Elementary Math Licensure Test: A Tale of Two States

The Massachusetts Story

The Massachusetts Board of Education was the first state board to decide that elementary teachers should be expected to know the math they teach in grades 4-6.  The teaching license for these teachers covers grades 1 to 6. So, that meant, to the Board, that prospective elementary teachers should be tested on one of their licensure tests for their knowledge of mathematics in grades 1-6.  Not how to teach math in those grades, something they should learn in their math methods classes and something they should practice in their student teaching while being supervised by their “cooperating” teacher and their math methods instructor in their teacher preparation program.  But the same deep understanding of the math they are expected to teach their elementary students in grades 1-6.

In December 2006, the Massachusetts Board of Education approved the development of a 40-item stand-alone licensure test for prospective elementary and elementary special education teachers.  The test is based on the reasonable principle that these teachers-to-be should be expected to demonstrate without the use of a calculator a deep understanding of the same mathematical concepts underpinning what they teach their students, also without the use of calculators.

The test went into effect in 2009. The startlingly low pass rate on the first test administration was a shocker.  The Massachusetts Board of Education was faced with approving a cut score for a new elementary mathematics licensure test that meant that only 27% of the test-takers would pass. The official minutes of the May 19, 2009 meeting containing the paragraph below don’t capture the tension (or the exact words spoken) at this point in the meeting.  The heart of the discussion was about the effect on diversity of a recommended pass score that meant getting only 60% of the items correct on a 40-item test of elementary mathematics. How could the Board ensure that diversity meant academic quality in its teaching force? 

The test had been developed and vetted by the state’s own mathematics educators and mathematicians. The state’s mathematics organizations were all in favor of the test as well as the recommended cut-off score. At least, no one in these organizations testified against the test or recommended cut-off score. But one Board member tried to raise an argument against them.

Vice Chair Chernow said her concern is that teacher licensing is already overly complex and bureaucratic, and additional tests complicate it further. She said that 46 states use the Praxis test and asked whether using a different test [the new test] impedes reciprocity and teacher mobility. Dr. Howard said he appreciated the expressions of concern about the diversity of the teaching pool, but standards should be set based on what students require from their teachers. Dr. Stotsky said the Praxis test [for elementary teachers] is weak on mathematics (pp. 6-7).

Dr. Howard, whose undergraduate and graduate degrees were both from Harvard University, said something in my recollection of the discussion closer to: “Quality comes first.  Then look for diversity.” (I attended this meeting in my position as a state school board member.)

He is an African American, and there was no further discussion of this issue by the Board. The recommended cut score and test were approved unanimously. About 50 percent of the test-takers across test administrations on average have passed the test since the Board’s decision. No information is available on race or ethnicity (pp. 66-67).

In sum, the pass rate has hovered around 50 percent on average over test administrations, suggesting how needed such a test was (only 60 percent of the items need to be correct for a passing score).  The only overt opposition to the board’s vote of approval for this pass score (in 2009) was from the state affiliate of the National Education Association. State affiliates of a variety of mathematics organizations approved it, and the one minority member of the board (at the time) indicated that academic quality came first, diversity second, in response to concerns expressed by two white board members about the minority pass rate.  There is no information available on what happens to test-takers who never pass the mathematics portion of the test, no matter how many times they take it. (They can get their college degree but not a teaching license.)  The numbers who fail this test are very high. The numbers who fail this test when they re-take it are also very high.  

To address the competition from the Bay State, ETS also developed a separately-scorable mathematics test for prospective elementary teachers (also with 40 test items), available in 2011.    Pass rates are not easily comparable across states since states set their own pass score on a PRAXIS test.  

The 40-item Massachusetts test of elementary mathematics knowledge in its General Curriculum test (03) and the 40-item PRAXIS test of elementary mathematics knowledge in its Multiple Subjects test (5031) are the only two stand-alone elementary mathematics tests available.  But we do not know how strong the PRAXIS test is. The ETS website provides no information on how strong the test is or who developed the guidelines for test items on mathematics. In contrast, the mathematicians and others who developed the Massachusetts test are identified in Guidelines for the Mathematical Preparation of Elementary Teachers. These 2007 Guidelines also indicate what topics should be taught in mathematics coursework for prospective elementary teachers (pp. 67-68). The Guidelines strongly suggest that teacher preparation programs in Massachusetts should require all prospective elementary teachers to take 2-3 elementary math courses (not math ed courses).  Information on how many do is not available.

On to North Carolina

The story does not end here.  It continues in North Carolina.  The argument over this test in North Carolina, as noted in a local newspaper article, concerns, first, the number of prospective elementary teachers who have failed the test since its inception in 2013. That number is still unclear. However, the gist of that article (and subsequent articles on the topic) was that the problem was likely the test. To understand why the test was likely not the problem and that the root of the problem was more likely test-takers’ lack of preparation for the test, it is necessary for policy makers on the North Carolina Board of Education to know something about the test itself.

Clearly, there is good reason to think that any group of prospective teachers would have already obtained a strong understanding of pre-algebra mathematics.  After all, whether in Massachusetts or North Carolina, they had completed elementary, middle, and high school themselves—years earlier—and had been required to study mathematics at most grade levels. 

In 2012, upon recommendation from the North Carolina Department of Public Instruction, the North Carolina state board of education voted to adopt the Massachusetts test. The North Carolina superintendent of public instruction at the time likely helped to set the pass score (cut-off score) for the state. It may or may not be identical to the pass score in the Bay State; we don’t know. The testing company provided a practice test similar to the 100-item practice test it provided in the Bay State to suggest how difficult the actual test may be.  

We also do not know if all teacher preparation programs in North Carolina require prospective elementary teachers to use it or take the recommended elementary mathematics coursework described in the Guidelines for the Mathematical Preparation of Elementary Teachers developed in the Bay State to help teacher preparation programs. These guidelines also indicate what topics should be taught in mathematics coursework for prospective elementary teachers, preferably by people with advanced degrees in mathematics. How this conflict will play out remains to be seen. But the animus against licensure tests for teachers seems to be growing, as a 2017 article about Florida suggests. 

The basic question about licensure tests for teachers has yet to be discussed publicly. Are such tests to have relatively low cut-off scores (i.e., high passing rates) to protect the putative right of any adult who wishes to be a teacher, or are they to have high enough cut-off scores to protect students from academically incompetent teachers?  (Or are states to have no teacher licensure tests at all and implicitly facilitate nepotism or racial quotas in school hiring practices?)  The authors of the 1966 Coleman report on teacher quality and its influence on academic achievement implicitly suggested that schools need to protect minority students from academically inadequate teachers.

Why did most legislation (in re-authorizations of the Elementary and Secondary Education Act or ESEA) and in most grant programs talk about “failing” or “low-performing” schools and do little if anything to strengthen academically our teaching corps?  Perhaps it has been easier to claim that teachers were not teaching some students as well as they should because teachers were bigoted and held low expectations for some students.  It was certainly unclear how general undergraduate academic demands could be strengthened for teachers at a time when political leaders claimed they wanted to see more college graduates, not fewer. The higher the bar, the fewer will likely pass it.  But we should find out if the bar is being lowered while at the same time the claim is that it is being raised. All students, including minority students, need academically stronger teachers, whether in North Carolina or Massachusetts.