Collegial collaboration when planning and preparing lessons: A large-scale study exploring the conditions and infrastructure for teachers ’ professional development

(cid:1) Using an extensive survey, this study explores teachers' perceptions on planning and preparation in relation to aspects of structured collegial collaboration and professional development. (cid:1) There are shortfalls in infrastructure on teachers' planning and preparation in relation to structured collegial collaboration. (cid:1) There is a profound impact of a supportive collegial structure on teachers' descriptions of their working conditions. However the local infrastructures do not seems to support ef ﬁ cient formative teaching. (cid:1) Our ﬁ ndings indicate that schools are typically not organised to support systematic formative teaching and assessment.


Introduction
The insight that teachers can make a difference in student learning is well established.Hence, in Sweden (Kools, 2015), as in many other countries, significant efforts have been made by authorities (Organisation for Economic Co-operation and Development [OECD], 2020) and research communities alike to develop and enhance teachers' professionalism (Chetty et al., 2014;Lefstein et al., 2020).In Sweden, there are some major trends in improving teaching.Most e if not all e teachers have undertaken information and communication technology (ICT) training (Player-Koro, 2012), and they are familiar with formative assessment for supportive feedback and adaptive teaching (e.g., Black, 2015).They have participated in collegial learning with their peers, most often in cross-curricular settings (Harvey & Teledahl, 2019).Some of them have been enrolled in learning studies (Cheung & Wong, 2014) or employ other models to assess how their teaching contributes to student learning.During the last decade, there has also been a growing governmental interest in reforming career services for teachers (Alvunger, 2015).
Measures to improve teaching are investments expected to make a long-term impact.However, the transfer from initiative to improved teaching is not a straightforward process and takes time to establish (Boeskens & Nusche, 2021;Bryk et al., 2010).What seems to be crucial for a lasting effect is the active involvement of the local school organisation to develop supportive infrastructure for teaching improvement (Jarl et al., 2017;Timperley et al., 2009).This article draws attention to specific teacher activities vital for teaching improvement but often neglected in school improvement projects.
Teachers' planning and preparation of lessons (PaP) is the backbone of teaching.PaP are outside-of-classroom activities that teachers do to arrange for learning to happen: that is, everything from relating aims to content, selecting and constructing tasks, assignments and tests; preparing the classroom, adjusting for students' individual needs, choosing ways to explain, describe, challenge and support students; to reflecting upon outcome of chosen methods and students' achievement and preparing for formative feedback and summative response (John, 2006;Yinger (1979).PaP activities are directed to specific lessons and learning activities and, as such, often the individual teacher's responsibility.Hence, projects for school improvement rarely directly target the role and functions of PaP within the school organisation (Feldhoff et al., 2016;Lindahl, 2011;Levine & Marcus, 2010).Just as a city requires infrastructure to function (such as transportation), school organisations rely on systems that enable them to work.However, as Miyakawa and Winsløw (2019) note, a school's infrastructural systems are often abstract and administrative.While PaP has a specific praxis and activities, they are integrated into a complex infrastructural system that combines both given factors (such as time and official responsibility) and local praxeologies (such as traditions, expectations, and knowledge).PaP is central to teachers' everyday work, but the surrounding infrastructural support is often vaguely defined, so it may be crucial to pay attention to the local PaP praxis when implementing school improvement.Research indicates that PaP can be a central driver for change when it is regarded as a collegial rather than an individual responsibility and when this consideration is matched with specific infrastructural support (Boeskens & Nusche, 2021;Hixson et al., 2013;Wiggins & McTighe, 2005;Yuan & Zhang, 2016).
Using an extensive survey, this study explores how Swedish (compulsory and upper secondary school) teachers' perceptions of PaP interrelate with their perceptions of scheduled collegial collaboration and professional development.The aim is to explore teachers' working conditions for lesson PaP.Two specific questions have guided our work: 1) How do teachers perceive the conditions for their lesson planning, both individually and through collegial collaboration? 2) Do the possibilities for collegial planning have an effect on teachers' perceptions of infrastructural resources and pedagogical possibilities?
The first question is explorative, and the second is analytical.Together, they will provide a status report in the Swedish context that, in many ways, has general similarities with the OECD's (2019,2020) survey averages on working conditions for teachers.However, this survey is unique in that it focuses specifically on PaP in terms of teachers' preparation and follow-up work, as well as their individual and collegial work.It provides an analytical basis for examining how teachers' perceptions of their PaP affect both how they view their basic pedagogical activities and their pedagogical development.We will also examine how the working conditions for teachers' PaP affect how they perceive their work environment, pedagogical support and organisational support from principals and lead teachers.

Swedish teachers' working conditions in comparison with the OECD average
Swedish teachers' working hours are similar to teachers in most OECD (2019) countries.Swedish teachers teach for a similar amount of time during the working week, and PaP time does not differ considerably either.Also, like most teachers in the developed world, Swedish teachers regard teaching and learning as the most important occupational task.On average, they spend about 40% of their time each week on teaching activities, following the same pattern as in other comparable countries.They find this amount of teaching time appropriate, even though they note that the expectations associated with the quality of teaching and learning are rising.This puts pressure on PaP time.Even though PaP time for Swedish teachers does not differ from that of teachers in comparable countries (i.e.25% of weekly working hours), they find that this time may be reduced due to other activities, such as administrative tasks, student counselling and engagement with parents (Skolverket, 2015(Skolverket, , 2019)).Thus, PaP time becomes fragmentally scattered into small time slots during the school week, and the individual teacher has to handle and protect that time.This fragmented PaP time seems to enhance the individual approach to PaP since teachers bring this activity home as an outside-of-school activity (Skolverket, 2015), thereby affecting opportunities for collegial PaP time.

Positioning the research
There is strong evidence that schools that manage to create collaborative cultures among teachers are also supportive of student learning (Bryk et al., 2010;Lomos et al., 2011;Ronfeldt et al., 2015;Vescio et al., 2008).Well-functioning teacher collaboration is characterised by a professional orientation towards meeting the goals of the organisation (e.g.Jarl, Blossing, & Andersson, 2017;Stoll et al., 2006).However, building a successful and sustainable milieu of collaboration demands endurance, as it takes time to establish.It therefore needs systematic support to integrate such collaboration into daily practice (Jarl et al., 2017;Lefstein et al., 2020;Lindahl, 2011;Stoll et al., 2006).Such support can be in the form of inputs of knowledge and expertise, as a positive correlation exists between teachers' professional development and research cooperation (Hopkins et al., 2014;Tan & Caleon, 2016;Taylor et al., 2014).Perhaps most importantly, the collaboration should be organised with a clear focus on student learning (Hattie, 2008;Mertens et al., 2010).
Hence, implanting collaborative models focusing on the quality of teaching and learning goes beyond in-lesson activities, as they are embedded in PaP (Hattie, 2008;Ronfeldt et al., 2015).Nevertheless, the precondition for PaP is most often implicit or even neglected in the discourses of professional development and collegial collaboration (Lefstein et al., 2020).These activities are typically individual responsibilities for teachers (Darling-Hammond & Rothman, 2011;Ellegård & Vrotsou, 2013).Teacher collaboration is often characterised by the use of self-regulated learning processes by teachers to identify their deficits and determine how they may improve their instruction (Timperley et al., 2009).Thus, outside-of-lesson activities need support from the curriculum level up to the local organisation level (Miyakawa & Winsløw, 2019).
Arguably, teachers need individual PaP time to provide highquality teaching (Merritt, 2016).However, collegial planning can help teachers to further develop their skills in structuring lessons (Strickroth, 2019) and to evolve their pedagogical content knowledge (Smit et al., 2018).The possibilities for teachers to design, enact and reflect on lesson ideas vary among countries and schools (see, e.g.Bieda et al., 2020;Boeskens & Nusche, 2021).Dever and Lash (2013) have shown that the organisation of collaboration is important.They followed teams of middle school teachers who devoted common planning time to both interdisciplinary and subject-oriented teams.The interdisciplinary team did not create or share any specifics about academic lessons, and their conversations mostly concerned students' social behaviour rather than their learning.In the subject-specific team, the teachers constructed common plans, and their conversations were proactive and supportive so that the members could improve their teaching.This could also be said to be true from the students' perspective: As Wheelahan (2007) emphasised, generic competencies are not sufficient if students need to go beyond learning knowledge only as a product and instead understand the process by which knowledge is derived.This indicates that the time spent by teachers on subjectspecific PaP is important in the development of high-quality teaching.Yan and Wei (2019) used Teaching and Learning International Survey (TALIS) data to explore the relationship between professional development and teacher efficacy, and they found that content-and inquiry-oriented collaborative learning, along with instructional skills and school management, were statistically and positively correlated with the overall scale and subscales of teacher efficacy.
Evidently, schools still continue to experience problems in identifying strategies to make professional development and teacher efficacy a reality in their everyday work (Darling-Hammond et al., 2017;OECD, 2014;Mourshed et al., 2010).In the United States, for instance, some improvements have been evidenced at different schools scattered around the country.However, …the structures and supports that are needed to sustain teacher learning and change and to foster job-embedded professional development in collegial environments fall short (in the US).The time and opportunities essential to intense, sustained professional development with regular follow-up and reinforcement are simply not in place in most contexts (Darling-Hammond et al., 2009, p. 27).
Other international and national studies indicate a similar picture to that of the United States (OECD, 2014(OECD, , 2019)).The OECD TALIS is an important international survey examining teachers' and school leaders' understanding of teacher professionalism in 48 countries.From the TALIS (2019) analyses of the collaborative culture among teachers, teachers' collaborative practices are most likely to be about the development of specific students (61% of teachers on average) and exchanging teaching materials with colleagues (47%), while "far fewer teachers engage in the deeper forms of professional collaboration, which involve more interdependence between teachers, with only 9% of teachers saying they provide observation-based feedback to colleagues."(p.16) As presented in the results, our in-depth data from the Swedish context show that the general trend is the same: Most teachers do not participate in systematic professional PaP collaboration.The TALIS (2019) is an important point of reference.However, while the TALIS focuses on general teacher collaboration, our focus is on PaP.This design allows us to analyse the PaP as both individual and collegial activities.Further, it allows us to explore the impact of organised collaboration on how teachers perceive their pedagogical possibilities and organisational resources.
Previous research offers an essential baseline for the theoretical framing and methodological decisions of our study.We, therefore, focus on exploring the conditions for teachers' PaP time, collegial collaboration and research-related opportunities so as to develop their teaching and improve student learning.

Theoretical framing
We understand teachers' lesson PaP as crucial for both student learning and pedagogical development, as expressed by Yinger (1979, p. 163): Of the many different things that teachers do in the preactive phase of teaching, planning is probably the most important.The wealth and variety of instructional materials available, the emphasis on meeting school or district objectives, and the wide range of student aptitudes in most classrooms are but a few of the factors that virtually necessitate thinking and planning for the term, coming weeks, or even the next day.
To explore PaP as an explicit and qualified part of teachers' work, we reserve the term PaP as a distinct outside-of-lesson activity.In this context, we regard teachers' adjustments, revisions when something arises and mini-planning during class as part of teaching or in-lesson activity.Tyler (1950) described PaP as a four-step activity: specifying objectives, specifying knowledge and skills, selecting and sequencing learning activities, and evaluating outcomes.These basic features should be understood as activities that take place in a cyclic process that also considers student participation (Cruickshank, 2018;John, 2006;Wraga, 2017).In practice, the preactive phase, such as deciding on learning activities, and the post-active phase of marking and assessing student work, correspond, and they are not as linear as Tyler described them.In this study, we include both plannings for upcoming teaching activities as well as formative and summative assessments in the concept of PaP.Our survey is guided by three interrelated themes.Below is the brief rationale for framing these themes.a) Teachers' conditions for PaP.The term "infrastructure" refers to both formal and informal structures and praxis that ensure that basic functions within an organisation can be performed (e.g.Miyakawa & Winsløw, 2019).Time is an essential resource and constraint for PaP with both given limits and a praxis.To explore how teachers experience their PaP efforts, we distinguish between the time they use to plan lessons and evaluate them.
Studies on teachers' working hours have shown that the planning time tends to be more scattered and fragmented during the working day and week, while the time used to assess student achievement is more cohesive (Ellegård & Vrotsou, 2013).We have further explored several other components of PaP, such as reflecting on the methods and results of formative assessment and working conditions (Black, 2015;John, 2006;OECD, 2019;Wraga, 2017).b) Effects of collaborative infrastructure.Effectiveness related to teacher instruction is often associated with teaching modelling, time management, their role in making the classroom a productive learning environment, and a systematic formative relationship between assessment and planning (Darling-Hammond et al., 2017;Muijs et al., 2014).Using the exploration of infrastructure and how it affects the time for collaborative PaP also allows for the analysis of its effects on teachers' working conditions in general.c) Aspects of systematic formative teaching and PaP.We based our understanding of formative teaching on Wiliam (2018).The term "systematic formative teaching" refers to the regular use of teaching models and the recurring evaluation of student performance to help teachers further develop their teaching (Christodoulou, 2016;Wiliam, 2018).

Method
This study was based on data collected via a questionnaire survey seeking to explore the infrastructural conditions of PaP as a resource for teaching development.The design of the questionnaire was partly inspired by former studies (e.g.OECD, 2014OECD, , 2019)); however, as we are more explicitly targeting teachers' PaP as a part of school organisation and the possibilities of collegial cooperation, the design also draws on broader theoretical considerations, as outlined above.A pilot study was conducted within the same main population to validate the items; that is, with compulsory (6e16 years old) and upper secondary school teachers.The questionnaire was improved based on the results of the pilot study.In addition, experts from Statistics Sweden examined the included items (Statistiska Centralbyrån, 2018).The idea for the study was then subjected to ethical evaluation (Forskningsetiska Kommitt en, 2018).
The questionnaire investigated teachers' PaP, organisational aspects and forms of school development (Nordgren et al., 2019).In addition to the questionnaire data, register data were collected (i.e.gender, age, teaching position, head of school [municipal or independent], school size [number of students] and municipality) from Statistics Sweden.Background information collected in the survey consisted of each teacher's main teaching subject, the total number of working years as a teacher, the number of years working at the present school, the teacher's position (i.e. in Sweden, teachers can hold positions of lead teacher, senior lecturer, in charge of school development) and whether the teacher held a tertiary-level degree (e.g. a PhD).

Sample and population
The survey was distributed during February and March 2017, and two reminders for completion were provided.In total, the population included 108,370 individuals, out of which a stratified sample (based on gender, age, head of school [municipal or independent]) of 4988 individuals (approximately 5% of the population) was drawn in order to get a representative sample.A stratified random sample means that all objects within a stratum have the same probability of being included in the sample.The response rate was 46% (n ¼ 2285).The empirical distributions regarding gender, age, and head of school in the sample are similar, almost identical to the distributions in the actual population (Statistiska Centralbyrån, 2018).The analysis of the external non-responses showed no systematic patterns with respect to central background variables within the two school forms (compulsory and upper secondary).

The questionnaire and data processing
The questionnaire was structured into four themes: planning, evaluation and development of lessons; scheduled time for collaborative work; systematic and formative developmental work; and perceived support from principals and lead teachers in teaching development.

Planning, evaluation and development of lessons
This study specifically analysed questions addressing teachers' opportunities in terms of the time they devoted to planning and evaluating their lessons.In particular, the following specific survey questions were used to measure teachers' time for planning and evaluating their lessons: "How often can you prepare lessons without being disturbed within working hours?", "How often can you evaluate lessons without being disturbed within working hours?", "How often do you plan lessons outside of regular working hours?", "How often do you evaluate lessons outside of regular working hours?", "How often can you reflect on the fitness of a method for reaching a lesson's goals?" and "How often can you reflect on a lesson's results for planning the next lesson?"The following response options were offered: "Never", "On some occasions each semester", "On some occasions each month", "On some occasions each week", and "On several occasions each week".

Scheduled time for collaborative work
In addition, a set of questions on teachers' scheduled time for collaborative work was used.Thus, the following survey questions were used to measure the occurrence of teachers' scheduled time for collaborative work: "How often do you designate time for planning lessons with your colleagues?", "How often do you designate time for the evaluation of tasks and exercises used in lessons with your colleagues?",How often do you designate time for the development of tasks and exercises with your colleagues?" and "How often do you designate time for the development of subject-specific tasks and exercises with your colleagues?"The following response options applied: "Never", "On some occasions each semester", "On some occasions each month", "On some occasions each week", and "On several occasions each week".

Systematic and formative developmental work
To explore aspects of collegial systematic and formative developmental work, two survey questions were used to measure responses to this topic: "How often do you and your colleagues use a pedagogical model for teaching development, for example, lesson studies, collegial conversations, or research circles?" and "How often do you participate in the development of teaching tasks and exercises based on the mapping of students' results?"The first question was used to measure teachers' engagement in collegial teaching development based on research models and proven experience.The second question was used to measure teachers' engagement in formative processes, that is, developing tasks and exercises for teaching based on students' prior achievements.The following response options were applied: "Never", "On some occasions each semester", "On some occasions each month", "On some occasions each week", and "On several occasions each week".

Perceived support from principals and lead teachers in teaching development
Three survey questions were used to measure the perceived support from principals and lead teachers in teaching development: "What is the extent of the organisational support from your principal?","What is the extent of the pedagogical support from your principal?"and "What is the extent of the support from your lead teacher (or equivalent) towards developing your teaching"?The following response options applied: "No support at all", "Not much support", "A fair amount of support", and "A lot of support".

A composite measure of organised collaboration
Since it is not possible to measure different aspects of organised collaboration using a single item, a composite measure was constructed to investigate how time designated for organised collaboration was linked to other aspects of teaching.The following questions were provided to the respondents: "How often do you designate time for planning teaching with your colleagues?", "How often do you designate time with your colleagues for the evaluation of assignments and exercises used in teaching?","How often do you designate time with your colleagues for the development of tasks and exercises in specific subjects?"and "How often do you organise (schedule) collaboration with other teachers to develop tasks and exercises?"The following response options were offered: "Never", "Occasionally in each semester", "About once a month", "About once a week", and "Several times a week".A composite measure of organised collaboration was then constructed.This construction was guided by psychometric analyses based on the Rasch model (Rasch, 1980(Rasch, /1960) ) and comprised the summation of responses across the items.
To facilitate a contingency table analysis and logistic regression analyses, the composite measure was further divided into three equal-sized categories based on percentile values.The first category (concerning the least amount of time for collaboration) was labelled "Less time", while the second and third categories (involving moderate and the highest time availabilities for collaboration) were labelled "Moderate time" and "More time", respectively.

Psychometric analyses
Since the measurement invariance property is an integral part of the Rasch model, it is possible to test if the items work invariantly across different classifications of individuals who are to be compared statistically (Andrich, 1988).This means that if the data conform to the Rasch model, the use of person measures based on the summation of raw scores across items is justified.Given that the data fit the model, analysis within the Rasch framework transformed the non-linear raw scores to person values on a linear interval (equidistant) logit scale suitable for parametric statistical analyses.Based on these analyses, the composite measure of organised collaboration showed acceptable psychometric properties with good internal reliability consistency (Person Separation Index ¼ 0.77 and Cronbach's alpha ¼ 0.82).All psychometric analyses were conducted using the RUMM2030 software (Andrich et al., 2013).

Descriptive analysis
In addition to the contingency table analysis, significance tests between proportions via two-tailed z tests and binary logistic regression analyses were conducted.

Sample characteristics
Seventy percent of the participants were females, and the remainder were males.The participants' ages ranged between 21 and 64 years, with the mean age being 47.6 years (SD ¼ 10.2 years).
The majority of the participants were employed in municipal schools (81%) and the remainder in independent schools, and a larger proportion worked in compulsory schools (72%) than in upper secondary schools (28%).On average, the participants had 16 years (SD ¼ 9.7 years) of experience working as teachers (range ¼ 1e43 years).

Conditions for planning, evaluating and developing lessons
As an introduction to the theme "Time for planning and preparing", the teachers were asked to estimate if the amount of time available for planning their lessons was reasonable.This question revealed a clear split: approximately 54% of the teachers stated that the time for planning lessons was not reasonable considering their workload, while 46% stated the opposite.When asked if they had a reasonable amount of time for evaluating and assessing their teaching, the results showed that the teachers seldom found time for these activities: approximately 70% stated that the time available was not reasonable given their workload.In general, a dichotomy exists in the teacher community with regard to teachers' perceptions of reasonable working conditions.It is also clear that teachers differentiate between planning lessons and the evaluation of the outcomes of those lessons.
Table 1 shows the distribution of responses to the questions on the working conditions e namely, those related to planning, preparing and reflection e as well as whether teachers tended to undertake work outside of working hours.The first four questions were related to working conditions at the school, namely whether it was possible to plan and prepare lessons without being disturbed and to what extent teachers undertook their tasks outside of working hours.The last two questions were related to professional aspects, namely, how often teachers could reflect on their teaching methods, selecting and sequencing learning activities, and evaluating outcomes.Regarding planning, approximately 60% of the teachers reported that they were able to plan lessons without being disturbed within working hours some or several times a week.The corresponding figure for the evaluation of lessons within working hours was 45%.Furthermore, 68% of the teachers reported conducting their planning outside of working hours some or several times a week, and 62% evaluated lesson outcomes outside of working hours some or several times a week.Regarding teachers' time to reflect on the alignment of teaching methods with lesson goals, approximately 46% of the respondents reported success with this aspect some or several times a week, while 52% reported that they considered a previous lesson's outcomes for planning the next lesson.
The survey also addresses the extent to which collegial meetings are actually scheduled for PaP (Table 2).The results indicate that such formal organisation seems crucial to ensure sustainable collegial work and support professional development.However, it is difficult to estimate what may be a reasonable amount of time for scheduled meetings.Therefore, the extent that scheduled meetings occurred with regularity was considered to reflect continuity and organised support for collegial cooperation.About one-third of the teachers reported having scheduled meetings for planning lessons on some or more occasions each week.Regularly scheduled time to evaluate teaching was even more scarce.Even when the time was scheduled, it was typically not used for developing tasks and exercises for upcoming lessons.The overall pattern indicated that more than half of the teachers rarely or never attended collegial PaP meetings and that evaluating or developing tasks was not a commonly implemented activity when the teachers were provided with time for collaboration.
To qualify teachers' attitudes towards collegial work, the survey also contained items elaborating on "cost-benefit" issues, where the respondents were asked to prioritise between planning and preparing lessons individually or together with their colleagues.The results showed that 29% (626) preferred to work by themselves, 22% (479) worked with colleagues teaching the same students, and 37% (791) worked with their colleagues in subject-specific teams.We also offered an open alternative, which was chosen by 12% (258) of the respondents; they explained that they preferred having all the above opportunities (not shown in the table).
The survey contained questions regarding the frequency of teaching modelling and the formative and systematic teaching approach used by the participants (Table 3).The general pattern indicated that high teaching effectiveness does occur, albeit not in everyday teaching.Approximately 9% of the respondents reported using pedagogical models for developing teaching some or several times a week, while roughly 27% stated that they never employed such models.More than 36% of the respondents reported that they never participated in formative work developing tasks and exercises based on mapping students' results, whereas approximately 4% of the teachers reported that they participated in such exercise development on a weekly basis.
Last, we also focused on teachers' experiences with the support provided to them by principals and lead teachers or equivalents.Table 4 shows that approximately 60% of the teachers perceived nil or little organisational support from their principals with regard to developing their teaching.The reported pedagogical support from principals was even poorer.Support from lead teachers (or equivalents) did not compensate for this shortcoming.Hence, the results showed that many teachers indicated a lack of support from school leaders in connection to teaching development.

Organised time for collaborative work
To further investigate whether the time allocated for organised collaboration affects teachers' perceptions of PaP-based support, working conditions and teaching, a composite measure was constructed.This measure allowed for the examination of the relationship between the amount of collegial work and the qualitative aspects of teachers' working conditions.
The overall patterns regarding scheduled collegial work with respect to PaP revealed a consistent pattern: The more scheduled collegial work the teachers had, the more satisfied they were with the amount of time available for planning and preparing lessons.For instance, among the teachers experiencing "Less time" for scheduled collegial work, a smaller proportion (34%) were satisfied with the amount of time available for planning, while the corresponding figures for "Moderate time" and "More time" were higher at 43% and 57%, respectively.
Table 5a presents the results of binary logistic regression analysis, and it confirms the results of the contingency table analysis shown above.Thus, teachers located in the "Moderate time" category of the trichotomised composite measure of scheduled collegial  work were almost twice as likely to be satisfied with the time available for PaP compared to those in the "Less time" category.The corresponding odds ratio (OR) for the teachers located in the "More time" category was almost three (2.72) times as large as that of the "Less time" category.Table 5a also shows that male teachers were more satisfied with the time available for PaP compared to their female counterparts, that teachers employed in compulsory schools were less satisfied than those in upper secondary schools, and that teachers in municipal schools were less satisfied than those in independent schools.These patterns generally match those of the time allotted to reflect on and assess lessons.Teachers with more time tended to be more satisfied.However, it is worth noting that it is challenging to allocate time for assessments; among the teachers in the "Less time" category of the trichotomised composite measure of scheduled collegial work, a smaller proportion (18%) reported being satisfied with the time available for reflection and assessments.These proportions were higher for those in the "Moderate time" (25%) and "More time" (38%) categories.Table 5b presents the same variables subjected to a binary logistic regression analysis along with adjustments for gender, school level and school type.Teachers located in the "More time" category were more than three times as likely to be satisfied with the amount of time available for reflection and assessments compared to those in the "Less time" category.The same comparison for teachers in the "Moderate time" category resulted in an OR of 2.07.
Organised time for collaborative work depends on whether school leaders formalise this time on a daily basis.Does this formalised approach affect teachers' views on leadership for school development in general?The results indicated that it did.Teachers with more scheduled collaborative time generally reported stronger organisational as well as pedagogical support from their principals and lead teachers.For instance, 55% of the teachers in the "More time" category of the composite measure of scheduled collegial work reported "Very strong" or "Fairly strong" organisational support, while pedagogical support and support from lead teachers stood at 47% and 42%, respectively.Decreasing scheduled time for collaborative work led to a reduction in the number of teachers perceiving "Very strong" or "Fairly strong" support.Thus, given the same response options, the following results were recorded for the "Less time" category: organisational support (24%), pedagogical support (20%) and lead teacher support (18%).
Table 6 shows the results of the binary logistic regression analysis after subjecting it to the variables of organisational, pedagogical and lead teacher support and with adjustments for gender, school level and school type.The general pattern observed here implies that increasing scheduled time for collegial work also leads to a greater likelihood of teachers experiencing "Very strong" or "Fairly strong" support.Regarding principal organisational support, the pattern clearly indicated that more scheduled time for collegial work is connected to perceptions of stronger organisational support.Thus, teachers belonging to the "More time" category of the composite measure were approximately four times as likely to perceive "Very strong" or "Fairly strong" organisational support from their principals compared to those in the "Less time" category.The corresponding OR for the "Moderate time" category was approximately 1.6.The patterns for pedagogical and lead teacher support were very similar.Table 6 also reconfirms that male and female teachers experience the availability of support similarly and that the patterns are relatively independent of school level.However, wider differences were observed regarding the school type.Teachers at municipal schools perceived that there was significantly less support from their principals and lead teachers compared to those at independent schools.
Thus far, the results indicate how the infrastructural settings can be a resource or constraint for teachers to develop their teaching.Notably, the composite measure also made it possible to examine how the amount of scheduled time for collaborative work influences teachers' perceptions also on their individual conditions for PaP.
Table 7 presents a contingency table analysis of the relationships between the composite measure and PaP-related items.The results indicate the proportion of the participants who can fulfil the requirements on a weekly basis.Interestingly, the amount of time teachers devotes to scheduled collegial work seems to influence their perception of individual conditions for PaP and teaching.Teachers with more scheduled collegial work were much more likely to report that they could work without disruptions when planning (68.9%) and evaluating (56.7%) than teachers with a moderate amount of time (60.9% and 42.0%, respectively) and less time (50.6% and 32.8%, respectively).The results were arrived at using the two-tailed z tests of differences between proportions and by considering significance at p < 0.05.
We observed that planning and assessing activities were mainly performed individually and often outside of the working day.Nonetheless, the time for collegial work affects this pattern and seems to offset the teachers' workloads to some extent.This indicates that when teachers have the possibility to collaborate around PaP, this also seems to have a broader effect on how they

Table 5a
Binary logistic regression analysis of teachers' time for planning and preparing lessons (reasonable amount of time: "yes" compared to "no") with adjustments for gender, school level, and school type.Log odds (b) and Odds Ratios (OR) with 95% confidence intervals (95% CI) provided.

Table 5b
Binary logistic regression analysis of teachers' time for reflections and assessments (reasonable amount of time: "yes" versus "no") with adjustments for gender, school level, and school type.Log odds (b) and Odds Ratios (OR) with 95% confidence intervals (95% CI) provided.Binary logistic regression analysis of support from the principal and lead teacher, contrasting the "Very strong support"/"Fairly strong support" to the "Not much support"/"No support at all" categories, with adjustment for gender, school level, and school type.Log odds (b) and Odds Ratios (OR) with 95% confidence intervals (95% CI) provided.perceive their working conditions.The same pattern can be observed with regard to the results on teachers' time for reflection.Teachers with more scheduled time for collegial work reflected significantly more often on methods related to lesson goals (63.7%) and lesson results for future planning (69.1%), indicating that they were able to follow a more coherent planning approach for enacting and evaluating their lessons.

Variables
As already noted, the teachers reported limitations when working with aspects of high-efficiency teaching, such as using teaching models and conducting formative assessments.This pattern also applied to teachers with the most time for collegial collaboration.However, the results also indicated some differences.Teachers with more collegial time did use teaching models significantly more often (two-tailed z test; p < 0.05).About 16% of the respondents with more time reported using such models compared to 6.5% for those with moderate time and approximately 5% for those with less time.The same pattern, albeit a less obvious one, applied to formative assessments.Teachers with more collaborative time reported that they were engaged in formative processes, developing tasks and exercises for teaching based on mapping students' results some or several times a week (6.7%).This aspect was even rarer among teachers with moderate time (2.1%) and less time (1.3%).Clearly, this systematic and formative work is implemented extremely infrequently, regardless of teachers' relative collaborative time, suggesting that these tasks are very rarely undertaken and do not depend on the time set aside for collaboration.

Discussion
This study explored teachers' perceptions about their conditions regarding PaP as well as their attitudes towards a range of qualitydriving factors, with a focus on collegial collaboration and systematic and formative teaching.By combining teachers' perceptions of PaP and a range of infrastructural settings, we have been able to contribute with quantitative data that broaden our knowledge on the value of teachers' collegiality.These findings have relevance for anyone with interest in teachers' professional development as well as in infrastructural aspects of school development.We can establish that when there are possibilities for collaboration around PaP, this has an impact on how teachers validate their pedagogical practices and possibilities.However, our results show that schools generally do not organise collegial settings that are devoted to PaP activities.This result should, in itself, be alarming for anyone interested in the sustainability of pedagogic development as the infrastructure that surrounds PaP might determine a longterm implementation (cf.Hubers, 2020).As described by the respondents in our survey, schools seldom have an infrastructural system that works to support teachers to evaluate and develop their teaching and to integrate innovations and new knowledge when they plan their lessons or assess their students.PaP could be a part of such infrastructure and provide adequate epistemic communities that might manage and take local ownership over projects for improvement such as professional learning communities, learning studies and formative assessment.
We can further establish that access to collaborative PaP has a positive spillover effect on how teachers perceive their school leaders and their overall working conditions.This result further indicates that teachers' subject-oriented PaP should be of higher priority in collegial learning and research.
Our study highlighted PaP as a specific and qualified activity.It revealed the following results: First, this survey supports and adds to the results of previous studies both in the international and Swedish contexts; namely, that teachers perceive that they do not have sufficient time for either planning lessons or assessing students.The overwhelming majority stated that they took care of these chores in the evenings, on weekends and whenever they could.However, we extrapolated these results, as these findings correlated with teachers' experiences of deficiencies in their working conditions, with a lack of support from school management and with insufficient time to ponder over the choice of teaching methods and reflect upon the outcomes of lessons.Thus, from the teachers' point of view, schools are typically not organised in ways that can optimise the prerequisites for teaching.
A second conclusion is that teachers' attitudes do not merely indicate an inexpressible desire for more time and better working conditions.First, we can note that there is an interesting division among the respondents, where close to half of them stated that they had reasonable time for planning.A similar bisection also recurred in how teachers assessed their possibilities of working without disturbances and reflecting on the pre-and post-conditions of teaching.This calls for further research, as we could not relate this outcome to specific background variables.However, part of the explanation for this is the stable correlation between teachers' attitudes and their access to time for collegial collaboration: When the time for PaP is collegially structured, teachers can perceive factors directly related to teaching (such as time to reflect upon teaching methods and outcomes) as well as external aspects (such as the work environment and support from the school management).Our study did not permit us to determine causality; we cannot say if collegially structured PaP actually causes teachers to appreciate their workplace or if the reverse causality applies.However, the positive link is pervasive and clear.Previous studies (e.g.Konu et al., 2010) focused on broad categories and indicated that teachers' wellbeing correlates with the school level (compulsory, lower secondary and unified).Dinham and Scott (1998) showed that teachers are most dissatisfied with "system-level/societal factors" (such as status), "workload and impacts of change", and "merited promotion and local hiring".They are more ambivalent about "school-level factors" (such as organisation and leadership) and are most satisfied with factors related to the "core business of teaching" (such as student achievement).
Our research design was successfully used to examine the effects of infrastructural factors that directly affect teachers' core business, and thus, our method is perhaps more effective than one focusing on broad categories and systems that are difficult to change.Teachers with access to collegial time value the opportunity provided to them to develop good teaching to a much higher level than those teachers with individual PaP tasks.Nevertheless, more than two-thirds of the teachers stated that they could not devote reasonable time to PaP, given their workload.
However, even though the collegial time for PaP seems to point to a positive effect on teaching quality, our third conclusion is that this time is not likely to be sufficient enough to ensure high-quality and formative teaching.Time for collegial collaboration only exerts a limited positive impact on how teachers validate the frequency of systematic formative teaching, participation in the organised sharing of lesson plans and usage of teaching models.Most teachers stated that the schools they worked in did not ensure these necessary conditions on a regular basis.Thus, to create a condition for teachers to systematically engage in formative processes such as developing tasks and exercises based on mapping students' results seems to require more than setting aside time for collegial collaboration and learning.Scheduled time is a vital but not sufficient factor.
This conclusion supports the results of Darling-Hammond et al.'s (2017) meta-study of teachers' professional development.Their work indicated that to ensure high-quality teaching, both teachers and principals require regular and frequent access to coaching and expert support as well as opportunities for feedback and reflection.
Lastly, our study provides relevant insights for policymakers in that not only should educational infrastructure settings and teachers' working conditions be improved given the role they play in PaP, but due attention must also be paid to the required diversities of these tasks.PaP is not a standalone working unit.Assessing students' performance requires time on a continuous basis, while planning seems to occur when time permits.Fragmented time is arguably more difficult to use effectively for reflection and collegial collaboration.It could also be difficult to guard this planning time against interruption by other important tasks, such as ensuring contact with students and parents.Our results indicate that collaboration over PaP needs to be supported by an infrastructure that is both integrated into the school's daily routine and open to the influx of research in education so as to increase the systematic application of formative models.This also indicates that both reforms and research on professional learning communities and pedagogical innovations need to pay closer attention to the quality of teachers' collegial PaP and to the surrounding infrastructure.

Limitations
It is important to recognise that the cross-sectional nature of the data used in this study prevents us from ruling out possible errors in the directions of the observed associations.Thus, it may be that the scheduled time influenced the perceived working conditions (e.g.support from principals and lead teachers); however, it may also be that the perceived working conditions influenced the scheduled time available for collaboration.To rule out this aspect, a future study should adopt a longitudinal design.

Conclusion
This study provides a descriptive and analytical overview of infrastructure settings available for teachers to prepare and evaluate their teaching to ensure systematic and formative teaching.The basic rationale involves a) exploring teachers' conditions with regard to PaP and b) assessing whether collegial collaboration affects teachers' perceptions about the conditions related to PaP.The survey was conducted among Swedish school teachers.This largescale survey on this topic is the first of its kind in the Swedish school context.School cultures and working conditions differ between countries.In many respects, however, the situation in Sweden is comparable to that in many OECD countries (OECD, 2019).Thus, the results and conclusions of this work are also relevant to schools in other OECD countries.
The deficiencies in the infrastructure correlated with the teachers' descriptions of the conditions they experienced, prevent quality-driven teaching.Our results showed that, in general, not only is the infrastructure deficient with regard to scheduling regular collegial meetings for PaP but that it is even less common for scheduled time to actually be used to develop high-efficiency teaching, such as using teaching models and conducting formative assessments.The overall pattern indicated that efforts to promote collegial cooperation and learning are most often not organised in ways that directly support teachers in developing their lesson planning and teaching.The role of PaP in teachers' day-today practice, as well as the infrastructure to support collegial PaP, seem to be overlooked in projects and research about collegial learning.
Teachers' PaP efforts are distinct outside-of-lesson activities that would benefit from a collegial setting involving subject-oriented planning teams (Darling-Hammond et al., 2017), and the implementation of such designated collegial collaborations needs the active support of the local school management (Blossing et al., 2014).Drawing on previous research on teacher collaboration and professional development surveys, this study explored Swedish school teachers' PaP.The findings provided a descriptive and analytical overview of infrastructural settings that seems to support teachers to prepare and evaluate their teaching in ways that ensure systematic and formative teaching.
The teachers' attitudes did not indicate a mere general desire for more time but were also correlated with their access to time for collegial collaboration.The overall pattern showed that conditions for individual-based PaP could be improved by scheduling more time for collegial work.The teachers tended to use pedagogical models when developing teaching if they had more scheduled collegial time.The results, therefore, direct our attention to the importance of a stronger infrastructural foundation for systematic and formative developmental work around teaching.Joint collegial time for PaP seems to boost teaching quality and outcomes.
However, while scheduling time for collegial collaboration and learning is vital, it is not a sufficient factor.In general, teachers are interested in further education to develop both their subjectrelated and general teaching skills.We, therefore, propose establishing an infrastructure that combines collegial time for PaP with an organised influx from researchers in the education field.By its very nature, PaP is an important and multifaceted educational activity, and thus, it needs undivided attention in the context of both education-related research and school-level improvements.

Table 1
Teachers' planning, evaluation, and reflections of their lessons, percentage (number of respondents).

Table 2
Time scheduled for collaborative work involving PaP, percentage (number of respondents).

Table 3
Systematic and formative developmental work around teaching, percentage (number of respondents).

Table 4
Support from principals and lead teachers for developing teaching, percentage (number of respondents).

Table 7
Time devoted for collaborative work in relation to teachers' own time for planning and preparing lessons and implementing formative work (on a weekly basis) (units: percentage; number of respondents in parentheses).