Pre-service teacher research: a way to future-proof teachers?

ABSTRACT Internationally, the attention to pre-service teacher research and inquiry in teacher education is growing. In the Netherlands, pre-service teacher research has been a compulsory component of primary teacher education for a decade. The assumption is that such research can help ‘future-proof’ teachers. This study examines the relationships among the quality of inquiry, the quality of teaching and pre-service teachers’ perception of this research. Scores for assessments of graduating pre-service teachers (N=650) and a survey (n=236) were used as measurements. The findings indicate positive perceptions of practitioner research and a positive correlation between the quality of inquiry and quality of teaching. Using these data, the study identifies four profiles of pre-service teachers, differentiated by their perceived learning outcomes.


Introduction
In recent decades, attention to various forms of pre-service teacher research in teacher education has grown internationally. The assumption is that knowing about research and conducting practitioner research by oneself strengthens the quality of teaching. Using practitioner inquiry as a professional learning strategy can help pre-service teachers, teachers and teacher educators become more aware of setting and achieving goals and substantiate their efforts by relying on scientific knowledge produced by others (Darling-Hammond 2017;Livingston and Flores 2017;Menter, Peters, and Cowie 2017). However, research shows that many pre-service teachers have a negative attitude towards conducting research (e.g., Joram 2007;Ponte, Beijard, and Ax 2004;Puustinen et al. 2018;Ulvik 2014) and offer several reasons for their negative attitudes: (1) preservice teachers would rather spend time practising teaching than conducting research; (2) preservice teacher research demands too much cognitively and therefore causes stress; and (3) it is a compulsory exercise that involves assessment. Moreover, Grossman's (2005) review study referring mainly to studies situated in the United States revealed 15 years ago that little empirical evidence suggests that doing research, as is taught in teacher education, leads to more effective (prospective) teachers.
More recently, studies about pre-service teacher research in various countries have shown value in pre-service teacher research (e.g., Aspfors and Eklund 2017 (Finland)

The purpose of pre-service teacher research
Pre-service teacher research in Dutch primary teacher education can be viewed as a noncollaborative form of practitioner research or inquiry. Extant studies of practitioner research and its purpose tend to assume that the continually changing society requires professional teachers who are lifelong learners, who can pose and address emerging challenges and problems that do not have existing answers Munthe and Rogne, 2015). Ideally, these professional teachers are engaged in practitioner research; they know how to integrate and link different kinds of knowledge to the complex problems of schools and classrooms. Practitioner research (or practitioner inquiry) 'refers to a variety of educational research modes . . ., including action research, teacher research, narrative inquiry, . . . and the use of teaching as a context for research' (Cochran-Smith et al. 2009, 18). In the Dutch language the distinction between research and inquiry does not exist, so 'Inquiry is a process of systematic, rigorous and critical reflection about professional practice, and the contexts in which it occurs, in ways that question taken-for-granted assumptions. Its purpose is to inform decision-making for action' (Reid 2004, 4). In contrast, the use of a variety of quantitative and qualitative research methods, and research literature, as well as the contributions to the knowledge base accessible to other researchers, is essential to research but optional for inquiry. Practitioner research shows vast similarities to the scholarship of teaching and learning (SoTL), though this focuses mostly on practitioners in higher education (Gilpin 2011). Both practitioner research and inquiry are a form of professional development aimed at understanding and improving practices within the teacher's local context from an insider perspective (Borko, Liston, and Whitcomb 2007;). It offers a range of potential outcomes, such as improving teaching practice (e.g., Ermeling 2010;Pareja Roblin et al. 2014), increasing teachers' knowledge and understanding of students (e.g., Butler and Schnellert 2012;Elm and Nordqvist 2019;Jacobs, Yendol-Hoppey, and Dana 2015;Willegems et al. 2018), and fostering teacher empowerment and transformation (e.g., Dana and Yendol-Hoppey 2019;Esposito and Smith 2006). Learning how to conduct practitioner inquiry in teacher education can lead to an increased critical, reflective and innovative attitude and a better understanding of scholastic culture; it can also prompt participation in professional learning communities (Cochran-Smith and Lytle 2009;Dobber et al. 2012;Willegems et al. 2018).
In addition to developing pre-service teachers' research skills, teacher educators can directly influence perceptions of and attitudes towards practitioner research (e.g., Aspfors and Eklund 2017;Baan, Gaikhorst, and Volman 2019b;Maaranen and Krokfors 2008;Munthe and Rogne, 2015;Van Der Linden et al. 2015). The main aim of teacher education is to educate future-proof teachers with an inquiry stance (Cochran-Smith and Lytle 2009; Dana and Yendol-Hoppey 2019), who can work in inquiry-based contexts and use literature or conduct practitioner research or inquiry to reflect on their own practices or those of their school organisation (Baan et al. 2019a). Inquiry-based working teachers show an inquiry habit of mind and contribute to a culture of inquiry at the school and classroom levels (Uiterwijk-Luijk et al. 2019). Characteristics of this inquiry habit of mind include being critical, curious and willing to share, as well as wanting to achieve deep understanding and improve one's own practice (Earl and Katz 2006;Van Der Rijst 2009;Van Katwijk et al. 2019a).
Although most teacher educators endorse the value of pre-service teacher research, a considerable number of pre-service teachers seem sceptical of its relevance for, and direct use in, the teaching profession (Puustinen et al. 2018;Reis-Jorge 2007;Ulvik 2014;Van Katwijk et al. 2019a). The intention to conduct practitioner research in a future job is related to perceptions and attitudes towards it (Van Der Linden et al. 2015;Vrijnsen-de Corte et al. 2013). Van Der Linden et al. (2015) state that beliefs about research are equivalent to perceptions of its value, which are more influential than knowledge and therefore strong predictors of future behaviour.

Perceived and actual learning outcomes
Extant studies do not explicitly focus on actual learning outcomes of pre-service teacher research. Yet four general findings of perceived learning outcomes emerge, mostly based on self-reports. First, most authors consider professional and personal development in general a learning outcome (e.g., Aspfors and Eklund 2017;Ion and Iucu 2016;Niemi and Nevgi 2014;Råde 2014). Taylor (2017) explicitly mentions the development of a teacher researcher identity, when the teacher educator uses narrative, pedagogical stories and take the role of teacher researcher. Others mention the development of an inquiry stance, including critical reflection, curiosity and wanting to share findings (e.g., Råde 2014;Uiterwijk-Luijk et al. 2019;Ulvik, Riese, and Roness 2017;Van Katwijk et al. 2019a). Second, studies often refer to knowledge about research and professional knowledge on various educational topics (e.g., Gray 2013;Kowalczuk-Walędziak et al. 2019;Munthe and Rogne, 2015;Van Katwijk et al. 2019a). Third, studies commonly consider the development of research skills, such as academic writing, beneficial (e.g., Aspfors and Eklund 2017;Baan, Gaikhorst, and Volman 2019b b ;Maaranen 2009). Fourth, Baan et al. (2019 a ) and Kowalczuk-Walędziak et al. (2019) also mention the application or use of research to improve practice, a concept closely related to the development of teaching competences such as improved support for student learning and introductions of innovative teaching strategies (e.g., Ion and Iucu 2016;Niemi and Nevgi 2014;Råde 2014;Taylor 2017).
However, Kowalczuk-Walędziak et al. (2019) and Volk (2010) find little evidence that practicing teachers use learned research skills and knowledge after completing their teacher education. In summary, though teacher educators endorse a considerable number of highly ambitious purposes for pre-service teacher research, pre-service teachers' perceptions and actual learning outcomes regarding the quality of teaching are unclear.

Teaching quality of pre-service teachers
The quality of teaching is assumed to be a core tenet in teacher education programmes; preservice teacher research also has come to represent a substantial part of those programmes. Next, pre-service teacher research is assumed to contribute to better teachers (e.g., Menter et al. 2017). However, the exact relationship between these two learning outcomes is complicated, and empirical data on the relationship are lacking. Teacher quality and teaching quality are closely related; teaching quality is the most important indicator of teacher quality (e.g., Darling-Hammond 2017). A highly qualified teacher can manage groups of pupils and knows what and how to teach -all observable indicators of teaching quality. However, highly qualified teachers also know about learning and make decisions informed not only by their own classroom and school evidence but also by theory and research ). Similarly, in the framework for effective teaching, Stronge (2018) emphasises six behaviours of effective teachers from the perspective of twenty-first-century teaching: professional knowledge, skills for instructional planning (including classroom management), skills for instructional delivery (including cognitive activation, differentiation and learning strategies), assessment for learning, creation of an adequate learning environment and professionalism. These behaviours are driven by key characteristics reflecting the teacher's dispositions, goals and beliefs, which directly affect teaching effectiveness (Kaiser and König 2019;Nilsen and Gustafsson 2016).

Research questions
Against this background, the current study aims to shed light on the perceived and actual value of pre-service teacher research for becoming a teacher by exploring the relationships among perceptions, teaching quality and pre-service teacher research. Pre-service teacher research is assumed to contribute to an inquiry stance and therefore to better teachers. Because the general idea is that pre-service students are more enthusiastic about teaching than about research, we aim to examine student profiles defined by teaching quality and research quality. The associated research questions we investigate are as follows: (1) What are pre-service teachers' perceptions of the value of pre-service teacher research?
(2) What is the most important learning outcome of pre-service teacher research, according to pre-service teachers?
(3) a. How does the quality of inquiry projects -required for graduation-relate to the quality of teaching exhibited by pre-service teachers?
b. How pre-service teacher profiles emerge, reflecting the correlation of research quality and teaching quality?
(4) How do these groups differ in their perceptions of pre-service teacher research?

Context
The context of this study is initial teacher education for primary schools, a bachelor's degree programme at a university of applied sciences in the Netherlands. The research component in the programme was introduced ten years ago. During the four years of teacher education (240 European credits [ECs]), pre-service teachers spend 70 ECs in practice and 25 ECs on pre-service teacher research-related learning activities (see Table 1). In the second half of their final year, the pre-service teachers teach their own group three days a week for 20 weeks. In this period, they also collect data for their preservice teacher inquiry project, which starts in the first semester and covers 15 ECs in total. Pre-service teachers' practical work is assessed by the supervising teacher and one teacher educator.
The purpose of pre-service teacher research in the Netherlands is to develop five connected competences: (1) research knowledge, (2) knowledge about research in the domain, (3) research skills, (4) application of findings of previous research in practice and (5) an inquiry habit of mind (Van Katwijk, Jansen, and Van Veen 2019b). This document analysis of pre-service teacher research in the Dutch programmes also shows that an inquiry habit of mind is the most important purpose. Pre-service teacher research is taught in a teaching-learning trajectory over four years, in which all research activities are closely related to practice. In the third year, pre-service teachers individually conduct a limited literature review and a research project in a team, commissioned by a primary school and supervised by a teacher educator. Pre-service teachers can choose their own topic for the final individual inquiry project, which is typically in the form of design research or action research. This inquiry project is assessed by two teacher educators using scoring rubrics about all components of practitioner research, such as use of literature, content of the chosen topic, form and the quality of reflection on the process.

Study design
We used a multi-methods approach with a questionnaire and a database of assessment scores. Our analysis strategy consisted of confirmatory factor analyses on Likert-scale questions, which led to four scales measuring pre-service teachers' attitudes to and perceptions of student research (RQ 1); a qualitative analysis of an open question about the most important learning outcome (RQ 2); calculating the correlation between the quality of pre-service teacher research and the quality of pre-service teacher's teaching, both based on assessment scores (RQ 3a); a cluster analysis to identify pre-service teacher profiles (RQ 3b) and, finally, descriptive statistics and an analysis of variance on the profiles and the perceptions of pre-service teacher research (RQ 4) (see Figure 1).

Participants
We used the assessment scores of 650 pre-service primary education teachers, collected during the period 2014-2018. All participants had just graduated from a university of applied sciences in the Netherlands. In total 236 pre-service teachers (15% male, mean age = 23.0 years) completed the questionnaire, titled 'Perceptions of pre-service teacher research', for a response rate of 36%, distributed evenly over the four cohorts. As part of this survey, 194 pre-service teachers (80% of respondents) formulated their own most important learning outcomes of pre-service teacher research in an open question.

Questionnaire
The questionnaire was based on two previous instruments about perceptions of preservice teacher research: Van Der Linden et al.'s (2012), (2015) questionnaire and the more general Research Acceptance in Vocational Education Questionnaire (Griffioen 2018). These instruments have four underlying dimensions, or scales, to measure students' attitudes, perceptions and self-efficacy.
(2) Value-cognitive attitude. This scale indicates if and how the pre-service teachers perceive the importance of pre-service teacher research (e.g., 'I think conducting research is a good way for me to improve my pedagogical skills'). We added some items about the perceived difficulty of conducting pre-service teacher research.
(3) Self-efficacy towards pre-service teacher research. This scale identifies the degree to which students believe they have learned to conduct pre-service teacher research (e.g., 'I have learnt enough to independently conduct practitioner research in practice').
(4) Behaviour in the future. The scale items pertain to intentions and expectations to conduct practitioner research or inquiry in a future teaching profession (e.g., 'I'll conduct practitioner research in my future job as a teacher to improve my professional teaching practice'). The final questionnaire consisted of 24 Likert-type questions. We measured all items with six-point Likert scales (1 = 'totally disagree' to 6 = 'totally agree'), deliberately eliminating a neutral option. We reverse-scored four negatively formulated items in the analysis (e.g., 'I don't think I'll conduct research to resolve a problem in future'). Appendix A shows the scales, items, factor loadings and reliability measures (Cronbach's alpha) of the final questionnaire.
We added open questions, connected to the questions about self-efficacy: 'What have you learnt about pre-service teacher research in the teacher education programme? . . . The most important learning outcome for me with respect to conducting practitioner research was . . . '.
To confirm the suitability of the four scales of perceptions about pre-service teacher research, we performed a confirmatory factor analysis (IBM Amos, version 25). For model fit, we used chi-square fit statistics along with their associated robust comparative fit index (CFI; should be between .90 and .95 for model fit), PClose (for model fit: >.005), standardised root mean square residual (SRMR; close to zero for model fit) and root mean square error of approximation (RMSEA; ideally <.05, but <.08 is acceptable). Because of the sample size, we used the normed chi-square (CMIN/DF), which indicates model fit if it is below 3.0 (Brown 2006).
The initial model did not result in acceptable model fit (normed χ 2 = 2.36; CFI = .89; PClose = .006 and RMSEA = .08). We evaluated the items in the model on factor loading (which should be >.40) and content and inspected modification indices to estimate which modifications we could make to improve model fit. We ultimately removed seven added items that were not used in the original questionnaires and diverged from the constitutive scales (e.g., 'I do not think I will conduct research in future because it is too timeconsuming'). We also correlated two errors ('I think pre-service teacher research is nice' and ' . . . is motivating'). The fit of the adjusted model was acceptable: CMIN/DF = 2.01; CFI = .95; PClose = .02; SRMR = .055 and RMSEA = .07. The Cronbach's alpha of the four scales indicated good internal consistency (Table 2).

Teaching quality
To determine the quality of the pre-service teacher research and teaching quality, we took the assessment scores on the final internship of all just-graduated pre-service primary school teachers (N = 650). We measured teaching -or teacher -quality in the final internship with scores on assessments of seven teacher competences: (1) pedagogical competence (safe and stimulating learning environment), (2) interpersonal competence (communication and relation with learners), (3) pedagogical content knowledge competence (teaching skills), (4) organisational competence (leadership and classroom management), (5) competence to collaborate with colleagues, (6) competence to collaborate with external parties such as parents and (7) competence in own professional development. In the Netherlands, the grading runs from 1 (lowest score) to 10 (highest score), with 5.5 for constituting a passing assessment score. In this study, only successfully graduated preservice teachers participated, so the scores we used are all between 5.5 and 10.
Supervisors' ratings of teaching are prone to be inaccurate and unreliable (Praetorius et al. 2014). Because these pre-service teachers were assessed by supervising teachers of the school, who were not specifically trained to assess teaching quality, a bias caused by confounding factors (e.g., halo effects) could easily occur (Creemers, Kyriakides, and Antoniou 2012). For example, the relationship (good or bad) between the supervising teacher and the pre-service teacher could have influenced the assessment score. To avoid this bias and increase reliability, we measured the teaching quality of 80 pre-service teachers in the final internship with the validated International Comparative Analysis of Learning and Teaching (ICALT) instrument (Van Der Lans, Van De Grift, and Van Veen 2018). The correlation between the ICALT scores and the assessment scores for the final internship was significant though moderate (r = .334, p = .003; two-tailed) (see Van Katwijk and Van Der Lans 2016). We deemed this as a fair indication for the use of the assessment scores as measures of teaching quality for the complete sample of 650 students.

Quality of pre-service teacher research
We measured the quality of pre-service teacher research by the scores (between 5.5 and 10, as noted previously) on the pre-service teacher research project, assessed by an inquiry report (N = 650). Examples of topics are: learning effects of using games in maths education, inquiry-based learning and pupil engagement, motivational effects of the application of the theory of multiple intelligence, and improvement of reading pleasure. The assessment consists of three main aspects: (1) content (weight 60%), including problem analysis, theoretical framework, research question, research design, results, conclusions and discussion; (2) form (weight 20%), including structure, language, academic writing, use of correct APA style and layout; and (3) process (weight 20%), including description of the research process, participation in research team and reflection on professional development. To improve and guarantee the reliability of these assessments, internal calibration as well as an external check on the quality of assessment by researchers of other institutes takes place regularly. The quality of the inquiry projects was also the focus of the accreditation process in 2015.

Data analysis
For the first research question, we used descriptive statistics, using SPSS 25 to determine the mean scores and standard deviations on the four scales of the questionnaire. Two researchers independently analysed the open-end question about the perceived most important learning outcome inductively. We compared the categories that emerged, discussed the few differences and came to consensus by applying previous research findings (Van Katwijk, Jansen, and Van Veen 2019b). In 18 of the 194 cases, the preservice teachers had mentioned two 'most important learning outcomes' instead of one. We decided to split those cases and use both expressions as most important learning outcomes. An independent third researcher analysed the answers once more, deductively with the determined categories. The interrater agreement was high, over 95% (Miles, Huberman, and Saldana 2014). We were able to link the data from the questionnaires to the scores of both the assessment of pre-service teacher practice and the pre-service teacher research. We calculated the Pearson's correlation between the scores of assessments of pre-service teachers' practice and the scores of assessments of pre-service teachers' research. The correlation is significant when p < .001 and the 95% bias corrected accelerated (BCa) confidence interval (CI) does not include 0 (Field 2018). To identify different profiles, using scores of pre-service teacher research and scores of pre-service teacher practice, we conducted a two-step cluster analysis in SPSS, using the data of all pre-service teachers (N = 650). The statistical cluster quality is good when the ratio of sizes, largest cluster to smallest cluster, is below 2 (Anderberg 2014). Because the indicator variables had different ranges, we standardised the scores.
After assigning the pre-service teachers to a cluster, we conducted an analysis of variance (ANOVA) with post hoc comparisons (Bonferroni) to investigate differences between the profiles on their perceptions of pre-service research and the perceived most important learning outcome. We analysed all data anonymously.
The study followed the general ethical standards approved by the scientific community and listed on the website of the Netherlands Organisation for Scientific Research (NWO) (2018). Participants gave their informed consent and their anonymity was guaranteed.

Perceptions regarding pre-service teacher research
We measured student teachers' perceptions of pre-service teacher research with four scales (Table 2). We determined two scales that were linked to the perception of the value of pre-service teacher research: the cognitive attitude (mean score: 4.5) and the affective attitude (mean score: 4.6). Both offer indications that students believe that pre-service teacher research is valuable.
The scale Self-efficacy indicates that pre-service teachers believe themselves able to conduct practitioner research after finishing teacher education. The mean of the Future behaviour scale is relatively low (M = 4.0), compared with the scales Value and Self-efficacy, but still above the scale mean. We observed no significant differences in perceptions of pre-service teachers across the cohorts, by age or by gender.
The pre-service teachers indicate a positive cognitive attitude towards research and regard it as important and a good way to become more professional. They also express a positive affective attitude and indicate that the pre-service teacher research project was interesting to do.

Most Important Learning Outcomes of Pre-Service Teacher Research
As mentioned previously, 194 pre-service teachers (80% of the respondents) formulated their own most important learning outcome, which we subdivide into seven categories (see Table 3 for examples): (1)To conduct research: pre-service teachers mentioned conducting research in general or focussed on research aspects such as the research cycle.
(2)To write academically: pre-service teachers mentioned that academic writing is difficult, something they have not encountered before. (3)To use literature: pre-service teachers noted that research involved using literature in new way, not just for passing an exam.
(4)Knowledge about the subject: pre-service teachers experienced becoming a specialist in a certain topic.
(5)An inquiry habit of mind: pre-service teachers mentioned several characteristics of an inquiry habit of mind (e.g., to be critical, to want to achieve), as well as an inquiry habit of mind itself.
(6)To link research and practice: pre-service teachers indicated that they experienced how to use research for improving practice.
(7)Miscellaneous: learning outcomes mentioned just once or twice (e.g., to be able to conduct research independently and not in a group).
Most responses indicated 'to conduct research' as most important learning outcome. Although 'to write academically' and 'to use literature' also fit into the category 'to conduct research', we decided to make separate categories for them because they were mentioned explicitly by more than 10 students. Over 40 pre-service teachers mentioned 'knowledge about the subject' or 'to apply research in practice'. Thirty-four pre-service teachers responded with 'an inquiry habit of mind' specifically or else mentioned one or more characteristics of an inquiry habit of mind.

Relationship between pre-service teacher research and teaching
To determine whether the quality of pre-service teacher research was related to pre-service teaching quality, we calculated the bivariate correlation between the scores of pre-service teacher research and the final assessment scores of the pre-service teaching practice (N = 650). We found a significant, positive correlation between the scores on pre-service teacher practice and the scores on pre-service teacher research (r = .224, 95% BCa CI [.148, .298], p < .001); pre-service teachers who achieved high scores for their pre-service teacher research achieved significantly higher scores for their teacher practice assessment than students who scored lower on pre-service teacher research. To gain a deeper understanding of this relationship and how various pre-service teachers perceive the value of preservice teacher research, we identified profiles by conducting a cluster analysis.
The cluster analysis performed on the scores for pre-service teacher practice and research (N = 650), was fairly good; the ratio of sizes, largest cluster to smallest cluster, was 1.66. We determined clusters by face validity and statistical cluster quality, resulting in four profiles (see Figure 2): Profile 1, Good practitioners (n = 207; 32%): pre-service teachers who score low on preservice teacher research (M = 6.3) and high on teaching practice (M = 8.5).

Perceptions of pre-service teacher research in the profiles
We analysed whether students in the four profiles differed in attitudes towards pre-service teacher research or their perceptions of the most important learning outcome. The ANOVA showed statistically significant differences between the profiles for all the scales (see Figure 3): Value affective F(3, 227) = 9.73, p = .000, Value cognitive F(3, 227) = 7.39, p = .000, Self Efficacy F(3, 227) = 8.53, p = .000, and Future F(3, 227) = 4.04, p = .008. The  Bonferroni post hoc test showed significant differences on the first three scales between the good practitioners (profile 1) and the high achievers (profile 3), and between the high achievers and the low achievers (profile 4). For the scale Future, we only found significant differences between the good practitioners and the high achievers. Profile 2 (average students) showed no significant differences with the other profiles. Figure 3 shows a bar chart of the mean scores on the scales divided over the four pre-service teacher profiles.
We observed no significant differences across the profiles in gender, age, cohort or previous education. However, the profiles showed different patterns of their perceptions of the most important learning outcomes (Figure 4).
The high achievers, pre-service teachers who achieved high scores on their research project as well as on their teaching assessment, indicated development of the inquiry habit of mind as the most important learning outcome. This is characterised by being critical and curious, wanting to share and wanting to achieve. This perceived learning outcome was mentioned less frequently in all other profiles. The average students (profile 2) indicated conducting research as the most important learning outcome. The good practitioners (profile 1) also indicated relatively often that conducting research was the most important learning outcome, but they indicated increased knowledge of the studied subject just as frequently. The low achievers (profile 4) indicated that they mainly learned to apply research findings in practice. Overall, we found significant and remarkable differences across the four profiles of pre-service teachers. The pre-service teachers in all four profiles noticed the value of linking theory and practice.

Conclusion and discussion
The aim of this study was to gain insights into the perceived and actual value of preservice teacher research for becoming a primary teacher. Positive relationships among perceptions, teaching quality and teacher research were found. Focussing on the relationship between the quality of the inquiry projects -required for graduation-and the quality of teaching of pre-service teachers, four profiles were identified with distinctive perceptions of the most important learning outcomes earned through pre-service teacher research. This section first focuses on the relationship between pre-service teacher research and teaching quality. It continues to discuss the findings about the most important learning outcomes and the value of pre-service teacher research in primary teacher education in relation to previous research. Finally, the paper concludes with limitations of the study and suggestions for further research.
Regarding the relationship between the quality of the pre-service teacher inquiry projects and their teaching practice, a small though positive relationship was found. Because no causality was established, this result should be addressed with caution. However, as the analysis of different profiles showed, the high achievers were more positive about the value of pre-service research. The positive relation between research and practice supports the assumption that pre-service teacher research contributes to better teachers (e.g. Menter, Peters, and Cowie 2017). These findings are in line with studies about improvement of teaching practices as result of practitioner research by in-service teachers (Ermeling 2010;Ion and Iucu 2016). Grossman (2005), who stated a lack of empirical evidence for this relationship in teacher education, suggested that the way pre-service teacher research is taught is of influence. Other studies about pre-service teacher research in master programmes also show a positive effect of pre-service teacher research to all professional competences (e.g. Kowalczuk-Walędziak et al. 2019;Niemi and Nevgi 2014). They emphasise the importance of authentic experiences with research studies and integration into classroom practices, which is the case in the Dutch primary teacher education context. The way pre-service teacher research is taught, with the aim to understand and improve teaching and learning, might be crucial for this positive relationship between research and teaching quality in both bachelor and master programmes.
Between the four profiles the differences in perceptions of the most important learning outcomes were remarkable. The high achievers' reported learning outcomes (i.e. an increased inquiry habit of mind) correspond to the most important purpose of preservice teacher research as described in the teacher education programme (Aspfors and Eklund 2017;Van Katwijk, Jansen, and Van Veen 2019b). The newly qualified teachers from this profile have developed an inquiry stance ), so they might be able to work in inquiry-based contexts and to contribute to a culture of inquiry in primary schools (Uiterwijk-Luijk et al. 2019). This concurs with teachers who are educated at a master level (e.g. Aspfors and Eklund 2017;Baan, Gaikhorst, and Volman 2019b;Ulvik, Riese, and Roness 2017). The average students and low achievers in our study might not have understood the recursive nature of research. The concerns of these students focus more on their teaching skills (Flores et al. 2014;Joram 2007). In addition, the assessment part with scoring rubrics might have encouraged a procedural understanding of practitioner research for these students ). Teacher education programmes might consider how they can shift the focus of pre-service teacher research for all students towards the main purpose; an inquiry stance. This purpose needs to be made more explicit and aligned with the other teaching activities and the forms of assessment (Biggs and Tang 2011). On the other hand, differentiation to the needs of the students in these profiles might improve the quality of teacher education. Teacher educators' knowledge of these profiles and of students' readiness and interests, could be used to appropriately differentiate content, process, product, and learning environment (Santangelo and Tomlinson 2012). High achievers might be challenged with more complex academic literature and research methods, while good practitioners and low achievers are encouraged to improve their practice by application of previous findings in their own context.
The perception of graduating pre-service primary education teachers towards the value of pre-service teacher research is positive, regardless of their assessment scores. Pre-service teachers indicate confidence in undertaking practitioner inquiry after finishing their research project. Although their self-efficacy for conducting research might be influenced by their recent completion of their research project, it still might contribute to their involvement in inquiry-based working in a future job (e.g., Baan, Gaikhorst, and Volman 2019b a ; Uiterwijk-Luijk et al. 2017). These positive perceptions are in line with some previous studies about master degree programmes (e.g., Aspfors and Eklund 2017;Kowalczuk-Walędziak et al. 2019;Ulvik, Riese, and Roness 2017), though other studies report negative perceptions among pre-service teachers, who stated they did not see the connection with their teaching practice (e.g., Puustinen et al. 2018). The explicit link between theory and practice and the possibility of applying findings from previous research to their own practice seem to contribute positively to pre-service teachers' perceptions. The positive affective and cognitive attitudes are important for their continuing professional development as a teacher (e.g., Van Der Linden et al. 2015). Despite these positive attitudes, not all pre-service teachers expect to continue conducting practitioner research in their future profession. The perspectives of in-service teachers regarding inquiry have a major influence on the pre-service teachers' research dispositions (Willegems et al. 2017;Kotsopoulos, Mueller, and Buzza 2012). Negative models, such as supervising in-service teachers without any research experience, could result in pre-service teachers classifying the research process as unnatural and as not contributing to their teaching practice. To this end, teacher education should provide scenarios where pre-service teachers can collaborate with inservice teachers, and provide compelling opportunities for them both to establish an inquiry stance and make evidence-based decisions in teaching and pupil learning: in other words, become an extended professional (Cochran-Smith and Lytle 2009; Darling-Hammond 2010). Collaborative (pre-service) teacher inquiry is still a scarce mode of pre-service teacher research and has therefore not been studied in great detail (Willegems et al. 2017). It aims to provide evidence-based answers to shared challenges in practice. Previous studies suggest that creating such an environment in teacher education helps pre-service teachers become responsible for approaching their classroom as a laboratory in which observation, inquiry and self-development are realised (Cochran-Smith and Lytle 2009). Partnerships with schools to create these spaces are essential for this.
This study has some limitations. First, all participants were pre-service teachers from one university of applied sciences who followed the same programme. Therefore, generalisations regarding other programmes should be made with caution. Second, all data were from recent graduates; pre-service teachers who did not pass one or both of the assessments on teaching or research might have decided not to complete the study, and their perceptions were not included in this study. Third, the research depends only on subjective, institutional criteria to determine both 'teacher quality' and 'research quality'. These criteria draw on national norms for these elements of performance in pre-service teacher education. Fourth, some results from the questionnaire highlight the need for further research. For example, the Future scale indicated the expectation of pre-service teachers to conduct practitioner research in their future profession, which might not be a reliable measure for their actual future behaviour (Fishbein and Ajzen 2011). Attitudes can vary with the context in which they are expressed. Therefore, examining the various contexts (e.g., different teacher education programmes, background information about the research culture at schools, internships where pre-service teachers obtain experience with practitioner research) would be worthwhile pursuits for further research.
Previous research has rarely taken the quality of the pre-service teacher research, assessed by teacher educators or researchers, into account; most studies are based on self-reporting methods such as questionnaires and interviews (e.g., Aspfors and Eklund 2017;Kowalczuk-Walędziak et al. 2019;Niemi and Nevgi 2014;Van Der Linden et al. 2015). However, a limitation of the present study is that the results are based on single assessment scores of pre-service teaching practice and of the inquiry project. We did not differentiate the quality of various aspects of the assessment of the inquiry project, such as theoretical framework or reflection on professional development teaching assessment; therefore, we cannot indicate which aspects of pre-service teacher research contribute to developing teachers with an inquiry stance. We also cannot indicate which specific aspects of the teaching competence are positively related to pre-service teacher research because of its dependence on the chosen topic and the method of the preservice teacher research. Moreover, longitudinal research into the relation between the quality of pre-service teacher research and ongoing professionalisation during the teacher career and improvement of practice would be an interesting next step. Finally, the addition of a qualitative study, focused on the 'why' and 'how' of the revealed patterns can lead to deeper insights.
Despite these limitations, our study provides insight into perceptions and actual learning outcomes of pre-service teacher research, which was introduced ten years ago in primary teacher education in the Netherlands. These insights likely are relevant for teacher education programmes with similar settings. Teacher education aims to develop future-proof teachers, such that they are effective in ensuring student achievement and focussed on ongoing professional development (Darling-Hammond 2017). These teachers are able to use and conduct research to evaluate and improve their practice and contribute to a culture of inquiry in their classrooms and school (Baan, Gaikhorst, and Volman 2019b a ;Uiterwijk-Luijk et al. 2019). The preservice teachers in the studied programme believe they have developed their research skills, inquiry habit of mind and research literacy and that they are able to apply previous research in their own practice. Due to pre-service teacher research, they are ready to do inquiry-based work and possess the abilities to adjust and improve their practice to changes in their classrooms and schools. .833 *Items that were reverse-scored