Chapter 1: Elementary and Secondary Education

Mathematics and Science Teachers

Strengthening the quality of teachers and teaching has been central to efforts to improve American education in recent decades (NCTAF 1996 and 1997). Research findings consistently point to the critical role of teachers in helping students to learn and achieve (Darling-Hammond 2000; Goldhaber 2002; Wright, Horn, and Sanders 1997). Today's teachers are being called on to provide the nation's children with a high-quality education and to teach in new ways (Little 1993). Many believe that professional development is essential to improving teacher quality and that changes in teaching practices will occur if teachers have consistent and high-quality professional training (Desimone et al. 2002). Although professional development can help improve the quality of teachers and instruction, its effectiveness is diminished if schools cannot keep the most successful teachers in the profession. The issues of teacher salaries and working conditions have come under increasing scrutiny in recent years because cumulative evidence suggests that these are two key influences on teachers' persistence in the profession and professional satisfaction (Hanushek, Kain, and Rivkin 2004; NCTAF 2003; Odden and Kelley 2002). This section uses data from various sources to examine important issues related to teaching in mathematics and science, including teacher quality, participation in professional development, pay, and working conditions. Indicators in this section traditionally have relied heavily on the Schools and Staffing Survey (SASS) of the U.S. Department of Education. The SASS: 2003–04 data collections have been completed, but these new data were not available when this chapter was prepared.

Teacher Quality

The NCLB emphasizes the importance of teacher quality and requires all public school teachers of core academic subjects to meet specific criteria in preparation for teaching by academic year 2006.[32] In recent years, many states have developed new standards for teaching and implemented policies to improve the quality of teaching (Hirsch, Koppich, and Knapp 2001; Potts, Blank, and Williams 2002) (see sidebar "State Education Policies Related to Teachers and Teaching"). Although there is substantial agreement that teacher quality is one of the most important influences on student learning, disagreement remains about what specific knowledge and skills constitute "quality" (Goldhaber and Anthony 2004; Greenberg et al. 2004; McCaffrey et al. 2003; Wilson, Floden, and Ferrini-Mundy 2001). The following indicators of teacher quality focus on traditional measures identified in the literature on teaching effectiveness (Darling-Hammond 2000; Hanushek 1996): the academic background of college graduates entering the teaching force and congruence between teacher preparation and their assigned teaching fields.[33]

Academic Background of Entering Teachers

Early research on the sources of teacher effectiveness often examined their academic background and skills because these attributes predict teacher subject mastery and verbal ability, two elements believed to be critical to effective teaching (Darling-Hammond 2000; Vance and Schlechty 1982; Weaver 1983). Measures of academic competence commonly used over the past two decades are standardized test scores (Henke et al. 1996; Murnane et al. 1991; Vance and Schlechty 1982; Weaver 1983). Based on test scores, research shows that college graduates who became teachers had less rigorous academic preparation than those who did not go into teaching (Murnane et al. 1991; Vance and Schlechty 1982; Weaver 1983). These findings are further supported by transcript data from the National Education Longitudinal Study of 1988 (NELS:88), which tracks student progress from middle school through postsecondary education. Among 12th graders in the high school class of 1992 who had earned bachelor's degrees by 2000, those who entered K–12 teaching trailed graduates in nonteaching occupations on a number of academic measures in high school and college: they took fewer rigorous academic courses in high school, had lower achievement test scores at the 12th grade, and scored lower on college entrance examinations (figure 1-14 figure.). The differences were particularly salient when comparing teachers with those who entered the fields of engineering or architecture; research, science, or technology; computer science; and health care (appendix table1-20 Excel table.). Teachers also were more likely to attend less-selective colleges and less likely to graduate from selective institutions, particularly when compared with those entering engineering, architecture, research, science, or technical fields and those working as editors, writers, reporters, or performers (appendix table1-20 Excel table.).[34]

Congruence Between Teacher Preparation and Teaching Assignments

Although almost all U.S. teachers hold at least basic qualifications (e.g., a bachelor's degree and teaching certification) (Henke et al. 1997), many are teaching subjects for which they lack adequate academic training, certification, or both (Seastrom et al. 2002). This mismatch, commonly termed out-of-field teaching, has been a major policy concern, and its elimination has become a target of federal and state reform initiatives (Ingersoll 2002, 2003). The discussion below focuses on two important credentials required by NCLB for a teacher to meet the law's definition of highly qualified: certification and a college major or minor in the subjects taught.

Certification in the Assigned Teaching Field. Teaching certification is generally awarded by state agencies to teachers who have completed specific requirements. These requirements vary across states but typically include completing a bachelor's degree, completing a period of practice teaching, and passing one or more exams (Kaye 2002). A teaching certificate in their assigned teaching field provides basic but essential documentation of teachers' academic preparation and teaching skills (Goldhaber and Brewer 2000) (see sidebar "National Board-Certified Teachers").

In 2002, 80% of public high school mathematics teachers had full certification in mathematics (table 1-11 table.); one-fifth were either not fully certified or certified in a field other than mathematics.[35] The percentage of public high school science teachers with full certification in their teaching field ranged from a high of 83% for biology teachers to a low of 72% for earth science teachers. Certification rates for public middle-grade (seventh and eighth grade) mathematics and science teachers were lower: 60% and 58%, respectively.

Certification rates of mathematics and science teachers declined from 1990 to 2002. The percentage of public high school mathematics teachers with full certification in mathematics decreased from 90% in 1990 to 80% in 2002. Declines also occurred among biology, chemistry, physics, and earth science teachers. At the middle-grade level, the picture is somewhat different. The percentage of mathematics and science teachers with full certification increased in the late 1990s but declined subsequently.[36]

Certification rates varied greatly across states, reflecting, in part, different state policies and licensing requirements. In 1999.2000, the percentage of public school teachers who taught mathematics to 7th to 12th graders and who had full certification in mathematics ranged from 100% in Rhode Island and West Virginia to 65% in Hawaii (appendix table1-21 Excel table.). Likewise, certification rates for public school 7th to 12th grade science teachers ranged from 100% in Idaho, Vermont, and Wyoming to 77% in Kentucky.

College Major or Minor in the Assigned Teaching Field. A growing body of research shows that teacher subject- matter knowledge is significantly associated with student learning (Greenberg et al. 2004; Hill, Rowan, and Ball 2004; Monk and King 1994), but what counts as "useful subject-matter knowledge" for teaching remains largely unspecified. One indicator used to gauge the breadth and depth of teacher subject-matter knowledge is whether they have a college major or minor in their teaching field (Ingersoll 2003). The assumption is that teachers acquire their subject-area expertise mostly in college, so a college minor in a subject is the minimum prerequisite for teaching that subject.

In 1999–2000, 71% of public school teachers who taught mathematics to 7th to 12th graders had a college major or minor in mathematics, and 77% of public school teachers who taught science in these same grades had a college major or minor in science (appendix table1-22 Excel table.). In other words, 29% and 23%, respectively, of 7th to 12th grade mathematics and science teachers in public schools had neither a major nor a minor in the subject they taught.

As with certification, the distribution of mathematics and science teachers with a college major or minor in their field was uneven across states. In 1999.2000, only in Arkansas did 90% of 7th to 12th grade mathematics teachers have a college major or minor in mathematics, and only in Minnesota and New Jersey did more than 90% of 7th to 12th grade science teachers have a college major or minor in science (appendix table1-22 Excel table.). More than 30% of teachers lacked even a college minor in their assigned teaching fields in 21 states for mathematics and 10 states for science.

Some recent studies suggest several reasons for the prevalence of out-of-field teaching. Demand for qualified teachers may exceed the supply, forcing school districts to hire less-qualified candidates to fill vacancies (Broughman and Rollefson 2000; Howard 2003). Also, schools may assign current staff members to out-of-field classes rather than expending administrator time and effort and school resources on finding and hiring new teachers in the field (Ingersoll 2003). Furthermore, the perception of precollegiate teaching as a female-dominated and easy-to-enter occupation not requiring a great deal of expertise, skill, and training may foster the belief that teaching credentials do not matter very much, thus out-of-field teaching is considered a tolerable practice (Wang et al. 2003).

Top of page.Top of page

Teacher Professional Development

Ongoing efforts to raise academic standards in mathematics and science require teachers to have knowledge and skills that many did not acquire during their initial preparation for teaching (NCTM 2000; NRC 1996). The changing and expanding demands of teaching jobs have prompted increased attention to the importance of professional development in providing teachers with opportunities to acquire new knowledge and keep abreast of advances in their field (Elmore 2002; Little 1993). For two decades, the U.S. government has made teacher professional development a component of its reform efforts (Porter et al. 2000). Many states also have developed and implemented policies designed to promote participation in professional development and to improve its quality (CPRE 1997; Hirsch, Koppich, and Knapp 1999, 2001). By 2002, 48 states had required professional development for teacher license renewal, and 24 had adopted professional development policies aligned with state content standards (figure 1-15 figure.). As of 2004, 37 states financed some professional development programs, 35 had standards in place for professional development, 27 provided professional development funds for all districts in the state, 16 required and financed mentoring programs for all novice teachers, and 13 required districts or schools to set aside teacher time for professional development (Editorial Projects in Education 2005; Potts, Blank, and Williams 2002) (see sidebar "New Models and Current Practices in Professional Development").

Effects of Professional Development

Research literature contains a mix of large- and small-scale studies, including intensive case studies of classroom teaching (e.g., WestEd 2000), evaluations of programs designed to improve teaching and learning (e.g., Banilower 2002; Weiss, Banilower, and Shimkus 2004), and surveys of teachers about professional development experiences (e.g., Choy and Chen 1998; Parsad, Lewis, and Farris 2001). Thus far, strong evidence of the positive effects of professional development is limited to teaching practices. Relatively few rigorous studies have directly linked teacher professional development to improved student outcomes (Elmore 2002; Guskey 2003). More research is needed following the advent of mathematics and science testing under NCLB. Several recent studies on the effects of professional development are summarized below.

  • In their longitudinal study tracking the experiences of mathematics and science teachers participating in various professional development activities, Desimone et al. (2002) found that professional development focusing on specific teaching strategies (e.g., use of technology, higher-order instruction, use of alternative assessments) increased teachers' use of these strategies in the classroom. Also, the effects on teachers' instruction were stronger when professional development included collective participation of teachers from the same school, department, or grade; active learning opportunities such as reviewing student work or obtaining feedback on teaching; and coherence such as linking to other activities or building on teachers' previous knowledge. The Consortium of Chicago School Research also found that "high-quality" professional development programs (those characterized by sustained and coherent training, collaborative learning, and followup support) had a significant effect on teachers' instructional practices (Smylie et al. 2001).

  • Studies conducted by NCES based on national data found that a majority of teachers who had participated in professional development programs on various topics relating to teaching and instruction reported that these programs were useful and improved their classroom teaching practices (Choy and Chen 1998; Parsad, Lewis, and Farris 2001; Smith and Desimone 2003).

  • Studies show that teacher participation in professional development affects teaching practice, which in turn affects student performance. For example, the National Staff Development Council examined the features of award-winning professional development programs at eight public schools that had made measurable gains in student achievement (WestEd 2000). The researchers observed that, in each school, the nature of professional development had shifted from isolated learning and occasional workshops to focused, ongoing organizational learning built on collaborative reflection and joint action. Wenglinsky (2002) found that higher student test scores in mathematics and science were linked with teachers' professional development training in higher-order thinking skills.

  • Based on an extensive review of studies on the effects of professional development on student achievement, Clewell et al. (2004) concluded that the content of professional development linked to subject-matter knowledge was more important than its format in terms of improving student achievement. Clewell and her colleagues cited the work of Kennedy (1998) and Cohen and Hill (2000) to support their conclusion. Based on 12 studies of professional development programs that reported effects on student achievement, Kennedy (1998) found that the programs showing the greatest effects were those that focused on subject-matter knowledge and on student learning in a particular subject. Cohen and Hill (2000) also reported that students of California elementary school teachers who attended curriculum-focused workshops and learned about the state assessment system had higher achievement scores on the assessment.

  • Numerous studies indicate that sustained and intensive professional development is an important factor in influencing change in teachers' attitudes and teaching behaviors (Clewell et al. 2004). For example, the amount of time teachers spent on professional development activities was positively related to their perceptions of these activities' usefulness (Parsad, Lewis, and Farris 2001). The more time teachers spent on professional development in using computers for instruction, the more likely they were to have their students use computers during class (Choy, Chen, and Bugarin forthcoming).

  • Based on data from the NSF-funded Local Systemic Change (LSC) project,[37] researchers found that participation in LSC professional development positively changed teachers' attitudes and teaching behaviors (Banilower 2002; Boyd et al. 2003; Weiss, Banilower, and Shimkus 2004). Changes were most evident among those who participated intensively (e.g., more than 60 hours or even more than 80 hours) in LSC professional development (Boyd et al. 2003; Weiss, Banilower, and Shimkus 2004). Other research also suggests that teachers typically need at least 80 hours of intensive professional development before they change their classroom behaviors and practices significantly (Supovitz and Turner 2000).

Top of page.Top of page

Teacher Salaries

Teacher salaries are the largest single cost in education, making compensation a critical consideration for policy-makers seeking to increase the quality of the teaching force. For many years, schools have tried to attract highly qualified and skilled people to teaching and to keep the most able ones from leaving the profession (Hanushek, Kain, and Rivkin 2004; Macdonald 1999). Evidence suggests that teacher salaries play an important role in determining both the supply of new teachers and retention of current teachers (Odden and Kelley 2002; Shen 1997). The indicators below review changes in U.S. teacher salaries and compare their salaries with those of teachers in other nations.

Trends in U.S. Teacher Salaries

The average salaries (in constant 2002 dollars) of all U.S. public school K–12 teachers decreased from 1972 to 1982, increased from 1982 to 1992, and remained about the same between 1992 and 2002 (figure 1-16 figure.; appendix table1-23 Excel table.). The net effect was that the average inflation-adjusted salary of all public school K–12 teachers was $44,367 in 2002, just about $2,598 above what it was in 1972. The average salary for beginning teachers followed a similar trend.

Teacher salaries are often lower compared with the salaries of other white-collar occupations (Allegretto, Corcoran, and Mishel 2004; Horn and Zahn 2001), but comparing teachers' annual salaries to those of other workers is complicated by some unique features of the teaching profession, such as a shorter work year. To control for differences in time worked, a recent study focused on the weekly wages of teachers from 1996 to 2003.[38] The results showed that teachers' weekly wages consistently and considerably lagged behind those of other workers with similar education and experience and that this gap had enlarged over time (Allegretto, Corcoran, and Mishel 2004).

International Comparisons of Teacher Salaries

After adjusting for the cost of living, U.S. teachers earn more than teachers in many other countries (OECD 2004). In 2002, the beginning, mid-career (after 15 years of teaching), and top-of-the-scale statutory salaries for U.S. public primary and secondary school teachers were all higher than the corresponding OECD averages (figure 1-17 figure.).[39] However, regardless of experience, teachers in Germany and Switzerland earned significantly more than U.S. teachers and the gaps seemed to increase with the level of schooling. Teachers with 15 years of experience in Japan and South Korea also earned more than their U.S. counterparts (appendix table1-24 Excel table.).

Statutory salaries may not capture all differences in salaries because teaching time varies considerably across countries. To control for this variation, an alternative measure of teacher pay is the ratio of annual salary to the number of hours per year the teacher is required to spend teaching Students in class (referred to as salary per instructional hour). When instructional time was taken into account, U.S. teachers did not fare well compared with teachers in other nations (appendix table1-24 Excel table.). The salary per instructional hour of U.S. teachers with 15 years of experience was lower than the OECD average at both the lower and upper secondary levels and was the same at the primary level.

Another way to compare teacher salaries across countries is to compute the ratio of salaries to the per capita gross domestic product (GDP). The resulting ratio compares teacher salaries with a country's overall wealth and may indicate a nation's financial investment in teaching as a profession. Appendix table1-24 Excel table. shows the ratio of teacher salaries after 15 years experience to per capita GDP. U.S. ratios were below the average for OECD countries for all three levels of education.

Top of page.Top of page

Attrition and Mobility of Mathematics and Science Teachers

In addition to salary, working conditions affect the career decisions of potential and current teachers and their professional satisfaction with teaching (Bogler 2002; Hanushek, Kain, and Rivkin 2004; Hardy 1999; Luekens, Lyter, and Fox 2004; Ma and Macmillan 1999; Shen 1997). Research shows that teacher effectiveness can be enhanced in environments that support and value their work and can be diminished by poor working conditions, lack of professional support, wide-spread student problems, and inadequate facilities and resources (Macdonald 1999; NCTAF 2003; Scott, Stone, and Dinham 2001). The following indicators examine the attrition and mobility of mathematics and science teachers, discuss their reasons for moving or leaving the profession, and examine their views on school working conditions.

Various studies, commissions, and national reports on teacher supply and demand have concluded that teacher shortages in mathematics and science are considerable (AAEE 2003; NCTAF 2003). Teacher attrition (teachers leaving the teaching profession) is a general contributing factor, whereas teacher mobility (teachers moving from one school to another) also creates staffing problems in individual schools. Between the 1999–2000 and 2000–01 school years, 7% to 9% of public school mathematics and science teachers left the teaching profession and 6% to 7% moved to a different school (figure 1-18 figure.). The attrition of mathematics and science teachers appears to be increasing over time: only about 5% of public school mathematics and science teachers left the profession between the 1987–88 and 1988–89 school years.

Reasons for Leaving or Moving

In 2000–01, both mathematics and science teachers and other teachers rated the following reasons as very or extremely important in their decision to leave teaching: pursuing another career, obtaining a better salary or benefits, and retiring (table 1-12 table.). However, mathematics and science teachers were more likely than other teachers to cite pursuing another career as a very or extremely important reason for leaving, whereas others were more likely to give retirement as a very or extremely important reason for leaving. These results suggest that retaining mathematics and science teachers can be particularly difficult because they may find more lucrative career opportunities elsewhere (see sidebar "Occupations of Former Teachers").

Teachers who moved to another school seem to have different motives from those who left the profession. Among the top reasons given by mathematics and science teachers who moved to a new school were dissatisfaction with support from school administrators (40% for mathematics and science teachers and 38% for other teachers) and dissatisfaction with workplace conditions (37% for mathematics and science teachers and 32% for other teachers) (table 1-13 table.). Mathematics and science teachers who moved were more likely to report changing schools to obtain a better salary or benefits (29% and 18%) but less likely to move for a better teaching assignment (26% and 42%).

Perceptions of Working Conditions by Teachers Who Moved, Left, or Stayed

In general, teachers who left or moved expressed less satisfaction with their schools' conditions than did those who stayed (appendix table1-25 Excel table.). Among public school mathematics and science teachers, those who left the profession were less likely than those who stayed to report satisfaction with the amount of autonomy and control they had over their classrooms, with teaching in their current or last year's schools and teaching overall, with the availability of computers and other technology for their classrooms, and with opportunities for professional development.[40] Mathematics and science teachers who left for nonteaching jobs appeared to be more satisfied with their new jobs (see sidebar "Former Teachers' Satisfaction With New Jobs Compared With Teaching").

Those who moved to a different school also appeared to be more critical of experiences and conditions in their former schools than those who stayed: they were less likely to report satisfaction with the subject they were assigned to teach, with the amount of autonomy and control, with feeling safe inside or outside the school, with job security, with the high caliber of professionalism, with school emphasis on academic success, with supportive administrators, and with uninterrupted class time (appendix table1-25 Excel table.).

Top of page.Top of page


Indicators in this section reveal both progress and ongoing challenges in strengthening the U.S. teaching force. Based on a number of measures, ranging from the rigor of high school coursetaking and achievement test scores at the 12th grade to college entrance examination scores and the selectivity of the institutions from which teachers enrolled and graduated, teaching appears to attract a higher share of college graduates with weak academic backgrounds. Although almost all public middle-grade or high school mathematics and science teachers held a bachelor's degree and teaching certification, many were teaching subjects for which they did not have certification or a college major or minor in the field. The distribution of out-of-field teaching in mathematics and science was uneven across states.

During the past decade, many states have developed and implemented professional development policies and increasing proportions of teachers participated in professional development programs. Although the characteristics of high-quality professional development have been identified, most teachers' professional development experiences were not of high quality. The dominant form of professional development in the late 1990s were still one-time workshops with little followup and most teachers attended programs for only a few hours over the course of the school year, far below the minimum of 60 to 80 hours some studies show as needed to bring about meaningful change in teaching behaviors.

Between the 1999 and 2000 academic years, 7%–9% of public school mathematics and science teachers left the teaching profession, and another 6%–7% changed schools. Those who left often reported they planned to pursue another career. Those who moved cited various aspects of poor working conditions as reasons for changing schools. One-third of leavers found a job outside the field of education and many reported more satisfaction with their new job than with teaching.

Top of page.Top of page

Top of page.Top of page

Top of page.Top of page

Top of page.Top of page

Top of page.Top of page

Top of page.Top of page


[32] NCLB defines a highly qualified elementary or secondary school teacher as someone who holds a bachelor's degree and full state-approved teaching certificate or license (excluding emergency, temporary, and provisional certificates) and who demonstrates subject-matter competency in each academic subject taught by having an undergraduate or graduate major or its equivalent in the subject; passing a test on the subject; holding an advanced teaching certificate in the subject; or meeting some other state-approved criteria. NCLB requires that new elementary school teachers must pass tests in subject-matter knowledge and teaching skills in mathematics, reading, writing, and other areas of the basic elementary school curriculum. New middle and high school teachers either must pass a rigorous state test in each academic subject they teach or have the equivalent of an undergraduate or graduate major or advanced certification in their fields.

[33] Teaching experience is another indicator of teacher quality and was examined in the 2004 edition of Science and Engineering Indicators. Because of a lack of national data, that indicator cannot be examined in this edition. Other factors may also play important roles in teacher quality, including ability to motivate students, manage classroom behavior, maximize instructional time, and diagnose and remedy students' learning difficulties (Goldhaber and Anthony 2004; McCaffrey et al. 2003; Rice 2003). These characteristics are rarely examined in nationally representative surveys because they are difficult and costly to measure.

[34] Other research has found that teachers tend to have higher undergraduate GPAs than other graduates (Frankel and Stowe 1990; Gray et al. 1993; Henke et al. 1996). However, grades are not standardized among or within institutions, which makes it difficult to compare teachers' academic performance with that of other graduates.

[35] Full certification refers to a state's regular, standard, advanced, or probationary certificate. It does not include temporary, alternative, provisional, or emergency certificates granted to those who have not fulfilled requirements for licensing. These teachers are referred to as "not fully certified."

[36] Researchers often cited teacher shortages as a major reason for this decline, claiming that increasing student enrollment, reduction of class sizes, high rates of teacher turnover, and lack of qualified candidates have created teacher shortages, which in turn have forced schools and districts to hire less-qualified candidates to fill vacancies (Boe and Gilford 1992; Howard 2003). This explanation, however, has not been empirically demonstrated.

[37] The purpose of the LSC project is to improve the teaching of science, mathematics, and technology by focusing on the professional development of teachers within whole schools or districts. Each participating teacher is required to have a minimum of 130 hours of professional development over the course of the project. The training focuses on preparing teachers to implement designated exemplary mathematics and science instructional materials in their classrooms (Weiss, Banilower, and Shimkus 2004).

[38] Data on weekly pay of teachers come from the Bureau of Labor Statistics' Current Population Survey (CPS). Weekly earnings were either reported directly by respondents or estimated using the number of weeks worked and annual, monthly, or biweekly earnings.

[39] Statutory salaries refers to salaries set by official pay scales. These figures should be distinguished from the actual salaries teachers receive. The 2002 U.S. salaries were estimated from average scheduled salaries from the 1999.2000 SASS. The 1999–2000 figures were adjusted for inflation by 3.8% for 2000–01 and an additional 2.9% for 2001–02 (OECD 2004).

[40] Differences in other items also appear large, but are not statistically significant, because of large standard errors associated with mathematics and science teacher leavers.

National Science Board.