Physical education preservice teachers’ perceptions of the subject and profession: development during 2005–2016

ABSTRACT Background: During the socialization process when becoming a physical education (PE) teacher, the knowledge, perceptions and expectations of what it means to work as a teacher are developed. In this socialization, the initial acculturation phase is shown to be of the most importance, since individual PE teachers’ experiences during this phase are shown to have a long-lasting influence on their approach to and perception of the subject and the profession. Furthermore, research shows that most physical education teacher education (PETE) programmes are ineffective in altering these initial perceptions and beliefs during the programme. This inertia to change may resemble Bourdieu’s concept of habitus. Purpose: The purpose of this study was to analyse the background of PE preservice teacher students (PSTs) and examine their embodied perceptions and beliefs related to the subject and profession when they enrol. Specifically, the study focuses on their background characteristics, perceptions of PE and PE teachers, and whether their background and perceptions changed between 2005 and 2016. Method: This study draws on a web-based questionnaire completed by 224 students (90 women and 134 men) enrolled in the PETE programme at a major university in Sweden between 2005 and 2016. The questionnaire used in this study addressed the PSTs’ experiences, views, beliefs and perceptions of PE and the PE profession, and it was completed during the first semester of respective students’ PE subject studies. Findings: PE PSTs are a homogeneous group of students with similar backgrounds, experiences and perceptions of PE and their future profession as PE teachers. Participants suggested that important characteristics for a good PE teacher include possessing subject knowledge, having pedagogical competence and being considerate. A good PE lesson should be fun and inspiring, consist of physical activity and be adapted to all. Important goals for PE are to develop pupils’ character and promote healthy behaviours. The PSTs’ background characteristics and perceptions do not seem to have changed during the studied period, in spite of the fact that the structure of the PETE programme did change. Conclusions: The homogeneous background among PSTs, with vast experience of sport and physical activity, implies that they will interact and engage with students with similar backgrounds and perceptions (i.e. habitus) during PETE. This may limit the potential influence of PETE and fail to prepare PSTs for the demands of their future profession. However, if the influences of acculturation were accounted for during PETE, the programmes could be better designed and better prepare PSTs for their future profession.


Introduction
Sport and physical activity are often seen as solutions to various problems in society, and the participation in them is viewed as an effective means of improving individual life quality and contributing to personal development (Coakley 2011). As part of this reasoning, physical education (PE) in school is also linked to various positive effects. According to Dagkas and Stathi (2007), PE teachers are central to pupils' experiences of the subject. The PE teacher and PE lessons are important in this regard as pupils will be able to acquire knowledge and experience of different sport and physical activities during PE lessons (Stuij 2015) and participate in a wide range of activities that could cause them to develop a lifelong interest in sport (Åström 2013).
In relation to PE teachers, it is widely recognized that individuals' experiences during upbringing from school and leisure-time activities play an important role in recruiting students to teacher education programmes (cf. Brown 2005;Curtner-Smith, Hastie, and Kinchin 2008). It is also recognized that these earlier experiences influence the development of preservice teachers' (PSTs') approaches to teaching. In fact, some researchers indicate that what is learned about teaching in school as pupils has a stronger impact on PSTs' approaches to teaching than formal teacher education programmes (Zeichner and Gore 1990). In the long term, these early experiences influence teachers' perceptions of the purposes of education and the teacher profession (Templin and Schempp 1989), which in turn influence how the actual teaching and lessons will be carried out.
Specifically writing on the topic of PE teacher recruits, Dowling (2011) and McCullick et al. (2012) point out that it is critical for those involved in physical education teacher education (PETE) programmes to identify and understand their students' backgrounds, experiences and perceptions when entering PETE in order to design effective PETE programmes. Or as Ralp and Mac-Phail (2015, 62) put it in an Irish context, 'PETE faculty can build upon these PSTs' positive perspectives to develop effective school PE programmes, acknowledging that they may need to deconstruct and/or supplement the PSTs' lived experiences with new activities and pedagogies that are suited to the current needs of schools'. This study answers this call by analysing the backgrounds and perceptions of Swedish PE PSTs.

Previous research on PSTs
In a Swedish context, Larsson (2009) states that PE teachers' views of PE and their future profession are influenced by those beliefs and perceptions that have been embodied during the socialization process of becoming a teacher. By participating in different social fields during this process, PE teachers develop what Bourdieu calls a habitus that influences their desire to enter the profession. The habitus will also have an impact on how the teacher education will be received (cf. Mordal-Moen and Green 2014) and ultimately how the PE teachers will act in the future. Following Bourdieu's theoretical framework, an individual's habitus is formed by earlier experiences from participating in different social fields and is defined as '[a] system of durable, transposable dispositions, structured structures predisposed to function as structuring structures' (Bourdieu 1990, 53). An individual's habitus is relatively stable over time (but may develop) and is among other things expressed through a teacher's perceptions and behaviours. On this topic, Brown (2005) and Green (2002) argue that when students enter PETE, they arrive with a particular habitus as well as particular views and dispositions regarding the purpose and teaching of PE.
The habitus is mainly developed during childhood (Bourdieu 1990), when the primary social fields are the family, school and leisure-time activities, and where important social agents influence the development of the habitus. Similar to international research on background characteristics of PETE recruits (cf. McCullick et al. 2012;Richards, Templin, and Graber 2014;Rossi, Sirna, and Tinning 2008;Templin and Richards 2014), Larsson (2009) shows that Swedish recruits have vast experience of participating in competitive sport settings. PE PSTs are also often from families who are largely involved in sport (Lawson 1984), and their career choice has been influenced by important social agents such as former coaches and PE teachers, parents and peers (Annerstedt 1991;Curtner-Smith, Hastie, and Kinchin 2008;Spittle, Jackson, and Casey 2009). No major changes seem to have occurred in this respect over time, and PSTs can therefore be said to consist of a relatively homogeneous group of students with similar background and experiences (McCullick et al. 2012). Pike and Fletcher (2014) together with Woods, Richards, and Ayers (2016), however, indicate that there has been a change in the type of students admitted to PETE programmes, with more students interested in the area of health and fitness. On the same topic, McCullick et al. (2012) show a change in recruits' motivations for pursuing the PE profession, with more recruits expressing an interest in physical activity participation over traditional sport. With this in mind, faculty members at PETE programmes must take into account that the PSTs of today may have an alternative view of PE and their role as PE teachers than those of previous generations. Zeichner and Gore (1990) indicated that teacher education programmes in general do not effectively alter the beliefs and dispositions that recruits have acquired before entering the programmes. Research has found that recruits enter PETE programmes with well-grounded dispositions about what it means to be a PE teacher, and PETE programmes fall short in changing these beliefs (Adamakis and Zounhia 2016; Mordal-Moen and Green 2014). Brown (2005) even indicates that PETE instead may reinforce these initial dispositions. PSTs' former experiences and impressions thus have a traceable impact and strong influence on teachers' future beliefs and practices (Curtner-Smith, Hastie, and Kinchin 2008). However, not all teacher education programmes are ineffective, and some studies have found that, under the right set of circumstances, PETE can successfully challenge PSTs' beliefs about teaching PE (cf. Graber 1996;Levin and He 2008;Richards, Templin, and Graber 2014). On this topic, McCullick et al. (2012) state that in order to alter PSTs' dispositions and design effective PETE programmes, it is therefore essential to have knowledge of PSTs' backgrounds, beliefs and perceptions (cf. Dowling 2011 for a similar reasoning in the case of Norwegian PETE).

Theoretical framework
This study draws on occupational socialization theory as delineated by Lawson (1984) and Templin and Schempp (1989), who describe the general socialization of teachers in three phases: acculturation, professional socialization and organizational socialization. Through these phases, teachers develop knowledge, and embodies a habitus, which has implication on their perceptions and expectations on what it means to work as a teacher. Studies that focus on socialization in relation to the teaching profession refer broadly to 'that field of scholarship which seeks to understand the process whereby the individual becomes a participating member of the society of teachers' (Zeichner and Gore 1990, 329). According to McCullick et al. (2012, 180), using the occupational socialization theory is useful since it allows 'researchers to consider how prior familial and educational experiences inform majors' beliefs and conceptions of teaching PE and the possible implications on teacher education programs'. In the following, all three phases of the socialization are described briefly, although the paper mainly focuses on the outcome of the acculturation phase.

Acculturation
Acculturation is the phase prior to the formal decision to pursue teacher education (Lawson 1984;O'Bryant, O'Sullivan, and Raudensky 2000). During this phase, early experiences form the basis of role identities and the ways individuals envision themselves as teachers. This is an important phase of socialization in the development of PE teachers and has an impact on the shaping of future teachers' perspectives and beliefs (Pike and Fletcher 2014;Richards, Templin, and Graber 2014). Influenced by significant social agents, prospective teachers during this phase learn and develop understandings of what it means to be a good PE teacher . The perspectives and beliefs are embodied in the PSTs' habitus, and since the habitus is long lasting and resistant (but not impossible) to change (cf. Brown 2005), it has influence on teachers' perceptions of the purposes of education and the teacher profession (Richards, Templin, and Graber 2014;Templin and Schempp 1989).

Professional socialization
Professional socialization begins when PSTs enrol in a teacher education programme (Lawson 1984). During this phase, recruits learn about the important knowledge and skills demanded by the profession of PE teaching (Lawson 1984;Richards, Templin, and Graber 2014;Sirna, Tinning, and Rossi 2010). Most PETE programmes, however, are shown to be ineffective in changing PSTs' previous perceptions and that PSTs' embodied beliefs (i.e. habitus) instead remain unchanged (Graber 1996).

Organizational socialization
Organizational socialization refers to workplace socialization and begins when PSTs enter the work force as teachers. This socialization phase is ongoing throughout a teacher's career and has a powerful impact on teachers' development as teachers and their practice (Richards, Templin, and Graber 2014). When entering this phase, teachers may experience reality shock in their encounters with the actual requirements of the profession, and the lessons learned during PETE are washed out (cf. Christensen et al. 2018). However, as Richards, Templin, and Graber (2014) point out, there has to be something to wash out, which in light of previous research showing that PETE is ineffective may not always be the case.

Knowledge learned from teacher socialization research
As noted by Templin and Schempp (1989), occupational socialization is a dialectical process, meaning that individuals have an active role in the actual socialization (cf. Richards, Templin, and Graber 2014). Drawing on the concept of habitus, this could be understood as the new experience and knowledge gained during the professional socialization phase, for instance, being either accepted or rejected depending on how it fits the embodied dispositions of a PST. In summary, researchers on occupational socialization in PE suggest that it is important for the PE profession (cf. McCullick et al. 2012) and may, if considered, have an impact on the design and outcome of PETE programmes (cf. Dowling 2011; Pike and Fletcher 2014). According to Richards, Templin, and Graber (2014), many of the studies on the backgrounds of PE PSTs are several years old. The results of these studies are not absolute, as school cultures and contexts differ and change over time. It is therefore possible that PE now attracts students with different backgrounds and perspectives from those studied in the past. Mordal-Moen and Green (2014) also point out that there is a lack of studies in this area from a Scandinavian perspective. Consequently, McCullick et al. (2012) and Richards, Templin, and Graber (2014), among others, indicate the need for new studies on teacher socialization that illuminate the new conditions. In addition, Adamakis and Zounhia (2016) and Pike and Fletcher (2014) conclude that most studies on socialization of PE PSTs have been qualitative with a lack of longitudinal perspective, and they highlight the urgent need for quantitative studies. This study contributes to the research field through a longitudinal quantitative study conducted in the Swedish context.

Contextualization of PETE in Sweden
PETE at the large Swedish university in the study consists of a five-and-a-half-year programme (330 European Credit Transfer System [ECTS]), during which PSTs study general education courses, two school subjects and sequential practicums and conduct a final research project (Romar, Åström, and Ferry, forthcoming). Between 2005 and 2010, when the first PSTs in this study were enrolled, the first semester of the programme consisted of PE subject studies. Due to a change in the structure of the teacher education programme, those PSTs who enrolled between 2011 and 2106 started the PETE with a first semester focusing on an introduction to the teacher profession and general education studies (30 ECTS). During this semester, courses were taken together with PSTs with other subject orientations.

Purpose of the study
The purpose of this study was to analyse the backgrounds of PETE students and their embodied perceptions and beliefs related to the subject and profession when they enrol. The specific questions that the study intends to answer are . What background characteristics do students enrolled in PETE have? . What perceptions about PE and their future profession as a PE teacher do PSTs have at the beginning of the PETE programme? . Have PSTs' backgrounds and perceptions changed between 2005 and 2016, and if so, how?

Methodology
Data for this study derive from a larger longitudinal research project that addresses the background, experiences and beliefs of newly enrolled sport students at a major university in Sweden. The project uses a web-based questionnaire completed by students enrolled at a sport science and a PETE programme between 2005 and 2016. This specific study draws on the PETE students' answers to the questions considering their experiences, views, beliefs and perceptions of PE and the PE profession. In total, 224 PE PSTs answered the questionnaire (90 women and 134 men).
The questionnaire was answered by students during the first weeks of their first PE subject matter courses. Between 2005 and 2010, this was during their first semester in the PETE programme (n = 153); after a new PETE programme structure was implemented, between 2011 and 2016 this was during their second semester (n = 71). From the large questionnaire, this study uses the questions with predefined answers addressing the PSTs' backgrounds and former and present experiences in sport and physical activity. In addition, the question asking how important 19 specified goals would be in their future teaching in PE is used in this study. The assessment of how important the 19 goals would be in their future profession was answered on a 4-point scale from not important at all (1) to very important (4). Finally, the study included two open-ended questions highlighting the PSTs' perceptions about PE and their future profession, which asked the PSTs to describe characteristics of a good PE teacher and a good PE lesson. The questionnaire was not part of a graded assignment, was answered anonymously and began with provision of consent to participate in the research project.

Data analysis
The data were analysed using IBM SPSS Statistics version 24.0 for calculation of frequencies, chisquare tests, Fisher's exact test and t-tests. Since not all questions were answered by all PSTs, slightly different numbers of answers were included for each analysis. The statistical analysis of the 19-specified goals was before calculations first thematised into 6 main themes based on the meaning of the goals. Figure 1 shows the thematization of the 19 questions into the 6 themes.
Before calculations, the open-ended questions were the first subject to content analyses following three steps. In the first step, all answers were read thoroughly a number of times to form an opinion of the material and to find regularities and recurring themes (Miles and Huberman 1994). Stake (1995) describes this as an intuitive process that requires the researcher to determine what information is linked, determine what regularities in the information can be used to form categories and, after the categories are created, fill them with content. After this first rough categorization, the material was further processed during the second step. During this step, the categories were either merged to make their meaning somewhat broader or clarified to reduce their meaning (Miles and Huberman 1994). The main purpose of this process was to make the emerging categories independent of each other, exhaustive and sharply different and to ensure that as few responses as possible would be added to the 'other' category. In the final step, the categories that had emerged were related to existing research and other contextual issues in PE and PETE.

Findings
The findings begin with a description of PSTs' background characteristics, followed by a presentation of their perceptions of a good PE teacher and PE lesson, and assessments of a number of specified goals for PE in the future.

PSTs background
Unlike Swedish teachers in general, but consistent with national statistics on upper secondary PE teachers (Skolverket 2017) and previous research on Scandinavian PETE recruits (Dowling 2011;Larsson 2009), Table 1 shows that a larger proportion of the students enrolled in the programme are men, with no difference between the cohorts (p > .05). Furthermore, similar to Dowling's (2011) and Larsson's (2009) results, the PSTs are relatively young when they enrol (m = 22 years), and women are significantly younger (m = 21.3, s = 2.61) than men (m = 22.5, s = 4.13). The relatively young age among PSTs enrolled in the PETE programme implies that they probably have gained little work experience before  entering PETE. This could mean that a large part of their beliefs and perceptions of PE and the future profession has been developed and incorporated in their habitus through experiences in school and leisure-time activities during their upbringing. That is, during a time in their lives that according to Zeichner and Gore (1990) has a strong impact on PSTs' approach to teaching. Following earlier results on PE recruits' background characteristics (cf. Green 2002;Larsson 2009;McCullick et al. 2012;O'Bryant, O'Sullivan, and Raudensky 2000;Pike and Fletcher 2014;Richards, Templin, and Graber 2014;Rossi, Sirna, and Tinning 2008;Templin and Schempp 1989), PSTs in this study had vast experience participating in sport and physical activity (Table 2). Almost all of the participants had been members in a sport club during their upbringing, and nearly 70% were still members when they enrolled in the programme. This applies to both male and female PSTs and does not differ between the cohorts (p > .05). About half of the PSTs had participated in school sport, which in a Swedish context means that they had elite sport ambitions during their upper secondary school education (Ferry, Meckbach, and Larsson 2013).
Looking closer at the kinds of sport and activities PSTs participated in while enrolled in the programme, there are no major differences between the two cohorts. The largest share of PSTs exercised in strength/weight training on a regular basis, followed by team sport activities and exercise activities (such as running and dance). This result follows Swedish sports habits in general, and previous research showing that team sports are particularly popular among PE PSTs (Larsson 2009;McCullick et al. 2012). In relation to sex, there were some differences in sport and exercise habits among the PSTs. In total, a significantly larger proportion of the male PSTs were participating in individual sport (37%) compared to women (23%); meanwhile, a larger proportion of women PSTs were participating in exercise activities (64%) and group-training activities such as aerobics, step-up and body pump (41%) in comparison with men (52% vs. 23%). These significant differences in sport and exercise habits between men and women are primarily found among PSTs enrolled in 2005-2010, but they remain true for group-training activities for those enrolled 2011-2016.
In relation to background characteristics, the results describe a homogeneous group of PSTs enrolled in the PETE programme in the study, with extensive experience of participation in sport in a competitive setting. Through the participation in these activities, the PSTs beliefs and perceptions on PE have developed and been an embodied part of their habitus, and thus are influential on their future profession as PE teachers (Mordal-Moen and Green 2014; Templin and Richards 2014). The results also show that the participation differed in relation to sex, following traditional gender patterns. This could, when not addressed sufficiently during PETE, result in a further reinforcement of the existing gender norms in society and PE (Brown 2005;Green 2002).

Perceptions of a good PE teacher
PSTs could enter at most three answers to the open-ended question about important characteristics for a good PE teacher. Through the analysis, eight categories were generated (Figure 2). A good PE teacher should: have subject knowledge (mainly theoretical), have pedagogical competence (e.g. ability to explain to everyone, didactical skills), be considerate (e.g. be emphatic, see and listen to all pupils), be a motivator (e.g. engage pupils, motivate all pupils), be passionate for PE (e.g. committed to PE and sport, motivated), have social competence (e.g. good social ability, easy to talk to), be a role model (e.g. set an example, practice what you preach) and possess a group of other characteristics (interesting, flexible, fun, kind, etc.).
The analysis reveals that there are no significant differences in the overall patterns between the two cohorts (p > .05), and the highest proportion of PSTs indicated that a good PE teacher should possess subject knowledge (48%), have pedagogical competence (47%) and be considerate of all pupils' needs (47%). In comparison with earlier studies, these three characteristics were not mentioned as important for PE teachers in Pikes and Fletcher's (2014) review of the research in the area; however, having a passion for PE, serving as a role model and other characteristics such as being creative and flexible and being fit were mentioned. The reason to why PSTs in this study value these specific characteristics higher than previous studies is not obvious, however, the results coincide with O'Bryant, O'Sullivan, and Raudensky's (2000) qualitative study on a non-traditional group of prospective PE teachers (i.e. older, more diverse backgrounds), who were more committed to teaching children PE and to motivating all pupils to participate. The focus on subject knowledge and pedagogical knowledge also coincide with Dowling's (2011) results, where PSTs emphasized these as attributes of a good PE teacher. One possible explanation for why these forms of knowledge are valued highly among PSTs in this study could be attributed to the Swedish context, where PE during recent decades has developed to also contain some theoretical lessons, which requires subject knowledge and pedagogical skills among teachers. This could imply that the PSTs have experienced this during the acculturation phase and thus embodied this perception.
In terms of the proportion of answers categorized in each category, there are significant differences (p < .05) between the cohorts, with a larger proportion of answers belonging to the category of pedagogical competence among those PSTs enrolled in 2011-2016 (58%) compared to PSTs enrolled in 2005-2010 (42%). One possible explanation for this difference could be attributed to the changed PETE structure at the university, where the importance of pedagogical competence has been addressed (at least briefly) during the initial semester focusing on general pedagogy and the teacher profession. For the other categories, there are only minor differences in the proportion of answers (p > .05). When comparing women PSTs' answers with those of their male peers, there is a significant difference in the proportion of answers belonging to each category (p < .05). The most common answers among men belonged to the categories of pedagogical competence (52%), considerateness (50%) and subject knowledge (47%), and among women the most common answers were ability to motivate (49%), subject knowledge (48%) and considerateness (46%). In order to fully explain the differences in relation to sex, further investigations are needed that go beyond the purpose of this paper.
The answers given by the PSTs on characteristics for a good PE teacher provide information on their perceptions of their future profession (cf. Dowling 2011). A large proportion of these characteristics are not specific to PE teachers, but instead apply to teachers in general. The small differences in views of good PE teachers between PSTs in the two cohorts indicate that they are a homogeneous group of students and that the changed programme structure of PETE had only minor effects on these perceptions.

Perceptions of a good PE lesson
On the question about characteristics of a good PE lesson, the PSTs could also enter at most three answers. The analysis generated eight categories: a good PE lesson should be fun and inspiring, consist of physical activities, be adapted to all pupils, develop pupils' skills, be well planned, offer theory and reflection, consist of various content, and be creative and imaginative (Figure 3). These characteristics mainly coincide with the results of earlier studies on the purpose of PE (cf. Green 2002;Larsson and Redelius 2008;Larsson 2009;O'Bryant, O'Sullivan, and Raudensky 2000;Pike and Fletcher 2014).
There are no significant differences between the two cohorts in the proportion of answers to this question categorized in each category (p > .05). The largest proportions of answers belong to the categories fun and inspiring lesson (52%), physical activity (40%) and adapted to all pupils (30%). That PE lessons should be fun and inspiring is also put forward as characteristic for a good lesson by Green (2002) and presented in Larsson and Redelius's (2008) review of Swedish research in PE. This could be seen as a result of PSTs' experiences of PE and a desire to teach future pupils to consider lessons positive and interesting. The large proportion of PSTs indicating that a good lesson should consist of physical activities is quite obvious, considering the subject's historical origin as a practical subject, with a focus on doing physical activities (Larsson and Redelius 2008). The reason that a large part of the PSTs answering that a lesson should be adapted to all pupils could be explained by it being a requirement written in the Swedish curriculum for PE and something PSTs probably have been told during their own schooling, while they themselves have experiences that it has not always been the case in their PE lessons.
There are no differences in the characteristics of a good PE lesson between women and men in total or among PSTs enrolled in 2011-2016 (p > .05). However, among those enrolled in 2005-2010, there is a significant difference in the number of PSTs who named varying content as characteristic of a good PE lesson, with a larger proportion of women (21%) in comparison to men (6%). The reason for this difference could perhaps be that women are attracted by a wider range of activities compared to men, however, to ensure this result an in-depth analysis that goes beyond the scope of this paper is required.
Overall, the small differences are shown in relation to perceptions of a good PE lesson also indicate that PSTs are a homogeneous group of students and that the perceptions not have changed during the studied period of time.

Importance of specific goals in PE
The PSTs' assessments of how important the 19-specified goals will be in their future profession as PE teachers are presented in the 6 themes based on the general meaning of the question earlier described in Table 1. The result shows that there are no significant differences between the two cohorts in PSTs' assessments of how important the different themes will be (p > .05). Table 3 shows that the most important theme is to develop pupils' character (m = 3.74, s = 0.30), which implies that the main goal is to develop and influence their future pupils' character in a positive direction. Within this theme, the specific goals that are most valued are to increase pupils' selfconfidence (m = 3.9), a goal also put forward as important in previous research (cf. O'Bryant, O'Sullivan, and Raudensky 2000;Pike and Fletcher 2014). Also, highly valued in this theme is the goal of developing pupils' social competence (m = 3.7). The goals within this theme are not specific to the PE subject, as they aim to develop students more generally and develop characteristics that students may benefit from for their whole lives.
Similar to McCullick et al.'s (2012) finding, the PSTs also assessed the theme of health promotion as important (m = 3.66, s = .28), which could either affect pupils in the moment or prepare them with a perspective for the future. The specific goals in this theme that ranked highest were to let pupils experience the pleasure of movement (m = 3.9) followed by encouraging and developing pupils' lifelong interest in physical activity (m = 3.8), both of which have been presented in earlier research by Pike and Fletcher (2014). In general, the goals associated with this theme are also prospective.
The themes of facilitating education in general (m = 3.57, s = .38) and activity-centred goals (m = 3.45, s = .35) were rated slightly lower than the two previous themes. Within the theme of facilitating education in general, PE is seen as a support to schoolwork in general (cf. Green 2002)  are meant to come immediately and are not directly intended to influence pupils' future attitude and development. The specific goals assessed highest in this theme are to offer pupils an opportunity to participate in recreation and have fun (m = 3.9). The theme of activity-centred goals (m = 3.45, s = .35) is partly about developing the specified skills described in the curriculumswimming (m = 3.7) and teaching pupils how to perform outdoor life (m = 3.6)and partly about providing pupils with experience and knowledge of different sport and physical activities that can easily be performed in leisure time and later in life.

and the effects
The theme of curriculum goals in general (m = 3.04, s = .52) partly coincides with activity-centred goals; however, with the distinction that the specific goals in this theme are more general: achieving curriculum objectives (m = 3.4) and creating interest in environmental issues (m = 2.6). In this theme, there are differences between women in the two cohorts with a significantly higher mean value among women enrolled in 2011-2016 (m = 3.4) than those enrolled in 2005-2011 (m = 3.0). This could perhaps also be attributed to change PETE structure, where those PSTs enrolled later have treated the importance of curriculum during their initial semester. There is also a significant difference between women and men enrolled in 2011-2016, with a higher mean value among women (m = 3.4) compared to men (m = 3.0). The reasons for this difference are not obvious, but require further analysis. The goals connected to this theme relate specifically to developing knowledge clearly linked to the PE subject.
The theme of elite sport goals was assessed by the PSTs as the least important for their future profession (m = 1.86, s = 0.64). Within this theme, the two specific goals PSTs assessed were developing prospective elite athletes (m = 1.8) and teaching pupils to compete (m = 1.9). One explanation for the lower assessment of this theme may be due to the Swedish sports model, which means that there is a distinction between health-oriented PE in schools and competitive sports activities in voluntary sports clubs during leisure time (Ferry, Meckbach, and Larsson 2013). In this theme, there was a significant difference between men and women in how important the goal was going to be in their future profession, with a higher mean value among men (m = 2.0) compared to women (m = 1.7). One possible explanation of this difference could be due to the different activity patterns between men and women earlier presented in the result, with higher participation rates in competitive sports among men. When comparing the cohorts, this significant difference existed among those PSTs enrolled in 2005-2010 but not among those enrolled in 2011-2016. The goals in this theme are not specific to competence in PE or to physical activity in the future; instead, these goals are related to the competence pupils need to participate in competitive sport activities.
The results of PSTs' assessments of the specified goals in their future profession show that there are only minor differences between the two cohorts. The most highly valued themes are not specific to knowledge or competence related to the subject of PE. Instead, they are expressions of competencies required to be physically active in a lifelong perspective and to develop pupils' attributes or facilitate education in general (cf. Green 2002). That is, goals connected to education in general and/or problems to which sports are seen as a solution (cf. Coakley 2011). The only exception is those goals specifically linked to the Swedish curriculum, however, those goals were generally ranked lower by the PSTs.

Conclusions
The results of this study show that PE PSTs at this Swedish University are a homogeneous group of students with similar backgrounds, perceptions of PE and their future profession as PE teachers and that no major changes have occurred in this regard during the studied period. This in spite of the fact that the structure of the PETE programme changed and therefore the PSTs' previous knowledge of the teacher profession and general education therefore ought to have changed. This result can be explained by the strong influence habitus have on PSTs perceptions, indicating that these are mainly based on their previous experiences of PE and sport and that PETE only have limited influence.
Drawing on Green's (2002) notion of PE teachers, PE PSTs' perceptions of the subject and their future profession represent a combination of their previous embodied dispositions and the practical context they are a part of. The homogeneous backgrounds among PSTs imply that during PETE, they will interact and engage with students with similar backgrounds and experiences, and thus with similar perceptions and habitus. Since PSTs will mostly interact with other students with similar experiences and perceptions during the professional socialization phases, it is possible that these embodied perceptions will not be challenged and instead be reinforced (cf. Adamakis and Zounhia 2016). This may limit the potential influence of PETE and fail to prepare PSTs for their future profession, where they will have to deal with pupils from a large variety of backgrounds and experiences. In conjunction to this, Dowling (2011) emphasizes that society and thus teachers' work have undergone major changes in recent years, and that PE teachers therefore are confronted by new demands and challenges resulting from today's multicultural society.
In order to prepare PSTs for the demands and challenges of their future profession, PETE programmes therefore need to be better designed and take into account the effect of acculturation on PSTs' perceptions and beliefs. To change the recruitment patterns to PETE programmes in Sweden and revising the design and content is a long process. However, one direct suggestion would be that PSTs during PETE have multiple opportunities to discuss and reflect on their background and experiences, and the views and dispositions they bring to the education. By doing this, it would be possible to have a larger influence on PSTs' development and prepare them to 'meet the needs of tomorrow's PE teachers' (Dowling 2011, 203).
Finally, based on the result of this study, and with the mixed results of previous presented studies on the impact of PETE in mind (cf. Mordal-Moen and Green 2014; Richards, Templin, and Graber 2014), there is a need for further research conducted during the professional socialization phase to illuminate the development of PSTs' perceptions of characteristics of a good PE teacher, a good PE lesson and the purpose of PE in school. The knowledge created in those studies could also be used to develop and improve existing PETE programmes.

Disclosure statement
No potential conflict of interest was reported by the author.