Do New Exams Produce Better Teachers? States Act While Educators Debate
It took less than a minute for Mario Martinez to finish the first six questions of the algebra exam that his professor, Ivan Cheng, had just handed to him. The high school-level test was supposed to be a good example of an exam, so that the graduate students in Cheng’s math methods course at the California State University, Northridge’s school of education would better understand what rigorous high school-level questions look like, and how to write tests for their own lessons.
By the end of the first page, Martinez had already learned an important lesson: “Beware of redundant problems,” he scribbled on the side of his paper before flipping it over to finish the problems on the back.
Martinez has until the fall to hone his skills before he will be sent into a classroom to practice as a student teacher. And he has at least a year before he will have to prove that he can not only teach math, but also create tests and analyze student results. It is a skill that many educators say is a sign of a good teacher, and one so important it was included in a lengthy exit exam that all aspiring teachers must take before they receive a teaching credential from the state.
Aspiring teachers videotape themselves teaching a lesson and write several lengthy reflections. California introduced the performance assessments in 2001 to adhere to a 1998 state law. Teachers must pass them in order to receive certification.
Every teacher preparation program in the state must choose one of three versions for students to take, each of which centers around the teaching and self-reflection activity. The Performance Assessment for California Teachers, or PACT, is the test of choice for Northridge and more than 30 other teacher preparation programs in the state, and many classes, like Cheng’s math methods course, design curriculum around the assessment to ensure students are prepared to pass.
Although it is largely untested and debated amongst educators, the PACT has served as a model for a national exam, known as the edTPA, that at least 25 states are introducing. Developed by 12 California institutions in 2001, the PACT was put on hold when the state suspended the performance assessment requirement in 2003. Three years later, the requirement was reinstated, and in early 2007 the state’s Commission on Teacher Credentialing approved the assessment.
The multi-part test, which often takes a semester to complete and results in dozens of pages of essay reflections, tries to assess whether an aspiring teacher is able to teach multiple learners in real classrooms. It has been tapped as a nationwide model because supporters say it presents a complex picture of a candidate’s strengths, weaknesses, and classroom readiness.
But many educators hesitate to say that the new performance assessments are creating better teachers or that passing them is a sign a teacher will be effective, partly due to the lack of more evidence.
Martinez says that Cheng’s class has spent extra time on designing and grading tests for lessons they have created because it is typically “the part of the PACT that math teachers do the worst on.” While some say this practice of designing teacher preparation curriculum around the PACT bears resemblance to K-12 teachers “teaching to the test,” many educators at Northridge say the PACT is focused on critical areas of good teaching, like planning lessons with strong student assessments, and modifying lessons for English language learners and students with disabilities, and that it therefore only reinforces what candidates should learn anyway.
“The PACT certainly has forced us to give greater attention to certain aspects of what it means to teach and to deliver a lesson more effectively,” said David Kretschmer, chair of the Department of Elementary Education at California State University, Northridge. “We are churning out a better product, if you use that expression, than before we adopted PACT.”
That was the intent of the creators of the PACT, educators from teacher preparation programs in California who wanted to ensure that all pathways to teaching in the state were centered on research-based teaching practices that would produce better teachers.
Some research has found that high scores on performance exams like the PACT may signify that a teacher will be more effective in the classroom. One study out of Stanford University, which helped design the PACT, found that for each additional point an English Language Arts teacher scored on the exam, which is scored on a 44-point scale, students averaged a gain of one percentile point per year on California standardized tests. But the study only looked at 14 teachers and their 259 students.
If passing the PACT means teachers are prepared for the classroom, then by pass rates alone it would indicate that programs using assessment are, for the most part, producing teachers ready for the challenges of the classroom. In the 2009-10 school year, 33 percent of aspiring teachers in the state applying for their credential took the PACT. Ninety-four percent of them passed all sections of the exam on the first try.
But according to the California Commission on Teacher Credentialing, the PACT’s pass rates are much higher than those on the California Teaching Performance Assessment (CalTPA), taken by a majority of teacher candidates in the state, and the Fresno Assessment of Student Teachers (FAST), taken only by candidates at California State University Fresno. The CalTPA had the lowest pass rate, with only 77 percent of candidates passing all sections of the exam on the first try. The FAST has a first-time pass rate of 87 percent.
The high pass rates have skeptics wondering if the performance assessments are rigorous enough. All three versions of the assessments are usually scored by the institutions themselves, and students can retake them if they fail the first time.
Linda Darling-Hammond, a professor of education at Stanford University who helped design the edTPA, the national test, says the high pass rate on the PACT is expected. Teachers in California take up to three standardized tests, including a basic skills assessment and several subject matter tests, even before they take the PACT or one of the other two versions of it. Darling-Hammond says each exam knocks out about 10 percent of the aspiring teacher pool in the state. (In the 2009-10 academic year, 78 percent of candidates passed the state’s basic skills assessment, and 81 percent of applicants passed the reading instruction exam.)
“This [pass rate] is only the people who’ve made it through all those gauntlets, that managed to get into the program, and haven’t caved when they were asked to do the PACT,” said Darling-Hammond.
She added that the preparation programs that use the PACT, including the University of California system, Stanford, and several schools within the California State University system, have the highest selectivity in admissions to their preparation programs. “If this were statewide,” added Darling-Hammond, “the pass rate would certainly be much lower.”
A report from the state’s Commission on Teacher Credentialing cautioned against comparing the pass rates. Unlike the PACT, which is taken at the end of the preparation program, candidates take the four sections of the CalTPA at different times throughout their programs. Some programs counsel students out before they take the performance assessment, meaning only the top students may end up taking the exam.
Opponents of performance assessments say that preparation programs, and the state, are missing the point by relying on an assessment to determine if teachers are prepared for the classroom.
Ann Schulte, associate professor at California State University, Chico, says that preparation programs should be focused on working with and assessing teacher candidates in the field, so they receive frequent observations and feedback during their student teaching experiences from someone with extensive knowledge of their abilities and classrooms.
Schulte cited research that found alignment between the results of those who pass the PACT and the observations of educators supervising those candidates in the field. “It begs the question then, ‘why are we doing it?’” Schulte said.
Elsewhere in the country, some educators and students have asked the same question, and subsequently refused to administer or take the national version of the assessment. In 2012, all but one student in the secondary-teacher training program at the University of Massachusetts Amherst refused to participate in the exam, arguing that mentors who observed them in a student teaching setting for months would be better judges of their teaching ability than Pearson, the education company administering the exams.
More concerning to some schools is the idea that pass rates on performance exams could be used to determine the quality of teacher preparation programs. Since 1998, the federal government has attempted to increase the accountability for preparation programs by requiring states to collect and report information about the programs, including completion rates, average scores on state and national teaching tests, and the number of student teaching hours required.
California includes pass rates from performance assessments in its own annual analysis of this data, and uses that data as one of many measures that determines if a school of education is “low-performing.”
In California, there is general consensus that the performance assessment, which encourages students to focus on how they would teach a variety of students, has at least created more thoughtful teachers, even if the research isn’t clear that the tests are improving the quality of the teaching force.
“It’s hard to imagine that the exercise isn’t raising their expertise level,” said Julie Gainsburg, associate professor at Northridge. Gainsburg says that the assessment is requiring students to reflect on their teaching and planning in ways that are more sophisticated than before the PACT.
“Does PACT make a better teacher? No,” said Nancy Prosenjak, a professor at Northridge. “But I think we have a substantial program that’s research based, we have the PACT,” she added. “So with all of those, maybe we have better teachers.”
Vol. 32, Issue 32