The Relationship Between Pre-Service Benchmarks and Student Teaching Performance

From K-12 schools to higher education institutions, policy makers, parents, teachers, and other education stakeholders are concerned about the state of our nation’s public schools. Issues of reform, accountability, retention, paths to licensure, and preparation of teachers populate education news and research. In the era of accountability, it is important for teacher preparation programs to look within themselves to ensure the structure and requirements of their program help prepare teacher candidates for student teaching and careers. While some insist on raising admission requirements, such as grade point averages (GPAs), others are concerned about the role of high stakes admission and licensure requirements in teacher preparation programs. The focus of this thesis was to examine one Secondary English Language Arts teacher preparation program within a large, Research I institution in the northeast. A correlational study was conducted to determine the relationships between the various preservice benchmarks and candidates’ student teaching performance. Regression models were used to determine if any of the pre-service benchmarks were predictors of other preservice benchmarks or predictors of student teaching performance. Findings from this study reinforce existing literature on correlational relationships between pre-service benchmarks. Findings from the regression models add to the literature in the field. The results and implications of this study offer similar programs potential areas of reform.

iii

ACKNOWLEDGMENTS
The idea, research, and preparation would not have been possible without the tremendous help and support from the University of Rhode Island's School of Education staff and faculty.
To Dr. Diane Kern, who has provided countless advising meetings both in her office and at her kitchen table for the last five years, thank you for your endless belief and support. You have inspired and encouraged me to pursue opportunities beyond my wildest expectations. Though this milestone is important, I think we can both agree it is not nearly as impressive as my technology-enhanced, social justice based, station activity during my first middle school observation.
To Dr. Julie Coiro, thank you for your thoughtful feedback during the proposal phase of this research. You challenged me to dive deeper into the research mindset and experience and laughed with me as I reminded you I was not nearly as "intense" as the Ph.D students with whom you normally work.
To Dr. Jay Fogleman, thank you for putting up with my mouth and trouble making tendencies for the past six years. I apologize for the gray hairs I may have caused. I appreciate the countless time we have spent dreaming and planning for the schools and students of the future. Thank you for convincing Mrs. Fogleman to bring Emmy in so many times just so I could hug her fluffy head.
To Dr. Stephen M. Barber, whose influence and impact on every aspect of my life cannot be quantified or put into words. Thank you for sharing your heart and mind. I will continue to strive for agency in life. Above all, you have taught me to choose love.
iv To the Office of Teacher Education, specifically Janet Deignan, Daniel Riley, and Dr. Diana Marshall for their help in inspiring the topic of this thesis and for their assistance in the data collection process. You are the nicest people I know. Janet, thank you for always telling me everything would be okay. Dr. Marshall, thank you for your patience and life talks. Dan, thank you for your friendship and endless bravery. I love you all.  policymakers about the quality of schools and the teachers within them (Ingersoll & Collins, 2017;Klein, 2017). Policy makers are quick to conclude teacher preparation programs need to be reinvented to address these concerns (Hayes, 2002). The teacher retention problem contributes to these concerns. Riggs (2013) notes 9.5% of teachers will leave the field before the end of their first year. Broadening the scope, Ingersoll (2012) adds 40% and 50% of teachers leave the classroom within the first five years of their career. The large percentage of teachers leaving the field directly relates to the concerns about the quality of schools and teachers in them. In a school with a high turnover rate, administrators are stuck in a revolving cycle of continuously searching for new teachers (Zhang & Zeller, 2016).
A report from the National Council for Accreditation of Teacher Education (NCATE) notes the solution to address these concerns is to "reform teaching so that more who enter will stay in the school building rather than be propelled back out the revolving door" (Wise, 2005, p. 2). These reports indicate an interest in reforming education to allow more invested teacher candidates into the field. Ironically, the solution many departments of education at the state and national level call for is to raise the standards candidates must meet in order to be admitted into a teacher preparation program or receive licensure, thus eliminating a wide range of potential teacher candidates. This phenomenon is part of a growing focus on accountability and reform efforts to improve the processes for admitting teacher candidates into teacher preparation programs (Jamil, Sabol, Hamre, & Pianta, 2015).
Researchers and college faculty are rightfully concerned about the role of highstakes admission and licensure in undergraduate education (Moser, 2014;Petchauer, 2012;Thomas & Loadman, 2001;Warren & Curley, 1998;Watras, 2006). As a result of these concerns, educational institutions are interested in the level of preparedness, effectiveness, and quality of teacher candidates (Williams & Alawiye, 2001). From the initial requirements for admittance into a teacher preparation program to the final licensure testing requirements and successful completion of student teaching, some education policymakers insist higher test scores for teacher candidates will produce high quality teachers. The higher cut scores on tests impact education majors immediately since the initial requirements for admittance into a teacher preparation program are often the first to rise.
In this era of accountability, it is essential for policy makers, university faculty, and other stakeholders to reexamine teacher education programs. Those concerned with educational accountability and reform must understand what aspects of teacher education programs lead to the development of successful student teachers who, after completing their program, enter the teaching workforce highly qualified (Leathwood & Phillips, 2000). Therefore, it is important for teacher preparation programs to look within themselves to better understand the needs of their teacher candidates (Kornfeld, Marker, Rudel, 2003). By doing so, teacher preparation programs can understand the relationship between aspects of their own program and the development of highly qualified teachers.
Current concerns about teacher education and the researcher's experiences as an undergraduate teacher candidate inspired the concept of this study. Throughout this thesis, independent variables are referred to as "pre-service benchmarks." This title was chosen because it accurately captures both the timeline and nature of the assessments I used in this study.
This study explores the relationship between pre-service teacher benchmarks and student teaching performance. The purpose of this study is twofold. First, the purpose is to determine if there is a relationship between the various requirements undergraduate education majors must meet. Second, this study aims to determine if the various requirements are predictive of effective student teaching performance. The results of this study will be shared with the School of Education faculty and administrators so they may be better able to determine which variables indicate greater student teaching success, which will inform teacher candidate advising and program revision.

Limitations
There are several limitations to this study the researcher recognizes. First, the convenience sampling from only one institution in the state and country is not necessarily generalizable. The researcher will attempt to address this concern by comparing data points to national averages, when possible. Another limitation of the study is the exclusion of other content areas (e.g. History/Social Studies, Mathematics, Sciences). The students of other disciplines have different cut scores for their Praxis content tests. Thus, the findings of this study may not be applicable to students of other content areas. Addressing such a concern is beyond the scope of this study, but future research should investigate pre-service benchmarks and successful student teaching for other content areas. Though a few issues around reform are mentioned in the introduction, this research only looks at the preparation of teacher candidates.
Further research should include: qualitative information addressing how educational stakeholders would reform programs, a longitudinal study about the effectiveness of a program's student teachers throughout their career, the number of teacher program graduates who remain in the field after three, five, and seven years, and the relationship of pre-service benchmarks and teacher performance in non-traditional or alternative teacher licensure program.

Significance
The main audience for this research will be faculty teaching in English teacher education programs, although this study may also be of interest to additional groups, such as the university, the School of Education, faculty and staff within the program, teacher education candidates, curriculum reformers, and even policy makers. Due to the clinical nature of the teacher education program in this study, the results may impact public schools in the state, where teacher candidates complete pre-professional field experiences and student teaching. Should the results of this study find relationships between pre-service benchmarks and student teaching performance, program faculty might revise the nature of the student teaching experience and relationship with local schools. If a relationship is found, teacher education programs might place more emphasis on field experiences, which would affect local public schools and their students. Further research would need to be done to determine the impact teacher candidates have on student achievement in public school classrooms. If relationships are found, program faculty and policy makers might revise the structure of their program in order to better prepare teacher candidates to be successful student teachers or rethink the purpose of pre-service benchmarks which may be obstructing students from student teaching experiences.
A correlational study by Wilson and Robinson (2012)  Such findings would add to the existing field of research regarding teacher candidate programs and their components. Further, findings might inspire teacher candidate programs to reflect on the value of some pre-service benchmarks in preparing teacher candidates for successful student teaching.
If there is a relationship between one or more of the variables considered to be a pre-service benchmarks and successful student teaching, the results may be important for professors within the program to understand about how to better serve their teacher candidates.

REVIEW OF LITERATURE
The expectations teacher candidates must meet during their programs have a strong research history. Many aspects of student teaching have been researched as well.
Literature relevant to the pre-service benchmarks under analysis and student teaching performance mentioned below informed and inspired this thesis.

Pre-Service Benchmarks
Efforts to increase teacher quality and preparedness have focused on increasing admission and graduation standards for students in teacher preparation programs (Hall & West, 2011). The pressure to raise standards for teacher candidates impacts education majors from the start of their undergraduate study. The position of these exams establishes the education major as "fundamentally different from others because students must test into it" (Petchauer, 2012, p. 252). Though the use of admission tests is unique to education majors, the use of high-stakes standardized testing to determine admittance and even graduation is not. It is common for many majors and programs (e.g., medical school, pharmacy, nursing) to use grade point averages as a way to ensure the quality of their students. Some policymakers and educational program administrators believe a passing score on a content area test, such as English or mathematics, is enough to be a qualified and effective teacher (Goldhaber, 2007).
A review of 123 studies by D'Agostino and Powers (2009) indicates standardized test scores did not strongly relate to teaching performance. Two prior studies (Ferguson & Womack, 1993;Guyton & Farokhi, 1987) note education coursework to be a better predictor of teaching success than other benchmarks such as content area coursework and overall GPAs. In an era of teaching influenced by continued efforts to raise standards and high-stakes assessments, it is appropriate to question the evidence supporting these changes and review the literature surrounding this issue (Wilson & Robinson, 2012).
A report from Ferguson and Womack (1993) offers an overview of the pressure teacher preparation programs face. In the 1980s, the debate between subject matter and education coursework was strengthened when academic and political interest groups united to "secure the adaptation of accreditation standards and legislative mandates prescribing increased content preparation at the expense of education coursework" (p. 55). Reform documents such as A Nation at Risk (1983) echoed these concerns. While there is little evidence to defend the idea of placing more importance on subject matter preparation to increase teacher performance, research and reports show evidence that education coursework has a positive effect on teaching performance (Ashton & Crocker, 1987;Darling-Hammond, 1991;Everston, Hawley, & Zlotnik, 1985;Ferguson & Womack, 1993, p. 55). Ferguson and Womack (1993) found education coursework, of the variables they measured, to be the strongest predictor of teaching performance. Conversely, they found a teacher candidate's subject area grade point average was not a significant predictor of teaching performance (p. 60). Ferguson and Womack (1993) call for an assessment of existing evidence on the effect of education and subject matter coursework on "teaching performance and student learning and further research on the subject" (p. 55). This study answers that call and adds to the field of literature on this subject.
A summary report from Wilson, Floden, and Ferrini-Mundy (2001) notes prior research has explored the relative importance of pre-service benchmarks in regard to teacher preparation, and further research needs to be done to explore the "relationship between components of pedagogical preparation and teacher effectiveness" (p. 17).
Since then, researchers have begun to explore the relationship between different assessment measures teacher candidates must pass throughout their program. Whether admission scores, education or content area grade point averages (GPAs), Praxis scores, or student teaching outcomes, studies have found mixed results in relationships between variables. Casey and Childs (2011) note few international studies have examined the relationship of admission criteria to teacher candidate preparedness to teach at the end of their program. Their report calls for further research to determine if entering GPAs have predictive value for successful student teaching performance.
This thesis may help fill the gap in the research. Hall and West (2011) analyzed relationships between variables such as GPA, American College Testing (ACT) scores, and Praxis exam scores. Their analysis found GPA and Praxis scores correlated significantly and positively with student teaching performance scores (Hall & West, 2011). A multiple regression model consisting of Praxis scores and GPA variables explained sixteen percent of the variance in participants' student teaching performance scores. Hall and West (2011) acknowledge these results can support the movement to raise standards in teacher education programs, they note the current model leaves eighty-five percent of variance in student teaching performance unexplained. Hall and West (2011) argue raising standards might not lead to more prepared teacher candidates. While these efforts have occurred in an attempt to better prepare teachers, there is little empirical evidence to support these efforts (Zumwalt & Craig, 2005).
Previous studies have found relationships between program requirements and student teaching performance, but with contradicting results. Guyton and Farokhi (1987) found GPA at the sophomore (typically the second year of college for a traditional undergraduate student) and upper level (usually checked prior to student teaching) were significantly correlated with teaching success. The researchers also found sophomore and upper level GPAs correlated significantly with teacher certification test scores and subject matter knowledge tests. However, their study showed subject matter test (like the Praxis II content test) scores were not correlated with teacher performance. Ferguson and Womack (1993) found education coursework accounted for 48% of the variance in teaching performance. Additionally, research found academic criteria including GPA and ACT scores failed to predict future student teacher performance (Byrnes, Kiger, & Shechtman, 2003).

Furthermore, the Praxis series of tests, developed by Educational Testing
Service (ETS), is one of the most widely used certification tests in the country. The Praxis I Core tests, which assess basic skills in reading, writing, and mathematics, are often used by teacher preparation programs as an admission requirement. After a few years in the program, teacher candidates must then pass the more advanced content knowledge and pedagogical knowledge tests. In programs like the one used in this study, teacher candidates must meet certain cut scores in order to advance in their program. These cut scores are determined by policymakers in each state. Often, they raise these cut scores in an attempt to strengthen the quality of their teachers. Schuls and Trivitt (2015) argue these policies operate under the assumption that a teacher candidate who "fails the exam by one question is not fit to teach, while the individual who earns a score equal to the cut score is deserving of a teaching certificate" (p. 653).
However, evidence from Goldhaber (2007)  Cooperating teachers, teacher candidates, and university supervisors all play a role in the development of effective teachers. There is ample research exploring the influence of teacher education programs on the development of their teacher candidates. Adams and Krockover (1997) found beginning teachers attribute their knowledge of student-centered instruction, general pedagogical knowledge, and pedagogical content knowledge to their teacher education program. Furthermore, courses in teacher education provided candidates with a framework with which to organize, understand, and reflect on their experiences in classrooms. Such reflection contributes to the development of successful teachers while in their program and after. Grossman and Richert (1988) found prospective teachers cite education coursework and fieldwork as influential elements of their teacher preparation program, noting fieldwork as an aide in the development of their teaching practices.
The development of effective teachers is a central goal of a teacher preparation program. These above reports provide research regarding the qualities and dispositions of effective teachers.
Existing studies regarding student teacher perceptions of teacher training programs and student teaching experiences provide insight into teacher candidates' opinions and beliefs about their development over the course of the program (Hayes, 2002;Morin, 1996;Pettus and Smith, 1991;Thomas & Loadman, 2001;William & Alewife, 2001). However, little research was found that analyzed existing data from cooperating teachers in the context of program and teacher candidate evaluation and with the goal of determining the qualities of effective teacher candidates. By understanding how teacher candidates are evaluated in schools by their cooperating teachers and university supervisors, teacher preparation programs can develop a sense of how their student teachers perform. These evaluations can offer critical insight from cooperating teachers or university supervisors to not only help a teacher candidate develop, but can be used to review the preparedness of teacher candidates in general, or in specific key areas of development such as classroom management. While such research would likely produce fascinating results, it is beyond the scope of this study.
Further research should determine the qualities and characteristics current teachers believe effective students teachers possess.
One goal of this study is to determine which pre-service benchmarks predict student teaching performance, an area of exploration that will contribute and expand upon existing research.
The following research questions guide this study: 1. Is there a relationship between the following pre-service benchmarks: 2. To what extent do the above pre-service benchmarks predict student teaching performance?
Data analysis expands upon the existing research in this field and may provide implications for teacher preparation programs.

METHODOLOGY
Existing studies provided the inspiration and foundation for this methodology.
Wilson and colleagues (2001) argue research reports should explicitly explore the relationship of "teacher knowledge, skill, and practice that are thought important for effective teaching" (p. 33). Existing research on pre-service benchmarks provided the foundation for the research design of this study (Wilson & Robinson, 2012;Hall & West, 2011;Sandholtz et al, 2015). In many ways, this research was inspired by a report by Ferguson and Womack (1993) which sought to determine the extent to which "education and subject matter coursework predict the teaching performance of student teachers" (p. 59). The study by Ferguson and Womack (1993) inspired the research questions regarding the relationships between pre-service benchmarks and student teaching and the predictability of those benchmarks on student teaching performance. If there is a relationship between one or more of the variables considered to be a pre-service benchmarks and successful student teaching, the results may be important for professors within the program to understand about how to better serve their teacher candidates.

Setting
This study was conducted on a dataset from the School of Education at a large, public, Research I institution in the Northeastern region of the United States.

Study Population
The study population was 2013-2016 program completers from the Secondary Only data of students who successfully completed the program were used for this study since the independent variable under consideration was student teaching performance, which is required for program completion and for teaching licensure.
Since the researcher had no contact with participants or the current cohort of student teachers, participants had minimal to no risk. To protect the anonymity and confidentiality of the participants, the researcher's major professor de-identified the data and generated pseudonyms before data were made available to the researcher.
The total number of participants was fifty-three (n = 53).
In this area, the researcher had two concerns related to the confirmability of this study. It is important to note as a graduate of this program the researcher does have a relationship with some of the participants in the study, especially those in the 2015 cohort of which the researcher was a member. Another concern was the researcher's own information was included in the data collected. To address these concerns, a School of Education faculty member generated the queries and the researcher's major professor de-identified the data. These actions greatly reduced any potential risk to participants since their anonymity was protected.

IRB approval:
The proposal for this study was submitted to the University of Rhode Island Institutional Review Board and was approved. Since the study is limited to analysis of de-identified existing data, it was not necessary to complete the full IRB application. Rather, a Secondary Data Analysis Worksheet was submitted to the IRB and approved.
2. Data collection: A request to access the data was approved by the Director of the School of Education. Next, the researcher's major professor accessed data available to her and also requested a query from the School of Education Outcomes Assessment Office (data located in TaskStream) and the Office of Teacher Education (data located in Filemaker). The researcher's major professor de-identified the dataset and shared data in an Excel file format.

Dataset description:
The quantitative data included eight variables. Six independent variables were collected regarding pre-service benchmarks, and three variables were collected and combined to create one sum score regarding student teaching, the dependent variable. The following six pre-service benchmarks functioned as independent variables: 1) grade point averages for

Data Analysis
To analyze the quantitative data, the researcher used Statistical Package for Social Science (SPSS) software version 24 to run correlations and regression analysis.
Correlational data analysis was conducted to determine relationships between the different independent variables (six pre-service benchmarks). Pearson r correlation coefficients were calculated to determine the association among pre-service benchmarks and student teaching performance at the specific school level (i.e. high school and middle school). Correlational data analysis was conducted to determine relationships between the pre-service benchmarks and student teaching performance (as the sum score and as individual variables).
The researcher ran a series of regression models to determine the predictability of pre-service benchmarks and student teaching performance. The data were analyzed through Pearson r correlation coefficients, descriptive and frequency statistics. Data were also analyzed through Enter method, forward, backward, and step-wise regression models. Data analysis occurred through LSD, Tukey HSD, Bonferroni, Tamhane, Dunnett T3, and Games-Howell procedures for multiple comparisons to test for significant differences between group means.
This analysis helped researcher understand the extent to which each independent variable predicts student teaching performance. Correlational analysis aided in the researcher's understanding of the relationships between pre-service benchmarks and student teaching performance. The results of this data analysis are discussed in the next chapter. Findings may be significant for teacher preparation programs.

FINDINGS
Various combinations of variables were analyzed using SPSS to answer the research questions. The first approach was to separate student teaching variables into three scores--high school cooperating teacher final evaluations, middle school cooperating teacher final evaluations--and university supervisor final evaluations.
Though this method did not acknowledge student teaching performance as one variable, it offered insight into the more specific relationships between pre-service benchmarks and student teaching performance at the individual level.
The researcher must address variables used in this study are not normally distributed (see Appendix O) and have a high level of skewness. Highly skewed variables might make regression models inappropriate to interpret in any meaningful way. Given skewness, the findings are hypothetical patterns that might be indicative of different kinds of relationships.

Is There a Relationship Between Pre-Service Benchmarks and Student Teaching
Performance?
The data were input into SPSS to run a correlation model (Appendix I, Table   20) to determine if there were any relationships between pre-service benchmarks and student teaching performance, considered as three separate variables (high school cooperating teacher's final evaluation scores, middle school cooperating teacher's final evaluation scores, and university supervisor's final evaluation scores).
Correlational analysis showed multiple significant relationships among the pre-service benchmarks. Descriptive statistics were run to provide further insight into the performance of this program's teacher candidates (see Appendix H, Table 19).
Frequency statistics were run to determine the number of participants per cohort (see Appendix A, Table 1).

Pre-Service Benchmarks
Education GPA at admission and English GPA at admission correlated (r = .486, p = .000). Education GPA at admission and Education GPA before student teaching correlated (r = .520, p = .000). Education GPA at admission and English GPA before student teaching correlated (r = .370, p = .006). Lastly, Education GPA at admission correlated (r = .317, p = .021) with teacher candidates' highest English Praxis content score. Education GPA did not significantly correlate with any student teaching performance variables.
English GPA at admission and Education GPA prior to student teaching correlated (r = .520, p = .000). English GPA at admission significantly correlated with English GPA prior to student teaching (r = .778, p = .000). English GPA at admission correlated with teacher candidates' highest PLT score (r = .539, p = .000) and with teacher candidates highest English Praxis content score (r = .583, p = .000).
Education GPA prior to student teaching correlated with both Education and English GPA at admission (r = .520, p = .000). Education GPA prior to student teaching significantly correlated with English GPA prior to student teaching (r = .683, p = .000). Education GPA prior to student teaching correlated with teacher candidates' highest PLT score (r = .362, p = .008). Lastly, Education GPA prior to student teaching correlated with teacher candidates' highest English Praxis content score (r = .362, p = .008).
Teacher candidates' highest Principles of Learning and Teaching (PLT) Praxis exam score correlated with Education GPA at admission (r = .539, p = .000). Teacher candidates' highest PLT exam score correlated with Education GPA before student teaching (r = .362, p = .008). Teacher candidates' highest PLT exam score correlated with English GPA prior to student teaching (r = .584, p = .000). Teacher candidates' highest PLT exam score significantly correlated with highest English Praxis content score (r = .625, p = .000).
Though Education GPA at admission and prior to student teaching had a significant relationship, Education GPA at admission did not have a strong relationship with PLT scores. However, Education GPA before student teaching did have a significant relationship with PLT scores (see Appendix J, However, significant relationships were found among the student teaching performance variables.

Correlational Analysis with Student Teaching Performance as a Sum Score
To determine if there were any relationships between pre-service benchmarks and student teaching performance as a sum score, a correlational analysis was run.
Since the six pre-service benchmarks remained the same in this analysis, the significant correlations between them are reflected above. However, correlational analysis showed there were no significant relationships between pre-service benchmarks and student teaching performance as a sum score (see Appendix M, Table   27).

To What Extent do the Pre-Service Benchmarks Predict Student Teaching
Performance?
A series of regression models were run to determine the predictability of preservice benchmarks and student teaching performance. The Enter Method was used to input all independent variables and then one dependent variable. The Enter Method was used multiple times to create a few different reports. First, the six pre-service benchmarks as the independent variables and high school cooperating teacher final evaluation as the dependent variable (see Appendix D, Table 4). Results showed preservice benchmarks accounted for two percent of the variance in student teaching performance at the high school level. Second, the six pre-service benchmarks were entered into the model as the independent variables and middle school cooperating teacher final evaluation as the dependent variable, when applicable since not ever teacher candidate completes the middle level student teaching experience (see Appendix E, Table 4). Results from this analysis found pre-service benchmarks accounted for fifteen percent of the variance in student teaching performance at the middle school level. Third, the six pre-service benchmarks were entered into the model as the independent variables and university supervisor final evaluation as the dependent variable (see Appendix F). Results from this analysis accounted for nine percent of the variance in student teaching performance as evaluated by university supervisors (see Appendix F, Table 5).
Enter Method was used with the six pre-service benchmarks as the independent variables and the sum score of the evaluations as the dependent variable (see Appendix G). The Enter Method models showed the pre-service benchmarks used in this study were not statistically significant predictors of student teaching performance.
The results of this method accounted for eight percent of variance (see Appendix G, Table 16).
For this particular test, forward, stepwise, and backward regression models were also attempted. SPSS would not produce a model for forward regression because the results were not statistically significant. Similarly, SPSS would not produce a model using the stepwise method. The researcher also ran a backward regression, which did not produce a model to determine the predictability of pre-service benchmarks and student teaching performance.
Interestingly, when SPSS produced the backward regression model and removed English GPA before student teaching, the variance did not change at all.
Furthermore, when SPSS removed variables there was no change in the r2 value.
Stepwise selection method was attempted but SPSS did not produce a model.
Enter method was used to determine if Education GPA at admission and prior to student teaching were predictive of Principles of Learning and Teaching (PLT) Praxis exam scores. Results showed Education grade point averages accounted for thirteen percent of variance in PLT scores (see Appendix N, Table 29). The Education GPA at admission was not a significant predictor (p = .968). However, Education GPA prior to student teaching was a significant predictor of PLT exam scores (p = .024) (see Appendix N, Table 31).
Enter method was used to determine if English GPA at admission and prior to student teaching were predictive of English content Praxis exam scores. Results showed English GPA accounted for forty percent of the variance in English content Praxis exam scores (see Appendix N, Table 33). The English GPA at admission was not a significant predictor (p = .137) but the English GPA prior to student teaching was (p = .021)(see Appendix N, Table 35).

CONCLUSION
In the educational era of accountability, it is now more important than ever for teacher preparation programs to look within themselves to determine if the requirements students must meet are necessary to prepare for successful student teaching experiences and beyond. The expectations teacher candidates must meet in order to continue through their program should prepare them for successful student teaching. Efforts to increase teacher quality and preparedness have focused on increasing admission and graduation requirements for teacher candidates (Hall & West, 2011, p.145). While outside pressure for raising standards continues, it is appropriate to turn the lens inward to teacher preparation programs. However, it is important to keep existing research in mind while reviewing programs. A study by Ferguson and Womack (1993) indicates teacher preparation program improvements "will not be achieved by raising requirements beyond the existing floor of quality point average (2.5 out of 4.0)" (p. 61). However, the Rhode Island Department of Education has raised GPA standards for undergraduate teacher preparation programs to 2.75 out of 4.0 and for graduate students a 3.0 out of 4.0 (RIDE, 2013).
This thesis was prepared to determine if there were relationships between preservice benchmarks and student teaching performance and if pre-service benchmarks were predictive of student teaching performance.
Existing research shows mixed results in answering these research questions. A review of 123 studies by D'Agostino and Powers (2009) indicates standardized test scores did not strongly relate to teaching performance. Correlational analysis in this study found similar findings, as neither the PLT Praxis exam nor the English Language Arts Praxis exam had significant relationships with student teaching performance. Ferguson and Womack (1993) and Guyton and Farokhi (1987) found education coursework to be a better predictor of teaching success than other benchmarks such as content area coursework and overall GPAs. Other research echoes the conclusion that education coursework has a positive effect on teaching performance (Ashton & Crocker, 1987;Darling-Hammond, 1991;Evertson, Hawley, & Zlotnik, 1985;Ferguson & Womack, 1993, p. 55). Ferguson and Womack (1993) found education coursework, of the variables they measured, to be the strongest predictors of teaching performance. Whereas the subject area GPA was not a significant predictor of teaching performance (p.60). Similar variables were used in this study and did not produce significant results to add to these findings regarding pre-service benchmarks and student teaching performance.
Hall and West (2011)  variables explained sixteen percent of the variance in participants' student teaching performance scores, this study found pre-service benchmarks explained eight percent of student teaching performance scores.
Though teacher licensure exam scores are among the most cited as having a positive relationship with teacher effectiveness, analysis in this study determined there were no significant relationships between licensure exams and student teaching performance. However, regression analysis did find Education GPAs prior to student teaching were significant predictors of PLT scores and accounted for thirteen percent of the variance in PLT scores (p = .024). Regression analysis also found English GPAs prior to student teaching were significant predictors of English Language Arts content exam scores and accounted for forty percent of variance on the Praxis exam (p = .021). These findings are logical since Education courses and English courses are designed to prepare students for their licensure exams.
Interestingly, admission GPAs in both Education and English were not significant predictors of later licensure exam scores. Therefore, the role of admission GPAs as gatekeepers preventing students from entering the major should be reconsidered by teacher preparation programs. These findings are consistent with research from Henry et al. (2013) who noted "new and better indicators of candidates' strengths on entry and performance during the program will be needed to guide reform and continuous improvement of teacher preparation programs" (p. 440). Further, these test and GPA requirements may be contributing to the lack of diverse teaching candidates since Lee (2002) has conducted research on racial, ethnic, and economic gaps in test performance and has shown that with the exception of Asian students, students of color and students in low socioeconomic households score worse on these tests than their White and economically advantaged peers.
Interestingly, this study found there was a stronger relationship between English GPA at admission and PLT scores (r = .539, p = .000) than Education GPA at admission and PLT scores (r = .193, p = .166). The strongest correlational relationship this study found was between English GPA the time of admission and English GPA before student teaching (r = .778, p =.000). These findings contradict the researcher's hypothesis that there would be a stronger relationship between Education grade point averages and the Principles of Learning and Teaching Exam since the PLT exam assess students' knowledge of educational theories, practices, etc.
The results of this study indicate the pre-service benchmarks collected at strategic points throughout the program are not significant predictors of student teaching performance. While significant relationships were found between the various preservice benchmarks, there were no significant relationships between any pre-service benchmark and student teaching performance. While significant correlational relationships were found between student teaching performance as evaluated by the high school cooperating teacher, middle school cooperating teacher, and university supervisor, these student teaching performance variables had no significant relationship with any of the pre-service benchmarks. Henry et al. (2013) note current indicators of progress and performance do not predict later effectiveness. Thus, "new and better indicators of candidates' strengths on entry and performance during the program will be needed to guide reform and continuous improvement" of teacher preparation programs (p. 440).
The development of successful teachers is a central goal of a teacher preparation program. Looking within at student teaching performance is one way to help determine if programs are developing successful teachers. Grossman and Richert (1988) found prospective teachers cite education coursework and fieldwork as influential elements of their teacher preparation program. By understanding how teacher candidates are evaluated in schools by their cooperating teachers and university supervisors, teacher preparation programs can develop a sense of how their student teachers perform. These evaluations can offer critical insight from cooperating teachers or university supervisors can be used to review the preparedness of teacher candidates in general, or in specific key areas of development such as classroom management.
Research from Shulman (1986Shulman ( , 1987  Further research may provide valuable information for teacher preparation programs in this area.
Measures of professional dispositions, or the collections of behaviors, attitudes, and teaching qualities have been seen as critical components of teacher preparation programs (Flowers, 2006

Limitations
The researcher recognizes there are several limitations to this study. First, the convenience sampling from only one institution in the state and country is not necessarily generalizable. The researcher attempted to address this concern by comparing data points to national averages, when possible. Another limitation of the study was the exclusion of other content areas (e.g. History/Social Studies, Mathematics, Sciences). The students of other disciplines have different cut scores for their Praxis content tests. Thus, the findings of this study may not be applicable to students of other content areas. Addressing such a concern was beyond the scope of this study, but future research should investigate pre-service benchmarks and successful student teaching for other content areas.
Highly skewed variables might make regression models inappropriate to interpret in any meaningful way. Readers should take caution in using the results of this study to guide reform. The student teaching evaluation rubrics may not be valid assessments since evaluators commonly rate students between three and five since giving a student a one or two pulls them from student teaching.
Further research should include: qualitative information addressing how educational stakeholders would reform programs, a longitudinal study about the effectiveness of a program's student teachers throughout their career, the number of teacher program graduates who remain in the field after three, five, and seven years, and the relationship of pre-service benchmarks and teacher performance in nontraditional or alternative teacher licensure program. Also, further research should explore the use of dispositions assessments in teacher preparation programs. Further research should also explore the use of content pedagogical knowledge assessments and technological content pedagogical knowledge in teacher preparation programs.

APPENDICES
Appendix A