How teacher-teacher and teacher-student cooperation link with achievement – evidence from international studies

Student achievement has been shown to be a very important factor in predicting ESL (e.g. Bushnik, Barr-Telford, & Bussière, 2004; Knighton & Bussière, 2006; Lessard, 2010; Marks, 2007). Therefore, if cooperation practices in schools influence student achievement they are (indirectly) important for the context of ESL. However, due to their cross-sectional design international large-scale assessments (hereafter ILSA studies) cannot investigate the causal effects of factors on student achievement or even ESL. With cross-national comparisons of student achievement in relation to the education system, school and student background, they offer a way to gain insight into educational processes and practices occurring in successful systems and schools.

In ILSA studies the investigation of cooperation practices in schools is carried out using background questionnaires. Several ILSA studies used teacher questionnaires to collect data on teacher characteristics and their teaching and cooperation practices in an effort to find those that are linked to (higher) student achievement. However, investigations of ILSA data were unable to clearly identify teacher characteristics and classroom practices that could be directly linked to an improvement in student learning. For example, using TIMSS 2003 data Akiba (2007) found that teacher variables like education and experience are positively related to student outcomes across 46 countries. Similarly, Liang, Zhang, Huang, Shi and Qiao (2015) used TIMSS 2003, 2007 and 2011 data to show a positive association between the national maths achievement and the percentages of students whose teachers had participated in professional development. In contrast, Luschei and Chudgar (2011) reported they found only limited evidence that teacher characteristics are related to student achievement.

As discussed by Schleicher (2016), teaching is a complex task that involves interactions with a great variety of learners in a wide range of different circumstances and so there is not one single set of teacher attributes and behaviours that is universally effective for all types of students and learning environments. The lack of a strong relationship between teacher characteristics and student achievement in the ILSA studies may be due to the complexity and diversity of teacher professionalism and not all the facets of teacher professionalism being equally efficient (ibid.).

This paper presents a literature review of studies using ILSA data to address the associations between teacher-teacher or teacher-student cooperation and student achievement. In the next section, we present the methods used in the search. The results of our search first present a review of conceptual understanding of the teacher-teacher and teacher-student cooperation that are then followed by research findings on the associations of each type of cooperation with student achievement. Finally, some conclusions are given.

The initial challenge in performing a literature search in this paper is a conceptual one. Due to their design, ILSA studies provide a broad overview of the state of affairs in education systems, making it is impossible to derive thorough definitions or measurements of individual background concepts. This is also the case for the concepts of teacher-teacher and teacher-student cooperation and therefore our search for literature was somewhat constrained. More specifically, for the literature search to produce results, the terms entered in the search related to teacher-teacher and teacher-student cooperation in the broadest sense, most notably including collaboration and teacher-student relations or otherwise accommodating the definitions from the particular ILSA study in question.

Literature searches for the review in this paper were performed on diverse search engines and bibliographical databases, library catalogues and websites, starting with international reports of the results of the above studies. Keywords used in the search were different combinations of the names and acronyms of the studies or terms “international assessments”, “international studies” etc. and “collaboration”, “cooperation”, “teacher-student relationship” etc. As mentioned, the search with the terms “early school leaving” or “dropout” proved to be too narrow and did not produce any results. The results described below therefore do not address ESL directly but provide a wider international context of associations between teacher-teacher or teacher-student cooperation and student achievement.

Among ILSA studies that have collected data about cooperation practices in schools, the largest sets of constructs can be found in the Teaching and Learning International Study (TALIS) conducted in 2008 and 2013 in connection with the Programme for International Student Assessment (PISA) with the latest cycle of 2012. Some data can also be found in the International Study of Computer and Information Literacy Study (ICILS) carried out in 2013, the Trends International Mathematics and Science Study (TIMSS) and the Progress in International Reading Literacy Study (PIRLS), both with the latest data collection in 2011.

In our literature review it became evident that research on teacher cooperation practices occurring in schools is embedded in the research of the concepts of teacher professionalism and professional learning communities. Based on the view that teacher quality is the key to student achievement, these concepts have gained research prominence in the past decades (Darling-Hammond & Lieberman, 2013). In their meta-analysis, Lomos, Hofman and Bosker (2011a) discuss how the professional community concept has had a long development process, starting around 1982 with the introduction of collegiality and collaboration and developed around 2000 into professional learning community. From their meta-analysis and drawing from a number of previous studies they derive the concept with five dimensions; namely, reflective dialogue, de-privatisation of practice or feedback on instruction, collaborative activity, shared sense of purpose, and a collective focus on student learning. [1] Using the TALIS data, the concept of professional learning communities was studied by Vieluf, Kaplan, Klieme and Bayer (2012).

A report by the OECD (2016) addressed teacher cooperation practices by conceptualising teacher professionalism as consisting of three major domains: 1) professional knowledge (formal teacher education, professional development opportunities, practitioner research); 2) teachers’ autonomy in decision-making (autonomy, school-based decision-making); and 3) high peer networks (mentoring, induction, professional development plans, peer feedback, professional learning communities). Within this concept, the authors emphasised that maintaining high professional practices takes various forms, most of which stress peer collaboration and networks of information exchange, knowledge sharing and collective standard setting (ibid.).

In TALIS it was recognised that the cooperation of teachers can take various forms; teachers may, for example, exchange instructional materials and meet regularly for discussions about individual students while more complex forms of cooperation, named in TALIS professional collaboration, include collective learning activities such as observing others and providing feedback as well as engaging in professional learning activities and joint activities across classes and age groups (e.g. OECD, 2009). Notably, the TALIS data consistently show professional collaboration practices are still relatively rare compared with practices that focus on coordination and the exchange of information and material (ibid.). Similarly, in reporting on teacher-teacher cooperation in the context of using ICT for teaching and learning the ICILS study (Fraillon et al., 2014) showed that collaboration with colleagues [2] was on average across countries reported by schools to represent less than half the student population. This confirms the TALIS findings that more demanding collaborating practices are not very common. In contrast, according to teacher self-reports from TIMSS 2011 (Martin et al., 2012; Mullis et al., 2012a) and PIRLS 2011 (Mullis et al., 2012b) it seems that almost all students have the benefit of teachers who collaborate with other teachers to improve instruction. Compared to the TALIS and ICILS data, these reports seem to be somewhat overstated. Finally, TALIS showed that the majority of the variance in constructs about cooperation between teachers remains at the individual level with the teacher. In other words, teachers differ from each other in their cooperation practices even within the same school (OECD, 2009; OECD, 2014).

An example of conceptual development of teacher-student cooperation can be found in Jennings and Greenberg (2009). Their model highlights the importance of teachers’ social and emotional competence for the development of a classroom climate that is more conductive to learning and promotes positive developmental outcomes among students (ibid.). In the ILSA studies, teacher-student cooperation is represented with indicators of teacher-student relations, school and disciplinary climate (e.g. Mullis et al., 2012a; OECD, 2013a). For example, in PISA students were asked about the student-teacher relationship in terms of whether students get along with their teachers, are teachers interested in their personal well-being, whether teachers take the students seriously, whether teachers are a source of support if a student needs extra help, whether teachers treat students fairly. In PISA, students generally reported good relations and differences between students in their views of teacher-student relations are mostly found within schools and much less between schools (OECD, 2013b). In TALIS 2013, principals and teachers also reported good relations between teachers and students and it is only in the area of providing students with extra support that any variation is observed (OECD, 2014). In relation to the concept of teacher-teacher cooperation, TALIS 2008 showed that teachers who exchange ideas and information and coordinate their practices with other teachers generally report more positive teacher-student relations at their school (OECD, 2009).

The above concepts provide a theoretical background for the results of our literature review regarding evidence on the link between cooperation practices in schools and student achievement that can be found among the ILSA studies.

In this section, we start by shortly discussing general evidence on the link between teacher-teacher cooperation and student achievement. After that, we present the results found specifically from the ILSA studies.

Lomos et al. (2011a) explain that empirical evidence on the relationship between teacher-teacher cooperation (more specifically, conceptualised in their study as professional community) and student achievement is diverse and has produced mixed results, ranging from small to medium impacts. Although relatively small, their meta-analysis shows that the relationship between professional community and student achievement is positive and significant. That well-developed professional learning communities have a positive impact on both teaching practice and student achievement was also confirmed in a review by Vescio, Ross and Adams (2008). A thorough analysis is reported by Akiba and Liang (2016) in which they used a longitudinal study design to show that teacher-centred collaborative activities to learn about mathematics teaching and learning, namely teacher collaboration and informal communication, are more effective in improving student mathematics achievement compared to learning activities that do not necessarily involve such teacher-centred collaborative opportunities. Examples include professional development programmes, university courses and individual learning activities.

Our literature search on the studies using ILSA data also provided evidence that student achievement is positively associated with teachers and schools working as professional learning communities. Lomos, Hofman and Bosker (2011b) applied cluster analysis and hierarchical linear modelling to the TIMSS 2003 data to investigate the relationship between the aspects of mathematics department units as professional learning communities (as described in the conceptual background) and student achievement in Dutch secondary schools. The study showed that those departments which worked as professional learning communities are associated with successful schools and higher student achievement. After controlling for background variables at the student and teacher/school levels, an additional 7% of the variance among schools was explained by the presence of the above-mentioned dimensions of mathematics departments as professional communities. However, only the focus on student learning showed a direct positive effect on achievement in the model which, according to the authors, indicated that higher expectations of teachers with respect to the success of their students are associated with higher levels of student achievement. While the effects of reflective dialogue or collaboration were non-significant in the model, the authors state that this may be because there are indirect effects of teacher interaction on student achievement.

Associations between teacher collaboration and student achievement were also investigated in the reports of TIMSS 2011 (Martin et al., 2012; Mullis et al., 2012a) and PIRLS 2011 (Mullis et al., 2012b). Generally, both assessments in the domains and populations studied show that students had essentially the same average reading, mathematics or science achievement irrespective of whether their teachers were categorised as “very collaborative” or “collaborative”. Arguably, this non-differential effect of higher collaboration on student achievement may be due to the small variation in teacher self-reports about their collaboration practices, as mentioned in the previous section.

There is also evidence from ILSA studies that indirectly suggests a link between cooperation practices among teachers and student achievement. Schleicher (2016) states that evidence from TALIS that also draws on other studies (Cordingley et al., 2015; Liaw, 2009; Puchner & Taylor, 2006) shows that collaboration among teachers may enhance teacher efficacy which, in turn, may improve student achievement and sustain positive teacher behaviours. As revealed by TALIS, the practice most strongly related to teachers’ self-efficacy is taking part in collaborative professional learning and, further, there is evidence of collaborative professional development being linked to a positive influence on student learning processes, motivation and outcomes (Schleicher, 2016).

That building a collaborative school culture is key to successful schooling can also be observed from PISA. The OECD (2013b) report discusses that the nature of the relationship between school autonomy in allocating resources and student performance depends on whether there is a culture of collaboration between teachers and principals in managing the school. In systems where teacher participation in managing the school was higher than the OECD average, students in schools with greater autonomy achieved higher than students in schools with less autonomy. Contrastingly, in systems where teacher participation in managing the school was lower than the OECD average, students in schools with greater autonomy achieved lower than students in schools with less autonomy (ibid.). School autonomy combined with a culture of participatory leadership tends to be associated with better learning outcomes (Schleicher, 2016).

The ICILS report indirectly addressed the link between teacher collaboration and student achievement by indicating that teachers were most likely to use ICT for teaching when they worked in school environments where there was collaboration and planning concerning ICT use, and where there were fewer resource-based obstacles to using ICT (Fraillon et al., 2014). [3] The report emphasises that these showed to be the conditions which supported teaching about computer and information literacy. In other words, this suggests that if students’ computer and information literacy is to be developed to the greatest extent possible, then teacher expertise in ICT use needs to be developed and supported by collaborative environments that incorporate institutional planning (ibid.).

Overall, it can be concluded that the ILSA studies provide evidence on teacher-teacher professional collaboration, such as observing others and providing feedback or joint activities across classes and age groups, positively influencing student achievement.

In the conceptual background section, we explained that for the purposes of our review the concept of teacher-student cooperation is characterised by indicators of teacher-student relations, classroom disciplinary climate and school climate. Although there are mixed results concerning whether good teacher-student relationships serve as a directly protective factor against ESL (e.g. Lessard, Pirier, & Fortina, 2010; Murray & Malmgrenb, 2005), there are findings showing that students, particularly disadvantaged students, learn more and have fewer disciplinary problems when they feel their teachers take them seriously (Gamoran, 1993) and when they have strong and affective bonds with their teachers (Crosnoe, Johnson, & Elder, 2004).

In our search, it was possible to find ILSA evidence that teacher-student relations and school climate are important for student achievement. For example, several studies used national or international PISA data to show that school (Azigwe, 2016) and classroom climate (Bove, Marella, & Vitale, 2016) as well as teacher-student relations more specifically (Kalender, 2015; Konishi, Hymel, Zumbo, & Li, 2010; Mikk, Krips, Saalik, & Kalk, 2015) are significantly associated with student achievement, even after accounting for other student, teacher and school characteristics. Mikk et al. (2015) revealed that the strongest (and a positive) correlation between teacher-student relations and student achievement exists at the school level, followed by correlation at the student level. At the country level, a negative correlation was found, a phenomenon also called ecological fallacy due to the different contexts in which students’ perceptions of teacher-student relations are developed. For Canada, Konishi et al. (2010) additionally discovered that, for boys, teacher-student connectedness acts as a buffer against the effects of bullying on their mathematics and reading achievement.

Disciplinary climate was found in PISA to be related to higher average performance at the school level. Although it can be recognised that schools with a more negative disciplinary climate tend to have a largely disadvantaged student population, have greater socio-economic diversity among students, and suffer from more teacher shortages, the data also show that even when comparing schools of similar socio-economic status students in schools with more disciplinary problems tend to perform worse than their peers in schools with a better disciplinary climate (OECD, 2013b).

In their analysis of the TIMSS 2007 and TIMSS Advanced 2008 data for Slovenia, Kozina, Rožman, Vršnik Perše and Rutar Leban (2012) showed that in secondary school all types of evaluations of the school climate (performed by principals, teachers and students) are significant predictors of mathematics and science achievement. The evaluation performed by the principals showed to be the strongest predictor, followed by the evaluation by the teachers and last that by the students. They found that the predictive power of school climate seems to increase from the elementary to secondary school level. The results also showed that a positive school climate correlates with higher student achievement in both elementary and secondary schools when evaluated by teachers and principals, and with low student achievement when evaluated by the students of secondary and elementary schools (possibly because teachers high expectations regarding students’ achievement were one of the characteristics of a positive school climate in the study). The finding that school climate as perceived by the school principals is the most influential factor on the achievement of students was also made by Ghagar, Othman and Mohammadpour (2011) for Singapore and Malaysia. The authors performed school-level analyses on TIMSS 2003 data.

Mistele and Louis (2011) built a path model on TIMSS 2007 data to understand the impact the teacher-student relationship may have on student mathematics achievement by exploring the associations between teacher attitudes as perceived by their students, student attitudes and student achievement. The authors conjectured that a crucial component for middle school students experiencing success and confidence, while engaged in rigorous mathematics classes, begins with the relationship between the teacher and the student. The results of path analyses suggested that teacher attitude is indeed a mediating factor between student attitude and student mathematics scores (ibid.).

ILSA studies provide additional evidence of associations between teacher-student relations and so-called non-cognitive educational results that have been shown important for student achievement. For example, PISA showed (OECD, 2013a) that teacher-student relations are strongly associated with students’ engagement with and at school and with students’ mathematics self-efficacy. Among students with equal performance and similar socio-economic status, those who attend schools with better teacher-student relations reported a stronger sense of belonging and greater intrinsic motivation to learn mathematics (ibid.). From the opposite point of view, PISA showed that students who are in schools where teacher-student relations and disciplinary climate are poor are more likely to have low levels of engagement with and at school. They are more likely to arrive late for school, skip classes or days of school, report a weak sense of belonging, and hold negative attitudes to school. A lack of engagement – on the part of either teachers or students – can have adverse effects on the entire school community (Schleicher, 2016) and research has established that lack of engagement is one of the warning signs for students that later become early school leavers (e.g. Lamb, Markussen, Teese, Sandberg, & Polesel, 2011).

In summary, students who learn in a school climate characterised by high expectations, classrooms conducive to learning, and good teacher-student relations tend to perform better than those who do not (OECD, 2013b). According to Bove et al. (2016), an orderly and peaceful atmosphere in schools and classrooms together with cognitively activating instruction help to transform existing interests into mathematics achievement.

The more interdependent the world becomes, the more society relies on collaboration and schools need to prepare students to successfully mobilise, share and link knowledge (Schleicher, 2016). In this article, we presented a literature review of ILSA evidence on the link between teacher-teacher and teacher-student cooperation with student achievement. Our review shows that when appropriately organised and targeted cooperation practices in schools of both types between teachers and between teachers and students may positively influence student outcomes. The main ILSA findings from our search may be summarised in more complex professional collaboration between teachers, teacher-student relations and overall school climate being associated with an improvement in student achievement, even when considering students’ different backgrounds. It may also be important that teachers who collaborate more with their colleagues generally report more positive teacher-student relations.

In conclusion, building a collaborative school culture is key to many dimensions of school life including student outcomes. However, in his analysis of the difficulties in building collaborative cultures in schools Hargreaves (2004) warns that when collegiality is directed from above it inhibits bottom-up professional initiative and results in teachers actually collaborating less or abandoning collaborative ways of working as soon as the urgency of creating or implementing a school improvement plan is over. Building a collaborative culture therefore needs to be a collaborative process.