ERIC EJ795880: Teacher Labor Markets in Developing Countries

2007·0.1 MB·English

by  ERIC

11 5564-7 Vegas.qxp 1/15/2007 10:20 PM Page 219
Teacher Labor Markets
in Developing Countries
Emiliana Vegas
Summary
Emiliana Vegas surveys strategies used by the world’s developing countries to fill their class-
rooms with qualified teachers. With their low quality of education and wide gaps in student
outcomes, schools in developing countries strongly resemble hard-to-staff urban U.S. schools.
Their experience with reform may thus provide insights for U.S. policymakers.
Severe budget constraints and a lack of teacher training capacity have pushed developing na-
tions to try a wide variety of reforms, including using part-time or assistant teachers, experi-
menting with pay incentives, and using school-based management.
The strategy of hiring teachers with less than full credentials has had mixed results. One suc-
cessful program in India hired young women who lacked teaching certificates to teach basic lit-
eracy and numeracy skills to children whose skills were seriously lagging. After two years, stu-
dent learning increased, with the highest gains among the least able students.
As in the United States, says Vegas, teaching quality and student achievement in the developing
world are sensitive to teacher compensation. As average teacher salaries in Chile more than
doubled over the past decade, higher-quality students entered teacher education programs.
And when Brazil increased educational funding and distributed resources more equitably,
school enrollment increased and the gap in student test scores narrowed. Experiments with
performance-based pay have had mixed results. In Bolivia a bonus for teaching in rural areas
failed to produce higher-quality teachers. And in Mexico a system to reward teachers for im-
proved student outcomes failed to change teacher performance. But Vegas explains that the de-
sign of teacher incentives is critical. Effective incentive schemes must be tightly coupled with
desired behaviors and generous enough to give teachers a reason to make the extra effort.
School-based management reforms give decisionmaking authority to the schools. Such reforms
in Central America have reduced teacher absenteeism, increased teacher work hours, in-
creased homework assignments, and improved parent-teacher relationships. These changes,
says Vegas, are especially promising in schools where educational quality is low.
www.futureofchildren.org
Emiliana Vegas is senior education economist at the World Bank. Rekha Balu provided excellent research assistance.
VOL. 17 / NO. 1 / SPRING 2007 219
11 5564-7 Vegas.qxp 1/15/2007 10:20 PM Page 220
Emiliana Vegas
D
eveloping countries in teacher hiring and firing, directly to schools.
Africa, Asia, and Latin The results in terms of student achievement
America are struggling, just vary widely, depending on the context and
as the world’s industrialized the country.
countries are, to fill class-
rooms with qualified teachers.1 But the chal- Teacher Labor Markets
lenges they face are even more complicated. The supply of teachers in developing coun-
Demographers have projected that develop- tries, as in developed countries, depends on
ing countries have the fastest-growing popu- working conditions and teacher salaries, as
lations of people aged six to twenty-four in well as on how salaries and entry require-
the world.2 The swelling ranks of school-age ments in the teacher labor market compare
populations are driving up demand for teach- with other labor markets. Many teachers
ers. In accordance with the Millennium De- work in schools that lack adequate teaching
velopment Goals set forth by the United Na- materials or basic infrastructure. Pupil-
tions, every country must ensure universal teacher ratios, as shown in table 1, can be
primary education by 2015.3 Although the large: an average of 43:1, for example, in sub-
majority of children in all regions of the Saharan Africa, though in some countries the
world except sub-Saharan Africa attend pri- ratio is even larger. Many teachers in devel-
mary school, the quality of education is low oping countries cite lack of resources, such as
and disparities in student learning outcomes adequate facilities, textbooks, and teaching
are large.4 Children in developing countries materials, as a primary obstacle to effective
have the lowest mean test scores in interna- teaching.6 Location also affects teacher sup-
tional assessments of student learning, and ply. In most developing countries, unlike in
they often show the largest variation in test the United States, working conditions tend to
scores as well.5 The severe challenges facing be better in urban schools and teachers pre-
the developing world are not unique, how- fer to work there.
ever. In many ways, in fact, they resemble
those facing the U.S. schools with the lowest- Large cities in the United States have only
income student populations. Could strategies recently begun using housing subsidies to re-
used by developing countries offer lessons to cruit teachers to difficult-to-staff urban
policymakers in the United States seeking to schools, but developing countries have long
improve their nation’s lowest-performing made use of housing incentives, especially for
schools? teachers in rural schools. In many poor coun-
tries, however, these subsidies have not been
Budget constraints and a lack of teacher effective, in part because most teachers are
training capacity have led developing nations women and most single women choose not to
to try a wide variety of reforms. Some are hir- live alone or to transfer to rural areas for
ing part-time, contractual, or assistant teach- safety-related reasons.7
ers. Others are using pay incentives to attract
and retain qualified teachers. Still others are Recent research in Pakistan, however, sug-
trying to attract more teachers and raise the gests that placing secondary schools in rural
quality of teaching by experimenting with areas—rather than the urban areas where
school-based management or the devolution they are now concentrated—may attract some
of decisionmaking authority, including teachers, as many female secondary graduates
220 THE FUTURE OF CHILDREN
11 5564-7 Vegas.qxp 1/15/2007 10:20 PM Page 221
Teacher Labor Markets in Developing Countries
Table 1. Teacher Characteristics, by Region
Teacher salary (percent
Pupil-teacher ratio Trained teachers (percent) of GDP per capita)
Region Primary Secondary Primary Secondary Secondary
Sub-Saharan Africa 43:1 24:1 69 78a 6.7
Middle East/North Africa 23:1 18:1 96 85b
Latin America and the Caribbean 26:1 19:1 87 77 1.4
South Asia 42:1 33:1 62a . . . . . .
East Asia 22:1 19:1 96 71a . . .
Eastern Europe and Central Asia 17:1 12:1 93a . . . . . .
OECD 16:1 14:1 . . . . . . 1.3
Source: UNESCO Institute for Statistics, Global Education Digest 2003—Comparing Education Statistics across the World (Montreal,
Canada: 2002). Countries with populations of less than 1 million are excluded.
a. Data are based on 10–25 ppercent of the total population of the country group or region.
b. Data are based on 25–50 ercent of the total population of the country group or region.
who aspire to teach higher grades can then ees, but much of this difference could be at-
teach in their native villages instead of moving tributed to the teachers’ higher levels of
to the cities.8 The strategy of recruiting local schooling. One study demonstrated that al-
teachers and assigning them to schools close though entry-level wages for teachers are
to home may also be effective in the United low, teachers are compensated as well as
States, where, as researchers have shown, other professionals who work in similar loca-
teacher labor markets are mostly local.9 tions or have similar levels of education and
experience.12 And in Bolivia, a study found
Turning to compensation, in Latin America, that the concentration of teachers in the pub-
at least, teachers do not appear to be severely lic sector and the influence of union-negoti-
underpaid compared with similar workers in ated contracts on teacher wages for the entire
other occupations.10 Lucrecia Santibáñez ex- country reduced geographic variation in
amined urban professional salaries in Mexico teacher salaries, which meant that teachers in
during the late 1990s.11 Controlling for edu- rural areas, in particular, were better com-
cation, experience, and hours worked, she es- pensated than other professional workers in
timated that the hourly wage premium in similar locations. In addition, union influence
1998 was 13 percent for male secondary set teachers’ wages in a way that minimized
teachers and 30 percent for female teachers. the salary differences by gender, ethnicity,
She also analyzed salary differences among and marital status that are apparent for some
states in Mexico and found that, on average, private sector professionals subject to market
teachers in more developed northern states wages.13
earned relatively less compared with other
professions than did teachers in the rest of In many developing countries, teacher
the country. salaries make up a large share of total public
education spending—as much as 95 percent
In Chile, teachers’ wages were higher, on av- of total education costs. Governments in
erage, than those of nonagricultural employ- countries such as Uganda, Kenya, and Tanza-
VOL. 17 / NO. 1 / SPRING 2007 221
11 5564-7 Vegas.qxp 1/15/2007 10:20 PM Page 222
Emiliana Vegas
nia, which have recently expanded access to study also found that professionalizing
primary schools, cite high spending on teacher training by elevating its status to a
teacher salaries as the biggest constraint on university degree had the paradoxical effect
improving the supply and quality of teacher of causing qualified teachers to move to more
recruits. Education advocates have suggested remunerative professions.15
a variety of strategies to minimize this con-
straint, such as de-linking teacher salaries In addition, a UNICEF strategy paper warns
from civil service salaries or changing the of the cost consequences of expanding alter-
pace at which teachers progress along the native teacher hiring. Even if alternative
salary scale. But research has centered pri- hires start at lower salaries or with fewer ben-
marily on subsidies and incentives, such as efits, the study finds, they will eventually de-
merit-based pay and supplementary al- mand or qualify for higher salaries. Salary in-
lowances (housing, transport), as discussed creases for a large pool of teachers can
below. financially strain the system over the long
term as much as hiring regular teachers
Alternatives to Hiring would.16
Regular Teachers
Many developing countries are upgrading the Contract Teachers
training credentials required of teachers at Many developing countries are addressing
the same time as their governments are re- shortages by turning to contract teachers—
sorting to hiring part-time, uncredentialed, or graduates of regular teacher training insti-
contract teachers to meet demand or cut tutes who receive lower wages than do regu-
costs. For example, in the Kyrgyz Republic, lar teachers (just 40 percent of civil salaries)
policymakers emphasized teacher training re- and no benefits. Togo recently reported that
quirements even as teacher training colleges as much as 55 percent of its teaching force
were being closed for lack of funding. One was contractual. The strategy, though, does
approach, which Tajikistan has tried, has been not appear to have been successful, as the ad-
to shorten the length of teacher training pro- vent of contractual hiring in Togo reduced
grams.14 To limit expenses while responding the supply of high-quality candidates, while
to increased demand in the 1990s, India hired also raising absenteeism and creating resent-
more than 200,000 “para-teachers,” while ment over unfair pay. A retrospective evalua-
Pakistan hired contract teachers who were ex- tion found that the performance of students
cluded from training and other benefits. taught by contractual teachers lagged behind
that of students taught by regular teachers,
As in the United States, research findings on even after controlling for prior achievement,
how teacher training programs affect the household characteristics, and school, class-
quality of education are, at best, inconclusive. room, and teacher variables.17 Not surpris-
A survey of case studies in Latin America ingly, schools whose limited budgets forced
found that several different methods of them to hire contractual teachers also had
teacher preparation and training—for exam- less pedagogic supervision and poor facilities.
ple, stopgap training that covers only missing
skills and competences—achieved consis- Assistant Teachers
tently poor results, thus sounding a caution- Some countries are also experimenting with
ary note for U.S. education programs. The hiring assistant teachers, who often have
222 THE FUTURE OF CHILDREN
11 5564-7 Vegas.qxp 1/15/2007 10:20 PM Page 223
Teacher Labor Markets in Developing Countries
fewer qualifications than do regular teachers ship programs have influenced teacher re-
and are paid at substantially lower rates. In cruitment, quality, and retention in develop-
rural areas in India, a remedial education ing countries.
program reached more than 15,000 students
who had not attained basic literacy and nu- Salary Levels
meracy skills by third grade by hiring as During the 1990s in Chile, teachers’ real
teachers young women from the community wages increased and the quality of applicants
who lacked teaching certificates. A random- to the teaching profession improved. Be-
ized evaluation found that after two years, tween 1990 and 2002 real salaries grew 156
the program had increased student learning
by 0.39 standard deviation, with the highest
gains among the least able students. This Even if alternative hires
finding suggests that it is possible not only to
start at lower salaries or
keep poorly performing students in school
but also to ensure that they do not fall be- with fewer benefits, the
hind. It is conceivable that a similar catch-up
study finds, they will
program could help U.S. cities maintain high
net enrollment rates in the grades when stu- eventually demand or
dents are most likely to drop out. Because
qualify for higher salaries.
the program in India hires local high-school-
educated girls to teach classes of approxi-
mately twenty students, the average cost is
less than $5 a year for each child—far less percent, while the government launched a
than the average cost of outfitting classrooms publicity campaign to encourage college stu-
with regular teachers.18 While these findings dents to become teachers and also created a
are compelling, the validity of the evaluation scholarship program for outstanding students
has been called into question on several to study pedagogy. Simultaneously, the gov-
points and it was also found that the chil- ernment allocated substantial additional re-
dren’s gains began to fade out within one sources to schools, in the process improving
year of leaving the program. overall working conditions for teachers. Al-
though the individual effect of each of these
Teacher Pay reforms on student outcomes remains un-
Although some disagreement exists about clear, during the period the number of
the importance of the absolute level of teacher education applicants increased 39
teacher salaries in attracting qualified peo- percent, and the average university entrance
ple to and retaining them in the profession, exam score of applicants to teacher education
there is broad consensus that teacher programs increased 16 percent. Even though
salaries influence the type of people who the number of applicants to other degree
enter the field and how long they remain in programs, such as engineering, also in-
it. At the same time, research indicates that creased, the average exam scores of these ap-
working conditions and regulations can plicants remained more or less constant.
counteract or amplify the influence of wages These patterns suggest that changes in salary
on teachers. In this section I describe how level can affect an individual’s choice to be-
salary levels, salary structures, and scholar- come a teacher.19
VOL. 17 / NO. 1 / SPRING 2007 223
11 5564-7 Vegas.qxp 1/15/2007 10:20 PM Page 224
Emiliana Vegas
Some evidence also suggests that salary levels pupil ratios; the study inferred, because there
and salary equalization for teachers can im- was no decrease in enrollment, that the gov-
prove student outcomes. In Brazil, a finance ernments hired new teachers.20 The share of
equalization reform that targeted redistrib- teachers who had completed only primary ed-
uted funds to teachers resulted in smaller ucation also fell dramatically, most noticeably
class sizes, fewer overaged children in pri- in Brazil’s poorer regions and in the earlier
mary and secondary schools, and a diminish- primary school grades, where higher shares of
ing gap between high- and low-performing teachers had previously been underqualified.
students. Brazil, like the United States, is a That the reform was introduced at about the
vast country characterized by large inequali- same time as legislation requiring teachers to
ties in educational spending and educational have at least a secondary education degree
outcomes not only among states but also complicates any assessment of the results. But
among different municipalities within each the 2005 study found that funds received
state. The Fundo de Manutenção e Desen- from FUNDEF were not significantly linked
volvimento do Ensino Fundamental e de Val- with the steep decline in underqualified
orização do Magistério (Fund for the Mainte- teachers, though FUNDEF revenue was used
nance and Development of Basic Education to train and educate teachers.21
and Teacher Appreciation, or FUNDEF) is a
federal fund that addresses spending inequal- The FUNDEF-related changes in educa-
ities within states. State and municipal gov- tional inputs have, in turn, generated changes
ernments contribute a share of their tax and in student outcomes. More students are now
transfer revenues to the fund, which then re- attending school in the poorer states of
distributes revenues to the state and munici- Brazil, particularly in the higher grades of
pal governments in each state on the basis of basic education. The reform is also linked
the number of students enrolled in their with lower levels of overaged students in the
basic education systems. The federal govern- classroom. Having qualified teachers thus ap-
ment also promotes funding adequacy across pears to help students stay on track in school,
all states by providing supplemental funding repeat grades less often, drop out and reenter
in states where FUNDEF revenues per stu- less often, and perhaps also enter first grade
dent are below a yearly established spending on time. Because low-performing students
floor. These “top-ups,” which have benefited suffer most from inequalities in per-pupil
the poorer states of Brazil, located primarily spending, finance equalization reforms that
in the Northeast, point to the importance of decrease these spending inequalities may
additional federal financing when state and also narrow the performance gap between
local revenues fall short. high-performing and low-performing stu-
dents and between white and nonwhite stu-
Unlike teacher incentive programs in several dents. Studies of school finance reform in the
states in the United States, FUNDEF ear- United States, however, have not shown con-
marks 60 percent of funds specifically for sistent effects on student outcomes.22
teachers, with funds going to hire new teach-
ers, train underqualified teachers, and in- Salary Structure
crease teachers’ salaries. A 2005 study found Although teacher pay in developing countries
that governments that increased mandated is seldom linked to teacher performance, a
per-pupil spending lowered average teacher- few countries have recently experimented
224 THE FUTURE OF CHILDREN
11 5564-7 Vegas.qxp 1/15/2007 10:20 PM Page 225
Teacher Labor Markets in Developing Countries
with performance-based pay to raise teaching board wage increase for “good” teachers and
quality and student outcomes (see the article may thus be expected to have led to an in-
in this volume on performance-based pay by crease in the quality of entering cohorts of
Victor Lavy). Because few large-scale pay- teachers in the past decade.
for-performance programs have been imple-
mented in the United States, programs in de- A study by Patrick McEwan and Lucrecia
veloping countries provide particularly Santibáñez examined how effective the Car-
valuable evidence of the extent to which in- rera Magisterial incentives were in improving
centives affect performance. students’ test scores.24The study compared a
The effect of performance-based pay de-
pends critically on how it is designed and Although teacher pay in
linked to teacher performance. Chile and
developing countries is
Mexico, for example, have instituted differ-
ent types of performance-based incentives seldom linked to teacher
for teachers. In Chile’s Sistema Nacional de
performance, a few countries
Evaluación de Desempeño de los Establec-
imientos Educacionales (National System of have recently experimented
School Performance Assessment, or SNED),
with performance-based pay
top-performing schools within predeter-
mined groups earn a financial bonus for stu- to raise teaching quality and
dent performance; the bonus is distributed
student outcomes.
among the teachers in the winning schools.
Initially Chile’s school-based bonus had no
effect on student performance, but a recent
study found that in schools that have some group of teachers who had participated in the
likelihood of receiving the prize in each of program but whose characteristics put them
the three years they apply, average student far below or above the threshold for a bonus
test scores increase slightly.23 payment with a small group of teachers who
were close to, but not assured of, receiving
Mexico’s Carrera Magisterial (Master Train- the bonus. The study found that the mean
ing, or CM) program, instituted in 1993, al- test scores of students of teachers in the lat-
lows teachers to move up consecutive pay ter “incentivized” group rose by a small to
levels based on year-long assessments of their moderate amount, roughly 0.15–0.20 points
professional development and education, (less than 10 percent of a standard deviation),
years of experience, a peer review, and, im- relative to teachers without the incentive.
portantly, their students’ performance. The The effect was robust to a variety of alterna-
awards are substantial—they can represent tive specifications and subsamples.
between 25 and 200 percent of the teacher’s
annualwage—and last throughout a teacher’s Although Mexico’s Carrera Magisterial and
career, just as a salary increase does. Since Chile’s SNED are both nationwide programs
1993, more than 600,000 teachers have re- involving most of the country’s teachers, only
ceived the lowest level of award. The Carrera a minority of teachers has any real likelihood
Magisterial reform resembles an across-the- of receiving a promotion (in the case of Car-
VOL. 17 / NO. 1 / SPRING 2007 225
11 5564-7 Vegas.qxp 1/15/2007 10:20 PM Page 226
Emiliana Vegas
rera Magisterial) or a bonus (in the case of Figure 1. Decomposition of Teacher Pay in
SNED) each time they apply.25 Thus, most Chile and Bolivia
teachers who apply have no real incentive to Percent
improve performance. To be effective, as Vic-
90
tor Lavy explains in his article in this volume, 80 Base pay/years
of service
an incentive scheme must give all or most
70 Education
teachers a reason to exert extra effort. 60 Difficult
conditions
50
Administration
The size of the reward relative to a teacher’s 40 Individual
incentives
base pay also matters. When a teacher’s base 30
Master
salary accounts for a large share of total com- 20 teacher
SNED
pensation, incentives for specific behaviors, 10
such as working in rural schools or serving 0
children with special needs, will be relatively Chile Bolivia
less powerful. Figure 1 portrays the share of Sources: Cristián Cox, “Las políticas educacionales de Chile en las
teacher pay that comes from education and últimas dos décadas del siglo XX,” in Políticas educacionales en el
cambio de siglo: la reforma del sistema escolar en Chile, edited by
training, years of service, and performance in Cristián Cox (Santiago, Chile: Editorial Universitaria, 2003); and
Miguel Urquiola and Emiliana Vegas, “Arbitrary Variation in Teacher
two Latin American countries, Chile and Bo-
Salaries,” in Incentives to Improve Teaching: Lessons from Latin
livia. In Bolivia’s pay structure—one common America,edited by Emiliana Vegas (Washington: World Bank Press,
2005).
in both developing and developed coun-
tries—by far the largest part of a teacher’s
salary depends on experience and education. schools served as comparison schools. Teach-
Chile has tried to increase the share of ers were given a camera with a tamper-proof
teacher pay that is related to performance, date and time function, along with instruc-
but even there more than 60 percent of pay tions to have one of the children photograph
continues to depend on characteristics, such the teacher and other students at the begin-
as years of service and education, that are un- ning and end of the school day. The time and
related to performance. The mixed findings date stamps on the photographs were used to
on the effectiveness of performance-based track teacher attendance. Salary was a direct
pay in Mexico and Chile echo findings from function of attendance.
the United States (again, see the article by
Victor Lavy in this volume) that make clear An evaluation of the program by Esther
the difficulty of designing an effective per- Duflo and Rema Hanna reported that it im-
formance-based pay policy. mediately reduced teacher absenteeism.26
The absenteeism rate, measured using unan-
Some incentive programs, however, have nounced visits in all 120 schools, averaged 42
shown some success. A nongovernmental or- percent in the comparison schools and 22
ganization (NGO) project in India used a percent in the schools under study. When the
simple financial incentive program to reduce schools were in session, teachers were as
teacher absenteeism and to stimulate teach- likely to be teaching in both types of schools;
ing and better learning. The NGO initiated the number of children present was roughly
the program in 60 informal one-teacher the same. The program also improved stu-
schools in rural India, randomly chosen out dent achievement. A year after its start, test
of a sample of 120 schools; the remaining 60 scores in the schools participating in the in-
226 THE FUTURE OF CHILDREN
11 5564-7 Vegas.qxp 1/15/2007 10:20 PM Page 227
Teacher Labor Markets in Developing Countries
centive program were 0.17 standard devia- wan to explain why the country’s teacher re-
tion higher than those in the comparison tention rate remains high beyond the five-
schools, and children were 40 percent more year threshold.29 Such programs imply that
likely to be admitted into regular schools. strong relative wages and subsidized costs
can have an important effect on the quality of
The study by Duflo and Hanna demonstrates teacher supply.
how random assignment studies can be used
to learn about program effects, though the School-Based Management
results may be specific to the context in Some developing countries have tried de-
which they are implemented. The scheme in volving directly to schools the authority to
India was tested in a small group of one- make decisions regarding teacher hiring and
teacher schools; the question is whether the other administrative matters that are usually
results would be similar in different contexts. made by local, regional, or central govern-
For example, would the teacher attendance ments. The idea behind such decentralization
effect be smaller in regular public schools or is to bring these decisions closer to the
in larger schools with many other teachers school, and thus to parents and students, to
who can substitute for an absent teacher? generate incentives and conditions to im-
Another issue of practical importance for prove teaching quality and student outcomes
public policy is whether the camera mecha- and make teachers and schools more ac-
nism would share the fate of other already ex- countable to the community.
isting formal mechanisms for punishing ab-
sentees: weak enforcement. Several countries in Central America have in-
troduced such school-based management re-
Scholarship Programs forms. In El Salvador, a retrospective evalua-
To attract talented students to teaching, sev- tion found that the Programa de Educación
eral countries in South America’s Southern con Participación de la Comunidad (Educa-
Cone have introduced scholarship programs. tion with Community Participation Program,
In Chile, a scholarship program for talented or EDUCO) has affected management prac-
students covers 100 percent of tuition up to 1 tices, teacher behavior, and student out-
million pesos in exchange for a commitment comes.30 A few important powers, most no-
to teach for three years. Priority goes to can- tably the ability to hire and fire teachers,
didates in natural sciences, mathematics, have been transferred to the school, but
English, language arts, and basic education.27 many other decisions continue to be made
In Uruguay, where teacher education is free, primarily by central authorities. Most of the
scholarships are provided to talented candi- local decisionmaking power has been given to
dates from disadvantaged backgrounds to parents rather than principals. The study also
cover their living expenses during the three finds important behavioral differences:
years of intensive training at regional teacher EDUCO schools have fewer school closings,
training centers.28 In Asia, during the 1990s less teacher absenteeism, more meetings be-
Taiwan offered free pre-service education to tween teachers and parents, and longer
those who had taught for five years. Re- teacher work hours than control schools.
searchers cite the combination of the scholar- These changes in teacher behavior, in turn,
ship and generous salaries and benefits for are related to higher achievement in Spanish
teachers relative to other professions in Tai- in EDUCO schools.
VOL. 17 / NO. 1 / SPRING 2007 227
11 5564-7 Vegas.qxp 1/15/2007 10:20 PM Page 228
Emiliana Vegas
Another retrospective evaluation finds simi- In contrast to PROHECO and EDUCO,
lar effects in Honduras’s Proyecto Hon- Nicaragua’s School Autonomy program (Au-
dureño de Educación Comunitaria (Hon- tonomía Escolar) was aimed initially at urban
duran Community Education Project, or secondary schools, in particular those with
PROHECO).31 Like EDUCO, PROHECO higher-than-average resources. Unlike their
is a school-based management reform for peers in neighboring El Salvador and Hon-
rural primary schools. Comparing PRO- duras, parent associations and teachers in
HECO schools to similar schools in rural Nicaragua’s autonomous schools report little
areas (using propensity score matching meth- decisionmaking power. A decade after the re-
ods to construct a credible comparison form began, autonomous and nonau-
group), the study finds that PROHECO tonomous schools continue to differ in much
teachers are less frequently absent because of the same ways as before reform. Differences
union participation, although they are more in student socioeconomic background con-
frequently absent because of teacher profes- tinue to explain most differences in student
sional development. Teachers in PROHECO achievement. The reform appears to have
are paid less and have fewer years of experi- had no systematic effect on student learning.
ence than comparison teachers. And, as in El Although on average students in autonomous
Salvador’s EDUCO program, teachers in schools outscore students in traditional
PROHECO teach more hours in an average schools in mathematics in third grade, by
week than comparison teachers; they also sixth grade they score lower on both Spanish
have smaller classes and assign more home- and mathematics tests. There is little evi-
work. In these examples, at least, decentral- dence that differences between autonomous
ized schools appear to encourage greater effi- and traditional schools are responsible for
ciency and teacher effort. these differences in test scores.32
Although the studies found little evidence Some Lessons for
that teachers in community-managed schools the United States
differ from their colleagues in conventional Although developing countries differ in many
schools in terms of their classroom processes, ways from the United States, the inequality
planning, or motivation, PROHECO stu- and poverty in some of their schools closely
dents score higher on math, science, and resembles conditions in some hard-to-staff
Spanish exams than students in similar non- U.S. schools. Because of their widely varying
PROHECO schools. This higher student circumstances, these countries have tried
achievement is, in part, explained by unique many and varied reforms, often on a large
qualities and characteristics of PROHECO scale. Their experiences with reform may
schools. Specifically, the more hours a week a provide insights for U.S. policymakers.
teacher works, the higher is the mean student
achievement in all three subjects. The fre- Clearly, educational reforms of many kinds
quency of homework is associated with can affect teaching quality and student learn-
higher achievement in Spanish and math. Fi- ing. Research evidence supports the intuitive
nally, smaller classes and fewer school clos- notion that teaching quality and student
ings are related to higher student achieve- achievement are sensitive to the level and
ment in science. structure of teacher compensation. For ex-
ample, as average teacher salaries in Chile
228 THE FUTURE OF CHILDREN