1.1. ECEC professionals’ negative emotions and their assessment of the

emotional climate among children The work of ECEC professionals is deeply intertwined with their interactions with children that evoke a wide range of emotions including also negative emotions, such as frustration, irritation, stress, anger, and nervousness (de Ruiter et al., 2019; Jennings & Greenberg, 2009). Emotions profoundly impact our learning, functioning, and attentional focus (Conway et al., 2013; Fredrickson & Branigan, 2005; Frenzel et al., 2021). While diverse emotional experiences and negative emotions are essential aspects of our well-being (Quoidbach et al., 2014), negative emotions tend to narrow one's focus and perception, as well as temporarily impede the ability to adopt flexible patterns of thought and action, both on the individual level and when interacting with others (Tugade et al., 2016).

The emotional distress of ECEC professionals may affect their ability to provide high-quality care and education (Jennings et al., 2020). Professionals who are emotionally exhausted or overwhelmed by negative emotions are at risk of becoming less caring, less involved, more rigid, and even cynical and callous, which can lead to burnout over time (Ansari et al., 2022; Frenzel et al., 2021; Jennings & Greenberg, 2009; Musson Rose & Loomis, 2024). For example, Buettner and colleagues (2016) found that pre-school teachers' psychological load (i.e., stress, depression, and emotional exhaustion) was associated with more negative responsiveness to children and lower levels of engagement in the profession.

In turn, Jeon and colleagues (2019) discovered that early childhood teachers experiencing higher levels of personal stress perceived their children to have more anger-aggression challenges. Additionally, higher levels of job-related emotional exhaustion were positively related to their perception of children's anxiety-withdrawal challenges and negatively related to their perception of children's social competence. The emotions that professionals experience and express in the child group influence the relationships they establish with the children (Frenzel et al., 2021; Glüer & Gregoriadis, 2017; Jennings & Greenberg, 2009; Paes et al., 2023; Shewark et al., 2018). In a qualitative study by Shewark et al. (2018), pre-school teachers considered their own emotions to be key determinants of child group climate.

Expressing negative emotions in the child group can be considered contrary to the “ideal” of professional competence, that is, competent professionals should be able to suppress their negative emotions in the child group to maintain a good emotional climate (Shewark et al., 2018), whereas caring and supportive relationships between professionals and children are considered important in establishing a safe, caring, and positive emotional climate in the child group (Jennings & Greenberg, 2009; Shewark et al., 2018). In addition, close relationships with professionals, positive interactions with peers, and a caring climate play a crucial role in shaping children's present and long-term social, emotional, behavioral, and academic development (Arace et al., 2021; Chen et al., 2020; Denham & Bassett, 2019).

In previous studies, different concepts have been used to describe features of emotional climate (e.g., Chen et al., 2020; Evans et al., 2009; Harvey et al., 2012; La Paro et al., 2004) in the child group. For example, in the observational tool, the Classroom Assessment Scoring System (CLASS), one domain of classroom characteristics is emotional climate, which includes the dimensions positive climate and negative climate (La Paro et al., 2004). These dimensions examine interactions and emotional connection between the teacher and children as well as among children (La Paro et al., 2004; Pianta et al., 2008). According to Pianta et al. (2008, pp. 3, 22), positive climate is evident through warmth, respect, and enjoyment in both verbal and nonverbal interactions.

In contrast, negative climate reflects the overall degree of expressed negativity–such as anger, hostility, or aggression–in the classroom. The frequency, quality, and intensity of teacher and peer negativity are central to this dimension. Partly aligning with CLASS, in the current study, we use the terms harmonious climate and discordant climate to describe the climate in relationships among children. A harmonious climate refers to a climate that is warm-hearted, humorous, and considerate of others; a discordant climate is characterized by a restless, quarrelsome, and tense climate. We considered the terms harmonious and discordant to better capture the essence of emotional climate than the terms positive and negative.

1.2. Negative emotions and emotional climate as daily phenomena

In several theories (see e.g., the sociodynamic model developed by Mesquita and Boiger [2014] and the appraisal theories of emotions from Moors [2014]), emotions are considered dynamic phenomena, which fluctuate in the flow of everyday activities, more so than other human characteristics, such as values, personality, or attitudes (Kuppens, 2015). However, in most empirical studies, teachers' emotions are considered trait-like constructs and, therefore, teachers' habitual tendency to experience specific emotions generally is assessed through cross-sectional surveys (e.g., using teacher emotion scales [TES]; Frenzel et al., 2016) or interviews, which means that trait-reports capture only differences between teachers in their levels of emotions (Frenzel et al., 2021; Keller et al., 2014a).

Far less research has operationalized teachers' emotions as state-like constructs that emphasize emotions as brief, situation-specific momentary experiences that enable the study of both changes within the same professional and differences between professionals over time (i.e., within-person processes and between-person differences) (Keller et al., 2014a; Lischetzke & K¨onen, 2020). Then, researchers have often employed intensive longitudinal repeated measures data collection techniques (e.g., daily diaries) with relatively short time intervals between the assessment points (Bolger & Laurenceau, 2013; see e.g., empirical studies by Becker et al., 2015; Chang & Taxer, 2021; Frenzel et al., 2015; Lavy & Eshet, 2018; Wang & Buri´c, 2023) to capture the situational specificity of emotions in elementary or secondary school teachers' everyday lives.

However, despite these advances, daily assessments of ECEC professionals’ emotions remain scarce in prior research. As ECEC professionals' daily work lives are highly interactional and active, they generate various emotions, thoughts, and behaviors influenced by professionals' own attributes and competencies, as well as work-related factors, such as children's behavior, group dynamics, activities, and interactions with children, colleagues, and parents (see e.g., Acar et al., 2017). To comprehend the dynamic nature of these emotions and experiences in ECEC professionals' work, intensive research methodologies are needed, as they can produce more ecologically valid results than trait-reports (Lischetzke & K¨onen, 2020, p. 2) and minimize the risk of retrospective biases (Keller et al., 2014b).

Furthermore, such approaches offer insights into different patterns of short-term dynamics of emotions, including the range of a person's emotional states over time, the magnitude of emotional changes from one moment to another, and the extent to which an emotional state at one moment predicts the same state at the next (Houben et al., 2015) as well as potential within-person transfer between emotional states and cognitive processes (i.e., spillover, Bakker & Demerouti, 2013). To the best of our knowledge, repeated assessment studies on ECEC professionals' assessments are lacking. However, it is reasonable to assume that the short-term dynamics of ECEC professionals' assessments of the emotional climate in the child group may follow similar principles as observed in previous research regarding emotional processes (Houben et al., 2015).

1.3. Continuity in negative emotions and emotional climate and spillover

between them While emotions change and fluctuate throughout the course of daily life (Kuppens, 2015), they may also show short-term continuity, as subsequent thoughts and actions of an individual can cause negative emotional states to linger and carry over from one day to the next (Leger et al., 2018). This is due to the tendency to process negative information more thoroughly than positive, and in general, negative emotions and experiences have more impact on us than positive ones (i.e., negativity bias, Baumeister et al., 2001). Additionally, the individual might repetitively think about the past event (i.e., rumination, e.g., Leger et al., 2018) or have ineffective emotion regulation or coping strategies (Jennings et al., 2020; Liu et al., 2025).

Together, all these factors may cause the daily negative emotions of the ECEC professional to show day-to-day continuity, meaning that a professional's previous-day levels of negative emotions positively predict similar levels of negative emotions the next day making the level of daily negative emotions remain similar. It can also be assumed that professionals' assessments of the emotional climate in the child group show day-to-day continuity, as despite some daily variations, assessments generally might not fluctuate significantly for the same group of children. For example, if a professional's underlying view of the climate is harmonious, then even if they perceive the climate in the child group to be more discordant than usual on a particular day, their underlying assessment may still remain consistent.

ECEC professionals' daily negative emotions and their assessments of relationships and the climate within the child group are closely intertwined and, thereby, might influence each other in profound ways, as suggested by research on elementary and secondary school teachers' lesson-specific perceptions of student performance, motivation and discipline, and their own emotional states (Frenzel et al., 2009). Emotions guide our judgment and thinking, meaning that, in general, individuals' judgments and thoughts reflect their current moods (Clore & Huntsinger, 2007; see e.g., the affect infusion model of Forgas [1995]). Due to the narrowing effect of negative emotions, individuals are more likely to perceive threats, focus on specific details rather than the bigger picture (Tugade et al., 2016; Zadra & Clore, 2011), and possibly invalidate their accessible cognitions (Clore & Huntsinger, 2007).

In practice, ECEC professionals' negative emotions can lead to a heightened focus on negative events and conflicts in the child group, potentially coloring and biasing their interpretations of children's behavior, interactions, and the emotional climate in the child group to appear more discordant rather than recognizing moments of harmonious climate. In other words, a professional's emotional state spills over, transferring negative emotions to their assessments of the emotional climate. This spillover effect can occur during the same day, as emotions can quickly guide and impact professionals' cognitive processes, such as thought, attention, perception, and judgment (Clore & Huntsinger, 2007; Zadra & Clore, 2011).

Moreover, spillover can extend from day to day, as indicated in a study examining spillover between family affect and work-related cognition (Du et al., 2018). Misinterpretations of the emotional climate in the child group may harm professional–child interactions (see e.g., Shewark et al., 2018). Likewise, professionals' perceptions and cognitive interpretations, whether biased or not, might be seen as determinants of their daily negative emotions (Yeo & Ong, 2024; see e.g., the appraisal theories of emotions from Moors [2014]). If the professional interprets the emotional climate among children to be more harmonious than discordant, this might elevate their levels of daily positive emotions.

Conversely, assessments of a more discordant climate might lead to higher levels of daily negative emotions. Professionals' assessments of the emotional climate might spill over to their daily negative emotions both throughout the same day, as people tend to experience emotions immediately upon perceiving emotionally evocative stimuli (Zadra & Clore, 2011), but also from day to day, as found in studies examining associations between teachers' emotions and their perceptions of student behavior among elementary school teachers (de Ruiter et al., 2020). This study suggests that professionals’ past perceptions and interpretations might influence their emotions for a longer period of time than just the same day.

1.4. The present study

The aim of this study was to examine daily processes related to the negative emotional experiences of Finnish ECEC professionals and their assessments of the emotional climate among children in their child group. The focus was on negative emotions due to their potential harmful effects on professionals' well-being, perceptions, and interactions, and most importantly, on children's well-being and development (see e.g., Jennings et al., 2020). It was considered crucial to study both harmonious and discordant emotional climates because they capture different dimensions of the phenomena, even though they are related to each other. Based on previous findings on the carryover of emotions (e.g., Leger et al., 2018) and assessments, as well as the interconnections and spillover among emotions, perceptions, and interpretations (e.g., Clore & Huntsinger, 2007; de Ruiter et al., 2020; Du et al., 2018; Yeo & Ong, 2024; Zadra & Clore, 2011), our research questions (RQs) and hypotheses (Hs) were as follows: (RQ1) Do professional's negative emotions, as well as their assessments of emotional climate in the child group, exhibit daily continuity when explored with the same ECEC professional?

We formulated a tentative hypothesis (H1a) suggesting that an ECEC professional's higher (lower) levels of negative emotions on the previous day predict their higher (lower) levels of negative emotions on the following day. Additionally, we set an exploratory hypothesis that there would be daily continuity occurring in the professional's assessments of the harmonious (H1b) and discordant (H1c) climate between consecutive days. (RQ2) Does ECEC professional's negative emotions spill over to their assessments of the emotional climate in the child group during the same day, and vice versa? We set a tentative hypothesis that on the same day, negative emotions are negatively associated with a harmonious climate (H2a) and positively associated with a discordant emotional climate (H2b).

In other words, lower (higher) levels of negative emotions within an ECEC professional are expected to be associated with a more (less) harmonious climate, whereas higher (lower) levels of negative emotions are linked to a more (less) discordant climate. (RQ3) Does ECEC professional's negative emotions spill over to their assessments of the emotional climate in the child group from day to day, and vice versa? We formulated the following tentative hypotheses. We hypothesized that ECEC professional's lower (higher) levels of negative emotions on the previous day predict assessments of a more (less) harmonious climate the next day (H3a), and similarly, higher (lower) levels of negative emotions on the previous day predict assessments of a more (less) discordant climate the next day (H3b).

Furthermore, we hypothesized that assessments of a more (less) harmonious climate on the previous day predict lower (higher) levels of negative emotions the next day (H3c), while assessments of a more (less) discordant climate on the previous day predict higher (lower) levels of negative emotions the next day (H3d).

2. Methods

2.1. Participants and data collection

This study is a part of the TUIKKU – emotional skills and participation in early childhood education and care development and research project, funded by the Ministry of Education and Culture in Finland. The project aimed to advance and explore social-emotional skills and possibilities for participation in ECEC from the perspectives of professionals, children, and their parents. The participants were recruited from municipal ECEC centers in Finland involving size, geographical, and socioeconomic variation in municipalities. In total, 12 municipalities were selected, of which four were large cities (>100,000 inhabitants), four medium-sized towns (20,000–100,000 inhabitants), and four small municipalities (<20,000 inhabitants).

This resulted in 37 child groups from 24 ECEC centers in which the professionals worked. From each ECEC center, one to three child groups participated in the study. In the current study, professionals from 33 child groups in 21 ECEC centers participated. In Finland, daily work in ECEC child groups is grounded in multiprofessional teamwork, typically involving a team of three professionals with different educational backgrounds (Kumpulainen et al., 2023). In Finnish ECEC the qualifications of ECEC professionals are defined by the Act on Early Childhood Education and Care (540/2018). The professionals working in the child group have a Bachelors' or Masters' degree from a university (ECEC teacher), a degree from a university of applied sciences (social pedagogue in ECEC) or vocational school training (ECEC childcarer).

In addition to professional qualifications, the Act on ECEC also constitutes staff ratios in ECEC centers and composition of team of professionals in a child group (i.e., 2/3 of professionals working in a child group need to be either ECEC teachers or ECEC social pedagogues). Daily work at ECEC is demanding as teaching is an emotional practice that requires professionals to manage their emotions during interactions with children (Jiang et al., 2016). Despite the centrality of emotions in daily ECEC work, there is a shortage of research on the linkages between ECEC professionals’ daily emotions and emotional climate among children, both internationally and, in particular, within the Finnish context.

Our research utilizes mobile diary data collected in January 2022 with the eKoutsi mobile diary tool, a web-based application adapted for mobile devices (R¨onk¨a et al., 2015). Professionals answered the same 24 mobile diary questions using a Likert scale, assessing daily emotions, emotional agency, relationship quality, and emotional climate once a day for five consecutive working days (from Monday to Friday). This kind of fixed once-a-day reporting is suitable for recalling experiences at work that occurred during the day (see Lischetzke & K¨onen, 2020). A daily invitation to answer questions was sent to the participants via an SMS as a web link. The SMS was sent at 3 p.m., but the participants were asked to answer the questions right after the end of their work shift.

The daily web link was closed at 10 p.m., and a reminder SMS was sent 1 h before the closing time in case the participant had not yet answered. As the first question of the diary, participants were asked whether they had worked in a child group during the particular day, and if not, they were instructed not to answer the rest of the diary questions on that day. In total, 77 professionals (74 females, 3 males) responded to the diary. They provided 312 diary entries in total, representing 81.0% of the possible maximum of 385 entries. On average, participants completed about 4 out of 5 daily questionnaires (M = 4.1, SD = 1.1). The proportion of participants who completed the questionnaire varied by day: Monday 68.8%, Tuesday 85.7%, Wednesday and Thursday 84.4%, and Friday 81.8%.

The proportion of missing data ranged from 14.3% to 31.2%. The participants worked in child groups with children aged 3–6 years. The participants were on average 43.7 years old (SD = 10.8, range: 23–62). Their profession, educational background, and ECEC work experience can be found in Table 1.

2.2. Measures and variables

Participants assessed their daily negative emotions by answering the question “To what extent did you feel (emotion) during your work shift today?” on a 7-point scale (1 = not at all, 7 = extremely). The three negative emotions were angry, nervous, and a sense of inadequacy. The angry and nervous items were chosen from the PANAS-X scale (Watson & Clark, 1994), and “inadequacy” was modified from the adjective “guilty” to better reflect the ECEC context in Finland (Ahola & Valkonen, 2024). However, “angry” was excluded from further analyses as it did not correlate with the inadequacy variable on Monday and Thursday. Participants evaluated daily emotional climate (i.e., climate in the relationships among children) in their own child group with three adjectives describing a harmonious (i.e., considerate of others, warm-hearted, and humorous) and three adjectives depicting a discordant (i.e., restless, quarrelsome, and tense) climate.

Participants were asked to rate the items: “The climate among children in our child group today was (adjective)” on a 7-point scale (1 = strongly disagree, 7 = strongly agree). These items were developed by the authors for this study to measure the characteristics of the daily levels of emotional climate. The two-factor structure of emotional climate was checked via confirmatory factor analysis and found to be satisfactory (see Appendix A). A mean score for negative emotions as well as harmonious and discordant climate was computed on each day. The descriptive statistics and reliability for each scale are shown in Table 2.

2.3. Data analysis

To investigate possible daily within-person processes between ECEC professional's negative emotions and their assessments of emotional climate in one's child group, we applied random-intercept cross-lagged panel modeling (RI-CLPM; Hamaker et al., 2015). This method separates variation in negative emotions and the harmonious and discordant emotional climate into betweenand within-person variation. This allowed us to capture more accurately the processes in daily negative emotions and emotional climate. Due to the small sample size, we simplified the analyses by conducting them based on mean scores for negative emotions (NE), harmonious climate (HEC), and discordant climate (DEC) for five consecutive days.

We analyzed emotional climate in two separate RI-CLPM models: Model A focused on NE and HEC while Model B examined NE and DEC. Each RI-CLPM included two latent factors on the between-person level (BP): the ECEC professional's NEBP and either HECBP or DECBP across time. They are random intercepts in which all factor loadings on the corresponding mean score variables were fixed to 1 (Hamaker et al., 2015). These random-intercept factors capture the systematic differences (i.e., rank-order stability) between ECEC professionals on their levels of NE and HEC/DEC over the diary week. To account for the overall associations between the constructs, the random-intercept factors were allowed to correlate (i.e., NEBP and HECBP in Model A and NEBP and DECBP in Model B).

At the within-person level, we specified 10 latent factors in each model (i.e., one factor per day for each variable) based on the residual variances in the mean score variables of NE and HEC/DEC that were not captured by the random-intercept factors specified at the between-level. These within-level factors measured the extent to which an ECEC professional deviated from their personal average on a given day (Hamaker Table 1 Participants’ work experience in ECEC, profession, and educational background (N = 77). Background characteristics Work experience in ECEC <5 years 16 (20.8 %) 5–15 years 31 (40.3 %) >15 years 30 (39.0 %) Profession Director in ECEC 4 (5.2 %) Teacher or social pedagogue in ECEC 34 (44.2 %) Childcarer in ECEC 39 (50.6 %) Educational background University degree 28 (36.4 %) University of applied sciences degree 8 (10.4 %) Vocational upper secondary qualification or other 41 (53.2 %) M.

Laakso et al. et al., 2015). Each model included at the within level all autoregressive paths denoting the continuity hypotheses (i.e., H1a, H1b, and H1c) and cross-lagged paths representing the predictive spillover hypotheses (i.e., H3a, H3b, H3c, and H3d), all being predictions from one day to the next. We did not include either continuity or spillover predictions over several days (i.e., lag-2, lag-3, and lag-4 paths) in our models, as it was assumed that effects would not directly transfer over several days (meaning e.g., direct predictive path from Monday to Thursday) in the ECEC context. Finally, we estimated the covariances between the Day 1 measures and same-day correlations between the residual variances of the within-level factors in each model.

These bidirectional associations denoted our same-day spillover hypotheses (i.e., H2a and H2b). As our hypotheses focus on the within-person level, we only report them in the results section. However, inclusion of the between-level in the modeling was statistically essential to obtain valid within-level estimates (Hamaker et al., 2015). The between-level results are presented in Appendix B. The sizes of the day-to-day effects are interpreted according to the benchmark values of .03, .07, and .12 that indicate small, medium, and large effects, respectively (Orth et al., 2024). The sizes of the same-day effects are interpreted according to Cohen (1988): an effect size of 0.2 is considered small, 0.5 medium, and 0.8 large.

Both modellings (Model A and Model B) began by estimating a fully constrained RI-CLPM (Model 0 in Table 3), in which all the aforementioned paths were constrained to be equal across time. We then tested a particular hypothesis by releasing a set of these time-invariance constraints (See Models 1, 2, 3a, and 3b in Table 3) and checking the global fit of the model and changes in the fit statistics. If the release of the constraints resulted in a significant improvement in fit compared to the fully constrained model, then the model including fewer constraints was favored. Otherwise, the fully constrained Model 0 was favored. After conducting all these phases for each hypothesis, the results showed that model 3b (Table 3) fit the data best in Model A examining the associations between NE and HEC: χ2(41, N = 77) = 67.65, p < .01, comparative fit index (CFI) = .843, Tucker–Lewis index (TLI) = .828, and root-mean-square error of approximation (RMSEA) = .092.

Also, in Model B (negative emotions and discordant climate), model 3b (Table 3) fit the data best (χ2(41, N = 77) = 50.78, p = .141, CFI = . 939, TLI = .933, RMSEA = .056). The goodness-of-fit of the estimated RI-CLPMs was evaluated using the χ2 test, RMSEA, CFI, and TLI. A model was considered to be acceptable if the χ2 test was nonsignificant, RMSEA <.08, and CFI and TLI >.90. The nested RI-CLPM models were compared using the χ2 difference test (Satorra & Bentler, 2001) and the change (Δ) in the RMSEA, CFI, and TLI (Chen, 2007). The statistically nonsignificant χ2 difference test, ΔRMSEA ≤.015, ΔCFI ≤.01, and ΔTLI ≤.01 support the model with more equality constraints over time rather than the less Table 2 Items of negative emotions and emotional climate, reliability (Cronbach's α/McDonald's ω), intra-class correlations, between-level means, and betweenand withinlevel variances for the variables (N = 77).

Variables and their items Reliability α/ω Intra-class correlation Between-level mean Between-level variance Within-level variance Negative emotions .81 .34*** 2.38*** 0.58*** 1.14***

1. Nervous

2. Sense of inadequacy

Harmonious climate .86 .43*** 4.92*** 0.38*** 0.49***

1. Considerate of others

2. Warm-hearted

3. Humorous

Discordant climate .72 .34*** 2.52*** 0.46*** 0.90***

1. Restless

2. Quarrelsome

3. Tense

Note. Response scale for negative emotions from 1 (not at all) to 7 (extremely) and for emotional climate from 1 (strongly disagree) to 7 (strongly agree). ***p < .001. Table 3 Goodness-of-fit indices and their comparisons for the different specifications of the models. Models (freely estimated paths) χ2(df) p value Δχ2 p value RMSEA ΔRMSEA CFI ΔCFI TLI ΔTLI Model A (negative emotions and harmonious climate)

0. Fully constrained model 81.353 (44)*** - .105 - .780 - .775 -

1. Continuity free

(NEt-1 → NEt; HECt-1 → HECt) 81.409 (38)*** .610 .122 .017 .744 .036 .697 .078

2. Same-day free

(NEt ↔ HECt) 77.656 (41)*** .279 .108 .003 .784 .004 .763 .012 3a. Cross-lagged paths free (NEt-1 → HECt) 76.740 (41)*** .192 .106 .001 .789 .009 .769 .006 3b. Cross-lagged paths free (HECt-1 → NEt) 67.654 (41) ** .000 .092 .013 .843 .063 .828 .053 Model B (negative emotions and discordant climate)

0. Fully constrained model 61.082 (44)* - .071 - .893 - .890 -

1. Continuity free

(NEt-1 → NEt; DECt-1 → DECt) 57.611 (38)* .489 .082 .011 .877 .016 .854 .036

2. Same-day free

(NEt ↔ DECt) 56.401 (41) .198 .070 .001 .903 .010 .894 .004 3a. Cross-lagged paths free (NEt-1 → DECt) 60.116 (41)* .763 .078 .007 .880 .013 .868 .022 3b. Cross-lagged paths free (DECt-1 → NEt) 50.781 (41) .001 .056 .014 .939 .046 .933 .043 Note. The chosen final models are bolded. In each model, only one set of parameters was left free to vary across time. Each model was compared to model 0. Δχ2 p values, ΔRMSEA, ΔCFI, and ΔTLI refer to comparison of each model with model 0. NE = Negative emotions, HEC = Harmonious emotional climate, DEC = Discordant emotional climate. Final Model A: HECt-1 → NEt free. Final Model B: DECt-1 → NEt free. Δχ2 p values: Satorra-Bentler χ2 difference test p value.

RMSEA: Root Mean Standard Error of Approximation. CFI: Comparative Fit Index. TLI: Tucker-Lewis Index. Δ: Change in the χ2 value, RMSEA, CFI, and TLI. *p < .05, **p < .01, ***p < .001. constrained model. All analysis were conducted using Mplus version 8.11 (Muth´en & Muth´en, 2017, 1998–2017). As our variables were skewed, we estimated the models using a maximum likelihood estimator with robust standard errors. Owing to the missing data, the number of ECEC professionals varied according to the daily number of respondents. Missingness was handled by using the full information maximum likelihood procedure, which employs all the information in the data without imputing the missing values (Enders, 2010).

The hierarchical nature of the data by ECEC center was considered by employing the COMPLEX option in Mplus to estimate unbiased standard errors.

2.4. Ethical considerations

The study was conducted according to the ethical guidelines of the Finnish Advisory Board on Research Integrity (2019). The TUIKKU study was approved by The Human Sciences Ethics Committee at the University of Jyv¨askyl¨a (no: 1520/13.00.04.00/2021). Prior to the mobile diary data collection, informed consent was obtained from the participants, who were told about their right to withdraw from the study. Participation was voluntary, and before the mobile diary week, the participants received an instruction letter informing them of the response procedure and timeframe.

3. Results

3.1. Daily continuity in negative emotions and emotional climate

assessments First, we studied whether negative emotions and assessments of emotional climate in the child group showed continuity from day to day. Our findings revealed a statistically significant autoregression in negative emotions, which recurred across the whole study week (Fig. 1). Namely, ECEC professional's negative emotions on the previous day positively predicted their negative emotions on the next day, meaning that ECEC professionals who reported higher (lower) negative emotions relative to their own expected mean level on one day were also likely to report higher (lower) negative emotions relative to their own expected mean level the next day. The effect size for these estimates was large.

However, we did not find continuity in the assessments of the harmonious climate or in discordant climate (Appendix C). Thus, our findings supported H1a, but not H1b and H1c.

3.2. Same-day spillover between negative emotions and emotional climate

Second, we investigated whether ECEC professional's negative emotions were related to their assessments of emotional climate in the child group during the same day (i.e., same-day spillover). Our findings showed that professionals' negative emotions and their assessments of a harmonious climate were statistically significant and negatively associated with each other every day except on Monday (Fig. 1). Moreover, professionals' negative emotions and their assessments of discordant climate were statistically significant and positively associated with each other every day during the study week. Thus, the results indicate that on the days when ECEC professionals reported higher (lower) negative emotions than usual, they also assessed a less (more) harmonious climate and more (less) discordant climate in the child group than usual.

The effect size for these estimates was medium. Our findings supported the same-day spillover hypotheses H2b and partially H2a.

3.3. Day-to-day associations between negative emotions and emotional

climate Finally, we investigated day-to-day reciprocal associations (i.e., predictive spillover) between ECEC professional's negative emotions and their assessments of emotional climate in the child group. We found two statistically significant predictive associations that both occurred from Monday to Tuesday. A harmonious climate on Monday negatively predicted negative emotions on Tuesday, while a discordant climate reported on Monday positively predicted negative emotions on Tuesday (Fig. 1). ECEC professionals who assessed a more (less) harmonious climate relative to their own expected mean level on Monday were also likely to report lower (higher) negative emotions relative to their own expected mean level on Tuesday.

Furthermore, ECEC professionals who assessed a more (less) discordant climate relative to their own expected mean level on Monday were also likely to report higher (lower) negative emotions relative to their own expected mean level on Tuesday. The effect size for these estimates was large. Furthermore, no associations between ECEC professional's previous-day negative emotions and their assessments of the next day's harmonious or discordant climate in the Fig. 1. Results of same-day effects and statistically significant day-to-day effects of negative emotions and emotional climate Note. *p < .05, **p < .01 ***p < .001. The nonsignificant same-day effect is indicated with a dashed line. Statistically nonsignificant day-to-day effects are depicted in

Author contributions

Mari Laakso: Writing – original draft, Project administration, Investigation, Conceptualization. Emmi Lindroos: Writing – review & editing, Writing – original draft, Visualization, Methodology, Formal analysis. Merja Koivula: Writing – review & editing, Writing – original draft, Project administration, Funding acquisition, Conceptualization. Eija R¨aikk¨onen: Writing – review & editing, Methodology, Formal analysis, Conceptualization. Kaisa Malinen: Writing – original draft, Methodology, Investigation, Conceptualization. Eija Sev´on: Writing – review & editing, Writing – original draft, Supervision, Project administration, Funding acquisition, Conceptualization.

Funding

Finnish Ministry of Education and Culture

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The data that has been used is confidential.

References

  1. 1Acar (2017). Examining the role of teacher presence and scaffolding in preschoolers' peer interactions. European Early Childhood Education Research Journal, 25, 866. https://doi.org/10.1080/1350293X.2017.1380884
  2. 2Act on Early Childhood Education and Care 540/2018. https://www.finlex.fi/api/media/statute-foreign-language-translation/687728/mainPdf/main.pdf?timestamp=2018-07-12T21%3A00%3A00.000Z.
  3. 3Ahola (2024). Työhyvinvointi ja pitovoima muuttuvassa varhaiskasvatustyössä [Professional well-being and retention in changing early childhood work]. Journal of Early Childhood Education Research, 13, 105. https://doi.org/10.58955/jecer.146503
  4. 4Ansari (2022). Preschool teachers' emotional exhaustion in relation to classroom instruction and teacher-child interactions. Early Education & Development, 33, 107. https://doi.org/10.1080/10409289.2020.1848301
  5. 5Arace (2021). Emotional competence of early childhood educators and child socio-emotional wellbeing. International Journal of Environmental Research and Public Health, 18, 7633. https://doi.org/10.3390/ijerph18147633
  6. 6Aulén (2022). Psychological detachment as a mediator between successive days' job stress and negative affect of teachers. Frontiers in Education, 7. https://doi.org/10.3389/feduc.2022.903606
  7. 7Bakker (2013). The spillover-crossover model.
  8. 8Baumeister (2001). Bad is stronger than good. Review of General Psychology, 5, 323. https://doi.org/10.1037/1089-2680.5.4.323
  9. 9Becker (2015). Antecedents of teachers' emotions in the classroom: An intraindividual approach. Frontiers in Psychology, 6. https://doi.org/10.3389/fpsyg.2015.00635
  10. 10Blewitt (2020). Strengthening the quality of educator-child interactions in early childhood education and care settings: A conceptual model to improve mental health outcomes for preschoolers. Early Child Development and Care, 190, 991. https://doi.org/10.1080/03004430.2018.1507028
  11. 11Bolger (2013).
  12. 12Buettner (2016). Teachers' Social–Emotional Capacity: Factors Associated With Teachers' responsiveness and professional commitment. Early Education & Development, 27, 1018. https://doi.org/10.1080/10409289.2016.1168227
  13. 13Chang (2021). Teacher emotion regulation strategies in response to classroom misbehavior. Teachers and Teaching, 27, 353. https://doi.org/10.1080/13540602.2020.1740198
  14. 14Chen (2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 14, 464. https://doi.org/10.1080/10705510701301834
  15. 15Chen (2020). Influences of teacher–child relationships and classroom social management on child-perceived peer social experiences during early school years. Frontiers in Psychology, 11. https://doi.org/10.3389/fpsyg.2020.586991
  16. 16Clore (2007). How emotions inform judgment and regulate thought. Trends in Cognitive Sciences, 11, 393. https://doi.org/10.1016/j.tics.2007.08.005
  17. 17Cohen (1988).
  18. 18Conway (2013). The broaden-and-build theory of positive emotions: Form, function, and mechanisms.
  19. 19Dang (2018). Teacher perceptions: The importance of being a reflective practitioner.
  20. 20de Ruiter (2020). Teachers' emotional experiences in response to daily events with individual students varying in perceived past disruptive behavior. Journal of School Psychology, 82, 85. https://doi.org/10.1016/j.jsp.2020.08.005
  21. 21de Ruiter (2019). Relevant classroom events for teachers: A study of student characteristics, student behaviors, and associated teacher emotions. Teaching and Teacher Education, 86. https://doi.org/10.1016/j.tate.2019.102899
  22. 22Denham (2023).
  23. 23Denham (2019). Early childhood teachers' socialization of children's emotional competence. Journal of Research in Innovative Teaching & Learning, 12, 133. https://doi.org/10.1108/JRIT-01-2019-0007
  24. 24Denham (2017). Early childhood teachers' socialization of emotion: Contextual and individual contributors. Child and Youth Care Forum, 46, 805. https://doi.org/10.1007/s10566-017-9409-y
  25. 25Denham (2012). Early childhood teachers as socializers of young children's emotional competence. Early Childhood Education Journal, 40, 137. https://doi.org/10.1007/s10643-012-0504-2
  26. 26Du (2018). Daily spillover from family to work: A test of the work–home resources model. Journal of Occupational Health Psychology, 23, 237. https://doi.org/10.1037/ocp0000073
  27. 27Enders (2010).
  28. 28Evans (2009). Differentiating classroom climate concepts: Academic, management, and emotional environments. Kōtuitui: New Zealand Journal of Social Sciences Online, 4, 131.
  29. 29Forgas (1995). Mood and judgment: The affect infusion model (AIM). Psychological Bulletin, 117, 39. https://doi.org/10.1037/0033-2909.117.1.39
  30. 30Fredrickson (2005). Positive emotions broaden the scope of attention and thought‐action repertoires. Cognition & Emotion, 19, 313. https://doi.org/10.1080/02699930441000238
  31. 31Frenzel (2014). Teacher emotions.
  32. 32Frenzel (2015). Teaching this class drives Me nuts! - Examining the person and context specificity of teacher emotions. PLoS One, 10. https://doi.org/10.1371/journal.pone.0129630
  33. 33Frenzel (2021). Teacher emotions in the classroom and their implications for students. Educational Psychologist, 56, 250. https://doi.org/10.1080/00461520.2021.1985501
  34. 34Frenzel (2009). Antecedents and effects of teachers' emotional experiences: An integrated perspective and empirical test.
  35. 35Frenzel (2016). Measuring teachers' enjoyment, anger, and anxiety: The teacher emotions scales (TES). Contemporary Educational Psychology, 46, 148. https://doi.org/10.1016/j.cedpsych.2016.05.003
  36. 36Glüer (2017). Quality of teacher–child relationship and preschoolers' pro-social behaviour in German kindergartens. Education 3-13, 45, 558. https://doi.org/10.1080/03004279.2016.1140802
  37. 37Hamaker (2015). A critique of the cross-lagged panel model. Psychological Methods, 20, 102. https://doi.org/10.1037/a0038889
  38. 38Harvey (2012). Mapping the classroom emotional environment. Teaching and Teacher Education, 28, 628. https://doi.org/10.1016/j.tate.2012.01.005
  39. 39Houben (2015). The relation between short-term emotion dynamics and psychological well-being: A meta-analysis. Psychological Bulletin, 141, 901. https://doi.org/10.1037/a0038822
  40. 40Hughes (2016). Teacher–student relationship and peer disliking and liking across grades 1–4. Child Development, 87, 593. https://doi.org/10.1111/cdev.12477
  41. 41Jennings (2015). Early childhood teachers' well-being, mindfulness, and self-compassion in relation to classroom quality and attitudes towards challenging students. Mindfulness, 6, 732. https://doi.org/10.1007/s12671-014-0312-4
  42. 42Jennings (2009). The prosocial classroom: Teacher social and emotional competence in relation to student and classroom outcomes. Review of Educational Research, 79, 491. https://doi.org/10.3102/0034654308325693
  43. 43Jennings (2020). Introduction to the special issue on early care and education professionals' social and emotional well-being. Early Education & Development, 31, 933. https://doi.org/10.1080/10409289.2020.1809895
  44. 44Jeon (2019). Early childhood teachers' stress and children's social, emotional, and behavioral functioning. Journal of Applied Developmental Psychology, 61, 21. https://doi.org/10.1016/j.appdev.2018.02.002
  45. 45Jiang (2016). Teachers' emotions and emotion regulation strategies: Self- and students' perceptions. Teaching and Teacher Education, 54, 22. https://doi.org/10.1016/j.tate.2015.11.008
  46. 46Keller (2014). Exploring teacher emotions: A literature review and an experience sampling study.
  47. 47Keller (2014). Teachers’ emotional experiences and exhaustion as predictors of emotional labor in the classroom: An experience sampling study. Frontiers in Psychology, 5. https://doi.org/10.3389/fpsyg.2014.01442
  48. 48Kumpulainen (2023). Occupational well-being and teamwork in Finnish early childhood education. Journal of Early Childhood Education Research, 12, 71. https://doi.org/10.58955/jecer.v12i2.119784
  49. 49Kuppens (2015). It's about time: A special section on affect dynamics. Emotion Review, 7, 297. https://doi.org/10.1177/1754073915590947
  50. 50La Paro (2004). The classroom assessment scoring system: Findings from the prekindergarten year. The Elementary School Journal, 104, 409. https://doi.org/10.1086/499760
  51. 51Lavy (2018). Spiral effects of teachers' emotions and emotion regulation strategies: Evidence from a daily diary study. Teaching and Teacher Education, 73, 151. https://doi.org/10.1016/j.tate.2018.04.001
  52. 52Leger (2018). Let it go: Lingering negative affect in response to daily stressors is associated with physical health years later. Psychological Science, 29, 1283. https://doi.org/10.1177/0956797618763097
  53. 53Li (2017).
  54. 54Lischetzke (2020). Daily diary methodology.
  55. 55Liu (2025). Teacher–student relationships and early childhood teachers' well-being in China: The mediating effect of emotion regulation. Current Psychology, 44, 6811. https://doi.org/10.1007/s12144-025-07648-8
  56. 56Mesquita (2014). Emotions in context: A sociodynamic model of emotions. Emotion Review, 6, 298. https://doi.org/10.1177/1754073914534480
  57. 57Moors (2014). Flavors of appraisal theories of emotion. Emotion Review, 6, 303. https://doi.org/10.1177/1754073914534477
  58. 58Musson Rose (2024). Teacher burnout, student-teacher relational closeness, and the moderating effect of work stress focused supervision. Early Childhood Education Journal.
  59. 59Muthén (2017).
  60. 60Orth (2024). Effect size guidelines for cross-lagged effects. Psychological Methods, 29, 421. https://doi.org/10.1037/met0000499
  61. 61Paes (2023). The relations between teacher-child relationships in preschool and children's outcomes in kindergarten. Journal of Applied Developmental Psychology, 86. https://doi.org/10.1016/j.appdev.2023.101534
  62. 62Panthi (2025). The challenges of challenging behaviour: Early childhood educators' understandings of child behaviour and impact on occupational wellbeing. Teachers and Teaching, 1.
  63. 63Pianta (2008).
  64. 64Quoidbach (2014). Emodiversity and the emotional ecosystem. Journal of Experimental Psychology: General, 143, 2057. https://doi.org/10.1037/a0038025
  65. 65Rodrigo-Ruiz (2016). Effect of teachers' emotions on their students: Some evidence. Journal of Education and Social Policy, 3, 73.
  66. 66Rönkä (2015). A mobile-assisted working model for supporting daily family life: A pilot study. The Family Journal, 23, 180. https://doi.org/10.1177/1066480714565333
  67. 67Satorra (2001). A scaled difference chi-square test statistic for moment structure analysis. Psychometrika, 66, 507. https://doi.org/10.1007/BF02296192
  68. 68Shewark (2018). Teachers' perspectives on the consequences of managing classroom climate. Child and Youth Care Forum, 47, 787. https://doi.org/10.1007/s10566-018-9461-2
  69. 69Taxer (2015). Facets of teachers' emotional lives: A quantitative investigation of teachers' genuine, faked, and hidden emotions. Teaching and Teacher Education, 49, 78. https://doi.org/10.1016/j.tate.2015.03.003
  70. 70Tugade (2016). Positive emotions.
  71. 71Türktorun (2020). Psychological detachment and work-related rumination in teachers: A systematic review. Educational Research Review, 31. https://doi.org/10.1016/j.edurev.2020.100354
  72. 72Wang (2023). A diary investigation of teachers' emotional labor for negative emotions: Its associations with perceived student disengagement and emotional exhaustion. Teaching and Teacher Education, 127. https://doi.org/10.1016/j.tate.2023.104117
  73. 73Wang (2019). Do you want to play with me today? Friendship stability among preschool children. European Early Childhood Education Research Journal, 27, 170. https://doi.org/10.1080/1350293X.2019.1579545
  74. 74Watson (1994).
  75. 75Yeo (2024). Associations between cognitive appraisals and emotions: A meta-analytic review. Psychological Bulletin, 150, 1440. https://doi.org/10.1037/bul0000452
  76. 76Zadra (2011). Emotion and perception: The role of affective information. WIREs Cognitive Science, 2, 676. https://doi.org/10.1002/wcs.147