|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 14
1
Introduction
In recent years a number of studies have expressed concern about
current and prospective shortages in the nation's available supply of pre-
college science and mathematics teachers. Some studies claim that severe
shortages currently Ernst; other studies find that, while current shortages are
not severe, future shortages are likely; and still others find that, although
there is no quantitative shortage, there is a gap between the quality of
current teachers of science and mathematics and the quality needed to
ensure effective instruction. Most, but not all, of the studies have focused
on teachers at the secondary level, for which more information by discipline
is available.
Although the panel is not charged with determining whether a shortage
of precollege science and mathematics teachers either exists now or is likely
in the future, the mandate to specie types of data needed to understand
that issue requires the panel to examine the demographic and employment
patterns affecting supply and demand in particular labor market areas. Thus
we are concerned about the forces associated with changes in precollege
enrollments in science and mathematics courses, including both changes in
the demographic configuration of children in the relevant age ranges and
changes in state or district requirements specifying the number of science
and mathematics credits needed for high school graduation. We also look
at the principal determinants of the total supply of teachers, including the
demographics of the teacher corps.
A major concern is to understand the appropriate characteristics of
teacher qualifications and teaching quality, since supply and demand for
teachers come into equilibrium through adjustments in quality. Quality
cannot be monitored unless the characteristics associated with it can be
specified. Thus, our basic concern is to identify the types of data needed
14
-
OCR for page 15
INTRODUCTION
15
to understand quality in order to evaluate how it is changing. In the
course of the effort we examine some of the available data that have led
many to conclude that the quality of science and mathematics training
in the United States is not satisfactory. Specifically, we examine data
on student performance from studies carried out under the aegis of the
International Association for the Evaluation of Educational Achievement
and from the National Assessment of Educational Progress~ata that have
raised questions about the quality of teaching in that country.
Finally, we have considered the relationship between teacher~training
and preparation, teacher instructional activities in the classroom, and stu-
dent outcomes. Although it is certainly true that unsatisfactory outcomes in
terms of student understanding of important concepts and topics in science
and mathematics can be due in part to deficiencies in the academic back-
ground or pedagogical training of science and mathematics teachers, it does
not follow that poor outcomes can be attributed squarely to deficiencies in
these areas.
Many factors could contribute to poor student understanding. Unsat-
isfactory outcomes could be due to the structure of the science or mathe-
matics curricula; they could be due to insufficient emphasis on science and
mathematics topics in the allocation of time during the school day; they
could be due to the manner in which schools and classrooms are organized
with respect to opportunities for interchange among teachers, the amount
of time available to teachers for planning and preparation, the availability
of inservice training opportunities, and so on. Poor outcomes could also
be due to the fact that children receive less time and attention from par-
ents in home environments than was true in the past, or due to changes
in parents' expectations, beliefs, and behaviors related to learning science
and mathematics that influence children's developmental outcomes. It is
thus the panel's conviction that to understand the supply and demand for
precollege science and mathematics teachers, and to understand the quality
characteristics of teacher supply, we must go beyond a narrow mandate to
examine the adequacy of the available data from which teacher supply and
demand models could be constructed. However, the panel's mandate is not
so broad that it requires us to prescribe policies whose effects might be to
change either supply, demand, or quality.
THE MEANING OF SHORTAGE
In everyday parlance, when most people speak of a shortage of precol-
lege science and mathematics teachers, they are likely to mean that they are
dissatisfied with the quality of people teaching science and mathematics,
rather than to mean that there are insufficient numbers of teachers to staff
science and mathematics courses. In technical terms, it is hardly possible to
OCR for page 16
16
PRECOLLEGE SCIENCE AND MATHEh[4TICS TEACHERS
have either a shortage or a surplus of particular kinds of precollege teachers,
or indeed of teachers generally, since school systems typically have neither
classes without teachers to teach them (excess demand/short supply) nor
employed teachers without classes to teach (excess supply/short demand).
Thus a quantitative shortage fewer teachers teaching science and mathe-
matics than there are science and mathematics classes to be taught will
not be observed except in those cases (which may be frequent but not well
documented by data) in which a course or class is cancelled because a
teacher cannot be found with the appropriate credentials/qualifications.
What actually takes place is an equilibrating process that is expressed
in the short run by quality adjustments in the criteria for hiring next year's
teachers. In the long run, salary is the equilibrating factor for supply and
demand. While the quantity of people teaching science and mathematics
will almost always be equal to the quantity of science and mathematics
teaching offered, tendencies toward either surplus or shortage will surface
as adjustments in quality. In planning for the next school year, if there are
not enough applicants with science and mathematics credentials to teach
science and mathematics classes, a district will either undertake aggressive
recruiting or a teacher will be drafted from inside (or hired from outside)
and provided with emergency certification to teach the course. If there is
a potential surplus, qualified science and mathematics teachers will end up
either teaching some other subject or not teaching at all. In the fanner
case, if school systems do not recruit aggressively, they may have to dip
down far into the pool of teachers less experienced or qualified in science
and mathematics to fill the available positions. If the premise is true that
quality is positively associated with experience and training, their average
quality will tend to decline.1 In the latter case, depending on institutional
rules or practices, only the best qualified (or the most senior) science and
mathematics teachers will get the available science and mathematics classes,
and other teachers will have to go elsewhere or teach something else. If the
available classes go to the best qualified teachers, on average the quality
will tend to increase. However, if the available classes go to me most senior
teacher, which is the policy in most systems, the erect on quality is difficult
to assess.
1 Some economists define commodities By listing their attributes, of which quality is one. This way
of thinking about commodities, however, is a very special usage. In fact, in the situation described
the district did not get first-rate mathematics and science teachers, and therefore experienced
a shortage of such teachers. But it is rare that this would be revealed through questionnaire
responses, since questions ask only whether the district found people who were certified in the
relevant field to fill a vacancy. The dimensions in which equilibrium takes place, including quality,
are relatively unobservable. The concept of shortage does not suggest a strategy for measuring
shortage. Although we could refer to a "shortage of teachers of desired quality" throughout the
report, for simplicity we have chosen simply to refer to a "shortage."
OCR for page 17
INTRODUCTION
17
While these observations are straightforward and almost self-evident,
they do account for the fact that some studies of the supply-demand bal-
ance in precollege science and mathematics have concluded that there is
a considerable shortage of teachers, while others have concluded that no
shortage Busts at all. The former type of studies have defined shortage as
the absence of sufficiently qualified teachers to staff the relevant classrooms
and have judged that many classrooms are staffed by inadequately qualified
teachers (as examples, see National Education Association, 1988; Weiss,
1987; Akin, 1986~. The latter type of studies, asking whether schools have
been unable to hire teachers to teach science and mathematics courses,
have found that school systems are able to hire such teachers (as ex-
amples, see National Center for Education Statistics, 1985a; Feistritzer,
1988b). Thus the importance of the general proposition that, although
quantitative gaps between supply and demand are not generally identified,
quality adjustments ensure that supply and demand are equal is that un-
derstanding both the quantitative and the qualitative dimensions of teacher
supply and demand is essential to understanding the supply and demand
for teachers. 1b do otherwise is to miss a significant part of any potential
problem.
FACI ORS AFFECTING DEMAND
A data system able to track changes in the demand for precollege
science and mathematics teachers must as a minimum be able to assess
demographic factors, which include changes in student enrollment, in the
ratio of male to female students in science and mathematics courses, and
in the proportions of minority students in science and mathematics courses,
as well as changes in policy variables, such as graduation requirements
mandated by the state, entrance requirements of colleges and universities,
and changes in acceptable pupil-teacher ratios. Demand also depends on
the number of vacancies resulting from the creation of new positions and
from teacher attrition. All these factors affect the demand for classes in
science and mathematics.
As noted in the panel's interim report, the most accurate data used
in current supply and demand models are probably the demographic data
for projecting demand. For the precollege student population, projections
of the total will be extremely reliable for all K-12 grades for at least five
years into the future, since students starting kindergarten will already have
been born about five years ago. Thus, even birth rate projections have
only a small influence on demand projections, unless the projections go out
further than five years. And total enrollments in grades 7-12 (the point
at which specialized science and mathematics courses typically begin to be
OCR for page 18
18
PRECOLLEGE SCIENCE AND AL4THE~4TICS TEACHERS
50
40
en 30
o
. _
._
20
10
o
_
Grades K-12
Grades K-8
Grades 9-12
Projected
1972 1977 1982 1987 1992 1997
Year
FIGURE 1.1 Enrollment in grades K-12 of public schools, with projections: Fall 1972 to
1997. Source: National Center for Education Statistics (1988g:14~.
Offered) are known at least 12 years in advance since the children have
already been born.
The demographic base for projecting demand is not quite so solid as
the above paragraph suggests, even at a national level. Both in-migration
and out-migration occur among school-age children in the United States
as a whole. But at the more relevant regional, state, or local level, there
is obviously some migration of school-age children that must be taken
into account statistically. Thus, even very good national models need to
be substantially augmented with accurate subnational migration data to
produce useful demand projections at the relevant school district level.
School enrollment itself is projected to rise somewhat during the
next half-decade. For example, from 1978 to 1987, enrollment in public
secondary schools (grades 9-12) fell from about 14 million students to
about 12 million; but from 1988 to 1995 secondary enrollment is projected
to increase to about 13 million. Over the same period, elementary school
enrollment is projected to increase from a little over 28 million students to
about 31 million (see Figure 1.1~.
A number of forces currently under way suggest the need to track more
refined characteristics of the demographics and to add to them data relating
to mandated state requirements.For example, a potential exists for greater
OCR for page 19
INTRODUCTION
19
demand for science and mathematics training for female children in the
K-12 age range, simply because of persistent changes in attitudes toward
appropriate sex roles for men and women and the associated changes in
the career aspirations of young women. It is already evident that more
young women are planning to enter science and engineering fields than was
true 20 years ago. In fact, the number of undergraduate women majoring
in science and engineering has risen dramatically since the mid-1970s. For
example, the physical sciences showed an increase from 30,900 women in
1976 to 38,100 in 1984, then decreased to 36,500 in 1986 (National Center
for Education Statistics, 1988b:167~. Similarly, in 1976, 28,800 women
majored in engineering. This number increased to 74,800 by 1984 and
declined slightly to 71,200 in 1986. Though trends have attenuated for the
present, it is important to monitor the enrollment of women as science
and mathematics majors at the postsecondary level. Currently, 28 percent
of all physical sciences majors are female; because of the potential for
further increase, female enrollment in science and mathematics should be
monitored.
Similarly, it seems likely that the movement toward equal opportunity
will generate an increased demand for science and mathematics training
on the part of minority youngsters. The evidence here, some of which
is shown in Tables 1.1 and 1.2, is not easy to interpret. For students
who were high school seniors in 1980, the 1980 data indicate that blacks
actually took more semesters of mathematics than whites, and only Chicanos
and Native Americans (especially the latter) are markedly lower than
average. For science, black seniors were well below whites in number of
semester hours in 1972, but in 1980 black seniors had nearly caught up
with whites, while all the other minority groups except Asian-American and
Puerto Rican students were below whites. These data are for seniors, and
high school dropout rates are much higher for minority students than for
whites. Moreover, the data noted above do not standardize for the level of
science and mathematics courses. Minorities other than Asian-Americans
are historically more likely to be found in remedial mathematics than in
the more challenging mathematics courses and in general science courses
than in physics and chemistry (Office of Technology Assessment, 1988:45~.
A recent report by the Educational Testing Service (ETS) drawn from
a research paper on course-taking patterns in the 1980s by Goertz (1989),
compares students' course-taking patterns in 1982 and 1987, using data from
the High School and Beyond study by the National Center for Education
Statistics (NCES) (1982 graduates) and the High School Transcript Study
by Westat, Inc. (1987 graduates). This report (Educational Testing Service,
1989:20) finds significant gains in course-taking by black and Hispanic
students between 1982 and 1987. Some of the gains were impressive,
others modest. For example, 29 percent of black graduates in 1982 had
OCR for page 20
o
oo
c~
e~
-
o
· -
4~
-
· -
:^
-
L~
o
. -
CQ
:^
-^
o
oc
~ -
e~
1
o
m
_
a.
._
ct
J
C~
-
o
o
U)
._
w
o
a~
4
oo
V)
.
m
o
00
ba
V
U)
U]
m
O
O-~- ~c
~ g
._
bw ~
~ C~
~ .
U]
U)
-
._
io
-
~ ·C~6
~ O
._
~ ._
V] ~
* ** * *
~P ~ C~ CO 00 ~ CO
O ~oo ~ ~ O (D
· . .. . . . .
O O OO O O O O
1
U'
e~ a
~ CO
C~ _t
a' co
~ ~ O
C`4 oo c~ ~ a,
~ ~ ooa) ao ~ ~ oo
. . .. . . . -
~ ~ ~ e~ _1 ~
co ~ ooe~ co co ~ ~
0 0 e~ ~ c~ 0 a'
. . . . . . . .
~ ~ ~CO CO
c61
oo e ~ ~ c~
0 -4 ~ C9
e~ ~_ c~
_ _
oo oo CO~ ~ 0 ~ ~
. . .. . . . .
~ _1 ~ ~ ~4 ~ ~
CO ~ (D~ O a~ e~ oo
~ ~ 00C~ 0O 0 ~ C~
. . .. . . . .
CO CO COC~ C~
,, C
Q4 ~
O ~ ~ c: ~ ~ ~ ~ ~
it m :z v ~ o
20
a
o
P~
L.
:^
w
o
s"
_`
1
CD
-
_ R
~d a~
U, ~
O ~..
O ~ U,
._ oo
O
~q, _
_ ~Q _
~ 4,
O
~O ~
C ~p. O
oo . .
~n
_ ~ ~
~ ~o
* ~1 v:
OCR for page 21
o
w
i`
-
p~
f"
· -
~:
IY
-
w
·
b~
-
o
· _
U]
o
-
e~
1
o
a:
W
L.
C~
._
U]
-
o
o
~C
V]
b.
._
~4
o
0e
4)
U]
C~
-
m
b~
o
00
I
_ C~
b ~t-
oC O)
V
cn
m
b~
-
o
o
~C
U]
O ~C
00 t~
<53 ·-
~:
U)
-
W
._
.
bO
o
e~ · - °
W
U) ~
~ 8
.~
W ~
~ W
~ .
b~
U. ~
W
W
U] ~
~ ._
W ~
~ b85
~ .>
W
cn a
w
*
e~
o
e~
00
e~
*
e~ c~ 0 oo 0
e~ 0c~ 0 ~ ~ ~
. .. . . . .
O OO O O O O
~ ~ 00 0 ~ ~
co 0~ e~ ~ ~ u,
CDe~ ~ e~ ~ co
O
-
O ~ {D O
. . . . ...
_ e~ _ _ _ e ~_
CD ~ U ~e~ u~ ~ ~ eq
, co -
. . . . . . . .
a~ co o ~a~ co c~ co ~
b~|
~ ~ ~O O CD - 00
o ~ ~chq u~ co oo cO
o oO ~ o'
- - -
e~ C5) -
00 00 ~u:tD ~CS, 00
. . . ... . .
_ _ _I ___ _
_ ~ e ~u~ CO ~ 0 e~
U3 ~CO 00 00
... . . ~. .
coa~co e ~c ~co
E~
o
m
o
a~
:^
al
o
V
._
V .- ¢
m
21
o
v
v
V
·- U
._ ¢
o ,
~ o ~
~_
~1
s"
~. -.
- o
w
1
~CD
- ,Q
W W
o oo
o . - U:
o Ld °°
W
o ~
~o
b ~
OO
0O
~q
-
*
V
o
. .
L.
o
CQ
OCR for page 22
22
PRECOl.~.EGE SCIENCE AND M'4THEMi4 TICS TEACHERS
taken geometry; by 1987, 44 percent had taken geometry, compared with 64
percent of whites. In calculus, gains were slight for blacks, from 1 percent
in 1982 to 2 percent in 1987, compared with 6 percent for whites. The
percentage of Hispanic graduates who had taken algebra I rose from 55
to 77 percent between 1982 and 1987, and by 1987 they were nearly even
with whites (at 78 percent). Minority gains in science course-taking were
similarly notable. However, blacks and Hispanics still lag behind whites
and Asians in their enrollments in the higher-level mathematics and science
courses.
In addition, school districts have changed their graduation require-
ments to include more science and mathematics training or credits required
for graduation from high school. In 1985, NCES surveyed a sample of 565
districts and asked for math and science requirements for high school grad-
uation in 1982, 1985, and the expected requirements in 1988. Between
1981-82 and 1984-85, for example, nationally the average number of years
of course work required for graduation from public high schools increased
from 1.6 to 1.9 for mathematics, and from 1.5 to 1.8 for science (see
Able 1.3~. The National Commission on Excellence in Education (1983)
recommended 3.0 years for both science and mathematics.
In response to changes in the graduation requirements of districts,
states, and even in the entrance requirements of colleges and universities,
increased enrollments in high school science and mathematics courses have
been documented (Educational Testing Service, 1989~. The years between
1982 and 1987 have seen strong gains in science and mathematics course-
taking, except in physics and calculus, for which gains were modest or
nonexistent. To the extent that new state course requirements exceed
those already in place in the districts, the result can be a stronger demand
for science and mathematics training, given the same student population.
However, when local school district requirements already exceed new state
requirements, which they often do, new demand for teachers may not result.
Therefore it is important to monitor changes in course requirements at both
the state and district level to assess the effects on the demand for teachers.
In addition to changes in course requirements, a number of other
policy-related factors influence the demand for new science and mathe-
matics teachers. Changes in pupil-teacher ratio can result in changes in
demand. And a number of policy-related factors at the school, district, or
state level can influence the ratio-changes in budgets, class size policies, or
course requirements, for example. These changes should be monitored in
any data system that tracks changes in demand for science and mathematics
teachers.
Another major component of demand models is the pattern of attri-
tion for science and mathematics teachers-due both to retirement and
OCR for page 23
J
b4
. -
J
._
oo
v
._
CO
._
a
- o
o
J
VO
-
o
._
C.
~a
-
o
o
~C
v
U]
bO
._
Ll
s"
._
o
3 ~
oo
~V 1
C)
o
~-
oo
b.
e~
~ ,
~ _.
oo
°O C5)
m
al
CS ~
V
O ~
~n co
·. bO
L.
O
V
.m
co
-
o
o
v
U]
~C
~0
._
~0
c
._
v
~n
al
v
._
rC
h ._
r~
_
O ~
V `4
U] O
co oo a,0
. . ..
o
~1=1
0 ~ e~0
. . ..
O 0 0e~
CD 00 ~O
. . ..
CO CO CO~
~ ~ OO
. . ..
e~co
co a~ co°
. .. .
e ~o,
O O
oq ~
O ·E ~
._ ._
O C Vl
~1 ~ _ - O
e~ u~ oo ~ ~ ~ ·_
~ ~F i~ ~ ° ~ ~
23
oo
-
.
r. ~
~ o
~ co
u, ~
o co
o o
r. ~
~ v
4,
oo ~
a ~
r41 ~21
~o
oo
be V
. -~
~ v
'v
b4
~-
f"-
o ~
·-t.)
~- ~
-~
~ o
· - 'v
~ ~v
~ 3
.,4
o f.,
o
~ 3
oo
°v
o
:^
~C
Vl
~q
oo ~
o
C)
~ o
" a~
~ r.
L.
o
C al
V ~
._
~ V
._ ._
_ oo
o ._
o ~
"C _
V o
o
~o
_
oo -
o ~
~ o
b.
~V
J
-
~3
oo
~Q
bO
r.
~V
._
r.
._
e~
oo
, .
.e
Ld
~V
._
b.
-
- o
C~
~C
o
:^
O
,=
._ ~
_ ~V
~ . ~
r os
L.
~ ~ _
C~ ~ O
·
O
-0 bO
.
~ o ~
·
L. O
~- ~ ~
b.
i - ~ L.
c
~ v ~
L.
~5 oo
v c ~
~ ·
x v ~
·
~ v ~
X r#
c ~ ~
V ° a,
m C h
O O
a~
.s
OV O
V ~
C ~ J
C "C
._ ~
O
CS oo
C
r. ._
L. ~
ta
s" ~
4C'
V
q,)
O
~ ~1
-
~o
. .
C~
~o
-
._
al
._
U)
o
._
oo ~
V
C ~
L'
~q o
_ ~
r ~L
V
. .
V
O O
OCR for page 24
24
PRECOLLEGE SCIENCE AND MATHEMATICS TEACHERS
especially to leaving earlier in one's teaching career. It is essential for an
effective data system to be able to monitor attrition rates by subject as well.
Finally, research Is called for to identify other behavioral factors that
influence the demand for teachers: for example, patterns of dropping out
of high school, parents' choice of private over public schools, and the timing
of that choice.
FACTORS AFFECTING SUPPLY
Teacher supply can be examined in terms of retention rates for the
present stock of teachers, the flow of newly certified teachers from colleges
and universities, and the flow of returning teachers who have been absent
from the labor market, laid off during the past decade due to declining en-
rollments, or have come from other occupations or alternative certification
routes. As with demand, these factors include both demographic charac-
teristics (the age distribution of current teachers) and policy variables. In
our interim report, we noted that most of the existing supply models focus
on the flow of new graduates of education degree programs, despite the
fact that most of the new hires during recent years have come from other
sources.
The existing data, most commonly from the states and from periodic
surveys at a national level, should be examined in greater detail to estimate
future declines in the supply of available teachers, both for precollege
science and mathematics and for precollege teachers generally. 1b what
extent will there be a substantial decline in the overall teacher retention
rate, arising from the fact that large numbers of teachers will be entering
the age and experience combination at which teachers have often retired
in the past? One of the best-established relationships in the teacher supply
literature is the U-shaped relationship between age/experience and teacher
retention: in the early years, attrition rates are high either because many
entering teachers find that the occupation is not what they had thought,
have adverse experiences that result in withdrawal from the teacher corps,
or find more attractive employment opportunities. At the other end of
the spectrum, where the older and more experienced teachers are located,
attrition rates rise as retirement approaches. Thble 1.4 illustrates this
pattern for the state of New York. Both early and late attrition estimates
will be important factors affecting the supply of science and mathematics
teachers.
The current composition of the teacher corps is concentrated in an
age/experience cohort in which there will be many retirements starting in
the late 1990s. For example, the 1985-86 Survey of Science and Mathe-
matics Education conducted for the National Science Foundation (Weiss,
1987) found some indication that the science and mathematics teaching
OCR for page 26
26
PRECOLLEGE SCIENCE AND AL4THEAL4 TICS TEACHERS
force is aging but did not predict an unusually large wave of retirees in
the near future. Monitoring relevant statistics related to factors associated
with choosing to stay or choosing to retire is crucially important for un-
derstanding future teacher supply, as is monitoring the effects of incentive
programs designed to encourage continuation and discourage retirement
(or vice versa).
One sign of an impending shortage of new teachers has been a decline
in the number of education degrees awarded. For example, the number
of bachelor's degrees in education fell from 108,000 in 1980-81 to 87,000
in 1985-86 (National Center for Education Statistics, 1986:134; 1988b:196~.
The number of master's degrees in education also declined, from 99,000 in
l9SO-81 to 76,000 in 1985-86. Among those enrolled as teacher candidates
in secondary education programs, the proportion majoring in mathematics
education held steady at about 25 percent between 1985 and 1988. How-
ever, the proportion of students majoring in science education has declined
from 21 percent of all enrolled in 1986 to 16 percent in 1988 (AACTE,
1989~. The shortage issue is complicated, since new teacher supply can be
fairly quickly adjusted as opportunities are perceived to arise.
Conventional teacher training institutions are not the only source of
new supply. In recent years new supply has come mainly from a broader
source of teachers that includes (1) graduates of other institutions who
enter the teacher supply with temporary credentials and later certification;
(2) the so-called reserve pool: past graduates of teacher training or other
institutions who did not enter teaching when they graduated but could
be attracted to teaching careers with the right incentives; and (3) former
teachers who return to teaching from another occupation or activity. In
short, monitoring the basic demographics of teacher age/experience, as
well as the potential supply of new graduates and returnees from other
occupations, will be crucially important to understanding the probable
evolution of teacher supply over the next decade.
QUALITY ISSUES IN SUPPLY AND DEMAND
Much of the impetus for concern over the supply, demand, and quality
of precollege science and mathematics teachers arises from the continuing
evidence that U.S. students do not appear to know as much science and
mathematics as their age peers in other countries. The most widely cited
such data come from the International Educational Assessment program
(IEA) and from the National Assessment of Educational Progress (NAEP).
The IEA administered science tests to fifth-grade and ninth-grade
students and to twelfth-grade students who were studying biology, chemistry,
or physics in the terminal grade in school in 17 countries in 1983 (1986 for
the United States). The results of these tests tend to show U.S. students'
OCR for page 27
INTRODUCTION
27
science performance declining from a middle position in fifth grade to quite
low by twelfth grade (IEA, 1988~. In science, U.S. 10-year-olds were eighth
among 15 countries ranked? U.S. 14-year-olds were fourteenth among 17
countries ranked, and of 13 countries ranked for twelfth-grade students
who were taking science courses, U.S. biology students were thirteenth,
chemistry students eleventh, and physics students ninth (Table 1.5~. In
general, although U.S. students did relatively poorly overall, they did worse
at the higher grades and better at the lower grades.
This may be explained in part by cross-nationa1 differences in science
curricula. The science curricula in the other countries participating in
this study generally require more years of science than are required in
the United States. The U.S. results for grade 12 generally correspond
to student achievement near the end of their second year of the subject;
students in the other countries generally would have completed three years
of the science by grade 12 (Jacobson and Doran, 1988~.
U.S. twelfth-grade college-preparatory mathematics students fared
poorly against their peers in both developed and less-developed countries
of the world in performance on mathematics achievement tests (McKnight
et al., 1987~. For example, for high school seniors taking mathematics, U.S.
students' scores ranked in the lowest quarter of the countries in three cate-
gories (number systems, algebra, and geometry) and were below the median
in the other three (sets and relations, elementary functions/calculus, and
probability/statistics). Eighth-grade students in the United States ranked
somewhat higher, scoring at the median in arithmetic, algebra, and statis-
tics; at the 25th percentile in geometry; and below it in measurement. The
mathematics data are shown in Bibles 1.6 and 1.7.
In the IEA mathematics study, the method used by the United States,
England, and Wales to obtain a sufficiently large number of cooperating
school districts, namely requesting participation of twice as many school
districts as were needed with the expectation of a 50 percent cooperation
rate, might be expected to produce a bias in achievement scores. However,
no evidence of bias has been found (Garden, 1987:133~.
Neither of the international comparisons is without its problems and
ambiguities. For example, it is not clear whether the student populations
tested in the IEA science study are fully comparable across countries.
Furthermore, it is sometimes argued that the tests themselves are biased,
since the U.S. curriculum in science and mathematics may be different
from the typical curriculum used elsewhere, and the tests may be heavily
weighted with items that are not covered in U.S. curricula. Data from the
IEA study on test validity and test relevance, however, do not support that
proposition (IEA, 1988:88-95~.
A similar picture is presented by the NAEP data on achievement
scores, which indicate that large fractions of U.S. students do not appear
OCR for page 28
28
PRECOLLEGE SCIENCE AND MATHEMATICS TEACHERS
TABLE 1.5 Rank Order of Countries for Science Achievement at Three Levele of
Schooling
10-Year 14-Year Grade 12/13
-Old~, -Olds, Science Students a Non
Grade Grade Science
4/5 8/9 Biol- Chem- Phys- Students
ogy istry ice
Australia 9 10 9 6 8 4
Canada
6 4 11 12 11 8
(English speaking)
England 12 11 2 2 2 2
Finland 3 5 7 13 12
Hong Kong 13 16 5 1 1
Hungary 5 1 3 5 3 1
Italy 7 11 12 10 13 7
Japan 1 2 10 4 4 3
Korea 1 7
Netherlands - 3
Norway 10 9 6 8 6 5
Philippines 15 17
Poland 11 7 4 7 7
Singapore 13 14 1 3 5 6
Sweden 4 ~8 9 10
Thailand - 14
U.S.A. 8 14 13 11 9 -
Total
Number of
Countries 15 17 13 13 13 8
a Students taking biology, chemistry, or physics in the terminal grade in school.
Source: International Association for the Evaluation of Educational Achievement
(1988:3~.
OCR for page 29
INTRODUCTION
29
TABLE 1.6 Mathematics Achievement Comparisons: Twelfth Grade United States
and International, 1981-82 (Percentage of Iteme Correct)
United States
Pre
International
(15 Countries)
calcu- Calcu- 25th 75th
lus lus Percen- Percen
Topic Classes Classes Total tile Median tile
Sets &
relations 54 64 56 51 61 72
Number
systems 38 48 40 40 47 SO
Algebra 40 57 43 47 57 66
Geometry 30 38 31 33 42 49
Elementary
functions/
calculus 25 49 29 28 46 55
Probability/
statistics 39 48 40 38 46 64
-
Source: McKnight et al. (1987:23)
to meet minimal standards of literacy in science and mathematics. The
NAEP Science Report Card of September 1988 indicated that, despite
gains over the past four years, particularly among minorities, a majority
of high school students "are poorly equipped for informed citizenship
and productive performance in the workplace" (National Assessment of
Educational Progress, 1988b:5~.
A problem with both NAEP and the IEA tests is the limited extent
to which they assess higher-order skills. Although some test materials
administered by NAEP and IEA involve hands-on exercises, much more
research and development activity is needed to construct free-response
materials and techniques that measure skills not measured with multiple
choice tests. Current improvements in mathematics and science curricula
are focused on learning of "conceptual knowledge, process skills, and
the higher-order thinking that scientists, mathematicians, and educators
OCR for page 30
30
PRECOLLEGE SCIENCE AND MATHEMATICS TEACHERS
TABLE 1.7 Mathematics Achievement Comparisons: Eighth Grade, United States
and International, 1981-82 (Percentage of Items Correct)
United
States
(Percentage 25th
Correct) Percentile Median Percentile
Topic
International
(20 Countries)
-75th
Arithmetic 51 45 51 57
Algebra 43 39 43 50
Geometry 38 38 43 45
Statistics 57 52 57 60
Measurement 42 47 51 58
Source: McKnight et al. (1987:21~.
consider most important" (Murnane and Raizen, 1988:63~. It is not clear
what the relative standing of U.S. students would be on a test that assessed
higher-order thinking skills more fully.
Despite all the caveats that have been and can be made with regard
to these comparisons, the evidence is that U.S. high school students cannot
be judged to perform well in science or mathematics by any reasonable
standard, or at least not as well as society seems to expect.
Evidence from TEA that young people who concentrate heavily in
science and mathematics do not perform especially well implies even worse
outcomes for the great majority of American youth who take very little
science and mathematics in high school. From the perspective of employers,
for example, what matters at least as much as the quality of instruction for
high school students who are potential scientists and engineers is the quality
of technical or quantitative training for the great majority of high school
students who will not go on to these types of careers but will enter the
work force after graduation. Concern over the ability of young people to
function effectively in today's technical environment, given the inadequacy
and often the total absence of science and mathematics training with any
degree of rigor, looms as a major societal concern and is the subject of
numerous recent reports.
OCR for page 31
INTRODUCTION
31
Both low test scores and the generally low level of scientific literacy
underpin the concern with the quality of science and mathematics training,
and with the prospective shortage of qualified science and mathematics
teachers. Poor outcomes have thus spurred a deep concern with the quality
of teaching and the qualifications of teachers of science and mathematics.
Since it is through adjustments in quality that the supply and demand for
precollege science and mathematics teachers reach equilibrium in the short
run e.g., the next school year an examination of possible statistics to
measure quality is of central concern to the panel.
At least two different sets of factors are relevant to an assessment
of teaching quality. One set relates to the teaching environment and
includes school, district, and state policies and practices that enhance or
impede one's ability to secure the right teaching assignment and to teach
effectively. Such factors include time spent on science and mathematics,
teaching burden, textbook use, district decisions about recruiting and hiring
teachers, and inse~vice education policies.
Another set of factors relates to the background and qualifications of
the individual teacher. These include type of certification, relevant courses
taken in the past and currently, and measures of cognitive ability. The need
for better data on these kinds of factors, both for monitoring supply and
demand and for modeling purposes, is discussed in Chapter 5.
It should be kept in mind that even if all the comparison data were valid
and indicated that U.S. students have low absolute and relative achievement
in science and mathematics, it would not necessarily follow that the problem
lies solely or even mainly with the training of U.S. teachers of precollege
science and mathematics. Educational outcomes are a complex function of
student and family inputs, teaching inputs, educational curricula, school and
community environment factors, and student behaviors, including student
such as doing homework, attitudes toward science and mathematics, and
scientific habits such as objectivity, skepticism, and replication of results
(Murnane and Raizen, 1988~. Poor outcomes can clearly be due in part
to the inadequate training of teachers, but they can also be due to factors
that have little or nothing to do with the training and ability of the teacher
corps.
For example, there has been a continuing dispute among mathematics
teachers about curricular issues, which are seen by some as having a strong
influence on the level of performance of U.S. students in standardized tests
of mathematics skills. It Is alleged that mathematics skills in U.S. schools
are typically taught in a layered or "spiral" curriculum, whereby students
are taught a number of concepts in grade t, and are then taught slightly
augmented but basically similar concepts in grades t+l, t+2, .... It is
argued that students are thus introduced to relatively little new material
each year through grade 8; that most of what is done constitutes review
OCR for page 32
32
PRECOLLEGE SCIENCE AND MATHE3L4 TICS TEACHERS
of materials previously taught, and that as a result students become bored
with the constant repetition and never really master many of the key ideas
involved in the development of mathematical skills. It is also judged by
people who hold this view that part of the problem is that mathematics
textbook producers try to widen the appeal of their product to as many
school systems as possible; they end up including small segments on a varieW
of topics and intensive treatment of few, if any, of these topics. Since the
basic text is the primary resource used by most precollege mathematics
teachers (Weiss, 1987:31, 39), and since the text usually favors breadth and
facts over depth (Office of Technology Assessment, 1988:30-34), the result
is that a significant fraction of students master few if any of the topics.
A different line of argument, which could in principle be resolved
more easily and might make a substantial difference to outcomes, is that
U.S. students have inadequate skills in science and mathematics simply
because teachers, especially elementary school teachers, do not spend much
classroom time on science and mathematics topics. Research indicates a
great deal more time is devoted to reading than to mathematics (Cawelti
and Adkisson, 1985; Weiss, 1987~. Observations of actual classroom time
spent on mathematics also have found very large differences between
students in Minneapolis, Minnesota, and those in Taipei, Taiwan, or Sendai,
Japan (Stevenson et al., 1986~: U.S. students spend far less time on
mathematics than do Asian students.
To the extent that the performance of U.S. students on science and
mathematics tests and the level of their skill in these areas is simply due
to the emphasis on language arts found in U.S. classrooms and/or to the
smaller amount of time spent either in school or in school-related activities
at home, both the interpretation of the problem and the solution are
relatively simple provided school systems can be encouraged or induced
to change the structure of their curricula. But if that is the basic problem,
then the issue again is not one of inadequacy of preparation or academic
training on the part of teachers of science and mathematics in the United
States, but simply one of relative emphasis within the curriculum. In that
case, the question should be raised as to why fewer hours are spent on
science and mathematics in American classrooms.
Of course, it is possible that one reason U.S. students spend less time
on science and mathematics is that many U.S. elementary school teachers
are much less comfortable in teaching science and mathematics than in
teaching language arts, and that part of the reason for the curricular
emphasis is a preference on the part of teachers and/or administrators
derived in turn from their own training. It appears unlikely that specialist
teachers of mathematics in the early grades would be motivated to shorten
class time spent on mathematics, and use of such teachers is more common
in Japan, China, and Taiwan than in the United States in the early grades.
OCR for page 33
INTRODUCTION
33
There also appear to be differences in the nature of the pedagogical
training given to U.S. and Japanese mathematics teachers. The IEA math-
ematics study (McKnight et al., 1987) reported that among mathematics
teachers at the eighth- and twelfth-grade levels, the U.S. teachers had taken
more mathematics courses and fewer mathematics pedagogy courses than
their Japanese counterparts (p. 64~. American teachers also have much
less nonteaching time scheduled during the day, compared with their Asian
counterparts (Stevenson, 1987:32~. And the degree of teacher autonomy is
different: U.S. teachers are often on their own after the first year, while
in Asian classrooms younger teachers are typically under the tutelage of a
senior teacher for a number of years (Lee et al., 1987; Stevenson et al.,
1988; Stigler et al., 1987; Stevenson and Bartsch, in press).
Home environment is another factor that affects student outcome.
There is considerable evidence that learning and training for young children
take place in the home as well as in the school, and that the relative
importance of training in the home is much greater when children are
young. The home environments in which children are being raised in the
United States are considerably different now from the way they were several
decades ago. The proportion of children raised in single-parent households
is much larger now than in the past, and the proportion of mothers who
work full- or part-time is much higher now than in past decades. These
realities can create problems for children, especially for minority children,
many of whom are raised in single-parent households for a substantial
portion of their developmental years (Hill et al., 1987~. Although we
cannot be certain that the amount of time and attention parents pay to
young children's development is necessarily less because there are fewer
"parent hours" available in the aggregate, it is certainly plausible to suppose
that fewer total parent hours will result in fewer developmental hours spent
by parents on children. There is some evidence that working mothers
largely trade off leisure time and sleep for work hours, not for time spent
with their children (Hill and Stafford, 1985~. In any event, demographic
characteristics have a potentially serious influence on the process of skill
development in young children, and part of understanding educational
outcomes is surely to understand how these home environment factors
relate to these outcomes.
In addition to the demographic differences in home environments,
there also appear to be substantial differences in the practices, beliefs,
and expectations of parents in American households compared with those
in other countries. Again, the best-documented evidence comes from a
comparison of American and Asian households. As a generalization, Asian
mothers are less satisfied with the school performance of their children
than American mothers (despite the fact that their children are generally
doing better), they are more likely to attribute success in school to hard
OCR for page 34
34
PRECOLLEGE SCIENCE AND MATHEML4 TICS TEACHERS
work rather than to native ability, and they are less likely to be satisfied
with the way the schools are performing than their American counterparts
(Lee et al., 1987~.
The implication of the issues just discussed is not that the solution
to poor performance on standardized educational outcome tests, and pre-
sumptively in the level of skill development in science and mathematics
for American students, are to be found in factors other than either the
quantity of precollege science and mathematics teachers or the quality of
their training characteristics or classroom methods. Rather, it-is that poor
student outcomes are not uniquely correlated with, nor necessarily caused
by, inadequate quantity or quality, but could easily be due to factors that
are largely unrelated to teacher or teaching quality. It would thus be a
mistake, in the panel's view, to jump to the conclusion that poor science
and mathematics outcomes on the part of students necessarily reflect in-
adequacies in the background, training, or ability of their teachers and to
seek the remedy for the problem only by enhancing either the numbers
or the quality of precollege science and mathematics teachers. That could
turn out to be the case, but many other factors, such as the structure of the
curriculum, the practices of both K-12 school systems and teacher training
institutions, the amount of time spent on science and mathematics topics
in schools, and the influence of home environments on development out-
comes, all need to be understood before we can expect either to understand
the problem or to devise appropriate remedies.
TlIE PANEL'S WORK AND ORGANIZATION OF THE REPORT
During the course of its work, the panel broadened its understanding
of the flow of teachers through school systems by direct contact with 39
public school districts across the country. These school systems ranged
from the largest metropolitan systems to the most isolated small school
districts and represented a wide geographic range and a variety of labor
market conditions.
Six of the 39 districts were the subject of in-depth case studies, con-
ducted in 1987 and 1988, of supply and demand issues regarding science
and mathematics teachers. Two of the districts were in California and near
one another geographically: one was a large urban system whose ability
to attract talented science and mathematics teachers was affected by a
history of budgetary constraints and teacher-organization or school-district
provisions, while the other, a small, wealthy district, was able to exercise
greater autonomy in attracting and keeping talented teachers. Two other
districts one in California and one in Utah both with large enrollments,
were selected for their growing populations and rising enrollments. Hiring
OCR for page 35
INTRODUCTION
35
in one of them was severely limited by fiscal constraint as well as strong re-
ligious and community standards; the other was growing in both enrollment
and economic base. Finally, two contiguous school districts in Maryland
that were expected to hire from the same labor market were visited a
large urban district coping with school closures, leadership changes, and
traumatic layoffs as the student population has moved to the suburbs and
a medium-sized, stable, semirural school district nearby.
The in-depth case studies furnished invaluable context without which
statistics portraying supply and demand would be seriously incomplete.
Such context showed the role of the individual personnel administrator
and his or her ability to maneuver or use informal networks to attract
science and mathematics teachers. It showed the effects of competing
labor markets, teacher-organization provisions, budgetary constraints, and
other external factors.
The six in-depth case studies were supplemented by 27 additional mini
case studies, conducted by telephone interviews and follow-up question-
naires, in order to test the representativeness of the findings. The mini
case studies were conducted over the period June through December 1988.
Finally, a conference of the chief personnel administrators of seven
large metropolitan school districts, representing over 5 percent of the
nation's total public school enrollment, was convened in May 1988.2 Issues
of supply, demand, and quality of science and mathematics teachers were
discussed, and the districts' statistical information systems were examined
for data relevant to supply and demand models. Appendix A provides more
information about each of these activities. Discussions in the chapters that
follow frequently draw on the experiences of the school district personnel
administrators who participated in these studies.
In the chapters that follow, we further examine the characteristics of
demand for precollege science and mathematics teachers (Chapter 2) and
issues relating to supply (Chapters 3 and 4~. Chapter 3 reviews projection
and behavioral models and the essential behavioral components of effective
supply models. It examines individual incentives to teach and school district
actions that influence supply decisions and mesh supply with demand. In
Chapter 4, data needed to monitor the supply pool along its various stages
in the teaching career are discussed.
We then turn to the role of quality adjustments in bringing supply and
demand to equilibrium. In Chapter 5 we look at the question of measuring
teacher characteristics and teaching quality. Chapter 6 contains the panel's
conclusions and recommendations: some of the recommendations deal
with specific data needed to better understand demand, supply, or quality
2nle number of districts in the three case study activities add up to 40, with one of the large
school districts participating in two of the projects.
OCR for page 36
36
PRECOLLEGE SCIENCE AND MATHEAL4TICS TEACHERS
factors, and other recommendations deal with the types of research needed
to better understand the linkages among demand, supply, teacher quality,
and student outcomes and ways to facilitate this research.
Representative terms from entire chapter:
precollege science
