Few theoretical frameworks specifically address interprofessional collaboration within an educational context (Glavin and Erdal, 2018). One model that does account for this context is the Perceptions of Interprofessional Collaboration Model (PINCOM) (Ødegård, 2006). The PINCOM model systematically organizes key aspects that characterize perceptions of interprofessional collaboration. Rather than serving as a complementary model, PINCOM provides a foundation for a holistic understanding of interprofessional collaboration within educational settings. The development of this model has been instrumental in shaping the examination of interprofessional collaboration in the EPS setting. According to the PINCOM model, interprofessional collaboration can be reflected in three levels: individual, group, and organizational. Each level comprises specific factors influencing collaborative processes (Ødegård, 2006). These factors were selected based on a review of their theoretical and empirical significance within the interprofessional collaboration literature, as well as findings from a pilot study (Ødegård, 2016).

At the individual level, PINCOM highlights personal attributes and professional competencies that impact collaboration. This includes professionals’ role perceptions, professional power, personality traits, and motivation, all of which shape their engagement in interprofessional collaboration. At this level, trust and professional identity are critical in determining how individuals interact within collaborative frameworks. At the group level, the model emphasizes the dynamics of interprofessional collaboration, including the quality of relationships, group leadership, social support, and communication among professionals. At this level, the significance of shared goals, mutual respect, and clearly defined roles in fostering effective collaboration are emphasized. At the organizational level, the model addresses structural and systemic factors that either facilitate or hinder interprofessional collaboration. Elements such as organizational culture, institutional goals, professional domains, and environmental conditions are predicted to significantly influence collaborative practices. Additionally, policies, leadership support, resource allocation, and formalized collaborative structures play an essential role in shaping the effectiveness of interprofessional collaboration across professional domains.

By integrating these three levels, the PINCOM model provides a framework for understanding the complexities of interprofessional collaboration in educational settings. However, as the model is not exhaustive, further exploration of additional factors influencing interprofessional collaboration is necessary. Ødegård (2006) model offers insight into the perceived complexity of interprofessional collaboration. Still, the aspects involved in interprofessional collaboration processes are extensive, and the PINCOM model only highlights some of them (Ødegård, 2006). Therefore, it is of interest to explore how other relevant factors, such as knowledge about collaboration and affective commitment, are related to EPS counselors’ evaluations of interprofessional collaboration. This study contributes to the field by incorporating the perspectives of EPS counselors and examining factors such as knowledge about collaboration, perceived trust, affective commitment, time pressure, positive and negative interprofessional collaboration, job satisfaction, and intention to leave.

In this study, we operate with three sets of variables (predictors, mediators, and outcomes) and eight latent variables that are defined within them. The EPS counselor’s knowledge about collaboration, perceived trust, affective commitment, and time pressure are predictors. Positive and negative interprofessional collaboration are mediators, while job satisfaction and intention to leave are outcome variables. In the following, we will review previous research on the associations between these factors. To our knowledge, there is limited research on EPS counselors specifically focusing on the associations between these factors. Therefore, we have drawn on studies from other professions, such as teachers and nurses.

Building on the reviewed literature, a hypothesized model was developed to examine the relationships between the presented variables in the context of EPS counselors. Although some of these associations are derived from studies of other professional groups (e.g., teachers and nurses), they can still provide a relevant foundation for investigating these relationships among EPS counselors, as these professions operate within a human service context and may exhibit similar patterns in job satisfaction and intentions to leave. Prior research has demonstrated that knowledge about collaboration (Nooteboom et al., 2021), perceived trust (Wei et al., 2022; Ødegård, 2006), affective commitment (Elstad et al., 2016; Ndibu et al., 2019), and time pressure (San Martín-Rodríguez et al., 2005) all influence interprofessional collaboration. Consequently, we hypothesize that these factors will shape how EPS counselors evaluate interprofessional collaboration. We hypothesize that knowledge, trust, and affective commitment will be positively associated with positive evaluations of interprofessional collaboration and negatively associated with negative evaluations of interprofessional collaboration. Conversely, we hypothesize that time pressure will be negatively associated with positive evaluations of interprofessional collaboration and positively associated with negative evaluations of interprofessional collaboration. Additionally, we expect a positive association between knowledge about collaboration, perceived trust, and affective commitment, while time pressure is expected to be negatively correlated with these predictors. Given that interprofessional collaboration is considered a core component in the work of EPS counselors (e.g., Segal et al., 2017), we hypothesize that positive evaluations of interprofessional collaboration are positively associated with job satisfaction and negatively with intention to leave. In contrast, we expect negative evaluations of interprofessional collaboration to be negatively associated with job satisfaction and positively associated with the intention to leave. Finally, we hypothesize that job satisfaction is negatively associated with the intention to leave (Skaalvik and Skaalvik, 2011) and that there is a negative relationship between positive and negative evaluations of interprofessional collaboration, indicating that positive collaboration experiences reduce negative evaluations and vice versa (Figure 1).

This study was conducted in Norway, where all municipalities and county authorities are legally required to establish an Educational Psychological Service (EPS). This population-based study invited all EPS offices and employees working with children of compulsory school age or younger to participate. The organization of EPS in Norway varies, as no standardized national guidelines dictate how the service should be structured or staffed (Moen et al., 2018). Our final sample comprised 637 EPS counselors (92.9% female, 7.1% male) from 151 EPS offices. The average age of the participants was 44.6 years, and they had an average tenure of 8.5 years within the EPS.

In Norway, approximately 2,200 EPS counselors work with children of compulsory school age. Thus, this study captured responses from one-third of all EPS counselors nationwide. The study achieved broad geographic representation, with all 11 counties in Norway included, and participation rates across counties ranging from 52 to 93%.

Data was collected through a digital questionnaire consisting of 223 items. The invitation to participate was initially emailed to EPS leaders, who then distributed the survey link to the EPS counselors within their offices. Three email reminders were sent to EPS leaders to maximize response rates, followed by a final round of phone calls to offices that had not yet responded. Data collection was concluded after these follow-up efforts. Participation in the study was voluntary, and the Norwegian Agency for Shared Services in Education and Research reviewed and approved all procedures related to data collection, storage, privacy, and confidentiality.

The survey scales were primarily derived from previously validated scales. Some measurements were adapted to align with the EPS context, while others were newly developed for this study.

We conducted data exploration and preliminary analyses using SPSS, followed by structural equation modeling (SEM) with Mplus 8.5 for more complex model testing (Muthén and Muthén, 2012). The dimensionality of the constructs was assessed through confirmatory factor analysis (CFA). To evaluate the overall model fit, we used chi-square statistics along with additional fit indices provided by Mplus 8.5 (Muthén and Muthén, 2012), specifically: Comparative Fit Index (CFI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR). To evaluate the goodness of fit, and taking the model complexity into account, the following cut-off criteria were used to evaluate a good model fit after reviewing the literature: CFI ≥ 0.95, RMSEA ≤ 0.05, and SRMR ≤0.06 (Hu and Bentler, 1999; Marsh et al., 2004; Schermelleh-Engel et al., 2003).

As an initial step, we examined the psychometric properties of all variables. Descriptive statistics, including correlations and reliability estimates, were analyzed using SPSS (see Table 1). A basic measurement model was tested using confirmatory factor analysis (CFA) with all latent variables: knowledge about collaboration, perceived trust, affective commitment, time pressure, positive and negative interprofessional collaboration, job satisfaction, and intention to leave. The model properly fit the data: χ² (920) = 2067, p < 0.001; RMSEA = 0.044 [90% CI = 0.042–0.047]; CFI = 0.92; SRMR = 0.049. Factor loadings for most items ranged from 0.54 to 0.95, indicating adequate measurement properties. However, three items—item #7 (0.47), and item #12 (0.49) from the trust scale, and item #23 (0.47) from the interprofessional collaboration with kindergarten and schools scale—exhibited lower factor loadings. After evaluating each scale and the overall model, these items were maintained in their scales despite their low loadings.

Beyond assessing the psychometric properties, we examined potential multi-level structures in the data. Some factors exhibited intraclass correlations (ICC) greater than 0.05 at the EPS office level, suggesting a proportion of variance attributable to office-level clustering. Specifically, the following factors had ICC values exceeding 0.10: positive interprofessional collaboration with health services (0.13), positive interprofessional collaboration with child welfare services (0.13), negative interprofessional collaboration with child welfare services (0.23), job satisfaction (0.17), intention to leave (0.14) and time pressure (0.20). These findings indicate that EPS office level explained a significant proportion of the variance. Ideally, a multi-level analysis should be carried out (Heck and Thomas, 2020). However, our data included 102 clusters (EPS offices) with fewer than five counselors, which is below the recommended minimum cluster size for obtaining reliable parameter estimates in multi-level analysis (Clarke, 2008). Additionally, some rural EPS offices had only one EPS counselor, further complicating the feasibility of a multi-level approach. After scrutinizing the literature on this specific analytic challenge, we decided to use robust estimation methods to mitigate potential biases. Specifically, we employed Maximum Likelihood with Robust Standard Errors and Chi-Square correction (MLR, MLF) in Mplus to ensure unbiased parameter estimates at the individual level (Abadie et al., 2023; Clarke, 2008).

With the hypothesized model as a point of departure, we specified three structural models, one for each of the EPS collaboration partners: (1) kindergarten and schools, (2) child welfare services, and (3) health services. It should be noted that the only variation across these models was in the evaluation of interprofessional collaboration with the respective partners, while all other variables remained consistent. First, we tested the model for interprofessional collaboration with kindergartens and schools (model 1). Consequently, we regressed the two dependent variables (job satisfaction and intention to leave) on the independent variables (knowledge about collaboration, trust, affective commitment, and time pressure) and the intermediate variables (positive and negative evaluations of interprofessional collaboration with kindergarten and schools). In addition, we regressed the intermediate variables on the independent variables. Model 1 revealed a good fit to the data: χ²[563] = 1,407, p < 0.01; CFI = 0.92; RMSEA = 0.049, 90% CI [0.045–0.052]; SRMR = 0.055 (Figure 2; Table 2).

In this model, knowledge about collaboration was significantly associated with job satisfaction (β = 0.11, p < 0.01), positive interprofessional collaboration (β = 0.14, p < 0.01), and negative interprofessional collaboration (β = −0.12, p < 0.01). However, its association with intention to leave was not significant (β = 0.06). Perceived trust was significantly associated with job satisfaction (β = 0.17, p < 0.01), positive interprofessional collaboration (β = 0.23, p < 0.001), and negative interprofessional collaboration (β = −0.22, p < 0.001), but not with intention to leave (β = −0.04). Affective commitment was significantly associated with job satisfaction (β = 0.51, p < 0.001) and intention to leave (β = −0.35, p < 0.001), but it was not significantly associated with either positive (β = 0.02) or negative (β = −0.03) interprofessional collaboration with kindergartens and schools. Time pressure was negatively associated with job satisfaction (β = −0.23, p < 0.001) and positive interprofessional collaboration (β = −0.15, p < 0.01). In contrast, time pressure was positively associated with both negative interprofessional collaboration (β = 0.17, p < 0.001) and intention to leave (β = 0.26, p < 0.001). The paths from positive interprofessional collaboration to job satisfaction (β = 0.10) and intention to leave (β = 0.02) were not significant. Similarly, negative interprofessional collaboration was not significantly associated with job satisfaction (β = −0.04), while it was significantly associated with intention to leave (β = 0.16, p < 0.05). Additionally, two significant indirect paths were identified: (1) perceived trust influenced intention to leave indirectly via negative interprofessional collaboration, suggesting that negative collaboration mediates the effect of perceived trust on intention to leave and (2) time pressure also influenced intention to leave indirectly through negative interprofessional collaboration. This indicates that higher time pressure contributes to more negative collaboration experiences, which in turn increases the likelihood of considering leaving the profession. Altogether, the dependent variables (inclusive intermediate) explained 49% of the variance in job satisfaction, 23% in intention to leave, 13% in positive interprofessional collaboration, and 12% in negative interprofessional collaboration.

Overall, EPS counselors with higher levels of knowledge about collaboration, greater perceived trust, and lower perceived time pressure tended to evaluate interprofessional collaboration more positively. Affective commitment was not significantly associated with either positive or negative interprofessional collaboration, yet it emerged as the strongest predictor of both job satisfaction and intention to leave. Moreover, time pressure was a strong negative predictor of job satisfaction and a strong positive predictor of intention to leave, emphasizing its impact on EPS counselors’ professional well-being. The results also indicate a weak but significant relationship between negative interprofessional collaboration and intention to leave, suggesting that poor collaboration experiences may contribute to intention to leave. Following the initial model testing for interprofessional collaboration with kindergartens and schools (Model 1), we extended our analysis by testing a second model focused on child welfare services (Model 2). This model maintained the same structure as Model 1, with the only changes being the two interprofessional collaboration variables tailored to child welfare services. Model 2 also demonstrated a good fit to the data (χ² (563) = 1,377, p < 0.01; CFI = 0.93; RMSEA = 0.048, 90% CI [0.044–0.051]; SRMR = 0.054). Similarly, we tested Model 3, which assessed interprofessional collaboration with health services, adjusting the collaboration variables accordingly. This model also fit the data well (χ² (563) = 1,377, p < 0.01; CFI = 0.93; RMSEA = 0.048, 90% CI [0.044–0.051]; SRMR = 0.054). A comparative summary of Models 1, 2, and 3 is presented in Table 3. A visual representation of models 2 and 3 can be found in Appendices B, C.

When we compared the paths in the kindergarten and school model (Model 1) with those in the child welfare services (Model 2) and health services (Model 3) models, some differences were observed. Notably, some significant findings in the kindergarten and school model did not emerge in the child welfare and health services models, which may be attributed to the closer collaboration between kindergartens and schools compared to the other two services. Knowledge about collaboration and perceived trust were significantly associated with both positive and negative interprofessional collaboration in the kindergarten and school model. However, in the health services model, knowledge about collaboration was only significantly associated with positive interprofessional collaboration, and perceived trust showed no significant associations in either the child welfare or health services models. Affective commitment was significantly associated with positive interprofessional collaboration only in the child welfare services model but had no significant associations in the kindergarten and school or health services models.

A consistent finding across all three models was the significant association between time pressure and negative interprofessional collaboration, highlighting that time constraints negatively influence EPS counselors’ evaluations of interprofessional collaboration regardless of the service context. Additionally, time pressure was significantly associated with positive interprofessional collaboration in the kindergarten, school, and child welfare services models, indicating that it may be essential in shaping interprofessional collaborations. Only two significant associations were identified regarding job satisfaction and intention to leave: positive interprofessional collaboration was significantly related to job satisfaction in the health services model, while negative interprofessional collaboration was significantly associated with intention to leave in the kindergarten and school model. Beyond these findings, no strong associations were observed between interprofessional collaboration, job satisfaction, and intention to leave among EPS counselors.

Among the predictors examined, perceived trust was the strongest predictor of interprofessional collaboration between EPS counselors and kindergartens and schools. The most plausible explanation for this finding is the operationalization of the trust scale. The scale specifically captured EPS counselors’ perceptions of trust from kindergartens and schools, excluding child welfare and health services. Since respondents were asked whether professionals in kindergartens and schools trusted them in various work-related tasks, it is unsurprising that perceived trust emerged as the strongest predictor. This finding aligns with previous research demonstrating that trust is a critical factor in interprofessional collaboration (e.g., Wei et al., 2022; Ødegård, 2006) and is considered a key element in the development of collaborative practice (San Martín-Rodríguez et al., 2005). The findings might have differed if the trust scale was operationalized to include child welfare and health services. Consequently, future research should explore whether perceived trust holds the same significance for interprofessional collaboration between EPS counselors and other services.

Despite its importance, establishing trust in collaboration with schools has been reported as a challenge for EPS counselors (Kolnes and Midthassel, 2022). Our findings reinforce that trust particularly influences EPS counselors’ collaboration with kindergartens and schools. However, this contrasts with Elstad et al. (2016), who found that trust was not significantly associated with collaboration among teachers, suggesting that the role of trust may differ across professional groups and contexts.

Time pressure was significantly associated with interprofessional collaboration across all three collaboration partners—kindergartens and schools, child welfare services, and health services. This finding underscores the crucial role of available time and space for engaging in interprofessional collaboration, a factor previously identified as essential for successful interprofessional practices (San Martín-Rodríguez et al., 2005). Conversely, a lack of time has consistently been reported as a significant barrier to interprofessional collaboration (Villarreal, 2018). Similarly, Moen et al. (2018) identified time constraints and limited external professional resources as primary challenges in collaboration between EPS counselors and schools. The results of this study reinforce these findings, demonstrating that time pressure negatively impacts EPS counselors’ evaluations of interprofessional collaboration. Specifically, EPS counselors report more positive evaluations of collaboration when they perceive lower time pressure. This aligns with previous research emphasizing the importance of sufficient time availability in facilitating interprofessional collaboration (e.g., San Martín-Rodríguez et al., 2005; Villarreal, 2018).

EPS counselors’ self-assessed knowledge about collaboration was significantly associated with their interprofessional collaboration with kindergartens and schools, and health services but not with child welfare services. One possible explanation for this may be the frequency of collaboration. If EPS counselors engage less frequently with child welfare services, they may have fewer opportunities to apply their collaboration knowledge, making gaps in their expertise less noticeable. Additionally, interprofessional collaboration with child welfare services received the lowest ratings by EPS counselors, suggesting that systemic or structural barriers may hinder the application of collaboration knowledge in this context. While Friend and Barron (2024) emphasize that research on collaboration knowledge and skills remains limited, previous studies have established a link between knowledge about collaboration and interprofessional collaboration. For example, Nooteboom et al. (2021) found that knowledge and training in interprofessional collaboration act as key facilitators, while Roberts (2003) found that prior training in collaborative practices predicted the quality of collaboration. The findings of this study align with previous research, reinforcing the idea that EPS counselors’ knowledge about collaboration influences their evaluation of interprofessional collaboration. Given the scarcity of research on collaboration knowledge and skills (Friend and Barron, 2024), this study makes an important contribution by emphasizing the role of knowledge and training in fostering effective interprofessional collaboration between EPS counselors and their collaboration partners.

In this study, EPS counselors’ affective commitment was not significantly associated with interprofessional collaboration among EPS counselors. This result somewhat contradicts previous research, which has found positive associations between affective commitment and interprofessional collaboration. For instance, Ndibu et al. (2019) reported a significant relationship between these factors among mental health workers, while Elstad et al. (2016) found that affective commitment had a direct positive effect on collaboration among teachers. A possible explanation for this discrepancy is that interprofessional collaboration in the current study involves external professionals from services outside the EPS. This distinction may influence the relevance of affective commitment to the workplace, as EPS counselors primarily collaborate with professionals from other services rather than within their organization. Consequently, their commitment to the EPS may play a lesser role in shaping collaborative practices with external stakeholders.

Since interprofessional collaboration is a core component of EPS counselors’ work (e.g., Segal et al., 2017), we hypothesized that it would be associated with EPS counselors’ job satisfaction and intention to leave. However, this hypothesis was only partially supported. The results indicate that interprofessional collaboration is not strongly associated with job satisfaction or intention to leave, with two notable exceptions: (1) positive interprofessional collaboration was significantly associated with job satisfaction in the health service model, and (2) negative interprofessional collaboration was significantly associated with intention to leave in the kindergarten and school model. These findings partially contradict the study by Young et al. (2021), which found that EPS counselors identified effective collaboration as a key factor contributing to their job satisfaction. The results from the health service model align with this perspective, as EPS counselors may perceive collaboration with health services as more structured, specialized, and impactful, making them more satisfied when their collaboration with health services is successful. In contrast, negative interprofessional collaboration with kindergartens and schools appears to have a stronger influence on intention to leave. This may be because kindergartens and schools are EPS counselors’ closest and most frequent collaboration partners, meaning that persistent collaboration challenges in these settings could lead to more frustration and increased intention to leave.

The EPS counselors in this study reported high levels of job satisfaction, with an average score of 8.37 on a 10-point scale. This aligns with findings Young et al. (2021), who also found that most EPS counselors were satisfied with their work. The most dominant predictor of job satisfaction—by a considerable margin—was EPS counselors’ affective commitment to the EPS as an organization. Affective commitment, which refers to an employee’s emotional attachment to, identification with, and involvement in their organization (Allen and Meyer, 1990), was the most crucial factor in determining job satisfaction in this study. This finding reinforces previous research demonstrating robust links between affective commitment, job satisfaction, intention to leave, and job performance (Caricati et al., 2015; Gün et al., 2021; Ndibu et al., 2019). Notably, Galletta et al. (2016) found that higher levels of job satisfaction were associated with stronger affective commitment to the team, and this association was more potent when collective perceptions of positive collaboration were higher. The results of this study provide further support for this relationship, emphasizing that affective commitment is not just a contributing factor but perhaps the most influential driver of job satisfaction among EPS counselors. Additionally, the results revealed low levels of intention to leave among EPS counselors, with an average score of 2.51 on a 10-point scale. This contrasts with findings from Young et al. (2021), where 47% of EPS counselors reported that they were considering leaving the profession. This discrepancy may reflect differences in sample characteristics, organizational structures, or contextual factors influencing job retention in different studies.

Moreover, time pressure emerged as a significant predictor of job satisfaction among EPS counselors in this study. As predicted, EPS counselors who reported higher levels of time pressure were more likely to consider leaving their profession. These findings align with previous research demonstrating a strong correlation between job satisfaction and intention to leave, with lower job satisfaction systematically linked to increased stress and a higher likelihood of leaving the profession (Skaalvik and Skaalvik, 2011). Similar patterns have been observed among teachers, where excessive time pressure has been associated with decreased job satisfaction and a greater risk of burnout (Skaalvik and Skaalvik, 2011). Beyond time pressure, perceived trust from kindergarten and school personnel was also positively associated with job satisfaction. Since collaboration with these professionals is a central aspect of EPS counselors’ work tasks, experiencing that professionals in kindergarten and schools trust their expertise and contributions likely enhances their overall job satisfaction. This finding aligns with prior research demonstrating that trust is crucial to employee satisfaction (Roberts and David, 2020). These results highlight the importance of reducing time constraints and establishing trust-based professional relationships to support EPS counselors’ well-being and retention.