The need for teachers from diverse backgrounds has been well documented (Villegas & Irvine, 2010). Black and Latinx students who were taught by same-race teachers, or in schools where the diversity of the teachers mirrors the diversity of the students, achieved positive outcomes on a variety of measures when compared to those who were taught by white teachers or in schools where the staffing did not reflect the diversity of the student population. For students of color, these gains may be due to teachers of color serving as role models, holding higher expectations, and understanding the cultural backgrounds and lived experiences of the students. Similarly, teachers with learning disabilities (LD) could benefit students with LD. Teachers with LD who taught students with LD reported being able to: identify with their students, hold students to higher expectations, and serve as exemplars of success (Ferri et al., 2001). However, there are practices in education that work in counter ways, such as the standardized assessments required for public school teacher licensure that have shown to negatively impact efforts in the field to diversify the teacher workforce (Bennett et al., 2006; Cowan et al., 2020; Memory et al., 2003; Wakefield, 2003).

With the aim of improving outcomes on teacher licensure exams, some researchers suggest setting minimum test scores for acceptance into teacher preparation programs, since these measures have been correlated with licensure exam outcomes (Burke, 2005; Poelzer et al., 2000). However, setting these admittance thresholds may inadvertently screen out potentially successful candidates from diverse backgrounds. Indeed, some institutions that became SAT optional found their entering classes more diverse than previous classes (Baker & Rosinger, 2020). On the other hand, early indicators such as standardized test scores could provide insights into those who may be at risk of not earning their license and help these individuals adjust their academic plans. Thus, our college sought to identify whether there were metrics that could improve licensure exam outcomes. Some faculty wondered if we should set a higher admissions bar, while some wondered if we could identify students who should be counseled into majors with more achievable testing requirements, such as preschool teacher pathways. Others wondered whether we could identify interventions that improve chances of success. If the metrics were used for admittance or advising into particular majors, identifying the impact of the use of these metrics would be critical. Who would be counseled out of licensure programs for teaching grades 1-6 or special education? Would these decisions be appropriate, or could those individuals have been successful? Could these metrics reduce our college’s contribution to diversifying the teacher workforce?

Presented here is my analysis of our college’s student data to answer these questions, and how exploring our own data impacted our thinking. I examined 10 years of data on teachers seeking licensure (TSLs) in elementary or special education accepted to Curry College. The data set (N=103) included only those teachers who attempted the mathematics portion of the state teacher licensure exam. It should be pointed out that in Massachusetts several exams must be passed for licensure in elementary or special education, and the mathematics portion is typically one of the last exams attempted. The data set included traditional undergraduate students, continuing education students, and graduate students, including 5th-year master’s students. Ninety-four (94) of the TSLs submitted their SAT scores. I identified a mathematics SAT score below which the probability of passing the mathematics portion of the teacher licensure exam was substantially lower. I determined which of the TSLs scoring below this SAT score were ultimately successful on the mathematics portion of the state teacher licensure exam. I also examined college mathematics course-taking patterns.

Figure 1 shows that TSLs who scored 450 or below on the mathematics SAT were substantially more likely to not pass the mathematics portion of the state teacher licensure exam. Among the TSLs who reported scores of 450 or lower on the mathematics SAT, 39% did not pass the mathematics portion of the state teacher licensure exam, compared to only 5% of TSLs reporting scores above 450. However, if an admittance cut score of 450 had existed, 23 TSLs who were ultimately successful on the mathematics portion of the state teacher licensure exam would have been screened out, including TSLs identifying as Black or Hispanic or participating in the optional program for students with LD, as displayed in Table 1. While these numbers may seem small, our college is a small, majority white institution, so proportionally they are notable. Comparably, only 8% of teachers in Massachusetts are teachers of color (Jung, 2018). Note that in the table “passers” refers to TSLs who passed on their first attempt, as well as students who passed after failing one or more times. “Non passers” attempted the mathematics portion of the state teacher licensure exam one or more times but never passed. “Non passers” will often change tracks to pathways that require less rigorous testing, such as licensure for preschool through second grade.

Figure 1: Mathematics SAT scores of Teachers Seeking Licensure (TSLs) who never passed the mathematics portion of the state teacher licensure exam (non passers) compared to those who passed the mathematics portion of the state teacher licensure exam on their first or subsequent attempts (passers).

Table 1

Outcomes of Mathematics Portion of State Teacher Licensure Exam by Demographics and Mathematics SAT Score

Our college prides itself on welcoming a student body with a wide range of academic backgrounds and supporting them in achieving their academic goals. Given the problematic nature of cut scores, I was curious what interventions could increase the likelihood of TSLs’ success on the mathematics portion of the state teacher licensing exam. I compared TSLs who had taken at least one additional college level mathematics course beyond that required by their major, and controlling for mathematics SAT score, determined if there was a difference in their licensure exam outcomes. Of the 102 TSLs whose transcripts could be accessed, 36 TSLs took college-level mathematics courses beyond the mathematics content courses for elementary teachers required by the state; 31 of these students took more than one additional course. The most popular courses were Statistics (35 TSLs), College Algebra (21 TSLs) and Calculus (6 TSLs), though additional options have become popular with TSLs in recent years. Of the 19 TSLs who never passed the mathematics portion of the state teacher licensure exam whose transcript could be accessed, 17 (89%) took no additional mathematics coursework. Among the 36 TSLs who took one or more additional mathematics courses, 34 of them (94%) eventually passed the mathematics portion of the state teacher licensure exam. As shown in Figure 2, TSLs who took additional mathematics courses also failed the mathematics portion of the state teacher licensure exam fewer times, across a range of mathematics SAT scores. Notice that the white circles, representing TSLs who took additional mathematics coursework, tend to cluster lower vertically, representing fewer attempts on the mathematics portion of the state teacher licensure exam. The small circles, representing TSLs who never passed the mathematics portion of the state teacher licensure exam, are nearly all black; these TSLs typically did not take additional mathematics coursework. Perhaps most surprising to me as I conducted this analysis, the large white dots were spread across the various mathematics SAT scores. Originally, I suspected that students who were naturally skilled at or enjoyed mathematics or mathematics tests chose to take additional mathematics courses, and this predisposition to mathematics also led them to do well on the state teacher licensure exam. Instead, this graph suggests that students with a range of mathematical test taking skills chose to take additional mathematics courses, which then improved their performance on the mathematics portion of the state teacher licensure exam.

Figure 2: Number of Times Individuals Failed the Mathematics Portion of State Teacher Licensure Exam, by Mathematics SAT Score and Additional Mathematics Coursework.

These results were discussed with mathematics faculty, education department faculty, faculty who advise education majors, and the director of advising. The data changed the discussion from policies that would screen out TSLs to policies that would better support TSLs to succeed on licensing exams. I myself faced an unconscious bias: I typically recruited TSLs to take additional mathematics courses and pursue a mathematics education minor if they performed well on assessments in my mathematics-for-teachers courses. I had likely failed to actively recruit students who scored lower on the mathematics SAT, though they stood to tangibly benefit from additional mathematics courses.

As teacher educators and advisors, my colleagues and I were aware of the research on unconscious and systemic bias, persistence, and growth mindset. We knew of the need for teachers from diverse backgrounds. Yet, we were unknowingly entertaining policies that were not aligned with this need and could have further reduced the number of Black and Hispanic teachers or teachers with LD. Examining the quantitative impact such policies would have had on our students challenged us to rethink our assumptions and practices, particularly regarding admission to the education majors and recruitment to mathematics courses. Local data, in contrast to larger research studies, are powerful in this way.

As mathematics teacher educators, we have a unique skillset. We have familiarity with teacher licensing exams and other standardized assessments. As social science researchers and quantitatively savvy individuals, we can analyze, present, and communicate findings about data. Many of us also have an awareness of the historic exclusion of students of color and students with LD from higher education. The Association of Mathematics Teachers Educators’ (AMTE) Standards for Preparing Teachers of Mathematics calls upon programs to recruit teachers who share the demographics of the communities in which they will work (indicator P.5.2) (AMTE, 2017). The Standards also call upon mathematics teacher educators to prepare beginning teachers to challenge the status quo, hold themselves and others accountable, advocate for their students, and challenge unjust policies (indicators C.4.4 & C.4.5). Thus, when we as mathematics teacher educators hear of policy suggestions that could negatively impact those who have been historically excluded from higher education or reduce the diversity of the teacher workforce, we have a duty to use our knowledge and skills to investigate and advocate as well as to reflect on our own beliefs and practice. We investigate, advocate, and reflect not only to recruit a more diverse teacher workforce, but to model to our beginning teachers that which we expect from them. For both mathematics teacher educators and for beginning teachers, the work inside a classroom is only one part of our role in working toward a more equitable system of mathematics education.

References

Association of Mathematics Teacher Educators. (2017). Standards for Preparing Teachers of Mathematics. Available online at amte.net/standards.

Baker, D., & Rosinger, K. (2020). Test optional offers benefits but it's not enough. Education Next, 20(3), 67–70.

Bennett, C. I., McWhorter, L. M., & Kuykendall, J. A. (2006). Will I ever teach? Latino and African American students’ perspectives on PRAXIS I. American Educational Research Journal, 43(3), 531–575.

Burke, K. (2005). Teacher certification exams: What are the predictors of success? College Student Journal, 39(4), 784–794.

Cowan, J., Goldhaber, D., Jin, Z., & Theobald, R. (2020). Teacher Licensure Tests: Barrier or Predictive Tool? Working Paper No. 245-1020. National Center for Analysis of Longitudinal Data in Education Research. American Institutes for Research. https://files.eric.ed.gov/fulltext/ED609773.pdf

D’Agostino, J. V., & Powers, S. J. (2009). Predicting teacher performance with test scores and grade point average: A meta-analysis. American Educational Research Journal, 46(1), 146–182.

Ferri, B. A., Keefe, C. H., & Gregg, N. (2001). Teachers with learning disabilities: A view from both sides of the desk. Journal of Learning Disabilities, 34(1), 22–32. https://doi.org/10.1177/002221940103400103

Jung, C. (2018). Teaching force in Mass. public schools still overwhelmingly white. WBUR. https://www.wbur.org/news/2018/04/09/overwhelmingly-white-teaching-force...  

Memory, D. M., Coleman, C. L., & Watkins, S. D. (2003). Possible tradeoffs in raising basic skills cutoff scores for teacher licensure: A study with implications for participation of African Americans in teaching. Journal of Teacher Education, 54(3), 217–227. https://doi.org/10.1177/0022487103054003004

Poelzer, H., Zeng, L., & Simonsson, M. (2000, February). Hispanic preservice teachers: Variables that predict success on teacher certification exams. National Association of African American Studies & National Association of Hispanic and Latino Studies 2000 Literature Monograph Series. https://files.eric.ed.gov/fulltext/ED454006.pdf

Villegas, A. M., & Irvine, J. J. (2010). Diversifying the teaching force: An examination of major arguments. The Urban Review, 42(3), 175–192. https://doi.org/10.1007/s11256-010-0150-1

Wakefield, D. (2003). Screening teacher candidates: Problems with high-stakes testing. The Educational Forum, 67(4), 380–388. https://doi.org/10.1080/00131720308984587