Remote internships: The experiences of pre-service teachers during the COVID-19 pandemic

Abstract School-based internships are an important learning opportunity for pre-service teachers (PSTs); they were partly or fully obstructed by the COVID-19 pandemic. This study aimed to investigate how schools, PSTs, and the university succeeded in organizing and supporting school internships during the pandemic. Accordingly, 434 pre-service teachers (PSTs) were asked to complete an online survey at two time points (just before starting and just before completing the internship) about their internship experiences under pandemic conditions in 2020 (201 PSTs) and 2021 (233 PSTs). We investigated the effects of the internships on PSTs’ health, lesson planning competence, and teacher self-efficacy; we also compared these effects between the cohorts in 2020 and 2021. While PSTs had little contact with school pupils, university teacher educators quickly offered alternative supervision formats (especially in 2020). Internships contributed more to PSTs’ career choice certainty and self-efficacy in 2021, though there was no difference in self-assessed lesson planning competence. The central implications are that university teachers can create learning opportunities even without PSTs being physically present in schools; this potential should be developed in the future, despite the removal of pandemic-related restrictions. Furthermore, schools should remain aware of their educational obligations to PSTs even in challenging times.

Abstract: School-based internships are an important learning opportunity for preservice teachers (PSTs); they were partly or fully obstructed by the COVID-19 pandemic.This study aimed to investigate how schools, PSTs, and the university succeeded in organizing and supporting school internships during the pandemic.Accordingly, 434 pre-service teachers (PSTs) were asked to complete an online survey at two time points (just before starting and just before completing the internship) about their internship experiences under pandemic conditions in 2020 (201 PSTs) and 2021 (233 PSTs).We investigated the effects of the internships on PSTs' health, lesson planning competence, and teacher self-efficacy; we also compared these effects between the cohorts in 2020 and 2021.While PSTs had little contact with school pupils, university teacher educators quickly offered alternative supervision formats (especially in 2020).Internships contributed more to PSTs' career choice certainty and self-efficacy in 2021, though there was no difference in self-assessed lesson planning competence.The central implications are that

ABOUT THE AUTHOR
The authors are working empirically to explore learning opportunities in school-based internships for pre-service teachers.The focus is on learning support provided by different groups of teacher educators.Much of the work is embedded in a developmental context, particularly on in-service training concepts for schoolbased mentors.

PUBLIC INTEREST STATEMENT
School-based internships are an important learning opportunity for pre-service teachers (PSTs) and were obstructed by the COVID-19 pandemic.This study aimed to investigate how schools, PSTs, and the university succeeded in organizing and supporting school internships during the pandemic.Students were asked in an online survey at two time points about their internship experiences under pandemic conditions.We investigated the effects of the internships in terms of PSTs' health, lesson planning competence, and teacher self-efficacy; we also compared these effects between the cohorts in 2020 and 2021.While PSTs had little contact with school pupils, university teacher educators quickly offered alternative supervision formats.Internships contributed more to PSTs' career choice certainty and self-efficacy in 2021, though there was no difference in self-assessed lesson planning competence.The central implications are that university teachers can create learning opportunities even without PSTs being physically present in schools.

Introduction
The arrival of the COVID-19 pandemic in the spring of 2020 had a massive impact on education.During the pandemic, pupils and teachers were temporarily barred from school buildings, and switching to remote teaching was necessary.Among those affected by these changes, pre-service teachers (PSTs) often could not complete school-based internships, which are an important learning opportunity in teacher education (Gröschner, 2015;Lawson et al., 2015;Ulrich et al., 2020).During practicum periods, PSTs can acquire key competencies, such as lesson planning skills, for their future careers.They can also reflect on their own career decisions and expand their selfefficacy in relation to typical teaching activities.Empirical evidence shows that PSTs need professional support during internship periods to maximize learning opportunities.Thus, the fact that school-based internships could not be conducted as face-to-face placements at many schools in 2020 and 2021 is likely to have impacted PSTs' learning opportunities; such internships may have offered none of the traditional learning opportunities obtained from face-to-face teaching.
In response, teacher educators worldwide were obliged to develop alternative learning opportunities that could be taught either partly or fully remotely.Such learning opportunities were designed differently depending on the targeted competency.They included, for example, a) planning, implementing, and reflecting on lessons for a digital learning setting (e.g., video conferencing, online course); b) developing individualized assignments; c) providing feedback on assignments; d) planning, implementing, and reflecting on lessons without direct (digital) communication with school pupils; e) analyzing other people's (videotaped) teaching situations or lesson plans.
These learning opportunities for PSTs had to be developed very quickly due to the unpredictability of the COVID-19 pandemic; there was often little time for systematic reflection on how planning related to the respective competence goals.Thus, their effects on PSTs need to be clarified by research.Accordingly, various international studies on relevant experiences with such alternative learning opportunities, particularly in school placements, have been published (Allen et al., 2020;Carrillo & Flores, 2020;Flores & Swennen, 2020;Kidd & Murray, 2020;König et al., 2020;Nel et al., 2021;Perry et al., 2021;Sepulveda-Escobar & Morrison, 2020).The majority of these studies include details on relevant considerations when designing learning opportunities for a remote school (practicum) experience.Individual studies have also included qualitative research on experiences with appropriate learning opportunities.However, we are unaware of any quantitative studies that have followed such experiences across different years; the present study seeks to fill this gap.
To address this gap in the literature, we conducted a quantitative web-based questionnaire survey with two groups of PSTs (in 2020 and 2021) at two time points (just before starting (T1) and just before finishing (T2) their internship) to examine (1) how PSTs took advantage of learning opportunities in a long-term teaching internship during the COVID-19 pandemic, ( 2) to what extent they felt supported by school-based mentors and university teacher educators, and (3) how they were affected in terms of stress and general health.To further enquire into the effects of the pandemic, we compared learning outcomes reported by PSTs in 2020 and 2021.Thus, this study aims to provide insights into internship practice during the COVID-19 pandemic from a PST perspective.This is particularly relevant in that it can be used to make suggestions for including remote teaching and learning in school placements, even after societies worldwide have moved on from the COVID-19 pandemic.In what follows, we first present the theoretical background, state of the research, and research questions.Then we describe the methodology and present the findings.Finally, the implications and recommendations for future research and teaching practice are discussed.

Importance of internships -Learning opportunities for PSTs in practical phases for self-efficacy and lesson planning competence
In school-based practice periods, PSTs encounter learning opportunities for professional development (Gröschner, 2015;Lawson et al., 2015;Mattsson et al., 2011).Often, these learning opportunities are linked to school-based teaching, i.e., they include lesson planning, lesson performance, or lesson reflection.Below, we elaborate why school internships are important and that they can serve the development of self-efficacy and professional competencies.We focus on lesson planning as a core task for teachers (Department for Education, 2013).The importance of teacher educators (Crasborn & Hennissen, 2014;Staub et al., 2003) is highlighted and special conditions of internships under COVID-19 are examined.
School-based internships are conducted with specific objectives for the learning outcomes of PSTs.According to an established model of professional competence (Baumert & Kunter, 2006;Shulman, 1983Shulman, , 1987)), teachers need subject-specific, subject-didactic, and pedagogical aspects of knowledge and skills; additionally, they require affective skills, including the ability to self-regulate.
Lesson planning is relevant for teaching (König et al., 2020); it requires a relationship to concrete pupils but does not necessarily have to be conducted with the pupils in a physical classroom situation.For years, lesson planning has often been based on specific planning models or planning grids (John, 1992), which PSTs learn to use in teacher education.König et al. (2015) developed a model that can assess lesson planning competence using a combination of self-reports and assessment of lesson planning documents.The model has since been further developed (König et al., 2021).They described lesson planning competence in three dimensions.First, with respect to the specific composition of the learning group and diagnostics in relation to the specific characteristics of the pupils; second, in relation to the conception of the central learning task of the lesson; and third, in relation to the planning of the lesson in terms of its concrete implementation.Our questionnaire was partly based on the three dimensions.
In addition to this knowledge-based aspect, school internships are also intended to develop learning outcomes related to beliefs about one's work as a teacher.This is reflected, for example, in career choice confidence (König et al., 2013), but specifically in self-efficacy, which is a recognized and intensively studied construct (Michos et al., 2022).Although not all relevant studies come to the same conclusions, PSTs can increase their self-efficacy through teaching experiences during practicum (Colson et al., 2017).The support PSTs receive is critical in developing self-efficacy (Woolfolk Hoy & Burke Spero, 2005).

Design of internships -Support from school and university-based teacher educators
School-based internships are popular among PSTs; they are more effective with support from both school-and university-based teacher educators (Beckmann, 2020).School-based mentors have a major role; university-based teacher educators can also make an important contribution (Aspfors & Fransson, 2015;Hennissen et al., 2008;Pennanen et al., 2015;Ulrich et al., 2020).In addition to the design of individual learning opportunities such as pre-teaching meetings (which are specifically important; Kreis & Staub, 2011), social (and emotional) support throughout the period is crucial (Richter et al., 2011(Richter et al., , 2013)).Social support is relevant to the well-being of PSTs, including physical and mental health outcomes (Cohen, 2004).Thus, social support from peers and teacher trainers can counteract any stresses they may encounter during the internship.

Health and stress during internships
Being a teacher is stressful for many active teachers; the behaviors of pupils in face-to-face classrooms are a major source of strain (Eysel-Gosepath et al., 2012;Kunz Heim et al., 2019).However, less is known about the experience of pressure among PSTs during school-based practice periods, and the existing results are inconclusive.

Impact of Covid-19
While the goals of the internship remained during the COVID-19 pandemic, no or limited physical presence could occur.Carrillo and Assuncao Flores (2022) demonstrated that a high-quality virtual internship can be achieved with appropriate technology and online pedagogical resources.Simultaneously, they indicated that further research on this topic seems necessary, including to rethink the design of conventional internships.In this respect, we assume that there were still opportunities to plan lessons, but fewer opportunities to actually conduct and reflect them.Thus, opportunities to build self-efficacy were few.Therefore, students who were taught during COVID-19 are less self-efficient (Symes et al., 2023).Accordingly, an examination of self-efficacy and lesson planning skills seems promising to us.Further, we hypothesize that student attendance has changed and that a focus on health and career choice safety should be addressed owing to various special circumstances.

Research questions
The following four research questions were developed based on our study aims, theoretical considerations, and state of the research.For each question, we also compare the results between the 2020 and 2021 cohorts.
(1) In what ways and to what extent were PSTs involved in school-based practice during their pandemic-affected internships?
(2) In what ways were PSTs supported, and how did they rate the quality of this support?
(3) How did PSTs rate their health status, career confidence, and stress during the COVID-19 pandemic?
(4) How did PSTs' self-efficacy and self-assessed lesson planning skills evolve over the internship?

Research design
To address the research questions, we conducted a longitudinal study with two measurement points for PSTs in long-term internships (Crowther & Lancaster, 2012).The PSTs participated in the survey via an online questionnaire at the beginning (T1) and end of the internship (T2).The questionnaire used quantitative self-report scales.The responses were linked via an individual participation code across the measurement time points.Data analysis comprised descriptive statistics, t-tests, and regression analyses.The overall study design is summarized in Figure 1.

Participants background
In Germany, and in Lower Saxony in particular (where this study was conducted at a University with in total about 1500 PSTs in different programmes), teachers are professionalized in a staged study structure consisting of a 3-year bachelor's degree and a 2-year master's degree in two teaching subjects (e.g.German, math or art).After these studies, there is a period of preparatory service in a school.
The school-based long-term internship, which the students interviewed here complete, covers 18 weeks.It is oriented toward subject didactics, and the students receive intensive subject didactic preparation, support and follow-up by the university, and mentoring in the school.In addition to seminars, the students are visited in the classroom and receive feedback on their teaching.In addition, the students are in subject-related tandems at the school to ensure interaction between the students.
When face-to-face teaching did not occur, the students received feedback on lesson planning or digitally supported teaching.
In the Bachelor's degree, the students have already completed two short-term internships at schools, focusing on reflecting on other people's and their own teaching from a general didactic perspective.
Sampling method In 2020 and 2021, we sought participation in the online questionnaire from all PSTs in the master's degree programs in teaching at elementary schools and secondary schools who registered for the long-term internship from our university since a substantial part of the internship was completed under pandemic conditions.By default, participation in the survey was part of the tasks that needed to be completed during the internship.However, PSTs could actively object to participation, which had no negative consequences for their education.They were also told the data would be stored anonymously.The authors of the present paper were not involved in the PSTs' evaluation process, so there was no dependent relationship between us.A total of 434 responses were received (201 in 2020 and 233 in 2021).Almost all participants in the internship also participated in the survey.PSTs had a mean age of 24.93 years; 78% reported being female.The internship took place in the first year of their master's program.Prior to this, all PSTs had completed a 6-semester bachelor's program.

Research instruments-Questionnaire indicators
The questionnaire comprised 13 scales (46 items) measuring health and satisfaction, intensiveness of involvement in internship activities, support, and planning competence and self-efficacy.Previously validated scales and items were used whenever possible to enable comparison with other studies and to ensure content validity.Where possible, response categories were also adopted; this concerns the following scales: teacher self-efficacy, self-assessed planning competence, career choice confidence, assessment of support from mentors and teachers, and selfassessed health.Table 1 describes each of the 13 scales.Good to very good internal realities (.70 < Cronbach's alpha < .89)were obtained for all scales used.Due to the high speed of the changes and new requirements brought about by the COVID-19 pandemic, especially in 2020, no tested scales or items were available to measure the specifics of school involvement and individual burdens during the pandemic.Therefore, we developed items .73I know that I can manage to teach materials that are relevant to the exam to even the most problematic pupils.Schwarzer and Jerusalem (1999) Notes: 1 = four single items that are thematically related but do not produce a common scale.
related to these aspects.To ensure content validity for these newly developed items, they were discussed intensively among the authors and with PSTs.

Data processing
The survey was conducted using an online, standardized questionnaire that PSTs were invited to complete individually.After data export, the data were prepared and further processed in SPSS, Version 28.For all research questions, descriptive analyses were conducted and we applied t-tests to investigate mean differences between the cohorts.For research questions 3 and 4, multiple regression analyses were also conducted to calculate the percentage of explained variance of certain variables.

Results
The results are presented in relation to each research question.

1.
In what ways and to what extent were PSTs involved in school-based practice during their pandemic-affected internships?Were there differences between 2020 and 2021?
Table 2 shows how intensively PSTs were involved in various school activities during remote learning periods.
PSTs were most intensively involved in teaching-related activities (2020: M = 2.17; 2021: M = 3.12).They felt more intensively involved here as these activities included, for example, the planning and implementation of subject lessons in digital settings and the development of tasks and the provision of written feedback for school pupils.PSTs were least intensively involved in care activities (2020: M = 1.05; 2021: M = 1.25),where they were not involved at all or only very little.This included, in particular, providing emergency care for children who could not be cared for at home.These results correspond to the goals of the internship, in which the PSTs are supposed to be involved primarily in teaching activities.It is also noticeable that teaching-related activities involving direct communication with pupils apparently took place less intensively than activities without student communication.
There were statistically significant differences between the years 2020 and 2021.In 2021, the PSTs were significantly more involved, albeit remotely, in schools than in 2020.This applies to all areas of activity examined.There were medium to high effect sizes in relation to teaching-related activities.The effect was particularly high in relation to classroom activities without direct communication with pupils, 2020: M = 2.17, SD = 1.03; 2021: M = 3.12, SD = 0.83; t(298)=-8.08,p = .001,d = 0.97.Time use (hours per week) during remote periods also differed significantly between the two cohorts (Table 3).It can be seen that PSTs in 2020 invested a significant amount of time (M = 13.41 hours) in working on substitute tasks provided by the university, whereas this time use was very low (M = 1.93 hours) in 2021, t(287) = 10.37,p = .000,d= −1.31.Thus, in 2020, the university seems to have succeeded in providing appropriate substitute tasks in the short term.These could be, for example, fictional lesson planning, observation tasks on instructional videos, or participation in online offerings that were extracurricular but subject-specific.However, the standard deviation for these tasks in 2020 was very high (SD = 10.24), which may indicate that coordination on the scope of the tasks was not optimal.It seems unlikely that PSTs who invested less time in substitute tasks spent more time on school-related tasks, as the standard deviations were lower.
that in 2020 PSTs were predominantly occupied by substitute tasks from the university, while in 2021, there was more involvement in teaching and communication with pupils.
PSTs' communication with their tandem partners, e.g., joint lesson planning or exchange of observations as well as clarification of organizational issues, was constant across the cohorts at about 4 to 5 hours per week, t(295) = 1.93, p = 0.55; the relatively high standard deviation indicates that the intensity of tandem pairs' collaboration varied greatly.

2.
In what ways were PSTs supported, and how did they rate the quality of this support?Were there differences between 2020 and 2021?
Perceived social support from university faculty and school-based mentors is shown in Table 4. Support took the form of, for example, the ability to talk with mentors about everyday problems and regular exchanges of material with mentors; respondents perceived that mentors were a great help with teaching content and methods.
It is noticeable that support from school mentors (2020: M = 3.10; 2021: M = 3.46) was perceived as more intensive than that from university teachers (2020: M = 2.84; 2021: M = 2.81), although the mean values are very high overall (especially in 2020).Thus, PSTs were apparently very satisfied with the support they received (Table 4).
In 2021, support from mentor teachers was rated significantly higher than in 2020, t(391) = 5.27, p = .001,d= −0.53.This is another indication that at the beginning of the pandemic, university teachers were more involved in intensive exchange with PSTs, but by 2021 the mentor teachers in schools became more important.

How did PSTs rate their health status, career confidence, and stress during the COVID-19 pandemic? Were there differences between 2020 and 2021?
Ratings for overall job satisfaction, career choice confidence before (T1) and after the internship (T2), perceived strain from the COVID-19 pandemic, overall health, and number of hours working in a second job during the internship are shown in Table 5.
PSTs' career choice confidence in both cohorts was similar before (T1) the internship, t(400) =-1.09, p = .279,d=−.108.However, career choice confidence was higher at the end (T2) of the internship in 2021 (M = 6.44) than in 2020 (M = 6.12), t(397)=-3.05,p = .002,d=−.305.It should be mentioned here that PSTs prior to the internship in 2020 had undergone regular studies, while those in 2021 had their first opportunities for face-to-face studies with the internship phase after a break of about one year.
There were no significant differences between 2020 and 2021 in the other indicators.PSTs felt a relatively high level of job satisfaction.Covid-19-related stress was at a medium level, while PSTs felt healthy.PSTs worked about 10 hours per week in a second job, with a relatively high standard deviation.A regression model was calculated to examine whether membership in a particular cohort had a statistically significant effect on the development of career choice certainty (Table 6).
The results show that a significant percentage of the variance in career choice confidence at T2 (32%; R 2 =.324) could be explained.Both cohort membership and initial values at T1 had significant effects on PSTs' career choice certainty at T2.In 2021, PSTs' career choice certainty increased more strongly than in 2020.

How did PSTs' self-efficacy and self-assessed lesson planning skills evolve over the internship?
Were there differences between 2020 and 2021?Notes: Response options: 1 = does not apply, 2 = does not apply very much, 3 = applies to some extent, 4 = fully applies.In the following, we will examine to what extent PSTs perceived that they had achieved the goals of their internship.Both cognitive (lesson planning competence) and affective (teacher selfefficacy) aspects are examined.Table 7 shows the descriptive statistics and cohort comparison of the scales in question.
The results show that there is no statistically significant difference between the years 2020 and 2021 with regard to PSTs' planning competence on any of the three planning competence subscales.Lesson planning competence in terms of description of the learning group and learning task increased during the internship in both years.With regard to application and management, the values before and after the internship were almost identical for both years.
With reference to self-efficacy, it can be seen that while pre-internship self-efficacy was comparable in both cohorts, post-internship self-efficacy differed; it increased more in 2021, t(399) =-2.96, p = .003,d=−.296.To examine the extent to which cohort membership and initial measurements were statistically significantly linked to the development of self-efficacy, the results of a regression model are shown in Table 8.
The regression reveals that the variables explained a significant portion of the variance in selfefficacy at T2 (35%; R 2 =.353).PSTs' self-efficacy before (T1) the internship predicted their selfefficacy at the end (T2) of the internship.Similarly, cohort affiliation contributed to the explanation of variance.Thus, the internship in 2021 was more conducive to developing self-efficacy than that in 2020.

Discussion
The COVID-19 pandemic greatly impacted teaching in universities and schools (Carrillo & Flores, 2020).To promote social distancing, in-person teaching was largely discontinued and substituted with remote learning, which also had a large impact on PSTs, specifically their need for in-school internships.The key learning opportunities for PSTs during their in-school internships rely to varying degrees on in-person situations.While, for example, conversations between PSTs and mentors can be replaced relatively easily, especially by video conferencing, in-person teaching of school pupils, especially in elementary schools, is much more difficult to replace or demands alternative teaching-learning settings.Despite these pandemic-related changes in education methods, the learning goals for PSTs did not change significantly.These are described in this study as lesson-planning skills (König et al., 2017) and self-efficacy expectations (Schwarzer & Jerusalem, 1999).Our study sought to examine these learning outcomes during in-school internships in the pandemic and compare them between PST cohorts in 2020 and 2021.In what follows, we summarize and discuss the key findings, using the research questions as a guideline.
First, it should be noted that PSTs were more involved in classroom-related activities than care activities since the beginning of the pandemic.This is consistent with the conception of the internship as a teaching practicum (Gröschner, 2015).However, communication with pupils was significantly limited during the context.It is noticeable that in the second year of the pandemic, PSTs were significantly more intensively involved in school activities, even though they remained physically absent in schools.Accordingly, it can be assumed that schools learned a significant amount in relation to the involvement of PSTs as the pandemic evolved and identified new  Notes: Scale a : 1 = does not apply at all, 2 = does not apply very much, 3 = applies to some extent, 4 = applies completely; Scale b : 1 = does not apply at all, 2 = does not apply very much, 3 = applies to some extent, 4 = applies exactly, PC = Planning competence.
opportunities to involve PSTs by 2021.Further support for this thesis comes from the fact that in 2020 contact with school pupils was almost non-existent, and contact with mentors was also extremely limited; this suggested that schools were overwhelmed and struggled to involve PSTs at the start of the pandemic.At the same time, however, the university appeared to respond very quickly by taking on significant mentoring responsibilities and providing PSTs with substitute assignments.In the second year of the pandemic, such substitute tasks were not as necessary, as PSTs communicated more with mentors and spent longer on classroom-related tasks.Thus, as the pandemic evolved, a learning effect occurred at the schools and school-based involvement was successful in the second year of the pandemic.
Second, PSTs overall reported relatively good support from school and university teacher educators, an important component of their education (Beckmann, with school mentors being the most important source of support.Notably, in 2021, support from school-based mentors was rated higher than in 2020, while support from university-based teacher educators remained consistent.This is another indicator that the university quickly took over the school's responsibilities at the beginning of the pandemic, while mentors became more involved again over time (even if remotely).
Third, regarding their health and career choice confidence, it can be noted that in most cases (job satisfaction, career choice confidence, Covid-19-related stress, health status, and the number of hours worked in a second job), there were no differences between 2020 and 2021.Despite the differences between the baseline educational situations of the different cohorts-i.e., a very abrupt change to a remote internship in 2020 and a slow normalization of distance internships by 2021the PSTs reported high overall job satisfaction and career choice security and good general health.This corresponds with PSTs' general perceptions of internships as important sites of learning (Lawson et al., 2015).Notably, in 2021, career choice confidence increased more during the internship than in 2020-this is as expected given that, in 2021, the internship was accomplished in more typical educational conditions.Fourth, in terms of skill development, it is clear that PSTs rated their planning skills (König et al., 2015) relatively highly and that there were no differences between the years.This is probably explained by the fact that the substitute tasks provided by the university were able to cover lesson planning effectively.In terms of self-efficacy (Colson et al., 2017), the practicum in 2021 was more beneficial to PSTs than in 2020, which is consistent with the expectation that self-efficacy is developed primarily through independent actions (rather than the substitute tasks provided by universities in 2020).
Overall, it appears that PSTs were able to cope with the situation at the moment; they experienced support in both phases of the pandemic and made learning gains related to the aims of the internship.The university was quickly able to assume responsibility for the PSTs in 2020 and partially compensate for deficiencies in school attendance by providing support on substitute assignments.The schools were less successful in doing this at the beginning of the pandemic, but they adjusted over time to make greater involvement possible in 2021.Notes: Corrected R 2 = .353(p = <.001),**p < .01.

Recommendations for teacher education practice
The COVID-19 pandemic affected universities and schools in a range of unpredictable ways and had particularly strong effects on PSTs' in-school internships.Although the largest impacts of the COVID-19 pandemic have now diminished in most places, and education once again predominantly takes place in face-to-face situations, important consequences and recommendations for teacher education practice can be derived from our findings.
The first key implication is that university support for PSTs is indispensable and that university teachers (can) continue to support PSTs, especially in times of crisis when school-based support has largely been discontinued and schools have focused primarily on the core task of supporting school pupils.Secondly, the results imply that central teacher competencies such as lesson planning (the acquisition of which is linked to internships) can, from the PSTs' point of view, also be acquired to some extent without direct contact with school pupils.Apparently, individual reflection with teacher trainers could be just as effective here as the PSTs' own opportunities to try out new ideas.The third implication relates to the high standard deviations with regard to the extent of substitute tasks; this shows that accompanying seminars probably contained very different substitute tasks.Thus, our results imply that efforts should be made to create more coherence should a similar emergency situation arise.
In summary, while the transfer of internships to remote learning resulted in the loss of a variety of learning opportunities for PSTs, it also created new opportunities, which may be utilized in the future, such as 1) individualized (video) consultations on lesson planning, 2) micro-teaching in video-based formats, 3) analysis of instructional videos under general-didactic or subject-didactic foci, and 4) development of teaching-learning materials for schools in larger units, rather than materials focusing on individual lessons.We recommend that in future, these (and similar) formats should be further explored and used if proven effective for PSTs' learning, even if traditional face-to -face internships continue to be possible.

Recommendations for research and study limitations
This study also provides further recommendations for research.While the COVID-19 pandemic hit the world unexpectedly, and thus the move to remote learning did not arise from conceptual considerations, decisions can now be made about which portions of higher education (and, to a certain extent, school) offerings need to remain in-person and which can continue to be delivered partly or fully remotely.It can be assumed that education will largely be carried out in person.Researchers have the opportunity to continue to observe this development, conduct surveys in subsequent years and derive long-term perspectives for designing learning opportunities in practical phases.
In this respect, particular attention should be paid to the limitations and constraints of the survey in the present study.These limitations are particularly evident in: 1) the regionality of the survey, which was only conducted at one location.2) The rapid development speed of the items on remote learning opportunities without presence; further theoretical and empirical validations would be advisable.3) The use of self-assessments in relation to learning outcomes.To measure an actual increase in competence, it would be helpful to use tests as well.Furthermore, especially due to the high speed of development for the substitute tasks for PSTs, it could be assumed that these tasks differed significantly not only in the amount of time they took but also in terms of their activities and goals.To establish comparability across the survey-and ensure coherent teaching-it would be helpful if these tasks referred to a common framework.
Despite these limitations, this study brings new and important insights because 1) it approximates a full survey of PSTs in a long-term internship for a typical teacher education location, 2) it examines a longitudinal sample of PSTs before and after their internship, thus revealing student developments over the internship period, and 3) it compares two cohorts representing two very different groups with respect to the COVID-19 pandemic.While the first group was surprised by the sudden emergence of the pandemic just before their internship started, the second group were in a different situation one year after the outbreak began, in which all involved actors had gained experience and periods of attendance were possible again.In summary, this study contributes to the description of teacher education practice with regards to school internships and makes practical suggestions for teacher education practice.

Figure
Figure 1.Study design.