How cooperative mindsets and course climate relate to the perceived impact of digital cooperation on learning in higher education

In previous research, the definitory boundaries of the terms cooperation and cooperative learning as well as collaboration and collaborative learning have been fuzzy, and often all four terms have been used synonymously. However, cooperation is generally defined as two or more people working together towards a shared goal (e.g., Jeong and Hmelo-Silver 2016), while cooperative learning refers to the learning process that occurs during cooperation in educational settings (cf. Dillenbourg 1999). Cooperation and cooperative learning are often differentiated from collaboration and collaborative learning (Kozar 2010). During cooperation, students work separately with hierarchically divided tasks among group members (i.e., a larger task is divided into independent subtasks), aiming to combine them later into a single final product (Dillenbourg et al. 1996; Dillenbourg 1999). In contrast, collaboration refers to a direct, symmetrical, joint, and interactive involvement of students while working towards a shared learning goal, with tasks divided heterarchically (i.e., tasks are divided equally among group members into interdependent levels) (Dillenbourg et al. 1996; Dillenbourg 1999). Collaborative learning, like cooperative learning, refers to the learning process during collaboration which is an interconnected endeavour (e.g., Dillenbourg et al. 1996; Kozar 2010). However, as Jeong and Hmelo-Silver (2016) note, this differentiation is theoretically useful but often not applicable in real-life situations, as the boundaries between cooperation and collaboration are fluid: “The complexity of group work is such that members often need to collaborate as well as cooperate in the process of working together” (Jeong and Hmelo-Silver 2016, p. 248). Therefore, we adopt their definition of cooperation and refer primarily to ‘working together towards a shared goal’, regardless of whether the working process tends to be more cooperative or more collaborative. This process of working together can moreover take various forms in terms of group size, synchronous or asynchronous interactions, as well as digitally mediated or face-to-face forms (e.g., Jeong and Hmelo-Silver 2016). Within our study, cooperative group work took mainly place in smaller groups (usually two to six people) in (digitally supported) learning contexts.

Aside from the rapid integration of digital tools in the educational area due to the COVID-19 pandemic, the implementation of cooperative forms of learning in digital learning contexts has long been the focus of CSCL research (e.g., Stahl et al. 2014). CSCL comprises all cooperative forms of learning supported or mediated by digital devices (Stahl et al. 2014). Thereby,

“CSCL can occur synchronously, with learners interacting with each other in real time (e.g., a chat room), or asynchronously, with individual contributions stretched out over time (e.g., an e‑mail exchange). CSCL can be completely mediated by computers and networks, with individual learners in different buildings or even different countries; or CSCL can involve learners together in the same physical space using computational devices (such as handhelds or tablets) to facilitate their face-to-face communication” (Stahl et al. 2014, p. 479).

In general, cooperative group work in face-to-face educational settings offers several benefits, including, for example, improved academic outcomes (Gillies 2003), increased academic achievement and motivation (Bossert 1988; Gillies 2003; Stump et al. 2011), and enhanced social skills and self-efficacy (Mendo-Lázaro et al. 2018; Stump et al. 2011). Transitioning cooperation to digital learning environments introduces additional advantages, such as location independence and the possibility to work asynchronously together, providing students with greater flexibility and thus helping them to overcome traditional barriers to cooperation (Jeong and Hmelo-Silver 2016; Nokes-Malach et al. 2015; Vali 2023). However, digitally-supported cooperation also presents unique challenges and requirements (Nokes-Malach et al. 2015; Strauß and Rummel 2021), such as the need for appropriate technical equipment (e.g., Jeong and Hmelo-Silver 2016; Krainz and Csar 2022), new forms of communication (Jeong and Hmelo-Silver 2016; Shimizu et al. 2022), organising and coordinating digital resource sharing (Jeong and Hmelo-Silver 2016), and being able to engage socially and thus creating positive social interdependence despite physical absence (e.g., Jeong and Hmelo-Silver 2016; Shimizu et al. 2022; Strauß and Rummel 2021). Specifically, Jeong and Hmelo-Silver (2016) compile seven affordances that play an important role in the success of CSCL: (1) engaging in a joint task, (2) communicating, (3) sharing resources, (4) engaging in productive cooperative learning processes, (5) engaging in co-construction, (6) monitoring and regulating cooperative learning processes, and (7) finding and building groups. Janssen and Kirschner (2020), applying a Collaborative Cognitive Load Theory (CCLT) perspective, also identify various relevant factors related to successful CSCL, clustering them into four groups: student (e.g., self-regulation or collaboration skills), group (e.g., group experience or composition), task (e.g., complexity), and technological (e.g., scaffolds, tools) characteristics. The authors argue that future studies should examine these different aspects in more detail. Acknowledging this, the present study, taking on a social-psychological point of view, analyses student (cooperative mindsets) and group characteristics (course climate, although from an individual perspective) in relation to the impact of CSCL on students’ learning outcomes from their own perspective.

We do so because one crucial factor for the success of (digital) cooperation is students’ perception of “the cooperation process to be useful both for academic purposes and for acquiring social skills for future cooperative processes” (Muñoz-Carril et al. 2021, p. 2). The perceived usefulness of cooperation is moreover a crucial factor for students’ intentions to return to cooperative group work, which, in turn, is closely related to the perceived impact of digital cooperation on students’ learning outcomes (Muñoz-Carril et al. 2021). Thus, it seems to be crucial to encourage students’ perception of digital cooperation as having a positive impact on their learning. Additionally, ensuring positive, productive interpersonal relations among students and creating a learning environment that supports effective digital communication, resource sharing, and monitoring and regulating learning processes are essential for successful digital cooperation (Jeong and Hmelo-Silver 2016). These aspects are closely linked to the characteristics defining a positive course climate (see below), which in turn has been shown to be related to students’ cooperative mindsets (Jördens and Zander 2024).

Cooperative mindsets are beliefs and attitudes about cooperation that shape how individuals think about cooperation and can subsequently facilitate cooperative behaviour¹. Students with pronounced cooperative mindsets may think more positively about cooperation, value its potential benefits for their learning processes, enjoy cooperating with others, and thus readily welcome opportunities for cooperative learning or actively seek out such opportunities. Research has shown that cooperative mindsets are associated with a reduced prevention focus (i.e., avoiding failure or losses), which, in turn, is related to heightened creativity (Bittner and Heidemeier 2013). Students with a stronger cooperative mindset also tend to experience a stronger feeling of connectedness towards their partners during cooperation (Canevello and Crocker 2017). Thus, cooperative mindsets can affect the perception of the learning process itself and can also influence its outcomes. However, cooperative mindsets are not fixed and can be developed or encouraged through student-centred learning methods such as flipped classrooms including group work and problem-solving activities (Kwon and Woo 2018).

Furthermore, Muñoz-Carril and colleagues (2021) have shown that students’ attitudes towards cooperation, along with their perceived enjoyment of working digitally together, predict their perceived impact on learning in CSCL contexts. Thus, they found a connection between students’ cooperative mindset and their perception of how digital cooperation can affect their learning outcomes, considering students’ individual perspective rather than relying solely on external measures for learning outcomes such as grades or credits. Previous research has also shown that a pronounced cooperative mindset of students, or, in other words, their positive evaluation of cooperative group work, are related to a positive perception of the course climate in university courses (e.g., Ghaith 2003; Johnson et al. 1983; Summers and Svinicki 2007).

Previous research has found that students perceive their overall learning environment—often referred to as the climate—as more positive when they feel that their individual contributions in group work or a classroom are positively connected to those of their peers (Johnson et al. 1983; Summers and Svinicki 2007). This so-called social interdependence “exists when the outcomes of individuals are affected by their own and others’ actions” (Johnson and Johnson 2009, p. 366)². This means that the structure of the group’s goals determines the way individuals interact, which in turn determines the outcomes and the perceptions of the outcomes of these interactions (e.g., Johnson and Johnson 1995, 2009).

Classroom climate in school contexts or course climate in university contexts refers to the overall social, emotional, and academic learning atmosphere shaped by interpersonal relations and interactions between students, as well as between students and lecturers (e.g., Gazelle 2006; Khalfaoui et al. 2021; Wang et al. 2020a). A positive course climate is characterised by the presence of academic or social support (Eder 1996), cooperation (Johnson et al. 1983; Jördens and Zander 2024), connectedness, belongingness, enjoyment, and enthusiasm of the students (Khalfaoui et al. 2021). Course climate is a multidimensional construct that is influenced by various social and academic factors, such as interpersonal relationships and management and structuring by the lecturers (e.g., Frisby and Martin 2010; Khalfaoui et al. 2021), and, in turn, influences a range of social and academic students’ outcomes. Fostering a positive course climate is associated with students’ motivation (Wang et al. 2020b, 2023), academic performance (Lizzio et al. 2002), achievement (Reyes et al. 2012; Wang et al. 2020b), and several social aspects (e.g., Lizzio et al. 2002; Wang et al. 2020b; see also Rowe et al. 2010). Moreover, course climate is positively related to the opportunity to actually cooperate with each other within a university course (Jördens and Zander 2024). However, this relationship seems to apply more strongly to face-to-face cooperation than to digital-only cooperation (Jördens and Zander 2024), potentially due to the frustration that students often experience in digital learning settings (Capdeferro and Romero 2012). Capdeferro and Romero (2012), for example, point out that frustration in CSCL settings occurs due to asymmetric cooperation among students, difficulties in group organisation, lack of shared goals, imbalanced levels of commitment and quality of contributions, invested time, imbalance between individual and collective grades, and communication difficulties. Given the ongoing trend towards the use of digital methods in higher education, it is crucial to understand how cooperation can be successfully implemented in digitally supported learning contexts.

Furthermore, Summers and Svinicki (2007) found that students’ perception of whether they worked interactively together predicts their sense of classroom community (i.e., the course climate) (see also Vu et al. 2021). Positive attitudes towards group work, in turn, are strongly linked to the time students spent actually working together (Freeman 1996). These findings are first indications that a cooperative attitude or mindset might also predict the course climate.

Taken together, there is evidence that students’ cooperative mindset is linked to the perception of the course climate, and that students’ cooperative mindset is related to their perceived impact of digital cooperation on learning. These relationships suggest the presence of more complex interactions among these three constructs. Thus, the present study intends to answer the following question: How do students’ cooperative mindsets and the course climate influence how students perceive the impact of digital cooperation on their learning outcomes?

We aim to assess the circumstances under which students perceive that digitally-supported cooperation positively impacts their learning experience. Specifically, we investigate whether and how students’ cooperative mindset and the course climate influence how students perceive the impact of digital cooperation on their learning outcomes. Based on prior research (e.g., Freeman 1996; Jördens and Zander 2024; Muñoz-Carril et al. 2021; Summers and Svinicki 2007), we hypothesise the following:

The total sample consisted of n = 313 students (133 female, 140 male, 1 diverse, and 41 did not report gender; M_age = 22.80 years, SD_age = 2.98) of different study programmes of three German universities. We excluded missing values in the variables of interest in the total sample as well as data of all students who self-reported in terms of their perceived impact of digital cooperation on their learning that they did not engage in digital cooperation during their courses (see also Sect. 2.2.3). Thus, n = 181 were eligible for the mediation analysis. Information about the digital tools the students used for different forms of digital cooperation can be found in Table A and Table B in the supplemental materials.

Among the students in the final sample, 73 were enrolled in Educational and Social Sciences, 61 in Law and Economics, 24 in Natural Sciences, 22 in Humanities, 22 in Engineering Sciences, 4 in Environmental Sciences, and 2 in Art, Design, Music. It is important to note that students had the option to choose multiple study programmes. Of the students, 43 were aiming for a school teaching degree, while 132 were not. Additionally, 96 students were enrolled in Bachelor’s programmes and 85 in Master’s programmes. Moreover, 157 students stated German as their first language. 82 students identified as first-generation university attendees, meaning neither their parents nor grandparents attended university. Finally, 29 students had previously completed a vocational training programme.

Data were collected twice, at the start and end of the winter semester 2022/2023. However, only data from the second timepoint (T2) were included in our analyses to reduce its complexity. Moreover, our dependent variable was measured only at T2 as the impact of digital cooperation on learning in a specific course can only be assessed at the end of the semester. The courses where we collected the data were heterogenous regarding the course topic but relatively homogenous regarding the course format and size (i.e., courses with a seminar structure and a maximum of 40 students). Participants provided informed consent and completed an online questionnaire administered via LimeSurvey. Students took part voluntarily and did not receive any payment or course credit for their participation. The study consisted of the following parts: Participants were asked to rate the degree to which they agreed with statements about their cooperative mindset, about their perception of the climate within the specific course in which the data were collected, and about the impact of digital cooperation on their learning processes and outcomes. The survey also included other measures such as individual, academic and social factors that are relevant to various outcomes, such as learning, achievement, performance, or well-being in higher education contexts. At the end of the survey, participants were asked to provide demographic information as well as information about their studies.

Descriptive statistics and bivariate correlations for all variables of interest were computed using jamovi version 2.3.28.0 (The jamovi project 2022). We also checked for normal distribution of the data (Shapiro-Wilk test). The mediation analysis was performed using PROCESS model 4 in R (Hayes 2022), specifically version 4.2.1 of RStudio (R Core Team 2022). Before conducting the mediation analysis, we tested whether the assumptions for regression were given (i.e., linearity, normality, homoskedasticity, absence of multicollinearity, absence of autocorrelation of errors, and no outliers). Any violations of these assumptions were appropriately addressed in the analysis.

We used percentile-bootstrapping for all paths (n = 10,000 bootstrap samples, seed = 654321) and report standardised effect sizes for the indirect effect. In addition to unstandardised effect sizes for all paths and the direct effect (b), we also report standardised effect sizes (B). In our mediation model, the perceived impact of digital cooperation was included as the dependent variable, while cooperative mindsets served as the independent variable. Course climate was specified as the mediator.