Testing the Associations Among Pre-Service Teachers’ Sense of Preparation, Readiness to Engage in the Profession, and Self-Efficacy for Teaching: Validation of a Causal Framework

Abstract

This study investigated shifts in preservice teachers’ perceived preparedness, readiness to engage in the profession, and teaching self-efficacy before and after a culminating field experience within a two-year, master’s level initial teacher education program at a large Australian university. Employing the Pre-service Teacher Professional Experience (PTPE) scale, we examined pre-service teachers’ pre- and post-field experience changes for each construct. We then used structural equation modeling to test a hypothesized causal framework, confirming theorized relationships among preparedness, readiness to engage, and self-efficacy for teaching. Path analyses revealed significant shifts in how specific dimensions of preparedness contributed to preservice teachers’ readiness to engage in the profession, which, in turn, influenced their teaching self-efficacy. Findings support the utility of the PTPE and causal framework in future research on teacher preparation, particularly as a means for teacher education programs to assess pre-service teacher development in alignment with professional standards for teaching.

1. Introduction

Entry into the teaching profession is a difficult journey for beginning teachers. The position of fully credentialed teacher places great demands on new educators, including the wide range of knowledge and skills needed to effectively impact student learning (Loughran, 2013), the multitude of in-the-moment decision-making inherent in each school day (Coskun et al., 2021; Jackson, 1990; Phillips et al., 2021), and the emotional intensity of working interpersonally with learners who have varying support needs and motivations (Chen, 2020). Education researchers and teacher educators are interested in understanding how to best prepare pre-service teachers (PSTs) for a stable beginning of their careers. In this article, we analyze the results of a causal framework that hypothesizes potential associations among three critical variables that support beginning teachers’ development and inform professional actions deemed critical for success; (1) preparedness for teaching, (2) readiness to engage in the profession, and (3) teacher self-efficacy. All three variables were measured through the employment of the Preservice Teacher Professional Experience (PTPE) scale (Abraham et al., 2021), which was administered to PSTs before and after a culminating field experience within a large Australian teacher education program. The variables of preparedness and readiness to engage are aligned to the Australian Professional Standards for Teachers (AITSL, 2022a). This study is unique from the extant literature by analyzing these latent constructs collectively, in alignment with professional standards for teaching, and through the lens of a causal framework. The results demonstrate the usefulness of employing the PTPE and causal framework in future studies to assess and compare the outcomes of teacher education program designs.

2. Literature Review

2.1. Self-Efficacy, Preparedness and Teacher Education Program Designs

Much of the research on PSTs’ development in teacher education programs has focused on their efficacy for teaching (for thorough reviews of the teacher self-efficacy literature, see Tschannen-Moran et al., 1998; Zee & Koomen, 2016). Rooted in Bandura’s landmark research (Bandura, 1977, 1997), a teacher’s self-efficacy refers to their beliefs about their capability to teach. Typically, a teacher’s personal self-efficacy (as opposed to general efficacy) focuses on their capability to facilitate instructional strategies, classroom management and student engagement (Brown et al., 2015). There is a considerable research base on the benefits of PSTs developing their self-efficacy for teaching. Zee and Koomen’s (2016) review of this research helpfully explains the cyclical nature of self-efficacy. They assert that teacher self-efficacy can support quality teaching, which has the potential to support student academic achievement and teacher well-being, which in turn have the potential to feed back into an increase in self-efficacy. They report many positive outcomes across the literature from increases in teachers’ self-efficacy, e.g., increased likelihood of coping with stressful situations, engaging in student-centered teaching, and developing a commitment to the profession. Increases in self-efficacy have been shown to contribute to a higher teaching performance during student teaching (Brown et al., 2021). Moreover, a teacher’s self-efficacy has been shown to correlate to their perceptions of a school’s collective efficacy (Stephanou et al., 2013), and collective efficacy is a significant predictor of increases in student achievement (Donohoo et al., 2018; Hoy et al., 2002).

Given its importance for beginning teacher development, a key question is, “What variables help develop teachers’ self-efficacy?” It is typically not explained by personal characteristics, for example, age and gender (Darling-Hammond et al., 2002) or the education level of teachers’ parents (Olczak Filipowicz, 2023). Rather, prior research points to preparation for teaching as a key predictor. Darling-Hammond et al. (2002) found that self-efficacy was differentiated among teachers by such factors as teaching grade level, ethnicity, teaching in area of certification, and years of experience, but their perceptions of preparedness was the strongest predictor of self-efficacy when those variables were controlled. Likewise, Schepens et al. (2009) found that PSTs self-efficacy at graduation was predicted by preparation.

If preparation positively impacts self-efficacy, which sets up a beginning teacher for potential success as they enter the profession, have teacher education programs been found to promote PSTs’ preparedness and self-efficacy? While PSTs will not necessarily feel equally prepared across different areas of preparation (Koehler et al., 2013), the initial teacher education program can have a considerable impact. Hulme and Wood (2022) found that beginning educator’s teacher education programs were the most important predictor of professional development needs at the end of the first year of teaching. Field experiences (termed ‘professional experience’ in the Australian context) play a particularly substantial role in PSTs’ teaching development. It is during these experiences, for which pre-service teachers (PSTs) observe and participate in a grades K-12 classroom within their licensure area, that PSTs often feel they learn the most about teaching (Olczak Filipowicz, 2023). While it can be a vulnerable and stressful time (Caires et al., 2012), quality field experiences have been shown to correlate to higher ratings of PSTs’ self-reported instructional skills (Biermann et al., 2015) and help PSTs learn about the diverse needs of their students (Adegbola, 2022; Cook & van Cleaf, 2000). Studies have revealed the positive impact of field experiences on increasing a beginning teacher’s self-efficacy (Knoblauch & Hoy, 2008; Levi-Keren et al., 2022).

2.2. Current Methods for Analyzing PSTs’ Self-Efficacy and Preparedness

Acknowledging the general positive relationship between teacher education program field experiences and increases in perceptions of preparedness and teacher self-efficacy, psychometrically valid and generalizable insights remain elusive. There are several limitations with current analysis methods. Analyzing pre-service teachers’ survey responses item-by-item through the use of descriptive statistics, be it through a report of percentages (Bondar et al., 2021) or ranking individual item means (Adegbola, 2022), can report a broad snapshot of teachers’ perceptions but do not allow for hypothesis testing. A mixed-methods approach can add nuance to a descriptive analysis of survey items, but because it typically involves a small number of participants (Koehler et al., 2013), the findings lack generalizability.

An analysis of paired t-tests on sub-scales of instruments like the MPPTP (Measuring perceptions of pre-service teachers’ preparedness) and TSES (Teachers’ sense of self-efficacy, developed by Tschannen-Moran & Woolfolk Hoy, 2001), as conducted in a study by Brown et al. (2015), provides a helpful opportunity to investigate statistically significant changes before and after a field experience. However, employing t-tests without first conducting factor analysis means there is no guarantee that each sub-scale represents a distinct and valid construct within the given population. At the most sophisticated level, researchers have used separate instruments to engage in regression analysis to determine if one variable predicts another. For example, Brown et al. (2021) used multiple regression to predict the extent to which PSTs’ sense of preparedness and self-efficacy performance scores predicted their teaching performance scores from an evaluation instrument. Reynolds et al. (2016) used regression analysis and two separate instruments to assess whether PSTs’ participation in a particular program (with an extended number of field days when compared to the traditional program) was associated with pre-service teachers’ perceptions of preparedness, support, and induction while controlling for demographic variables. Compared to a sole reliance on means testing, regression analyses have the added benefit of testing associations among variables. Yet, after Zee and Koomen’s (2016) extensive review on self-efficacy research, they concluded that the teacher education field needs to move beyond correlation studies, arguing that more emphasis should be placed on structural equation models (SEMs), “in which differing pathways of influences can be compared, and the temporal precedence of predictors can be established” (p. 1010).

2.3. Preparedness Connections to Professional Standards for Teacher Education

As the field of teacher education has become increasingly professionalized, organizations and governments across the world have developed standards for teacher preparation to articulate the knowledge and skills that beginning teachers should develop during teacher education programs. Teacher standards from organizations in the U.S. include accreditation organizations, i.e., the Council for the Accreditation of Educator Preparation (CAEP, 2025) and organizations for teachers of specific subjects, i.e., the collaboration between the National Science Teachers Association and Association of Science Teacher Educators (Morrell et al., 2020). Countries often articulate such standards as well, e.g., the United Kingdom’s teacher standards (Department for Education, 2021). Similarly, standards for teacher registration and licensure have been prominent in the country of Australia. Call (2018) explains that they first appeared in Queensland in the 1970s based on qualifications, but the most recent iteration is a result of political concerns about international rankings on standardized tests. The Australian Institute for Teaching and School Leadership (AITSL) was formed in 2009 (Dinham, 2013), leading to the creation of the Australian Professional Standards for Teachers (APST). These standards explain what Australian teachers should be able to do at different stages of their career—graduate, proficient, highly accomplished, and lead teacher (AITSL, 2022a). AITSL provides the requirements of initial teacher education programs to be accredited. Teacher education programs are expected to align their program components and assessments—e.g., a teaching performance assessment—to the graduate-level APST (AITSL, 2022b).

While a few studies investigating beginning teachers’ preparedness have aligned their survey items to professional standards for teachers (Brown et al., 2015; Reynolds et al., 2016), it is not a common practice. Moreover, to date we have not found a study that used factor analysis to validate sub-scales in alignment with professional standards. A commitment to establishing valid and reliable sub-scales for constructs like sense of preparedness, as opposed to the reliance of a single item for overall preparedness (Darling-Hammond et al., 2002) has great potential to provide deeper insights into the relationship between preparedness and self-efficacy.

2.4. Situating the Study

Prior to 2021, the field of teacher education was lacking an instrument that simultaneously accomplished the following objectives:

(1)

Facilitate the study of preparedness for teaching and self-efficacy within one psychometrically valid instrument

(2)

Propose potential mediating variables to explain how preparedness may predict the development of PSTs’ self-efficacy, and

(3)

Align the preparedness variable to initial teacher education standards

We designed such an instrument, the PTPE, and explained its development and validation in a previous article (see Abraham et al., 2021, for the entire scale). We now have the capacity to continue a deeper contribution to the teacher preparedness and self-efficacy literature by analyzing PSTs’ potential shifts for each construct before and after a field experience, and using SEM path analyses to test the causal framework inherently situated within the instrument.

Specific to the APST, we posited that standards associated with two of the APST domains—knowledge and practice—represent the fundamental concepts and skills that beginning teachers need to be prepared for. Confirmatory factor analysis validated four sub-scales that comprise the overarching variable of preparedness (Abraham et al., 2021). The two APST knowledge domain sub-scales are relation, focused on understanding students and how they learn, and knowledge, specifically knowing content and how to teach it. The two APST practice domain sub-scales are strategies, planning and implementing effective teaching, and environment, creating a safe and supportive learning environment for students.

We viewed the APST’s third domain, professional engagement, as distinct from the knowledge and practice domains. The professional engagement domain is not a list of discrete pedagogical concepts or skills to develop. Rather, its fundamental nature is about engaging as a teaching professional, which includes the phenomenon of learning on one’s own and collaborating with colleagues to participate in the full range of the job. We theorized that the development of PSTs’ conceptual knowledge and teaching skills would facilitate their readiness to engage as a professional. The PTPE’s engage sub-scale (aligned to the APST professional engagement domain) may therefore serve as a mediating variable to understand how PSTs’ self-efficacy is developed within initial teacher education. The alignment of the instrument’s sub-scales to the APST, and hypothesized causal framework is presented in

Table 1. Standards for teachers and construct names on the PTPE (Abraham et al., 2021).

Domains of Teaching	Australian Professional Standards for Teaching	Construct Name on the PTPE
Professional Knowledge	1. Know students and how they learn	Relation
	2. Know the content and how to teach it	Knowledge
Professional Practice	3. Plan for and implement effective teaching and learning	Strategies
	4. Create and maintain supportive and safe learning environments	Environment
	5. Assess, provide feedback and report on student learning 1
Professional Engagement	6. Engage in professional learning	Engage
	7. Engage professionally with colleagues, parents/carers and the community 1

¹ Standards not considered in this procedure as the descriptors were not deemed to be relevant to PSTs’ classroom experiences prior to and during their professional experience placement.

and Figure 1.

2.5. Purpose

The purpose of this study is two-fold. First, the study investigates PSTs’ perceptions of preparedness, readiness to engage in the teaching profession, and teaching self-efficacy before and after a final professional experience while completing an initial teacher education program, to understand whether and to what extent shifts are observed for each construct. Second, the study tests the potential associations among constructs proposed in the causal framework (Figure 1) before and after the final professional experience, to potentially elucidate paths of significance and shifts in predictive power pre to post. The causal framework has great potential to be helpful to the field of teacher education because (a) teacher education programs aligning their program designs to teacher education standards can reflect on those designs after examining PST outcomes on the PTPE, and (b) teacher educators can gain an understanding of paths within the causal framework that might explain potential development or regression in their PSTs’ self-efficacy.

3. Methods

3.1. Research Instrument

The PTPE scale (Abraham et al., 2021) measures six constructs (four predictor teacher preparedness variables and the two outcome variables—readiness to engage and teacher self-efficacy) of the proposed causal framework. On each item of all subscales except that for self-efficacy, PSTs rated themselves on a 5-point scale (1 = Not at all to 5 = Extremely well). The options for self-efficacy were on a scale where 1 = Strongly Disagree and 5 = Strongly Agree. The survey also collected demographic information of participants and included an opportunity for open-ended responses.

3.2. Participants

Participants of this study included a sample of PSTs in their final year of a master’s degree in initial teacher education at a large Australian university (235 PSTs completed the survey pre-field, and 195 PSTs completed the survey post-field). At this university, students can pursue a master’s degree in teaching through two pathways: (a) by completing an eligible bachelor’s degree and then enrolling in the School of Education’s master’s degree program or (b) by enrolling in the School of Education’s integrated program known as Pathways to Teaching that combines both degrees. The final professional field experience is a capstone subject in their two-year Master of Teaching program. This subject enables PSTs to proactively engage in real-life educational settings for 30 days to practice teaching and enhance their professional development as a future educator. PSTs represented three educational settings: Early childhood (birth to 5 years old), Primary (grades K-6), and Secondary (grades 7–12). The data were collected from PSTs from direct surveys which took approximately 20 min to complete.

This investigation gathered data from participants at two timepoints: before (Timepoint 1: T1) and after (Timepoint 2: T2) their final professional experience. Data collection was conducted simultaneously across the three educational settings, ensuring consistency in timing. Additionally, parallel wordings were employed for indicators that assessed the same constructs; therefore, it was deemed that the scores are comparable.

PSTs self-identified themselves into two gender categories, namely, female and male. There was no representation from other gender categories, although options were provided in the questionnaire. The gender distribution consisted of 29% male and 71% female during T1 (N = 235) and, 28% and 72%, respectively, at T2 (N = 195). See

Table 2. Pre-Service Teacher Participation at T1 and T2.

Program	T1			T2			Grand Total
	Male	Female	Total	Male	Female	Total
Early Childhood	4	34	38	2	25	27	65
Primary	18	73	91	19	64	83	174
Secondary	46	60	106	34	51	85	191
Total	68	167	235	55	140	195	430

for details.

3.3. Statistical Analysis

Data screening commenced with Missing Values analysis (Enders, 2010). Responses that had systematic missing data were handled through listwise deletion, following the approach suggested by Kline (2011). Only a small number of cases with randomly missing data were identified during this screening procedure, which was estimated by the full information maximum likelihood (FIML) imputation method (Enders & Bandalos, 2001).

The analytic procedures employed in the present study compared the perceptions of PSTs pre and post their final field experience. During these analyses, two main subgroups were compared to investigate whether there were any differences based on gender and program pathways. Descriptive statistics, such as mean and standard deviation, and inferential statistics, namely independent samples t-tests, were conducted using SPSS version-30 for comparing the means of two samples T1 and T2 (Hills, 2011). The statistical significance and the effect size are the two primary outputs of the t-test. Statistical significance indicates whether the difference between sample averages is likely to represent an actual difference between populations, and the effect size indicates whether that difference is large enough to be practically meaningful. A statistical significance level of 0.05 was set for hypothesis testing, as noted by Hills (2011). Following Cohen’s guidelines, effect sizes of 0.2, 0.5, and 0.8 were deemed as “small,” “medium” and “large” effects, respectively (Hills, 2011, p. 85).

The PTPE scale has been established as a theoretically sound, gender invariant and psychometrically valid instrument (Abraham et al., 2021). Nevertheless, the scale was re-validated for this study. The hypothesized factorial structure of PTPE was examined by assessing the model fit of a multi-factor Confirmatory Factor Analysis (CFA) (Byrne, 2009) at the different timepoints. The multiple fit items used for model verification and the accepted benchmarks (Byrne, 1998) are presented in

Table 3. Fit Indices for CFA Model for T1 and T2.

Fit Index	Accepted Benchmark	Timepoint		Type of Fit
		T1	T2	T1	T2
Root Mean Square Error of Approximation (RMSEA)	0.05 indicates a close fit	0.057	0.058	Good	Good
Comparative Fit Index (CFI)	greater than 0.90	0.928	0.937
Tucker–Lewis Index (TLI)	greater than 0.90	0.919	0.929
χ2/df ratio	less than 3	1.749	1.655

. The good fit indices demonstrated the factorial validity of PTPE at T1 and T2.

We also examined the proportion of variance in each factor that was explained by its indicator set. A value of 25% or higher is considered a large effect (Hills, 2011), and all subscales met the 25% threshold (see

Table 4. Average Variance Explained by the CFAs.

Subscales	T1	T2
Relation	26%	41%
Knowledge	26%	49%
Strategies	33%	50%
Environment	41%	59%
Engage	49%	43%
Self-efficacy	25%	26%

). Additionally, examination of the factor correlation values ensured that the condition of discriminant validity was satisfied (r < 0.85; Kline, 1998, p. 60) for T1 and T2 (factor correlations are provided in

Table 5. Factor Correlations for T1.

	Relation	Knowledge	Strategies	Environment	Engage	Self-Efficacy
Relation
Knowledge	0.701
Strategies	0.705	0.839
Environment	0.667	0.633	0.825
Engage	0.476	0.578	0.679	0.671
Self-efficacy	0.518	0.575	0.623	0.684	0.613

and

Table 6. Factor Correlations for T2.

	Relation	Knowledge	Strategies	Environment	Engage	Self-Efficacy
Relation
Knowledge	0.776
Strategies	0.766	0.879
Environment	0.740	0.785	0.864
Engage	0.564	0.693	0.746	0.717
Self-efficacy	0.613	0.654	0.723	0.629	0.643

). Gender invariance was tested by examining the variation in the Comparative Fit Index for the five nested models with increased parameter restrictions as recommended by Cheung and Rensvold (2002). These psychometric properties established the robustness and gender invariance of PTPE for both occasions.

Finally, internal consistency of the subscales measuring PTPE constructs was examined using Cronbach’s Alpha estimate, whereby an Alpha value greater than 0.70 suggests an acceptable benchmark (Hills, 2011). The results for T1 and T2 are presented in

Table 7. Internal Consistency of Subscales Across Timepoints T1 and T2.

Construct Name	Number of Items	T1 (N = 235)	T2 (N = 195)
		Alpha	Alpha
Relation	5	0.82	0.86
Knowledge	6	0.88	0.92
Strategies	8	0.92	0.94
Environment	6	0.90	0.93
Engage	7	0.94	0.92
Self-efficacy	4	0.79	0.83

.

The strength of associations among the constructs of the theorized research model were examined by means of Structural Equation Modeling (SEM) analysis (Kline, 2011) using AMOS 30 software. In this technique, the relative influence of the predictor variables on PSTs’ teaching self-efficacy was demonstrated through the path coefficients of the model. The criteria for assessing model fit using the multiple fit items are explained in Table 3. The path coefficients of the model were tested for significance, and the significance levels were set at 0.01 levels (i.e., T values of both beta and gamma should be greater than 2.58) (Kline, 1998). Furthermore, the variance explained by the model was used to assess the predictive validity of the model. A rule of thumb suggested by Hills (2011), where values around 1%, 9% and over 25% denote a “small”, “medium” and “large” proportion of variance explained, was considered in this study (p. 62).

4. Results

The results are divided into two sections. Section 1 presents the findings of the descriptive analysis, while Section 2 presents the results of the empirical validation of the causal framework.

4.1. Descriptive Analysis

The mean values of the Likert-type subscales demonstrated PSTs’ perception of the construct. For each scale with a range of 1–5 (higher values indicating positive perceptions), the mean value was somewhat above the scale midpoint of 3, indicating the students’ general endorsement of the scale items. With respect to indices of scale variability, the observed standard deviations were quite small (0.62 to 0.79) for the scales. This is likely a result of the higher degree of endorsement for the items on these scales. The mean and standard deviation for each subscale is presented in

Table 8. Descriptive Statistics for Sub Scales (Scale Mean = 3).

Constructs	T1 (N = 235)		T2 (N = 195)
	Mean	SD	Mean	SD
Relation	3.09	0.63	3.73	0.73
Knowledge	3.17	0.64	3.93	0.72
Strategies	3.22	0.65	3.94	0.71
Environment	3.28	0.71	3.89	0.78
Engage	3.40	0.79	4.11	0.76
Self-efficacy	3.65	0.62	4.22	0.62

.

The shift in PSTs’ perceptions across two the timepoints and statistical significance are presented in

Table 9. Significant Differences of Means between T1 and T2 (N = 430).

Constructs	Mean Values		Difference in Means (T2 − T1)	t-Value	df	Sig.	Effect Size
	T1	T2
Relation	3.09	3.73	0.64	9.77	428	0.000	0.68
Knowledge	3.17	3.93	0.77	11.68	428	0.000	0.68
Strategies	3.22	3.94	0.72	11.01	428	0.000	0.68
Environment	3.28	3.89	0.62	8.62	428	0.000	0.74
Engage	3.40	4.11	0.71	9.44	428	0.000	0.77
Self-efficacy	3.65	4.22	0.57	9.45	428	0.000	0.62

. An independent-samples t-test was conducted to compare T1 and T2 mean values. All constructs indicated a statistically significant improvement in mean values after their professional experience and the effect size was “medium.”

The results of differences in the mean values of the subscales for the total sample of two major subgroups (gender and study pathways), are presented in

Table 10. Mean Differences between T1 and T2 across Gender.

Constructs	Male (N = 123)				Female (N = 307)
	Mean Difference (T2 − T1)	SD	Sig	Effect Size	Mean Difference (T2 − T1)	SD	Sig	Effect Size
Relation	0.469	0.122	0.000	0.62	0.708	0.077	0.000	0.69
Knowledge	0.644	0.123	0.000	0.63	0.820	0.078	0.000	0.69
Strategies	0.550	0.123	0.000	0.62	0.790	0.078	0.000	0.69
Environment	0.405	0.134	0.000	0.74	0.706	0.085	0.000	0.73
Engage	0.519	0.141	0.000	0.68	0.789	0.089	0.000	0.81
Self-efficacy	0.365	0.112	0.000	0.57	0.651	0.071	0.000	0.64

and

Table 11. Mean Differences between T1 and T2 across Two Pathways to the Degree.

Constructs	Completed a Pathways Course (N = 282)				Did Not Complete a Pathways Course (N = 148)
	Mean Difference (T2 − T1)	SD	Sig	Effect Size	Mean Difference (T2 − T1)	SD	Sig	Effect Size
Relation	0.704	0.080	0.000	0.68	0.505	0.114	0.000	0.66
Knowledge	0.899	0.081	0.000	0.70	0.503	0.115	0.000	0.63
Strategies	0.834	0.080	0.000	0.68	0.471	0.114	0.000	0.65
Environment	0.694	0.088	0.000	0.75	0.430	0.125	0.000	0.71
Engage	0.841	0.093	0.000	0.79	0.456	0.132	0.000	0.74
Self-efficacy	0.612	0.074	0.000	0.64	0.468	0.106	0.000	0.59

.

Both subgroups displayed an improvement in mean values in relation to the constructs, and the differences were statistically significant. The effects sizes were medium, indicating the proportion of variances explained by the independent variables were meaningful in a practical sense.

4.2. Empirical Validation of the Causal Framework

Estimation of the model showed good fit indices to the hypothesized model on both occasions. For T1, the indices were χ²/df = 1.77, RMSEA = 0.057 (CI, 0.051–0.063), CFI = 0.924, and TLI = 0.914. For T2, the indices were, χ²/df = 1.615, RMSEA = 0.056 (CI, 0.050–0.060), CFI = 0.940, and TLI = 0.931. Thus, the theoretical model was empirically validated at different timepoints in this study.

According to Kline (1998), the path coefficients are considered more crucial than good fit indices in determining model validity. In adherence to Kline’s (1998) cautionary note regarding the removal of non-significant paths from structural equation models based solely on empirical criteria, those paths were retained in the models to avoid potential Type 1 or Type 2 errors. (See Figure 2 and Figure 3 as the validated models at T1 and T2).

Regarding path analysis at T1, significant paths were found between Environment and Engage (0.49) and Engage and Efficacy (0.44). These were found to be significant at the 0.01 level (T values were 3.171, and 7.687, respectively). The model demonstrated Engage can increase by 49 standard deviations given a one standard deviation change in Environment when the other predictor variables are controlled for. Likewise, Engage has a significant influence on Efficacy, increasing it by 44 standard deviations given a change of plus one standard deviation on itself. The remaining variables were nonsignificant in their influences on other variables when all variables are taken together. Significant positive correlations were found between all four predictor variables as expected due to the overlapping nature of the Standards (see Table 5 and Table 6). The model explained the variance of the endogenous variables effectively (45% of the variance of Engage and 56% of the variance of Efficacy were explained). Considering the contextual nature of these variables, this level is considerably high.

Regarding path analysis at T2, significant paths were found between Strategies and Engage (0.53), Environment and Engage (0.25) and Engage and Efficacy (0.76). These were found to be significant at the 0.01 level (T values were 2.628, 2.015 and 8.096, respectively). The model explained Engage can increase by 25 standard deviations given a one standard deviation change in Environment when the other predictor variables are controlled for. One standard deviation in Strategies can make a 53 standard deviation change in Engage. Likewise, Engage was found having a significant influence on Efficacy, increasing it by 76 standard deviations given a change of plus one standard deviation on itself. The remaining variables were nonsignificant in their influences on other variables when all variables are taken together. Significant positive correlations were found between all four predictor variables (see Table 5 and Table 6). Similar to T1 analysis, the validated model T2 also explained the variance of the endogenous variables effectively (28% of the variance of Engage and 42% of the variance of Efficacy were explained).

5. Discussion

5.1. Descriptive Analysis: PSTs’ Development Aligned to the APST

The PSTs’ perceptions across all three constructs were relatively positive at the beginning of the final professional experience, and then further developed in the positive direction at a statistically significant level by the end. These increases occurred for PSTs regardless of their reported gender and program pathway. These findings are very encouraging for the teacher education program in this study, pointing to a high level of support for these important constructs at two different time periods in the final year of the program. In light of the explicit alignment between the APST and the preparedness and readiness to engage constructs within the PTPE, the results are particularly useful from the lens of education policy, transparency, and accountability. The analysis allows the teacher education program to identify specific lines of evidence that the APST are connected to the program’s design, outputs with respect to PSTs’ development, and equity due to similar outcomes across gender and pathway subgroups. The goals reflected in the APST domains have broad appeal and are similar to teacher standards in other countries (CAEP, 2025; Department for Education, 2021). For any teacher education program that aligns their programs to the APST or similar standards, the PTPE and causal framework can be used to assess the extent to which those standards align to important PST outcomes.

Our instrument has similar sub-scales to preparedness subcategories tested in Brown et al.’s (2015) study, including knowledge (pedagogical content knowledge), environment (classroom management), and strategies (planning and preparation for instruction). In Brown et al.’s study and the present study, PSTs’ means for these constructs increased at a statistically significant level from the beginning to the end of the PSTs’ field experience. Whereas Brown et al. (2015) also found increased preparedness for promoting family involvement, we found an increase in PSTs’ preparedness for our instrument’s relation subscale. Cook and van Cleaf (2000) noted that first-year teachers who experienced student teaching placements in urban contexts reported a high mean score for a questionnaire item focused on their understanding of the sociocultural needs of their students. The increase in our multi-item relation subscale adds depth to that finding. Beyond understandings of students, the relation subscale’s focus on the utility and activation of those understandings for planning instruction reveals a deeper level of PSTs’ preparedness that increased during field experience. Due to the close relationship between a teacher’s understanding of students and subsequent decision-making about instructional strategies (Van Driel & Berry, 2012), increases in the PTPE’s relation subscale is a central goal for pre-service teacher learning.

5.2. Path Analysis via SEM: Insights on the Causal Framework

The findings of this study empirically validate the causal framework, with paths that explained a high degree of variance in the endogenous variables (specific preparedness variables explaining variance in readiness to engage, which in turn explained variance in self-efficacy). Other studies exploring similar variables report lower degrees of association, for example, Schepens et al.’s (2009) reporting that preparation combined with faculty and cooperating teacher support accounting for 8% of variance in PSTs’ self-efficacy, and Brown et al.’s (2021) reporting that self-efficacy and preparedness combined to explain 13% of variance in PSTs’ student teaching performance. The high degree of association in this study lends credibility to the causal framework’s employment in future studies.

Brown et al.’s (2015) study reported that student teachers were most efficacious about their classroom management skills at the end of their field experience, and made the greatest gains in their efficacy for implementing instructional strategies. Our study builds on those findings by presenting a path of association with a pivotal mediating variable—readiness to engage in the profession. The causal framework demonstrates that self-efficacy is not increased directly from increases in preparedness. Rather, engaging in the authentic work of a teacher during field experiences helps develop PSTs’ readiness to engage, which is the driver for increasing self-efficacy. Working a step backward, PSTs’ perceptions of preparedness can impact their readiness to engage. The validation of the causal framework supports further examination on the role of coursework and field experiences in supporting beginning teacher development, as discussed below.

At the beginning of the field experience, PSTs’ perceptions of their readiness to engage was explained by their preparedness to facilitate and manage the learning environment. At the end, readiness to engage was explained not only by the environment subscale but also the PSTs’ preparedness to facilitate instructional strategies. This finding can be understood through the lens of teacher concerns (Fuller, 1969). Beginning educators are especially concerned about classroom management very early in their careers, noting that establishing and setting up the learning environment is a foundational step for any learning experience to have a chance for success (Akdağa & Haser, 2016; van Tartwijk et al., 2017). Their perceptions of being prepared to engage in the profession are likely filtered through this primary concern when they, (1) might have few experiences as a teacher in the classroom, and (2) can conjure up images of classroom management from their prior role as K-12 students (Lortie, 1975). As they gain experience with designing and implementing learning experiences during their field experience, their developing preparedness for facilitating instructional strategies begins to explain their readiness to engage as well. The environment and strategy subscales represent the actions most noticeable and pressing to a teacher in the day-to-day activities of a classroom, and align to the professional practice domain of the APST. The results of this study highlight the critical role of preparedness for managing a learning environment and facilitating instructional strategies in developing a PSTs’ readiness to engage in the profession. To support PSTs’ developing self-efficacy, teacher education programs might seek out ways to further spotlight preparation in these two critical areas before and during field experiences to strengthen their association with engagement and self-efficacy.

And yet, the relation and knowledge subscales are also aligned to APST under the domain of professional knowledge. We argue that these subscales are important, as teaching is strengthened when teachers employ their pedagogical content knowledge (She et al., 2024; Ward et al., 2014) and understandings of students (Sadler et al., 2013) when facilitating learning experiences. Further investigation into the potential influence of these subscales on PSTs’ readiness to engage and ultimately self-efficacy is warranted.

Under the umbrella of professional knowledge (AITSL, 2022a), one might conjecture that the knowledge associated with the relation and knowledge sub-scales (content for teaching and understanding students) are most prominent during teacher education coursework, for example, courses that focus on content methods (Peercy et al., 2016), learners’ diverse backgrounds (Gorski, 2009), and learners’ various support needs (Rakap et al., 2017). In a study of PSTs in Singapore, the perceived relevance of teacher education coursework to their practica experience was significantly higher in the second practicum when compared to the first (Choy et al., 2013). By their second practicum, they had completed more coursework in education foundations and curriculum studies. PSTs have been found to rate their teaching skills higher during field experiences when they perceive an intentional linkage between theory and practice (Biermann et al., 2015). Reasoning from these findings, the content of coursework and its pairing with field experiences within a teacher education program deserve more attention. Coursework that intentionally focuses not only on knowledge of students and content, but the activation of that knowledge in authentic learning contexts, might potentially explain the PSTs’ readiness to engage during field experiences in future studies. Programs that intentionally integrate content methods courses and field experiences could investigate the potential associations between the relation and knowledge sub-scales on PSTs’ readiness to engage and whether those associations are paired with greater increases in self-efficacy.

In addition to coursework, the influence of the nature and type of field experiences can be further explored. Reynolds et al. (2016) reported that PSTs’ perceptions of success in meeting a particular professional teaching standard, “Teachers know their students and how students learn” (p. 460), was greater for those who completed more days in a field experience when compared to a traditional placement. They noted that this element was the only one with a difference between the two placement types and differences across standards were not particularly substantial. However, its alignment to our instrument’s relation subscale supports future studies investigating if increases in the relation subscale due to greater time in the field yields a pathway from relation to readiness to engage. In addition, the field experience setting could be further investigated. Studies reveal conflicting results on how PSTs’ self-efficacy may develop (or not) depending on the type of school. For example, Knoblauch and Hoy’s (2008) study found that the field setting (urban, suburban, and rural) did not reveal differences in PSTs’ self-efficacy. However, some studies suggest that PSTs in field settings with diverse populations (Kyles & Olafson, 2008) or a high percentage of learners qualifying for free and reduced lunch (Gomez et al., 2009) experience challenges in developing their self-efficacy or maintaining their interest in teaching. Employing the instrument and causal framework in teacher education programs that focus on different types of field settings will allow for meaningful comparisons of potentially different path analysis results.

5.3. Summary of Contribution

It is typical to study PSTs’ self-efficacy for teaching particular subject areas (Gibson, 2024; Ingram et al., 2024) or for implementing knowledge and skills in one domain of teaching, i.e., classroom management (Golubtchik, 2024). Due to its subject-agnostic approach and alignment to several domains commonly articulated in professional standards for teaching, this study’s causal framework provides another helpful avenue to study beginning teachers’ self-efficacy. Moreover, this study answers the call for increased use of SEM when studying teachers’ self-efficacy (Zee & Koomen, 2016), an approach that is highly valued but can be challenging because it requires a substantial number of participants (Depping et al., 2024). Finally, sense of preparedness (sometimes referred to as teacher readiness) is often analyzed as one overarching construct (Darling-Hammond et al., 2002; Li et al., 2023). The causal framework’s explicit delineation of the preparedness variable into four distinct sub-scales allows for a nuanced understanding of how distinct dimensions of PSTs’ sense of preparedness associates with their readiness to engage in the profession.

6. Conclusions

We investigated PST development in relation to three constructs: PSTs’ perceptions of preparedness, readiness to engage in the teaching profession, and teaching self-efficacy. Descriptive analysis of the PTPE allows teacher education programs to self-assess PST outcomes in relation to the APST and similar teaching standards. We then tested and validated a causal framework that elucidates helpful insights into the associations among critical variables for beginning teacher development. We learned that PSTs’ increased self-efficacy was largely explained by participating in the authentic work of teaching in the field (readiness to engage). In turn, their readiness to engage was explained by their perceptions of being prepared for facilitating the learning environment at the beginning of the professional experience and then also included their perceptions of being prepared to facilitate instructional strategies at the end. This shift demonstrates an important change in how self-efficacy was influenced during a critical juncture in a teacher education program. The results lead to new considerations for future teacher education program designs and associated research, to explore how programs might best facilitate preparedness for the learning environment and instructional strategies before and during field experiences, as well as program designs that could potentially support helpful associations between understandings of students and/or knowledge of content and readiness to engage. The causal framework allows future researchers to continue contributing new knowledge about beginning teachers’ development and how the teacher education community can support their success.