Predictors of Student Teachers’ ESD Implementation Intention and Their Implications for Improving Teacher Education

: Recently, the German state of Baden-Württemberg included ESD as a guiding perspective for all school curricula. Consequently, teacher education needs to empower and motivate student teachers to implement ESD in schools. In previous ESD research, however, the motivational dimen-sion rarely has been investigated. Psychological theories stress the importance of developing an intention for the actual realization of a behavior. Therefore, we investigated, in an online survey ( N = 366), which factors inﬂuence student teachers’ ESD implementation intention. Furthermore, we investigated how student teachers understand the guiding perspectives and what inﬂuences the actual ESD implementation from the respondents’ point of view. Via structural equation modelling, we found direct effects of subjective task value, expectation of success and ESD knowledge, as well as indirect effects of SD attitudes and ESD implementation beliefs on ESD implementation intention, but no effects for subjective norm and perceived costs. Analyses of open answers revealed complex understandings of ESD, but also misinterpretations mainly in terms of ESD as equivalent to environmental education. Furthermore, student teachers expected barriers (e.g., lack of time) and formulated support needs (e.g., teaching material) for implementing ESD in schools. Our ﬁndings stress the importance of advancing ESD implementation in teacher education.


Introduction
Our world is at a crucial crossroads: severe challenges such as the COVID-19 pandemic, the growing economic inequality and recent extreme weather events highlight the urgency of an extensive social transformation toward sustainability.Education is an essential prerequisite for advancing sustainable development, because "Education for Sustainable Development" (ESD) aims to empower and motivate individuals to create sustainable and just societies through education [1].
In order to implement ESD as comprehensively as possible in all areas of education, ESD has been established as a guiding objective of numerous international initiatives.Examples include the UN Decade of ESD (2005ESD ( -2014)), the Global Action Programme (GAP) on ESD (2015-2019), the 2030 Agenda for Sustainable Development with its 17 Sustainable Development Goals (SDGs) and the ESD for 2030: Towards Achieving the SDGs (2020-2030).In Germany, too, ESD is increasingly being anchored in educational landscapes.For example, the state of Baden-Württemberg has anchored ESD as one of the guideline perspectives in the new curriculum (see Figure 1) as well as a cross-sectional competence into teacher education [2,3].In its formulation, the guiding perspective focuses on understanding the multiple facets and the complexity of sustainable development, and on enabling students to deal with this complexity by developing future solutions as well as engaging actively in promoting sustainable development.Furthermore, the guiding perspective is rooted in the so-called "Beutelsbach Consensus", which has been formulated for political education in Germany [5].This consensus states that in schools, all education that contains political aspects has to adhere to three basic principles: 1.No overpowering: Overpowering students in any way in order to trick them into conforming to a certain opinion is not permitted.Teachers must support students to develop their own opinion towards a topic.The goal of all political education is the education of mature and responsible citizens.2. What is controversial in science and politics, has to be presented as controversial in school: It is important to describe different options and alternatives, in order to prevent indoctrination.This might even involve developing positions that might be alien to students in more detail.This also includes presenting those topics that are not controversial in science, such as human activity being the main reason for the current rapid climate change, as not controversial.However, this request of the Beutelsbach Consensus is sometimes misunderstood, and it is falsely assumed that climate change denial needs to be presented as a valid option, too.3. Students need to be enabled to analyze a political situation and their own interest, as well as to influence a political situation in terms of their own interests.This aligns with the guiding perspective, because it includes education about participation and engagements in order to pursue sustainable development.
Taken together, the idea behind the guiding perspective and the Beutelsbach Consensus comes close to what Vare and Scott [6] have called ESD 2, which is "building In its formulation, the guiding perspective focuses on understanding the multiple facets and the complexity of sustainable development, and on enabling students to deal with this complexity by developing future solutions as well as engaging actively in promoting sustainable development.Furthermore, the guiding perspective is rooted in the so-called "Beutelsbach Consensus", which has been formulated for political education in Germany [5].This consensus states that in schools, all education that contains political aspects has to adhere to three basic principles: 1.
No overpowering: Overpowering students in any way in order to trick them into conforming to a certain opinion is not permitted.Teachers must support students to develop their own opinion towards a topic.The goal of all political education is the education of mature and responsible citizens.

2.
What is controversial in science and politics, has to be presented as controversial in school: It is important to describe different options and alternatives, in order to prevent indoctrination.This might even involve developing positions that might be alien to students in more detail.This also includes presenting those topics that are not controversial in science, such as human activity being the main reason for the current rapid climate change, as not controversial.However, this request of the Beutelsbach Consensus is sometimes misunderstood, and it is falsely assumed that climate change denial needs to be presented as a valid option, too.

3.
Students need to be enabled to analyze a political situation and their own interest, as well as to influence a political situation in terms of their own interests.This aligns with the guiding perspective, because it includes education about participation and engagements in order to pursue sustainable development.
Taken together, the idea behind the guiding perspective and the Beutelsbach Consensus comes close to what Vare and Scott [6] have called ESD 2, which is "building capacity to think critically about what experts say and to test ideas, exploring the dilemmas and contradictions inherent in sustainable living" (p. 191).This idea of ESD goes beyond traditional environmental education, which Vare and Scott [6] call ESD 1, and consists of "promotion of informed, skilled behaviors and ways of thinking, useful in the short-term where the need is clearly identified and agreed" (p.191).
However, the leap from a theoretical mission to an actual implementation of ESD in schools can only succeed if teachers actually implement ESD in schools.Therefore, researchers investigate which factors support or hinder teachers' ESD implementation (e.g., [7][8][9][10][11][12][13]).While several ESD surveys among teachers have already been conducted (e.g., [7,8,10,[13][14][15]), there have been only a few large-scale ESD surveys among student teachers investigating which factors support or hinder their behavior to implement ESD in schools.Against the background that student teachers play a crucial role for the ESD implementation in schools in future [1], and that in several studies, teachers mentioned insufficient teacher education and teacher training as a barrier for the actual implementation of ESD in schools [7,8,10,[13][14][15], large-scale surveys might help to identify factors that support or hinder their intentions to implement ESD later in their careers as teachers.

The Role of ESD Implementation Intention
According to the findings of motivational psychology, an existing behavioral intention (here, the intention to implement ESD in school) represents a central basic prerequisite for the actual execution of this behavior (e.g., [16][17][18][19][20]).High ESD implementation intention among (student) teachers can thus be considered as an essential necessary condition for successful implementation of ESD in schools.Therefore, one goal of this study was to investigate what factors influence student teachers' ESD implementation intention.

Factors Influencing the Formation of Intention
First described by Ajzen in 1985, the Theory of Planned Behavior (TPB) is one of the most widely used theories to predict and explain behavioral intentions and behavior [21,22].According to the TPB [20], attitude, subjective norm and perceived behavioral control towards a certain behavior determine the intention to perform this behavior.In relation to our study, attitude describes the extent to which student teachers evaluate the implementation of ESD in schools as positive or negative.The subjective norm refers to the perceived social pressure that important reference people from professional or private life expect ESD implementation in their work as teachers.Perceived behavioral control describes the perceived control of actually being able to implement ESD in schools.
A central criticism of TPB is that the explanatory model is too narrow [23][24][25][26].According to Ajzen [20], in order to increase the predictive power of the model, the influence of potentially relevant background variables (e.g., general attitudes, emotions, gender, age, income, knowledge and experience) can be additionally examined.Furthermore, the model can be supplemented by additional determinants if they explain variance beyond the determinants of the TPB [27].Indeed, several studies already indicated that the TPB model can be extended by additional variables in order to increase the predictive power of the model (e.g., [28][29][30][31][32]).For example, the study by Weber and Fiebelkorn [32] showed that in addition to determinants of intention deducted from TPB, nature attachment and environmental concern also determine student teachers' intention to eat sustainably.Furthermore, the study by Cheung et al. [29] showed that beyond the TPB determinants of intention, college students' intention to recycle wastepaper was also determined by their knowledge.
Apart from TPB, the Expectancy-Value Theory (EVT) [33] also provides us with valuable insights on factors influencing an individual's intentions.According to the EVT, the personal expectation of success and the subjective task value influence the intention to show a certain behavior.In the context of the present research objective, the personal expectation of success describes the extent to which student teachers are convinced, on the basis of their perceived competences, that they can successfully implement ESD in everyday school life.The subjective task value indicates the value that student teachers attribute to the implementation of ESD in school.Eccles and Wigfield [33] specify four subcomponents of subjective task value: intrinsic value, usefulness, importance and cost [33].This means that student teachers experience a high task value if they expect to have fun and/or interest in implementing ESD in school (high intrinsic value), and if they expect that the implementation of ESD in school will conform with their own self-image, such as being responsible, justice-loving and nature-loving, (subjective importance).Furthermore, subjective usefulness indicates the extent to which student teachers are convinced that they can achieve long-term or short-term goals (e.g., career, reward and recognition of school leadership) by implementing ESD in schools.Perceived costs, on the other hand, refer to the negative aspect of the value component: they describe the extent to which student teachers believe that implementing ESD in schools involves a high degree of effort and/or sacrifice (e.g., fear of failure and less free time) [33,34].

Factors Influencing the Realization of Intention
Even though intention to perform a certain behavior seems to be a central prerequisite for the actual performance of this behavior, the findings of previous studies also indicate that about half of the individuals with behavioral intention do not succeed in acting according to their behavioral intention (e.g., [35][36][37][38]).In literature, such circumstances are described as intention-behavior gaps (cf.[39,40]).
The phenomenon of the intention-behavior gap raises the question why some people do not succeed in implementing their intention behaviorally.The Theory of Action Control by Kuhl [18] and the Rubicon Model of Action Phases by Heckhausen and Gollwitzer [19] offered a significant contribution to clarifying this question.These authors strongly argued for a distinction between motivational and volitional processes of an action course, because motivational processes are important for the formation of a behavioral intention and volitional processes for the realization of this behavioral intention, while volitional processes had remained unnoticed in motivation research for many decades [41].
Accordingly, intention-behavior gaps occur when problems, such as no time, intense emotions and lack of knowledge, arise in the realization of a behavioral intention (volitional processes) (cf.[18,19,42]).Therefore, the present study also investigated what barriers and support needs student teachers perceive with regard to the actual implementation of ESD in schools.

The Present Study: Research Questions and Hypotheses
The main goal of the present study was to identify relevant factors influencing student teachers' ESD implementation intention (motivational processes), because an existing ESD implementation intention is a central basic prerequisite for the actual implementation of ESD in schools as well as because the actual execution of the desired behavior (volitional processes), here to implement ESD in schools, is still in the future of the respondents.However, in order to gain insights into realization difficulties or realization facilitators, we investigated which factors might hinder or support the actual implementation of ESD from the respondents' point of view.Additionally, we investigated student teachers' ESD knowledge.With the goal of identifying and publishing misinterpretations regarding the ESD guiding perspective.
Regarding the ESD implementation intention, we formulated the following hypotheses:

Hypothesis (H1).
Based on the TPB [20], we expected that student teachers' ESD implementation intention is higher for participants with a positive attitude.

Hypothesis (H2).
Based on the TPB [20], we expected that student teachers' ESD implementation intention is higher for participants with a high subjective norm.

Hypothesis (H3).
Based on the TPB [20], we expected that student teachers' ESD implementation intention is higher for participants with a high perceived behavioral control towards ESD.

Hypothesis (H4a
).Furthermore, we expected that the motivational variables derived from EVT [33] have an additional influence on the ESD implementation intention.Implementation intention was expected to be higher for student teachers who attribute a high intrinsic value.

Hypothesis (H4b).
Based on the EVT [33], student teachers' ESD implementation intention was expected to be higher for student teachers who attribute a high subjective importance.

Hypothesis (H4c).
Based on the EVT [33], student teachers' ESD implementation intention was expected to be higher for student teachers who attribute a high subjective usefulness.
Hypothesis (H4d).Based on the EVT [33], student teachers' ESD implementation intention was expected to be higher for student teachers who attribute low perceived cost.
Hypothesis (H4e).Based on the EVT [33], student teachers' ESD implementation intention was expected to be higher for student teachers who attribute a high expectation of success towards ESD.

Hypothesis (H5).
Based on the research results of Weber and Fiebelkorn [32], we expected that student teachers' ESD implementation intention is higher for participants with high SD attitudes.

Hypothesis 6 (H6).
Based on the research results of Cheung et al. [29], we expected that student teachers' ESD implementation intention is higher for participants with high ESD knowledge.
The questions of how students understand the ESD guiding perspective as well what barriers and support needs they identify in order to implement ESD were addressed exploratively.

Participants and Design
The online survey included N = 366 student teachers (80.3% female, 18.9% male, 0.9% diverse, M age = 23.55,SD age = 3.07).Overall, 66.7% (n = 244) of the respondents were studying at a university of education, 31.7% (n = 116) at a university, and 1.6% (n = 6) indicated "other" at this point.Some 41.3% (n = 151) of the participants studied at least one ESD-related subject (biology, geography and basic sciences [8]).Basic science, in German called Sachunterricht, is a subject taught in elementary school in Germany that is intended to familiarize students with scientific and technical phenomena as well as with social, economic and historical conditions in their home region [43].Because of the advanced anchoring of ESD in the Baden-Württemberg higher education legislation Kultusministerkonferenz, this survey was targeted primarily at student teachers from Baden-Württemberg.Therefore, 89.6% (n = 328) of the respondents were matriculated in Baden-Württemberg and 10.4% (n = 38) in other German federal states.With regards to the gender distribution, this sample roughly equals the current student population [44].
Participants were recruited through teacher education centers, lecturers and social networks.Participation was voluntary.All subjects gave their informed consent for inclusion before they participated in the study.As an incentive, 20 vouchers were raffled among the participants.
The study was conducted as an online survey combining quantitative and qualitative questions.The quantitative questions were used to directly examine students' ESD implementation intention and the relevant influence factors by structural equation modelling.The open questions aimed at examining the understanding of the guiding perspective and the perceived barriers and support needs.The combination of both allowed us to better interpret quantitative data.

Questionnaire
The online questionnaire used in the present work was developed based on the TPB [20] and supplemented by the motivational variables of the EVT [33].In addition, participants were asked about perceived barriers and support needs regarding the actual implementation of ESD in schools as well as their interpretation of the guiding perspective.We chose the ESD questionnaire used by Waltner et al. [45] in the BUGEN-project as a base, because it allowed us to compare the results of our ESD survey with student teachers with the results of the ESD survey with in-service teachers.

Understanding of the ESD Guiding Perspective
In order to examine how student teachers understand the guiding perspective implemented in the curriculum, and to make sure that they are aware of the content of the guiding perspective, all participants were presented with the official guiding perspective on ESD at the beginning of the questionnaire.They were then asked to summarize the most important points in their own words.

Influencing Factors of ESD Implementation Intention
We used items from the ESD questionnaire of Waltner et al. [45] for measuring the determinants (attitude, subjective norm, perceived behavioral control) as well as a variety of potential background variables (general attitudes, age, gender, education, knowledge) of intention according to the TPB [20].For measuring the determinants of intention according to the TPB, we adapted N = 4 items for each scale (e.g., "ESD belongs in as many subjects as possible" for measuring ESD attitude; "ESD topics are important to schools" for measuring subjective norm; "Other topics would come up short if I would implement ESD" for measuring perceived behavioral control).For measuring the potential background variable general attitudes, we adapted the scale SD attitudes with N = 10 items (e.g., "It is still the case that politicians do far too little to protect the environment").Participants expressed their agreement for these scales on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree).For measuring the potential background variable knowledge, two items were adapted ("Did you know that ESD was included as a guiding perspective in the Baden-Württemberg education plans in 2016?." and "Have you attended any courses in the past four years that addressed ESD?").Participants expressed their agreement on this scale on a 4-point ordinal scale (1 = no to 4 = yes, I could spontaneously name goals/contents and 1 = no event, 2 = one event, 3 = two events, 4 = three or more events).
Furthermore, we adapted items from a study orientation questionnaire by Karst et al. [46] for measuring ESD implementation intention as well as for measuring the motivational factors according to the EVT (expectation of success, subjective task value: intrinsic value, importance, usefulness, cost [33]).For measuring ESD implementation intention, two items were adapted from Karst et al. [46] (e.g., "I am currently very determined that I will teach ESD in future") and one item was formulated independently ("Can you imagine teaching ESD?").Participants expressed their agreement on these items on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree).
For measuring the motivational factors according to the EVT, we adapted N = 3 items for each scale from Karst et al. [46], resulting in five scales: expectation of success (e.g., "I will teach the contents of ESD very well"), intrinsic value (e.g., "I will enjoy the implementation of ESD"), subjective importance (e.g., "Implementing ESD addresses issues that are important for me"), subjective usefulness (e.g., "The implementation of ESD will help me even more in my life") and perceived costs (e.g., "For the implementation of ESD I have to invest a lot of time").Participants expressed their agreement with these items on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree).
In addition, we used four control items to check the attention of the participants (e.g., "A small test for attention: Please tick the answer option 'strongly disagree'").

Perceived Barriers and Support Needs
An open response format was used for measuring student teachers' perceived barriers ("What difficulty or difficulties do you see in implementing ESD at school?") as well as perceived support needs ("What support/s would you like to have for the implementation of ESD at school?") regarding actual ESD implementation.

Coding of Open Answers
The data evaluation of the open questions was carried out according to the method of qualitative content analysis by Mayring [47] because this method allows for quantitative analysis once responses have been coded.

Understanding of the ESD Guiding Perspective
The process involved defining a categorization scheme based on inductive and deductive classification for open answers to the item: "What does the guiding perspective of ESD mean to you? Please summarize the most important points in your own words".The categories were mainly developed deductively from the guiding perspective on ESD of the federal state of Baden-Württemberg (see Figure 1).For this purpose, the guiding perspective on ESD was first paraphrased and then abstracted, to divide the content into 42 small units of meaning.These sense units were then reduced and finally transferred into 23 categories.From the results of the deductive generated categories, a coding guide including description, anchor example and keywords emerged.To check the categories, the participants' answers were sorted into the category scheme.Three categories had overlaps, these were transferred into other categories and sorted out.Responses that went beyond the categories were collected and analyzed by means of inductive content analysis.At the beginning, 8 inductively generated categories emerged, 3 of these remained central and were transferred into the final coding scheme.The inductively generated categories were added to the final coding guide with 23 categories in total (see Table 1).

Perceived Barriers and Support Needs
The category guides for the items measuring perceived barriers ("What difficulties do you see in implementing ESD at school?") and perceived support needs ("What support would you like to see for the implementation of ESD at school?") were obtained inductively from the participants' answers.First, the answers of the participants were paraphrased and clustered into units of meaning, followed by a reduction in the units if overlaps became apparent.The results were then combined into eight categories for perceived barriers and nine categories for perceived support needs.A first coding guide was developed.To ensure the validity of the coding manual, the category specifications were checked after approximately 30% of the statements and adjusted.The final coding guides including descriptions, anchoring examples and keywords can be found in Tables 2 and 3.

Interrater Reliability
To perform the interrater reliability check for all categories, 70 statements for each item (roughly 25%) were selected and coded by an independent rater.The range of Cohen's Kappa values of the interrater reliability test for the answers relating to the understanding of the guiding perspective of ESD ranged from 0.65 to 1, and more than half of the values were above 0.90.For the category "perceived barriers", the interrater reliability showed a range of 0.85 to 1, and more than half of the values were above 0.96.For the category "perceived support needs", the range of interrater agreement was 0.80 to 0.94, and the values of 5 out of 8 categories were above 0.89.Taken together, the interrater reliability was good to very good (based on [48]).

Procedure
The online survey was conducted with the online survey software LimeSurvey.Before the survey began, a qualitative pretest according to Döring and Bortz [49] was implemented with N = 5 student teachers in order to identify and eliminate possible processing difficulties.The online survey was available for participants from June 2020 to July 2020.The order of the items used in the survey was randomized to counteract potential halo or question series effects [50].With the exception of the demographic data items, which were placed at the end of the questionnaire, all items on a page had to be answered completely in order to continue with the questionnaire.With the aim to motivate student teachers to implement ESD in schools, participants had the opportunity to download an information sheet about ESD in schools after finishing the online questionnaire.

Economics, economic concept
The category guide was obtained by deductive and inductive content analysis according to Mayring [47].The guide is based on the guiding perspective of the state of Baden-Württemberg from "Educational Plans" by Ministerium für Kultus, Jugend und Sport Baden-Württemberg [4].a Statements that go beyond the content of the guiding perspective of the state of Baden-Württemberg.These categories were obtained inductively from the answers and added to the category guide."The school system as a whole is not very designed to promote reflective and sustainable thinking."School system, system, reformation The category guide was obtained by inductive content analysis according to Mayring [47].The category guide was obtained by inductive content analysis according to Mayring [47].

Understanding of the ESD Guiding Perspective
On average, in their summary of the guiding perspective, the student teachers described 4.3 aspects from the guiding perspective (SD = 2.2), with a maximum of 12 different aspects.Five student teachers did not produce a categorizable answer.
By far, the most frequent aspect that student teachers took out of the guiding perspective was the goal of environmental protection, with 57.9%, followed by the goal to enable students to make informed decisions (44.3%).Other common themes were the focus on future (time perspective) and the understanding of the concept of sustainable development (both 38.0%), as well as urging students to take responsibility (35.5%), acting sustainably (30.3%) and to participate and in public processes (27.6%).Interestingly, more than 20% of the respondents stated that raising awareness for sustainability issues was part of the guiding perspective, even though this is not mentioned in the text.Around 20% mentioned that ESD aims at improving communities in terms of good coexistence of everyone, which is also not mentioned explicitly in the guiding perspective.Almost 8% stated that the guiding perspective is about promoting certain values and norms.Competences such as empathy, creativity and democratic capacities were described altogether by 15%.Taken together, we found that the student teachers have a strong idea of ESD as environmental protection education, but are also aware of the broader perspective of understanding and dealing with complexity in the context of sustainable development.All results can be found in Figure 2.

Understanding of the ESD Guiding Perspective
On average, in their summary of the guiding perspective, the student teachers described 4.3 aspects from the guiding perspective (SD = 2.2), with a maximum of 12 different aspects.Five student teachers did not produce a categorizable answer.
By far, the most frequent aspect that student teachers took out of the guiding perspective was the goal of environmental protection, with 57.9%, followed by the goal to enable students to make informed decisions (44.3%).Other common themes were the focus on future (time perspective) and the understanding of the concept of sustainable development (both 38.0%), as well as urging students to take responsibility (35.5%), acting sustainably (30.3%) and to participate and engage in public processes (27.6%).Interestingly, more than 20% of the respondents stated that raising awareness for sustainability issues was part of the guiding perspective, even though this is not mentioned in the text.Around 20% mentioned that ESD aims at improving communities in terms of good coexistence of everyone, which is also not mentioned explicitly in the guiding perspective.Almost 8% stated that the guiding perspective is about promoting certain values and norms.Competences such as empathy, creativity and democratic capacities were described altogether by 15%.Taken together, we found that the student teachers have a strong idea of ESD as environmental protection education, but are also aware of the broader perspective of understanding and dealing with complexity in the context of sustainable development.All results can be found in Figure 2.

Data Analysis
In order to determine the influence of different variables on student teachers' ESD implementation intention, a Structural Equation Model was calculated.Before calculating this SEM, an Exploratory Factor Analysis (EFA) was conducted to empirically explore how the model of the TPB [20] can be extended by the determinants of intention according to the EVT (expectation of success and subjective task: intrinsic value, importance, usefulness and cost [33]).Based on these results and considering the internal consistencies for these scales, the final scales for all variables studied were then formed for the subsequent analyses.Descriptive statistics (sample mean, standard deviation, range, Cronbach's α, number of items, example items) of the final scales can be found in Table 4.

Exploratory Factor Analysis (EFA)
We performed an EFA, because we did not expect a certain number of factors.As the extraction method, we used the maximum likelihood method and as the rotation method, we used the VARIMAX rotation.Neither the scree plot nor the rotated factor matrix gave a clear picture of the number of factors.Therefore, we performed EFAs with six, five and four factors.The five-factor solution yielded the best result, explaining altogether 40.1% of the variance.Two items with loadings < 0.3 were excluded.We could confirm the scales expectation of success, subjective norm, as well as the factor perceived costs.In addition, the items measuring the variables intrinsic value, subjective usefulness and subjective importance, two items originally assigned to the scales ESD attitude and perceived behavioral control loaded on a common factor, which we labeled subjective task value.Thus, the items measuring perceived costs did not load with the items measuring intrinsic value, subjective usefulness and subjective importance on a common superordinate factor subjective task value, as assumed in the EVT by Eccles and Wigfield [33].This is in line with other studies (e.g., [51,52]) which also found that the items measuring the variables intrinsic value, subjective usefulness and subjective importance loaded on the superordinate factor subjective task value and the items measuring perceived costs loaded on a separate factor.Finally, the fifth factor consisting of three items we originally assigned to the scale ESD attitude and two items we originally assigned to the scale perceived behavior control, we labeled as ESD implementation beliefs.

Final Scales
Internal consistency, measured with Cronbach's alpha, was good (cf.[53,54]) for the scales of intention, expectation of success and subjective task value (see Table 4).For the scales of subjective norm and ESD implementation beliefs, values were slightly worse, but still acceptable (α = 0.66 and α = 0.68).However, since the exclusion of items would not have led to any improvement in the internal consistency of these two scales, the item assignment of these scales remained unchanged.The perceived costs scale was reduced by the item "I still have a lot to learn to implement ESD", as this resulted in an improvement in Cronbach's alpha from α = 0.65 to α = 0.68 for this scale.

Descriptive Statistics
On average, respondents rated their ESD implementation intention as rather high (M = 4.27, SD = 0.74, Range = 1.67-5.00,α = 0.81).The scale of ESD knowledge had the lowest mean value and the subjective task value scale had the highest mean value (see Table 4).This shows a general discrepancy between high ESD implementation motivation and low ESD knowledge.For more information, see Table 4. Five-point Likert scales (1 = disagree, 2 = rather disagree, 3 = partly agree/partly disagree, 4 = rather agree, 5 = agree) were used for measurement, with exception for measuring ESD knowledge (1 = no, 2 = yes, but I could not name goals/contents, 3 = yes, but I would have to think a bit before I could name goals/contents, 4 = yes, I could name goals/contents spontaneously and 1 = no event, 2 = one event, 3 = two events, 4 = three or more events).

Structural Equation Model (SEM)
We used full information maximum likelihood estimation for our analysis.All analyses were run with the lavaan package of the statistic program R. Based on our theoretical consideration as outlined in the hypotheses and the EFA described above, we formulated a model as presented in Figure 3, with subjective task value, perceived costs, expectation of success, subjective norm, ESD implementation beliefs, ESD knowledge and SD attitudes as predictors.Information on all correlations between the constructs used in the model are given in Table 5. ustainability 2021, 13, x FOR PEER REVIEW

Structural Equation Model (SEM)
We used full information maximum likelihood estimation for analyses were run with the lavaan package of the statistic program theoretical consideration as outlined in the hypotheses and the EFA des formulated a model as presented in Figure 3, with subjective task value expectation of success, subjective norm, ESD implementation beliefs, ESD SD attitudes as predictors.Information on all correlations between the c the model are given in Table 5.  ESD knowledge was operationalized through the item "knowledge on the ESD guiding perspectives" and the number of ESD-related courses that had been taken already by the student.For general SD attitudes, at the beginning, all 11 items were included.For SD attitudes, items with factor loadings in the measurement model below 0.40 were excluded; thus, the final number of predicting items for this factor was reduced to six.The resulting model showed acceptable fit indices: CFI = 0.915, RMSEA = 0.045 and SRMR = 0.048 (see [55]).Significant predictors of intention were ESD knowledge, subjective task value and expectation of success.ESD implementation beliefs were only significant in tendency.Subjective norm, perceived costs and SD attitudes did not have any direct predictive value for the intention (see Figure 4).ESD knowledge was operationalized through the item "knowledge on the ESD guiding perspectives" and the number of ESD-related courses that had been taken alread by the student.For general SD attitudes, at the beginning, all 11 items were included.Fo SD attitudes, items with factor loadings in the measurement model below 0.40 wer excluded; thus, the final number of predicting items for this factor was reduced to six.Th resulting model showed acceptable fit indices: CFI = 0.915, RMSEA = 0.045 and SRMR 0.048 (see [55]).Significant predictors of intention were ESD knowledge, subjective tas value and expectation of success.ESD implementation beliefs were only significant i tendency.Subjective norm, perceived costs and SD attitudes did not have any direc predictive value for the intention (see Figure 4).However, even though they did not directly predict intention, SD attitudes as well as ESD implementation beliefs had a medium correlation with subjective task value.Therefore, we developed a second model with a nested structure (see Figure 5).The model excluded perceived costs and subjective norms, and included ESD implementation beliefs and SD attitudes as predictors for subjective task value.The fit of the model was comparable to the first model: CFI = 0.909, RMSEA = 0.055 and SRMR = 0.053.However, even though they did not directly predict intention, SD attitudes as well a ESD implementation beliefs had a medium correlation with subjective task value Therefore, we developed a second model with a nested structure (see Figure 5).The mode excluded perceived costs and subjective norms, and included ESD implementation belief and SD attitudes as predictors for subjective task value.The fit of the model wa comparable to the first model: CFI = 0.909, RMSEA = 0.055 and SRMR = 0.053.A direct comparison of both models based on the AIC revealed that the second mode had a better fit with the data (AIC Model 1: 25,967.58 vs. Model 2: 20,746.01).Moreove when comparing the model fits of both models with an ANOVA, there was no significan difference between both models: Δχ 2 = 165.42,p = 0.25.Consequently, the model wit A direct comparison of both models based on the AIC revealed that the second model had a better fit with the data (AIC Model 1: 25,967.58 vs. Model 2: 20,746.01).Moreover, when comparing the model fits of both models with an ANOVA, there was no significant difference between both models: ∆χ 2 = 165.42,p = 0.25.Consequently, the model with fewer predictors, Model 2, was chosen to explain the data.This means that beliefs about the ESD implementation in school as well as general SD attitudes influenced the subjective task value, and that subjective task value, expectation of success and ESD knowledge together influenced the ESD implementation intention.However, contradicting our hypotheses, perceived costs as well as subjective norm did not play any major role in forming ESD implementation intention.Student teachers expected structural barriers, such as the lack of time (53.8%), to be most important, followed by barriers arising from the content, such as the complexity of the topics (27%).Roughly a quarter expected too little support by other teachers, and 18.3% felt that they were not sufficiently prepared by their teacher education.Smaller proportions expected to be hindered by lack of motivation of the students (16.4%) or protests of parents (7.1%).Interestingly, 10.1% found that the topics of ESD were too subjective to teach in school (see Figure 6).Student teachers expected structural barriers, such as the lack of time (53.8%), to be most important, followed by barriers arising from the content, such as the complexity of the topics (27%).Roughly a quarter expected too little support by other teachers, and 18.3% felt that they were not sufficiently prepared by their teacher education.Smaller proportions expected to be hindered by lack of motivation of the students (16.4%) or protests of parents (7.1%).Interestingly, 10.1% found that the topics of ESD were too subjective to teach in school (see Figure 6).Taken together, the student teachers expected little opportunity to teach ESD in the curriculum as well as little support from colleagues and headteachers.A fewer number feared that they might face opposition from parents, students and other stakeholders.Taken together, the student teachers expected little opportunity to teach ESD in the curriculum as well as little support from colleagues and headteachers.A fewer number feared that they might face opposition from parents, students and other stakeholders.

Perceived Support Needs
Student teachers saw the provision of appropriate teaching materials as the greatest source of support (39.6%).This was followed by about a quarter of student teachers who saw support opportunities at the structural level, such as fixed integration into the curriculum (27.6%), freedom for suitable forms of teaching (24.3%) and suitable education and training opportunities for (student) teachers (21.6%).Around one-sixth of the student teachers thought that support by colleagues would be helpful (15.8%), and another 17.5% wished for more financial and political support, as well as more collaborations with extracurricular partners (18.6%) and resources in everyday school life (19.9%).A small group (4.9%) asked for whole institution approach, in which the whole school acts as a role model.
In summary, the provision of materials and lesson plans was seen as the greatest opportunity, whereas structural support does not score as highly as might be expected in relation to Figure 7.

Discussion
Teachers play a crucial role in implementing ESD in schools.Therefore, ESD is being implemented more and more into curricula worldwide.In our study, we aimed at better understanding of how student teachers understand the ESD guiding perspective that was introduced by the state of Baden-Württemberg (Germany), and to examine what influences student teachers' intention to implement ESD in schools (motivational processes).In addition, we investigated perceived barriers and support needs regarding the actual implementation of ESD in schools to identify potential influencing factors for intention-behavior gaps (volitional processes).

Discussion
Generally, ESD implementation intention was high among student teachers.This result indicates that the central basic prerequisite for the actual implementation of ESD in schools among student teachers is fulfilled (cf.[16][17][18][19][20]).
Consistent with the EVT Model [33], high subjective task value and expectation of

Discussion
Teachers play a crucial role in implementing ESD in schools.Therefore, ESD is being implemented more and more into curricula worldwide.In our study, we aimed at better understanding of how student teachers understand the ESD guiding perspective that was introduced by the state of Baden-Württemberg (Germany), and to examine what influences student teachers' intention to implement ESD in schools (motivational processes).In addition, we investigated perceived barriers and support needs regarding the actual implementation of ESD in schools to identify potential influencing factors for intentionbehavior gaps (volitional processes).

Discussion
Generally, ESD implementation intention was high among student teachers.This result indicates that the central basic prerequisite for the actual implementation of ESD in schools among student teachers is fulfilled (cf.[16][17][18][19][20]).
Consistent with the EVT Model [33], high subjective task value and expectation of success towards ESD implementation in schools were found to be positive predictors of ESD implementation intention.Thus, these results indicate that student teachers' ESD implementation intention is particularly high if they attribute high subjective value, i.e., fun, interest, importance and usefulness, to ESD implementation in schools, and if they think that they can successfully implement ESD in everyday school life.These findings are in line with results of previous studies (e.g., [56][57][58]).For example, in a study by Bong [56], students' subjective task value and expectation of success influenced their course enrollment intentions.The superordinate variable subjective task value is based on the results of the present EFA, composed of the subcomponents intrinsic value, subjective importance and subjective usefulness.Eccles and Wigfield [33] note that the variable "intrinsic value" is closely related to the variable "intrinsic motivation" of the self-determination theory according to Deci and Ryan [59].Intrinsic motivation is defined here as a person's effort to do something for its own sake, because of pleasure or interest in the thing.A variety of study results show that intrinsic motivation positively influences the quality of learning processes and outcomes [60].For example, findings indicate that intrinsically motivated learners, compared to extrinsically motivated learners (motivated by external factors), experience more attention, curiosity and self-confidence toward a particular task, which translates into more sustained persistence, creative ability, better performance [61,62], increased identification with professional education [63], increased job satisfaction [64] and general well-being [65].Thus, ESD courses for student teachers should be designed to elicit enjoyment and interest in implementing ESD in schools.According to Deci and Ryan [59], the enjoyment of and interest in implementing ESD in schools increases by addressing needs competence (e.g., informative feedback, recognition, professional knowledge), autonomy (e.g., self-directed learning and teaching/choice, knowledge of agency) and social inclusion (e.g., supportive working and learning environment).In addition, content relevance (e.g., realism), instructional quality (e.g., clear structure) and perceived content interest of the teacher (e.g., commitment) also influence the enjoyment of and interest in implementing ESD in schools [59].Therefore, it seems profitable to consider these aspects when designing ESD courses.The subjective importance regarding the implementation of ESD in schools could be strengthened through ESD courses, for example, by considering individual interests, wishes and needs of the participants (cf.[19]).Subjective usefulness regarding the implementation of ESD in schools could be promoted in teacher education, for example, through recognition, such as a certificate of participation, or reward, such as ECTS credits.The expectation of success among student teachers can be promoted by teaching ESD-relevant professional knowledge (content knowledge, pedagogical knowledge and pedagogical content knowledge).This includes, for example, teaching knowledge about current problem situations as well as the goals and contents of ESD (content knowledge), fostering skills for changing perspectives and accepting a changed teacher role, i.e., as a learning supporter instead of a role model (pedagogical knowledge) and using appropriate teaching strategies, materials and media (pedagogical content knowledge) [66][67][68].
Moreover, student teachers' ESD implementation intention was particularly high if they were familiar with the goals and contents of the ESD guiding perspective, and if they had already attended ESD-related courses.In the present study, however, on average the respondents stated that they could not name any goals and contents of the guiding perspective ESD and attended either no or only one ESD course.In line with these findings, only 20% of the surveyed teachers in the ESD survey by Waltner et al. [45] stated that they were familiar with the ESD guiding perspective, most of them not being able to name concrete goals or content.Furthermore, only 15% of these surveyed teachers stated that they had attended an ESD-relevant training event in the past three years [45].These findings emphasize the relevance of greater ESD implementation in student teacher education as well as in in-service teacher training.Additionally, reflective essays or learning protocols that prompt (student) teachers to reflect about the connection of knowledge from ESD courses with knowledge (student) teachers gained in other courses or school practice might be promising, because they raise the availability of the knowledge and the perceived usefulness of the contents at the same time.
Finally, we found an indirect effect of SD attitudes and ESD implementation beliefs on intention through subjective task value.This means that the more important a sustainable development in general was to the student teachers, and the more they were convinced that it is principally possible to implement ESD in everyday school life, the more positively they evaluate ESD implementation in schools and, as a result, the higher their ESD implementation intention.Our findings are in line with the assumptions of EVT that the subjective task value as well as the expectancy of success are an important prerequisite for forming an implementation (e.g., [57,58,69,70]).In addition, the indirect effects of general attitudes have been documented by other researchers, including Weber and Fiebelkorn [32].ESD implementation beliefs can be fostered in ESD courses by addressing the diverse implementation possibilities of ESD in schools (e.g., excursions, projects, experiments, films, artistic design, working with texts).Teacher education could support a positive attitude towards ESD implementation in schools among student teachers, for example, by showing successful examples of ESD implementation (e.g., teaching units and school projects).In addition, communicating and reflecting on the relevance of ESD for a social transformation towards sustainability could play a key role for a positive evaluation of ESD in schools.
When asked about factors possibly supporting or hindering their implementation of ESD in school, student teachers described factors on the individual level (such as the importance of knowledge), but also structural factors on higher levels (integration in the curriculum, availability of resources) and on school level (such as support of colleagues and headteachers).Interestingly, the perceived barriers and support needs resemble the problems named by already teaching teachers, such as a lack of ESD teacher education and teacher training, more ESD teaching material, a lack of ESD knowledge, general overloading of the curriculum and a lack of ESD anchoring in the curriculum (e.g., [7,8,10,[13][14][15]).Thus, the perceived barriers and support needs among student teachers appear to be strongly in line with the actual barriers and support needs regarding ESD implementation in schools.These (perceived) barriers and support needs should be taken seriously in order to strengthen ESD implementation intentions as well as to avoid intention-behavior gaps, such as reported by teachers in previous ESD surveys [7,8,10,14,15].In order to enable student teachers to implement ESD in schools, a profession-oriented ESD teacher education-i.e., a teacher education that enables the networking of ESD-relevant content knowledge, pedagogical content knowledge and pedagogical knowledge-is necessary, because it generally helps to increase the transfer of knowledge [71].Following Waltner et al. [45], we therefore recommend an improvement in the quality and quantity of ESD teacher education as well as a presentation of high-quality ESD teaching materials and implementation methods in ESD courses.In order to increase the actual participation in ESD teacher education, there is also a need for appropriate incentives or obligations (e.g., ECTS points and certificates).Another possibility to address perceived barriers and to provide support to students are approaches based on the idea of communities of practice, in which students and teachers form groups, aiming at exchange about ESD in school and realizing and reflecting ESD-projects.A smaller way of realizing community-based approaches would be to invite more teachers to university in order to talk and exchange about successful projects.
Interestingly, student teachers showed a diverse interpretation of ESD.On the one hand, most student teachers associated ESD with environmental protection, which is more related to the simpler idea of ESD 1 [6] that ESD should generally raise awareness of sustainability issues, to convey knowledge and to promote certain values and norms.On the other hand, many student teachers summarized acting and taking responsibility in a complex world with a perspective on future developments as most important aspects of ESD.In order to promote a sophisticated understanding of the guiding perspective, it might be helpful to actively address the difference between ESD 1 and ESD 2, and support students in their reflection of their understanding.
Another interesting result is that there are also quite a few student teachers who perceived ESD and the related topics as too subjective.This could be due to a lack of understanding of the principles of the Beutelsbach Consensus, which explicitly give recommendations to deal with topics related to political actions and personal opinions.As formulated in the Consensus, ESD explicitly aims at empowering students to make their own decisions and to analyze and influence a political situation according to their own interests.Given the strong impact that non-sustainable development will have on young peoples' future, it should be even more important to empower them to understand current developments, rather than take it out of the curriculum.Again, these findings emphasize the importance of learning more about ESD in teacher education in order to avoid misinterpretations, such as being too subjective and understanding ESD as a mainly environmental education program aimed at convincing individuals to adopt certain behaviors.
In our view, the results from the open answers in general underscore the role of knowledge about ESD for the implementation in school.For example, the call for more ESD teaching material shows that little is known about the vast amount of publicly available material.Furthermore, the call for better curriculum integration stands somehow in contrast to the fact that ESD is already a guiding perspective.Furthermore, although the guiding perspective of ESD is an integral part of what should be learned in school and allows multiple ways to do so, student teachers requested additional time, such as project weeks, which shows that-other than what is expressed in the guiding perspectivestudent teachers perceived ESD as an add-on rather than as part of everyday teaching.This interpretation is also supported by the finding that quite a number of respondents called for engagement of external experts for teaching ESD-related topics.Again, this shows the need for improving the quality and quantity of ESD teacher education.
In contrast to our hypotheses, we found no significant effects for perceived costs and subjective norm.This means that a perceived high degree of effort as well as a perceived social pressure to implement ESD in their work as teachers had no influence on student teachers' ESD implementation intention.One reason for the absence of an effect could be the operationalization of this variable.In the present study, the items measuring perceived cost referred exclusively to the perceived (time) effort associated with ESD implementation.However, according to Eccles and Wigfield [33], the variable "perceived cost" includes all negative aspects, such as anxiety, associated with a particular task.Therefore, it would be interesting to examine in future studies whether perceived cost has a significant effect on student teachers' ESD implementation intention when this variable also measures subjective anxiety (e.g., fear of failure, fear of learner reaction).On the other hand, even though the subjective norm was no significant predictor in our models, the open answers indicate that there are at least some student teachers who anticipate little support from colleagues, little interest from pupils, and, moreover, negative responses from parents.This seems to be a kind of contradiction.One explanation might be that even though the student teachers in the sample voice concerns that can be summarized as a negative subjective norm, they do not believe at the time of studying that it will reduce their intention to implement ESD when they are in service, if they see at the same time the importance of sustainable development, feel that it is their job to implement ESD in school and expect a positive task value if they act in line with their own values.Another reason for the absence of an effect could be that for many student teachers, the implementation in school is still far in the future, and therefore, they cannot imagine the impact social norms may have on them.A significant effect of the subjective norm on the intention to implement ESD could, therefore, become apparent at a later point in their professional careers.It would be desirable to test this assumption in future studies.

Limitations
When interpreting these results, it should be considered that the mentioned intention to implement ESD is based on a prospective self-report, which might be influenced by social desirability.Accordingly, respondents' answers might have been positively biased by widespread social norms and expectations regarding sustainable development [49].In addition, it should be considered that voluntary participation probably could not avoid a certain selection bias towards ESD-receptive student teachers.To counteract these limitations, respondents were informed in advance of the study that participation in the study would be completely anonymous.
Moreover, a high intention does not immediately lead to an actual execution of this behavior.Instead, intention-behavior gaps can occur when problems with the realization of this intention (e.g., no time, lack of knowledge) arise (cf.[19]).Because the actual execution of implementing ESD in schools by student teachers is expected only in the future, it was not possible to investigate actual but rather perceived barriers and support needs which influence the realization of ESD implementation in schools.
Another limitation of the study is the sample, which studied different subjects as well as at different universities.Therefore, the structure of factors influencing implementation intention could be different for subgroups.In addition, the factors identified in the study explained only about half of the variance.Therefore, further studies are needed to gain more insight into other factors that influence student teachers' intention to implement ESD.
Despite these limitations, our study provides insights for the conception of ESD teacher education as well as further research needs for a successful ESD implementation in schools.

Conclusions
In summary, the results of the present study indicate that the intention of student teachers to implement ESD is significantly determined by cognitive and motivational aspects.In this regard, the present study revealed a discrepancy between motivation and knowledge.Although the respondents were motivated to implement ESD in schools, especially when they had a positive attitude towards sustainable development, their understanding of the content and goals of the ESD guiding perspective can be further improved.These findings are of direct relevance for the design of ESD teacher education to increase the likelihood that student teachers will implement ESD in their future work as teachers and thereby drive a societal transformation toward sustainable development.Specifically, ESD teacher education should include all domains of teachers' professional knowledge (content knowledge, pedagogical knowledge, pedagogical content knowledge), as this provides a coherent ESD knowledge base [72,73] and, at the same time, increases the motivational aspects that influence the actual implementation of ESD in schools.Professionoriented teacher education can be achieved by working on learning tasks that enable a coherent connection between content knowledge, pedagogical knowledge and pedagogical content knowledge.More specifically, the participants should be able to experience how they can successfully implement the goals and contents of the ESD guiding perspective in schools, i.e., an implementation that has an impact on students' sustainability competence development.
Thus, this study significantly contributes to the investigation of relevant factors influencing student teachers' intention to implement ESD in schools.The results also provide important information for the design and implementation of ESD teacher education.To advance actual ESD implementation in schools, future studies should investigate further ways to counteract the (perceived) barriers frequently reported by (student) teachers that hinder actual ESD implementation in schools (e.g., overload of curricula, teaching materials, financial and political support and more collaborations with extracurricular partners).The present study provides an important theoretical and empirical basis for such investigations.

Table 1 .
Coding scheme for the understanding of the guiding perspective on ESD of student teachers.
"This is about empowering people to form opinions and make decisions.""Studentsshouldlearnmoreabout(non-)sustainability, its impacts and their own possibilities for action.""Critical,conscioushandling of information."Formingopinions,weighingup,makingdecisions, critical thinking, consequences, information, knowledgeActingResponding to problem situationsStudents are enabled to respond to current and, above all, future problems and risks.Students should be able to deal with today's problems and risks."Problemsand challenges in relation to the world of tomorrow.""Enables people to reflectively deal with individual and societal problems and fields of action.""Short or long-term consequences.""Besides acquiring knowledge, empathy, commitment and dealing with risks play a major role.""Explaining consequences and risks of an unsustainable lifestyle."
"For this, the most important values and norms are taught."Moral,values,normsResponsibilityStudentsareencouragedto take responsibility for the consequences of their actions.Actions are taken with foresight to avoid negative consequences as much as possible."Participateindemocratic processes through responsible, thoughtful decision-making.""Studentsshould learn to show responsibility for sustainability and global justice.""Forme, the guiding perspectives also convey that each of us can and must participate.""Inmy opinion, ESD is closely related to value, democracy and peace education."Peace, peaceful

Table 2 .
Category guide for perceived barriers regarding ESD implementation in schools.Barriers are seen concerning the integration of ESD in cross-curricular teaching.Barriers can occur in the subject integration and due to the complexity of the topic."Referencemust be made to the subject taught.""[ . . .] applied in as many subjects as possible.""One must break down a very complex topic and present it in a simplified, but the correct way." "Too little support from colleagues, school management.""Some teachers just don't care enough to bring this subject properly to the students.""[ . . .] gives a lot of extra work for a teacher to do."

Table 3 .
Category guide for support regarding ESD implementation in schools.
Support that contributes ESD-related teacher education.ESD is to be promoted in general as well as through knowledge about teaching ESD in a subject specific, cross-curricular and didactic way (professional knowledge)."Furthertraining for teachers in the area of teaching ESD and bringing it closer to the students.""Extensive training and preparation of teachers.""Expert lectures."

Table 4 .
Descriptive statistics (sample mean, standard deviation, range, Cronbach's α, number of items, example item) of the final scales.

Table 5 .
Correlations of the variables of the final scales.