How well teacher candidates perform on the Praxis exam does not predict how well they will perform as teachers. 

Over the years, we have worked with countless students who have struggled to meet the minimum cut scores on the standardized tests aspiring teachers are required to pass. These students rarely struggle in their methods classes, clinical experience, or their student-teaching placement, but passing the test (in this case, the Praxis Core Academic Skills for Educators) is another matter. Here are just a few examples of students (with pseudonyms) who’ve had to spend significant amounts of time and money trying to clear this hurdle: 

Jay is a White man from a distant state who selected our university to earn his history education degree and compete as an athlete. He has taken the test five times and worked with Lisa, a senior faculty member, weekly for about seven months. Lisa encouraged him to get evaluated for learning disabilities, a process that took several months but resulted in a complex diagnosis of giftedness and disabilities that affect his writing. To receive test-taking accommodations, he will need to apply and wait for two to four months to be approved to take the test with extended time in a city two hours away. He has delayed his student teaching for a semester and is working in construction.  

Desiree is a White woman who came from a neighboring rural community to earn her elementary education degree. She worked extensively with Lisa on test-taking strategies for the reading test and learned that a yellow transparency helps her reading fluency. She has spent more than $1,000 taking the test over three years, with small improvements each time, and just recently passed her reading test on the sixth attempt. She recently volunteered that she does not want others to stop pursuing their dreams because of standardized tests.  

Victoria is a bilingual Latina majoring in elementary education. She started college in 2011, paused her studies for six years, and resumed them in 2017. She passed the reading and writing portions of the test, but her math score was extremely low. She has worked for five months with a tutor and requested accommodations from the Educational Testing Service (ETS). She had planned to take the test in late August but rescheduled it when the math test changed formats. It took six weeks for her new accommodations to be approved, and she recently passed the math test with extra time at the site two hours away. She has wanted to be a teacher forever and said the test “tore me up for months.” 

All of these students have strong grades and a good knowledge base. They exhibit the dispositions that our university department has indicated are important for teachers: responsibility, respect, collaboration and communication, a commitment to the complexity of content, a commitment to student learning, and reflection. Their scores on their initial clinical experiences are strong. But their quest to become teachers has been undermined by a test.  

Why test teachers? 

Critics have long decried the lack of perceived intellectual ability in teachers (Conant, 1963). This criticism became especially strong in the 1980s with A Nation at Risk (National Commission on Excellence in Education, 1983) and the perception that “current admissions policies allow just about everyone who applies to enter teacher preparatory programs” (Watts, 1980). In response, groups like the Carnegie Task Force on Teaching, the Holmes Group, and the National Commission on Teaching and America’s Future came together to raise standards and improve teacher education programs (Zumwalt & Craig, 2005). In 2001, the No Child Left Behind Act added teeth to the movement by requiring teachers to be highly qualified and certified in their subject area.  

All of these efforts raise the question: What makes for an effective, highly qualified teacher? For many critics of public education, the answer entails raising the academic profile of teaching by requiring standardized tests and higher college grade point averages (GPAs). As a result, these requirements have become part of the expectations that schools of teacher education have for aspiring teachers.  

But do these data points help us evaluate aspiring teachers? What if there is no correlation between GPAs and standardized test scores and candidates’ actual performance when student teaching? To find out, we examined the records of 218 graduates over five years.  

Admission criteria vs. student-teaching performance 

We sought to determine whether any of the academic criteria for admission to our university’s teacher education program predicted the scores teacher candidates would receive in the final evaluation of their performance as student teachers. We ran the study twice, once with 79 students from fall 2015 to spring 2017 and again with 139 students who graduated from fall 2017 to spring 2019. 

Admission into Minot State University’s teacher education program requires a minimum cumulative GPA of at least 2.75 (on a 4-point scale); a minimum 2.5 GPA on the communication portion of general education requirements (which includes courses in composition and public speaking) with no grade lower than a C; a minimum 2.5 GPA in the teaching major, and a passing Praxis score in reading, writing, and math. In addition, students must pass a background check and complete a clinical experience, which initially required 20 hours of classroom observation and then was changed in fall 2018 to require 10 hours of observation. These criteria are common in teacher education programs across the United States and Canada (Casey & Childs, 2007; Miller-Levy, Taylor, & Hawke, 2014). 

At Minot State, students who meet these criteria are admitted into the teacher education program and put on the track toward a student-teaching placement and eventual teaching licensure. Ideally, students are admitted in the fall of their second year after one or two clinical experiences, but students who struggle to pass the tests can be admitted as late as a few weeks before student teaching begins if that is when they pass the tests. Student-teaching placements are typically 6, 10, 12, or 16 weeks depending on whether students have one or two majors or a concentration (like kindergarten) that requires student teaching.  

We compared student performance on these admission criteria to their performance on the Student Teaching Observation Tool (STOT), which was developed over two years by a team of faculty from universities across our state and validated by a team at a partner university. All colleges and universities across the state use this tool for observing student teachers when they complete their      student-teaching placement(s) of 6-16 weeks. All criteria on the STOT are linked to the Interstate Teacher Assessment and Support Consortium (InTASC) standards and include criteria based on four broad categories: The Learner and Learning, Content Knowledge, Instructional Practice, and Professional Responsibility.   

Both the first and second study found no correlation between any of the six admission criteria — including performance on the three portions of the Praxis test and grade point averages overall, in communication courses, and in teaching — and performance on the STOT. This finding is consistent with other studies that show no correlation between basic skills tests and performance in a student-teaching placement (Gimbert & Chesley, 2009; Goldhaber, Cowan, & Theobald, 2017; Goodman, Arbona, & de Rameriz, 2008; Mikitovics & Crehan, 2002; Wright, 2015).  

Are tests doing their job? 

The purpose of admissions standards for teacher education programs is to ensure high-quality teacher candidates. However, this research suggests that standardized tests and GPAs do not predict who will be a good teacher. And our experience has shown us that the tests prevent some potentially good teachers from entering the field or cause them to delay entry as they work to pass the tests. Kimberly, for example, is a White woman who started her degree in 2003, paused her studies for three years, attended community college part time for two years, came to our campus part time for three years, paused her studies for three years because of the tests, worked at a restaurant, and returned again to take classes. She finished all of her classes but could not pass the math test so was unable to student teach last spring. Over the summer, she worked weekly with a math tutor and passed the test. She recently completed her student teaching assignment successfully.  

Our experience has shown us that the tests prevent some potentially good teachers from entering the field

In other cases, students never get the opportunity to demonstrate their potential or make an informed decision about whether teaching is right for them. For example, Jack, a deeply philosophical White man from a rural community who wanted to learn more about the world, attempted to take the tests multiple times but was unsuccessful despite tutoring from Lisa. He wanted to become a physical education teacher, but he did not complete a degree and now works in the oil fields. 

If neither tests nor GPAs tell us who is going to be an effective teacher, what purpose do they serve or should they serve? Linda Darling-Hammond (2006) argues that assessment of teacher candidates should be used not just for measuring their teaching abilities but also to help teachers improve. The fundamental purpose of assessment should be to see what our teacher candidates still need to work on and what they have successfully mastered, but test results reveal very little that genuinely helps teacher candidates become better teachers. Perhaps the only occasional benefit of the tests is that some college students get diagnosed with a learning disability and become more aware of accommodations and strategies that can help them. If tests are not effective at helping teachers improve, then the tests need to change or be eliminated. In short, we need to move from “externally generated single-measure tests to multi-pronged internal assessments of teacher performance and student learning” (Cochran-Smith et al., 2016). 

Ideally, assessments of teacher candidates would focus on their performance, via observations of their work in the classroom as well as in behind-the-scenes activities, such as planning, collaboration, and analyzing children’s strengths and challenges. Assessments should also include multiple opportunities for aspiring teachers to demonstrate the skills and knowledge they will use in practice. Our current clinical evaluations require such evaluations, and we use them to determine who needs further work. Current standardized tests do not include any of these features. In addition, the passing rates for many standardized tests used in teacher education vary widely by race (Gitomer & Qi, 2010; Goldhaber, Cowan, & Theobald, 2017; Michael Luna, 2016; Nettles et al., 2011), a fact that we need to be mindful of as we seek to create equitable admissions policies and foster greater diversity in the teaching workforce (Burke & Whitty, 2018). 

Some teacher education programs have begun to include interviews or screening instruments that look at prospective teachers’ knowledge, skills, and attitudes. One widely used interview format, the Haberman Star Teacher Selection Interview, has not been validated, but certain of its subscores have been found, in one study, to predict which students will graduate (Waddell & Marszalek, 2018). In another study, a group interview format was found to predict success on student-teaching evaluations and produced high inter-rater reliability scores (Byrnes, Kiger, & Shechtman, 2003). In a review of 13 studies on the relationship between practicing teachers’ performance and personality traits, Harrison Kell (2019) noted that many evaluations focus on the Big Five (agreeableness, conscientiousness, extraversion, neuroticism, and openness to experience) and urged further research into measuring and correlating these characteristics.   

Test results reveal very little that genuinely helps teacher candidates become better teachers.

Instead of focusing on blocking potential teachers from entering the field, these approaches to admissions aim to identify beliefs and dispositions that are likely to be problematic in the teaching environment and that those future teachers may need to change (DiGiacinto, Bulger, & Wayda, 2017). For example, admissions committees might ask candidates about their beliefs regarding inclusive education — if candidates respond by asserting that, say, dyslexic students cannot succeed in general education classrooms, this shouldn’t necessarily exclude them from the teacher education program (Pecek & Macura-Milovanovic, 2015); rather, it should be taken as an indication of what those candidates will have to learn about inclusive instruction to be effective teachers.  

A better approach to assessing teacher candidates 

In fall 2015, we began requiring an additional 170 hours of clinical experiences for teacher candidates, mostly in a school setting, before student teaching. Since making this change, we have found that fewer and fewer of our teacher candidates are having trouble in their student teaching because they decide much earlier in the program whether teaching is a good path for them. Many students complete their first clinical experience in their first semester, so if candidates see in their first year at the university that they do not like teaching, they select other majors. In addition, our cooperating teachers let us know right away when teacher candidates are not coming to their clinical experiences regularly or are not displaying the expected dispositions, allowing us to remove them from placements immediately. Finally, we have two checkpoints along the way, the early-level transition point and the mid-level transition point, at which we meet with the teacher candidates, review their evaluations, and help them identify areas of strength and challenges, as well as goals for their next stage. The additional time in schools, contacts with cooperating teachers, and transition points have enabled us to identify candidates who need additional assistance with the content, dispositions, or skills that will help them succeed. We then have those students complete additional clinical experiences before beginning student teaching.  

This change has enabled us to think of admission to our program in a new way, with students receiving support to help them succeed, rather than having to jump over hurdles designed to keep them out. Our two transition points, in which teacher candidates reflect on their strengths and challenges and set goals for the upcoming semester, provide opportunities to help them understand what skills are most critical for them to develop. Relying on ratings from cooperating teachers and teacher education faculty, we are able to give feedback that helps students assess their progress and make decisions about next steps.  

Those who want to stay in the program but are not meeting expectations — for example, for absences from clinical placements, inappropriate interactions, or dress code issues — are given clear and direct feedback. In some cases, we counsel students out of the program and into degrees that do not lead to teacher licensure. In other cases, we keep working with the teacher candidate to improve what needs to be improved. If we want teachers to give students multiple chances to succeed, we need to model that approach by giving extra mentoring to teacher candidates who need to learn how to be confident in front of the room or who are having trouble connecting their instruction to their assessments. We would much rather spend our time helping future teachers think about how to help their students learn rather than coaching them on how to pass a standardized test. After all, we can think of no times in a teaching career when students care about teachers’ abilities to pass a test.  

Standardized tests still play a role in our teacher education program, but it is quite limited. Most of the teacher candidates who enter our university take the ACT for admission, but they may also submit an SAT or take a free test for placement into English and math courses. Our state university system has established placement requirements, and based on these placement results, students either take and pass remedial classes before their required classes or go directly into the required classes. By looking at students’ placement test results and their GPAs for the education major, for the communications classes, and overall, we can see which teacher candidates are struggling academically and provide assistance for them to improve in areas they will need for their teaching. Students still need to pass two standardized tests, one on content and one on pedagogy, but it is rare for students to struggle on these. 

Our profession focuses on preparing students to be professional educators, yet our research reveals no clear value in requiring standardized entrance exams in the professional training of future teachers. As we consider the core activities that define our profession, perhaps it is time to reconsider the entrance exam model. Multiple teacher candidates in our program show strong signs that they will be excellent teachers, but they struggle to pass a standardized test that shows no statistical correlation with their performance as student teachers. Is that a reason to deny them access to the profession? Should we keep teachers with learning differences out of the profession when a key feature of our job is helping all students learn? Let’s treat teacher preparation criteria less like barriers to entry and more like guides for improvement.