Education in States and Nations: Introduction

Education in States and Nations: 1991

(ESN) Introduction and Overview

In 1983, when A Nation at Risk highlighted both the state of American education and its essential role in our nation's prosperity, the report's first piece of evidence was international comparisons of mathematics and science achievement. It appeared then that U.S. students were being outperformed by students in other countries, including some countries that educate their students at lower cost. This report from an independent commission appointed by the Secretary of Education suggested that, at a time when a nation's power and prosperity were more than ever before determined by the collective brain power of its citizenry, the U.S. education system seemed not to be performing as well as it could. (1), (2)

A few years later, in 1986, the National Governors' Association issued A Time for Results, a report similar to A Nation at Risk in tone, in the nature of its evidence, and in its recommendations. A Time for Results asserted even more strongly than A Nation at Risk that global economic competition meant that the most appropriate benchmarks for education system performance were now global as well. This report by a national association of state governors was at once an assertion that education was a national concern, and that it was still primarily a state and local responsibility.(3)

Since publication of A Time for Results, Americans have seen much activity on education policy at the interstices of authority between the separate branches and levels of government. The Federal government and the nation's governors joined their efforts formally at the Charlottesville, Virginia education summit in 1989; and the subsequently-formed National Education Goals Panel and National Council on Education Standards and Testing both included members from the Congress, the White House, the U.S. Department of Education, and the ranks of governors and state legislators. Agreement on six National Education Goals followed the Charlottesville summit. In 1994, Congress added two additional goals related to parental involvement and teacher professional development.

A commitment to reaching world-class education performance levels is explicitly expressed in National Education Goals 5 and 6. Goal 5 declares that U.S. students will be first in the world in science and mathematics achievement by the year 2000. Goal 6 asserts that every adult American will possess the knowledge and skills necessary to compete in a global economy.(4)

By joining efforts with the Federal government, the governors did not intend to share the management of the public schools. However, they did agree that the Federal government had an important role to play in the collection and dissemination of comparative data needed to manage the quality of American education.

In 1988, the U.S. Congress authorized the establishment of a Special Study Panel on Education Indicators for the U.S. Department of Education's (NCES). This panel was chartered in July 1989 and directed to prepare a report, published in 1991, Education Counts: An Indicator System to Monitor the Nation's Educational Health. The Panel's report recommended a variety of ways in which NCES'should increase its collection and presentation of indicator data. Among the many recommendations, the report urged NCES to: strengthen its national role in data collection and provide technical assistance to the states; improve its capacity to collect international data; and develop a mixed model of indicators - international and national indicators, state and local indicators, and a subset of indicators held in common.

Two of NCES's primary indicators projects include The Condition of Education and the National Assessment of Educational Progress (NAEP).(5) The Condition is an annual compendium of statistical information on American education, including trends over time, international country comparisons, and some comparisons among various groups (by sex, ethnicity, socioeconomic status, and others). However, the Condition contains very few state-by-state comparisons.

The National Assessment of Educational Progress (NAEP) is a congressionally-mandated assessment of the academic achievement of American students. Begun in the late 1960s, NAEP has been reporting assessment results state-by-state, on a trial basis, only since 1990. In that year, 37 states, the District of Columbia, and 2 territories participated in a Trial State Assessment program in eighth- grade mathematics. In the 1992 Trial State Assessments in 4th-grade reading and mathematics and 8th-grade mathematics, voluntary participation increased to 41 states, the District of Columbia, and 2 territories. The same number of jurisdictions participated in the 1994 Trial State Assessment of fourth grade reading.

At the same time that U.S. officials began looking outside our borders for education policy lessons and performance benchmarks, officials in other countries were doing likewise. The Organization for Economic Cooperation and Development (OECD), which had for years published indicators on macroeconomics, trade, industry, and agriculture, began an effort in the 1980s to develop and collect social indicators, starting with health care. Turning its attention next to education, the organization launched, in 1987, the Indicators of Education Systems project (INES) in its Center for Educational Research and Innovation (CERI).(6) CERI organized several international groups of experts to develop conceptual frameworks, to agree on definitions, and to execute pilot studies to determine the set of possible indicators that best illustrated the condition of education in the OECD countries. In 1992, the OECD published a set of indicators, employing data from the late 1980s, in Education at a Glance.(7) An updated second edition, Education at a Glance (Edition 1993), was published in December 1993, and a third edition was released in January 1995.(8)

The first edition of Education in States and Nations: Indicators Comparing U.S. States with the OECD Countries in 1988, produced in 1993, served as a logical next step and a U.S. companion volume to Education at a Glance, incorporating U.S. state-level data from the late 1980s. It not only allowed state-to-state and country-to-country comparisons, but state-to-country comparisons as well. For perhaps the first time, states could compare their support for education, the participation of their youth in the education system, or their educational outcomes with those of a number of industrialized countries, including some quite similar in size or wealth. In other words, on a variety of measures, education in U.S. states could now be compared internationally.

Why compare states to nations? In many countries, public responsibility for education is vested in the national government, in an education ministry (9) In the United States, however, public responsibility for education rests primarily at the state level. (3a) In 1992, state-level governments provided 46 percent of revenues for public elementary and secondary schools. This share of contribution ranged from 8 percent in New Hampshire to 90 percent in Hawaii. In many cases, the most valid American counterparts to other countries national ministries of education are our state education departments.

This edition, Education in States and Nations: Indicators Comparing U.S. States with Other Industrialized Countries in 1991, is much larger than its predecessor. This reflects both a greater availability of suitable international indicators and a greater effort to find relevant indicators, both domestic and international. The large size of this volume was not a goal in itself, but is coincident to others. Education in States and Nations/1991 has two goals:

To improve the quality of indicators, where possible, with better data; and
To expand the domain of indicators to encompass more topics pertinent to education policy.

With the addition of more topics and more and better sources of data, this second edition of Education in States and Nations offers more depth and breadth than did its predecessor.

The Content of Education in States and Nations/1991

Education in States and Nations/1991Education at a Glance (Edition 1993), from the International Assessment of Educational Progress (IAEP), and from several other contemporary sources of international education indicators. International indicators were selected for use in Education in States and Nations/1991 if they were relevant to states and if comparable state-level data on the indicators existed. The indicators are grouped into six categories:

Background;
Participation;
Processes and Institutions;
Achievement and Attainment;
Labor Market Outcomes; and
Finance.

Indicators were selected in an attempt to cover the domain of the educational enterprise. The background and finance indicators could be described as "stocks" or "input" measures. Both of these groups of indicators are richly represented, with background indicators relating to geographic, demographic, economic, and sociological factors, and with finance indicators presenting revenues and expenditures viewed several different ways. Similarly, the indicators for participation and for processes and institutions could be described as "flows" or "throughput" measures, which represent aspects of the size, character, and practices of the formal education system. Finally, the indicators for achievement and attainment and for labor market outcomes present the "product" or "output" of education systems, as measured by degree completion, educational attainment, and economic benefits.

The data come from a variety of sources. The data on countries come from the Indicators of Education Systems (INES) project of the OECD, the International Assessment of Educational Progress (IAEP), the National Science Board, the Luxembourg Income Study, Asia Pacific Economic Cooperation, the European Community, the World Health Organization, UNESCO, the American Federation of Teachers, and several other sources. The data on individual states come from NCES, the Department of Labor's Bureau of Labor Statistics, the Department of Commerce's Bureau of the Census, the Department of Health and Human Services National Center for Health Statistics, the National Science Board, the American Federation of Teachers, the Center for the Study of Social Policy, and Child Trends, Inc. All these sources are described in more detail in the "Sources of Data" section in the back of the report. In addition, results from the 1992 NAEP study of mathematics achievement of American 8th-graders have been statistically linked to results from a similar 1991 study of the mathematics achievement of 13-year-old students in various countries. This linkage allows comparisons of academic achievement between states and countries.

The presentation of each indicator includes an explanation of what it measures, why it is important, and key results from a comparison of countries and states. Throughout the report, comparisons are most often made in the text among "like-sized" entities: the United States to the other large and relatively wealthy countries that compose the so-called Group of Seven, or G-7 (Canada, France, Germany, Italy, Japan, and the United Kingdom); and U.S. states to all industrialized countries for which data are available, including the smaller and relatively less wealthy ones.

It should be kept in mind, however, that these comparisons are based on the data available. Not all countries are represented here. Some countries are not members of the international organizations which collected the data. Other countries are members, but did not participate in the relevant data collections. Some countries participated in the OECD's data collection but not the IAEP's, and vice versa. If there is any systematic bias in such "data driven" international comparisons, it is probably toward the inclusion of countries with a well-developed public data collection and management capability and the exclusion of countries without.

In addition to the explanations and key results, the presentation of each indicator includes separate tables for states and countries and a graph or set of graphs that display states and countries together. The graphs are, in most cases, simple bar graphs with the states and countries listed in order of highest value to lowest. This type of graph highlights the distributional aspects of the data - where countries and states stand in relation to one another and the magnitude of the differences between them. Where appropriate, notes on interpretation describe special circumstances affecting an indicator that warrant particular consideration in making comparisons. Data sources are listed at the bottom of each table and graph. Because some of the terms used in this report may not be familiar to all readers, a glossary is included in the back. Finally, appendices include supplemental and technical information on how various measures in the indicators were calculated.

In the remainder of the overview, we highlight some of the more important concepts and results from each of the six sections of the report.

Section 1: Background

Understanding the context in which education systems exist is important to proper interpretation of indicators. Each indicator in this report, while measuring one particular aspect of education, is affected by a host of other factors, some not directly connected to education. The first group of indicators in this report represent some of these other factors that make up the context in which education takes place. Indicators in this group are:

(1) Population and area;
(2) Youth and population;
(3) Labor force participation;
(4) GDP/GSP per capita;
(5) Percentage of population age 17 or younger in poverty;
(6) Births to teen mothers; and
(7) Youth violent death rate.

A complete comparative understanding of education would require a consideration of still more factors not represented here, such as: differences in the levels of development of education systems, national and state education priorities and strategies, and cultural differences. Nonetheless, the seven indicators presented in the "Background" section provide some understanding of the environments in which education programs are set and should be considered when evaluating data in the categories of participation, processes and institutions, achievement and attainment, labor market outcomes, and finance.

How closely do the states resemble other industrialized nations demographically and economically?
In general, the industrialized nations selected in this publication had higher population densities than the U.S. states. However, the U.S states tended to be wealthier, to have higher labor force participation rates, and to have greater proportions of youth (i.e., persons 5- to 29-years-old) in the overall population. For every indicator, one can find individual states closely resembling certain industrialized countries. For example:

Pennsylvania had a population just slightly larger than that of Hungary (Indicator 1), and had the same percentage of 5- to 29-year-olds in its population. (Indicator 2)
Texas, North Dakota, New Zealand, and Italy had similar labor force participation rates. (Indicator 3)
The gross product per capita in South Dakota was only marginally greater than that in Japan. (Indicator 4)

How closely do the states resemble other industrialized nations sociologically?
Thirty-eight of the U.S. states had higher percentages of children living in poverty than all 17 of the other countries to which they are compared. Births to teen mothers generally constituted a higher percentage of all births in the states than in many of the industrialized nations, but the range of rates in those nations was the same as that of the states. For the most part, a greater percentage of youth died violently from accidents, suicides, and homicides in the states than in the nations. As with the demographic and economic background indicators, a comparison can be found between individual states and nations for each sociological indicator included. For example:

With the exception of New Hampshire and Connecticut, the child poverty rate was higher in the states than in Italy, France, the former West Germany, the Netherlands, Sweden, the United Kingdom, and several other countries in some cases several times higher.(Indicator 5)

The poverty threshold used is an approximation of the U.S. average - 40 percent of median household income - and other countries data are adapted to it. These poverty rates are measured after taxes and transfers; that is, they account for the effect of taxes and of governmental aid programs to the poor. These data for nations come from the Luxembourg Income Study's collection of national household surveys.

The proportion of all births that were to 15- to 19-year-old mothers was similar in Alabama and Greece, 7.1 per 100 births. In 30 states, fewer than 6 out of every 100 births was to a teen mother, compared with 9 of the 12 European countries for which data are available. (Indicator 6)
While only 3 of the 30 countries for which we have data had violent death rates among youth higher than 500 per million, 19 of the U.S. states did. The rates of suicide and accidental death among youths aged 5 to 24 in Austria were almost identical to those of Wisconsin; the rates of homicide within the same age group were slightly higher in Argentina and slightly lower in the Soviet Union than they were in Oregon, Kansas, or Kentucky. (Indicator 7)

In summary, economic, demographic, and sociological characteristics of the U.S. states were similar in many cases to those of other industrialized countries. While these similarities between nations and states could almost always be found, some overarching trends differentiating states and nations are apparent. For example, the states tended to have lower population densities, greater wealth, and higher labor force participation rates than the other industrialized countries. Youths aged 5 to 29 typically composed a larger portion of the population in states than they did in other countries. This high proportion of young citizens in the states seemed to confront a relatively more negative social environment as well, manifested in higher rates of violent death among youth, of births to teen mothers, and of child poverty.

Section 2: Participation

Participation in formal education is influenced not only by demand - the number of persons who can and wish to attend school - but also by the supply - the number of places available. In terms of supply, preprimary (which includes both kindergarten and pre-kindergarten programs) and postcompulsory education are more available in some states and countries than in others. High participation can reflect a large public or private investment in education, a high valuation of education by society, or an economy dependent on a highly trained workforce. Measures of the degree to which young people participate in their state or country's education system are included in this section. Indicators in this group are:

(8) Participation in formal education;
(9) Enrollment in preprimary education;
(10) Secondary education enrollment;
(11) Entry ratio to higher education;
(12) Non-university higher education enrollment; and
(13) University enrollment.

(International comparisons based on levels of education can sometimes cause confusion because the levels do not always have the same entrance requirements or the same duration across countries. To aid in understanding such comparisons, an explanatory note is included in the supplemental notes [on page 231].)

How does participation in education change as people move from childhood to adulthood?
Two different measures of enrollment are used in this section: enrollment rates and enrollment ratios. Enrollment rates represent the percentage of students in a certain age group enrolled in a particular level of education. Enrollment ratios reflect the number of students of any age enrolled in a particular level of education per 100 persons in a reference age group, the ages typical of those enrolled at that level. Although enrollment rates are preferred to enrollment ratios, as they are not inflated by enrollments either outside the typical age of enrollment or by periods of enrollment longer than the typical duration, the requisite data needed to calculate enrollment rates - enrollment by age - are often unavailable.

For most countries and states, the ratio of persons enrolled in formal education (total enrollment divided by the population in the 5- 29 age range) was between 50 and 60 (Indicator 8).
Of the states, Nevada had the smallest ratio of persons enrolled in formal education, with a ratio of 52, which was higher than in 9 of the 22 other countries for which data are available. (Indicator 8)

Preprimary participation rates are affected by the relative value placed on early socialization of children in society, the availability of low-cost or public preprimary programs, and the degree of participation of women in the labor market. Enrollment rates in preprimary education at ages 3 and 6 varied greatly across states and nations. (Indicator 9)

In the G-7 countries for which data are available, 1991 preprimary education enrollment for 3-year-olds ranged from approximately 20 percent in Japan to almost 100 percent in France. In the United States, about one-third of 3-year-olds were enrolled.
None of the states had an enrollment rate higher than 39 percent among 3-year-olds, while 7 of 14 other countries did.

In the 50 U.S. states and in most industrialized countries, participation in primary and lower secondary education (the equivalent of grades 1 to 9 in the United States) has become almost universal, and in most cases is legally mandated. Upper secondary education (the equivalent of U.S. grades 10 to 12) encompasses the final stage of compulsory education in most industrialized countries. Because the age at which students can legally leave school typically arrives before their secondary education is complete, participation rates for those age 16 and older reflect the desirability and importance of secondary education credentials (like the high school diploma).

Furthermore, the nature of secondary education varies across countries. For example, in Germany and Austria, many vocational students obtain the equivalent of apprenticeship training in a basic skill while enrolled in secondary school. Some of them even return to secondary school later, after gaining several years work experience, to obtain a second credential, typically in a higher skilled trade. In the U.S. states, participation in secondary education was minimal beyond age 18, whereas enrollment rates for 20- and 21-year-olds were significant in some countries. (Indicator 10)

In 9 of the 19 other countries, over 20 percent of 19-year-olds attended secondary school; however, none of the U.S. states had enrollment rates above 10 percent among 19- year-olds. Likewise, among 21-year-olds, 7 of the 19 other countries had rates above 5 percent, while none of the U.S. states had rates above 3 percent at that age. (Indicator 10)

Participation rates continue to drop off as secondary students make the transition to non- university higher education (the equivalent of U.S. community colleges) and university education (4- year colleges and universities in the United States), although some countries and states are higher than others. For example, higher education enrollment rates are generally much higher in the United States and Canada than in other industrialized countries. (Indicator 8) When students are counted at the location of their higher education institution rather than at the location of their original residence entry ratios into higher education at the entry reference age ranged from approximately 74 percent in North Dakota to 15 percent in Turkey. (Indicator 11)

In some countries, higher education is highly career-oriented, and admission is often quite selective. In the U.S. states, however, the higher education system in general is less selective and is available to almost any high school graduate. Many U.S. students also enter higher education without focusing on a particular career, while their peers in many other countries focus exclusively on their area of specialization from day one of higher education.

Among 18- to 21-year-olds in 1991, the United States had relatively high full-time enrollment rates in non-university higher education (7.5 percent), as did Canada and France. (Indicator 12).
There was much variation in full-time enrollment rates of 18- to 21-year-olds in non- university higher education in both U.S. states and other countries. The range was wider across the states, however, than across the countries. The states ranged from 0.3 percent enrolled in the age group in South Dakota to 18.3 percent in Wyoming for a difference of 18 percentage points, while the countries ranged from 0.7 percent in Denmark to 14.0 percent in Belgium for a difference of 13.3 percentage points. (Indicator 12)
In university education, the U.S. states generally had higher full-time enrollment rates among 18-to 21-year-olds than did the countries for which data were available. Full- time enrollment rates exceeded 20 percent in 36 states, but did so in only 2 countries. The range of part-time enrollment rates among 18- to 21-year-olds was wider across the states than across the countries. Part-time enrollment rates were 6.2 percent in Alaska, and 2.3 percent in Australia, the country with the highest rate. (Indicator 13)

In summary, participation in formal education was virtually universal in every state and country for youths at the primary and lower secondary levels. Enrollment rates in early childhood education fluctuated across countries and states, with rates ranging from 0 to almost 100 percent for each age of preprimary enrollment. Early childhood enrollment in the U.S. states was most prevalent among 5-year-olds, with sparse enrollments among children aged 3 and 6 (most 6-year-olds in the United States are enrolled in primary school). Participation was nearly universal for only part of the upper secondary years; enrollment rates dropped dramatically in some countries beginning at age 16. Higher education participation rates were highest in Canada and the United States and more people enrolled in university than non- university higher education in every country except the Netherlands.

Section 3: Processes and Institutions

The indicators in this section measure two components of the instructional arena - the instructional process and the organization of personnel serving students' instructional needs. The instructional process involves both the time spent in the classroom - how students are taught and the tools used to teach them - and the effort required of students at home to reinforce classroom learning. Indicators in this group are:

(14) Staff employed in education;
(15) Number of schools and school size;
(16) Class size;
(17) Student use of technology;
(18) Student time spent doing homework and watching television;
(19) Instructional strategies in mathematics courses; and
(20) Time in formal instruction.

How does the amount of time students in the United States spend in the classroom compare to that of students in other countries? Do students in the United States spend more or less time doing homework or watching television than their international counterparts?
Although the number of days per year that U.S. students spend in school is generally lower than that in other countries, the hours of instruction per day often are greater. For the most part, the U.S. states had a higher average number of hours per year in formal instruction than the other industrialized countries. (Indicator 20)

The average hours of instruction per year in the United States (1,003) exceeded that of 13 of the other countries for which data are available. Only France, Taiwan, China, Switzerland, and Scotland had more instructional hours annually; the former West Germany and Israel had about the same.
U.S. states and most countries were fairly evenly distributed throughout the range defined by Ireland (931 hours of instruction per year) and China (1,276 hours per year). Nonetheless, 7 countries had less than 900 hours of instruction per year.

When not in class, however, lower secondary students in the United States reported doing less homework than did their counterparts in most other countries. Across the states, between 19 and 34 percent of public 8th grade students reported that they did 2 or more hours of homework each day. Instead, U.S. students spent more time watching television than did students in most other countries for which data are available. Across the states, between 72 and 90 percent of public 8th grade students reported watching 2 hours or more of TV daily. (Indicator 18)

The percentage of public 8th grade students in the states who reported doing 2 or more hours of homework daily was generally lower than it was for 13-year-old students in the other countries for which data are available. Twelve of 18 other countries had percentages above 40, whereas none of the states did.
Among the states, only Utah, Wyoming, and Colorado had less than 80 percent of 8th grade public school students report watching TV for 2 hours or more daily. However, 12 of the 18 other countries had percentages that low.

How do teaching strategies employed in mathematics classrooms differ across countries and states?
Similar resources can be applied in quite different ways to achieve desired educational goals. Sometimes the manner in which instruction is organized derives from tradition or some other cultural context; other times, it may result from an explicit policy decision to adopt one instructional strategy over another. For example, 8th grade mathematics classes in U.S. public schools were more likely to be organized by ability groups than their counterparts in other industrialized countries. Ability grouping was used more frequently only in England, Israel, Ireland, and Taiwan. It must be kept in mind, however, that ability grouping can occur at the school, in addition to the class level. School-level tracking (or streaming, as it is called in England) occurs both in countries that allow greater parental choice of schools and in those that assign students to either vocational or academic lower secondary schools based on their prior academic performance.

For the most part, a higher percentage of students were in math classes based on ability in the U.S. states in 1992 than in the other nations for which data are available in 1991. Fourteen of 19 nations, but only 1 state, had less than 40 percent of their students in math classes based on ability. (Indicator 19)

Another instructional strategy is to have students work in small groups within classes. In 1991, 49 percent of U.S. 13-year-olds reported working in such small groups in their mathematics classes each week. A higher percentage of students reported working in small groups in 8 of the 18 other countries for which data are available. (Indicator 19)

In 13 of 18 other nations, over 40 percent of 13-year-olds reported working in small groups in their math classes at least once a week. In only 4 states did 8th grade public school students report working in small groups that often.

Relative frequency of classroom testing is another form of instruction for which cross national data are available. U.S. 13-year-olds were more likely to take math tests or quizzes weekly than their counterparts in almost all of the other nations included - only Taiwan and Jordan had equal or higher frequencies. (Indicator 19)

In 11 of 18 other countries, 40 percent or fewer of the 13-year-olds reported taking math tests or quizzes at least once a week. In every state, at least 40 percent of public 8th grade students reported being quizzed that often. Louisiana, Taiwan, Mississippi, and Alabama had percentages greater than 80.

Are U.S. students more or less likely than their counterparts in other countries to use computers and calculators in the classroom?
Some educators argue that technology, effectively employed, can assist students in developing higher- order thinking skills. Two of the more common technologies utilized by teachers and students are calculators and computers. The use of calculators in class was relatively common in the United States in 1991, with 54 percent of 13-year-olds using them in school. Although this rate was about average for the countries, it was significantly lower than that in France, where 94 percent of the students used calculators in school. (Indicator 17)

In 1991, 90 percentage points separated the countries with the highest and lowest rates of in-school calculator usage among 13-year-olds: France at 94 percent and Korea and Brazil at 4 percent. Half of all the nations for which data are available reported percentages of less than 50 percent. Across the U.S. states in 1992, calculator usage rates among public school 8th graders ranged from at least 87 percent in Minnesota and Maine to 47 percent in Mississippi.

In every U.S. state, at least a quarter of the students used computers for homework or school work. Half of the nations reporting data had lower rates of computer use. (Indicator 17)

About a quarter of public 8th grade students in Tennessee reported that they use computers for school work or homework. Although this percentage was the lowest among the states, it was higher than in 9 other countries, including the former Soviet Union, Spain, and Taiwan. The students of Maine matched those of Slovenia in the highest rate of computer usage. (61 percent)

The instructional process is also affected by the way in which resources are organized in different education systems. Do the states and nations organize their instructional and non-instructional efforts differently? The organization of students and staff is the subject of the following three indicators: staff employed in education, class size, and the number of schools and average number of students per school.

How do the states and nations compare in their level of staffing?
A large proportion of the labor force employed in education reflects an extensive education system. Among the several industrialized nations for which data are available, teaching and non-teaching staff employed in education comprised between 3 and 7 percent of the total labor force. In the United States this proportion was 5.6 percent, slightly below France's 5.9 percent, but well above Japan's 3.1 percent. Countries vary, however, in the degree to which social and other non-instructional services are provided directly by the schools. In the United States, for example, school districts commonly pay directly for school-based health services, school cafeterias, pupil transportation, vocational and psychological counseling, building construction and maintenance, and administrative management of the schools. In other countries, many or all of these services are either provided by non-education public authorities (such as the Ministry of Health) or by the private sector. The United States had the largest non-teaching staff in education, as a percentage of the total labor force (2.9 percent), of the 7 countries reporting data. (Indicator 14)

The range across countries in the percentage of the total labor force employed in teaching was 3 percentage points: from about 2 percent in Turkey to over 5 percent in Belgium. This exceeded the range across the states of 1.4 percentage points: from 2.2 percent in Florida to 3.6 percent in Alaska.
For the 6 countries other than the United States for which data are available, teaching staff outnumbered non-teaching education staff. Teaching staff outnumbered non-teaching staff in 18 of the 49 U.S. states for which data are available.

How do the states and nations compare in their class sizes?
The number of students a teacher faces during a period of instruction - measured as average class size - is an indicator of the typical teacher's pupil load. Small classes may allow students to receive more personal attention from their teachers. Large classes, however, can be less expensive and do not necessarily hinder instruction. Depending on teaching style, student behavior, and other factors - such as the opportunity for students to meet with teachers outside of class large classes may function as effectively as small ones.

The countries reported a wide range of average class sizes, from 18 in Switzerland to 49 in Korea. That range is three times wider than the range across the states, from 19 in Wyoming and Vermont to 30 in Utah. (Indicator 16)

How do the states and nations compare in their school sizes?
School size may be determined by population density or a more deliberate organizational policy. The prevailing educational philosophy in the United States for the past three decades has been that large schools could offer more comprehensive curricula and a wider variety of programs at lower cost. Small schools, however, may have beneficial effects upon student participation, attendance, satisfaction, and achievement. (Indicator 15)

Students were organized into larger schools in the United States than they were in most other countries. Only Taiwan and Korea, among 12 other countries, had larger schools on average than did the United States at the preprimary through secondary level. Only Germany, Taiwan, and Korea, of 10 other countries, had larger schools at the higher education level.
The average number of students per preprimary through secondary school in Taiwan was 873, a figure more than five times greater than those of Finland or France, the countries with the smallest averages (at 156 and 166, respectively). For the most part, the schools in the U.S. states from the preprimary through secondary levels were larger than those in other countries: schools in 28 states, but only 2 countries - Korea and Taiwan averaged above 400 students.
The U.S. states generally had higher average numbers of students per school at the higher education level than did the other countries. Five states, but none of the countries, had averages above 6,000; whereas half of the other countries, but only 15 of the states, had averages below 3,000.

In summary, although students in the United States spent fewer days per year in school, they received a larger number of instructional hours per day than students in most other industrialized countries. U.S. students, therefore, received more instructional hours per year than did students in the majority of industrialized countries included here. The type of instruction students receive in class and the prevalence of student adoption of common instructional technologies varied across countries and states. U.S. lower secondary students were more often placed in math classes according to ability than were students in other nations. U.S. lower secondary teachers also tended to give math tests or quizzes more often than teachers in other countries; 68 percent of U.S. 13-year-olds reported taking a math test or quiz at least once a week. Work in small groups was also more common in lower secondary math classrooms in the United States than it was in math classrooms in other countries. Calculator usage was of average prevalence among U.S. math students (54 percent) compared to that among students in other industrialized nations, where, in 12 of 17 other countries, calculator usage was either above 70 percent or below 30 percent. However, the use of computers for homework and school work was more common among students in the United States than it was among their international counterparts. At least 25 percent of public school 8th-graders in each U.S. state claimed to use computers for school work or homework.

Outside of class, students in other nations generally reported spending less time watching television and more time doing homework than students in the United States. Only 29 percent of 13-year-olds in the United States did 2 hours or more of homework each day - a percentage lower than that in all but 4 other countries included here. Eighty-four percent of U.S. students watched TV for 2 hours or more daily.

In the United States, teaching and non-teaching staff employed in education accounted for 5.6 percent of the total workforce, an average proportion in comparison to that of other countries. The percentage of the total workforce employed as non-teaching educational staff, however, was higher in the United States than in any other industrialized nation included here. In no other country reporting data, but in almost two-thirds of the U.S. states, non-teaching staff outnumbered teaching staff. Compared to other countries, the organization of education personnel in relation to students resulted in larger schools for the most part (at both the primary- secondary and higher education levels) but smaller classes (at the lower secondary level).

Section 4: Achievement and Attainment

There are many outcomes of education. The six indicators in this section provide information on educational attainment; completion rates for programs of study; and exhibited academic skills and knowledge. They are:

(21) Educational attainment of the population;
(22) Educational equity for women;
(23) Secondary school completion;
(24) University completion; and
(25) Mathematics achievement (experimental).

The organization of levels of education in the United States is often quite different than it is in other countries. In most countries the end of compulsory education is the completion of lower secondary education which is roughly equivalent to 8 or 9 years of education. In the United States, compulsory education is described in terms of age or the completion of high school. For example, most states require young people between the ages of 6 and 15 to be enrolled in school. In many countries, upper secondary education is differentiated; that is, several different types of programs are available. Some programs are designed to prepare young people to work in a particular occupation; others are designed to prepare young people to pursue studies at a university. In the United States, almost all high schools (grades 9 to 12) are comprehensive, providing both academic and vocational courses; however, the latter is rarely of great depth.

Despite differences in the organization of education, it is useful to compare the educational attainment of the population in states and countries in order to compare the investment people in these states and countries have made in their own education.(10)

How well educated are the citizens of the states and the industrialized countries?
Although there was considerable variation among U.S. states, most had higher levels of educational attainment than most of the other industrialized countries. (Indicator 21) For the most part, the percentages of 25- to 64-year-olds who had finished high school in the states were greater than the percentages of 25- to 64-year-olds who had completed upper secondary education in other countries - for the purposes of international comparisons, high school completion is regarded as roughly equivalent to upper secondary completion. University completion rates (a bachelor's degree or higher in the United States) for this age group in the other industrialized countries ranged from 3 percent in Portugal to 17 percent in Canada, while the percentage holding this level of education in the states ranged from 14 percent in West Virginia to 31 percent in Massachusetts and Connecticut.

Included in the age range of 25 to 64 are many people who grew up in an era when educational opportunities in their countries, particularly for higher education, were less available than they are today. It is, therefore, illustrative to compare levels of educational attainment of older and younger members of the working-age population. For all countries and all but 3 states, high school (upper secondary) attainment levels were higher for younger people (25- to 34-year-olds) than for older people (25- to 64-year-olds). This indicates that over time larger and larger percentages of new cohorts are finishing high school or its equivalent. (Indicator 21)

Across the states, the percentage of 25- to 34- year-olds having attained at least an upper secondary level of education (high school or more) ranged from 77 percent in Mississippi to 93 percent in Minnesota and North Dakota. Across other countries, the distribution was wider, ranging from 22 percent in Turkey to 88 percent in Norway, Germany, and Switzerland.

The same trend is not as prevalent for college completion. In 2 of 21 countries and in 18 of the U.S. states, the proportion of persons in the older age cohort completing university education (a bachelor's degree or higher in the United States) exceeded that in the younger age cohort. (Indicator 21)

University completion rates were generally higher for U.S. states than for other industrialized countries. The percentage of 25- to 34-year-olds holding bachelor's degrees ranged from 14 percent in Nevada and West Virginia to 34 percent in Massachusetts, while university attainment rates in other countries ranged from 5 percent in Spain to 18 percent in Canada.

Is there a gap between the levels of educational attainment reached by women and men in the nations and states?
To illustrate whether or not women share in the educational opportunities available to their male counterparts in their nation or state, the percentage of various educational attainment groups who were women are compared across countries and states. Because women represented about 50 percent of 25- to 64-year-olds in each state or country, percentages above 50 percent suggest women were over represented in the group, and percentages below 50 percent suggest they were underrepresented in the group. In general, U.S. women seem to have fared better than women in other industrialized countries relative to their male counterparts in attaining upper secondary and university levels of education. Across all nations and states, however, women continued to compose a smaller proportion than men of the population having attained a university degree. (Indicator 22)

In 15 of the 20 other countries represented here, over half of women 25 to 64 years old had not completed upper secondary education. However, women comprised that large a proportion of high school dropouts in only 2 U.S. states.
In every country or state, women comprised less than half of 25- to 64-year-old university graduates (college graduates in the United States). In 14 of the 20 other countries represented here, the percentage of college graduates who were women was 43 percent or less. However, in only 3 of the U.S. states was the percentage who were women that small.

How well do American students compare to students of other nations in mathematics achievement?
To compare the performance of students in states and nations on mathematics performance, an experimental indicator was developed. The mathematics proficiency scores of participants in the Second International Assessment of Educational Progress (IAEP) were mapped to a scale used to report scores of U.S. students in the National Assessment of Educational Progress (NAEP). This cross-linking allows comparisons of the average and percentile scores of 13-year-old students in selected industrialized countries (not all of them OECD members) to 8th graders from public schools in selected U.S. states. (Indicator 25) The NAEP scale for mathematics ranges from 0 to 500. The supplemental note to Indicator 25 addresses the conceptual issues surrounding the task of linking two different assessments and the effects of alternative methods of linking assessments on the results.

Among the 7 largest countries (who assessed virtually all age-eligible children) the average proficiency score of 13-year-olds ranged from 262 in the United States to 285 in Taiwan. The average proficiency score was 273 in France and 270 in Canada.
The range in average mathematics proficiency across states was similar to the range across countries. Average proficiency scores for public 8th grade students in 1992 ranged from 246 in Mississippi to over 280 in Iowa, North Dakota, and Minnesota. Average scores for 13-year-olds students in 1991 ranged from 246 in Jordan to over 280 in Taiwan and Korea.
Over 25 percent of 13-year-olds in Taiwan and Korea scored above 300 in 1991, while about 10 percent of students of the same age scored above that level in the United States. However, in 4 states 25 percent or more of U.S. 8th grade public school students (who are generally older than 13 years) scored above this level in 1992.

To help interpret these differences, it is useful to consider another type of comparison: differences within the United States between the mathematics proficiency of better and poorer performers of the same grade level. The 10th percentile of mathematics proficiency among public 8th grade students in Mississippi was 201, and the 90th percentile was 291, a difference of 90 points, which is more than twice the 39-point difference between the average Taiwanese 13-year-old and Mississippi 8th grader. This suggests that variation among students within countries is far larger than variation in averages between countries.

In summary, the population of 25- to 64-year-olds in the United States generally had higher levels of educational attainment than did their international counterparts. The proportion of this age group that completed lower secondary education or less was smaller in the United States than it was in 18 of the 20 other countries included here. Inversely, of all the countries for which data are available, the United States had the second highest percentage of this age cohort that attained an upper secondary education, and the second highest proportion that attained a university education. However, much of the gap in educational attainment between the U.S. and other countries has narrowed considerably in recent years, as one can see by looking at the educational attainment rates in the younger age groups.

Section 5: Labor Market Outcomes

Although the four indicators in this section also measure educational outcomes, they focus on long- term outcomes, such as unemployment rates and earnings among graduates of various levels of schooling. and gender differences in earnings. The labor market outcome indicators are:

(26) Unemployment and education;
(27) Earnings and education;
(28) Gender difference in earnings; and
(29) New scientists and engineers.

What are the long-term economic effects of educational attainment in states and nations?
In general, higher levels of educational attainment are associated with lower rates of unemployment and higher earnings. In the United States in 1990, the unemployment rate for 25- to 64-year-olds who did not complete high school was 5 percentage points higher than for high school graduates. In 19 countries and all 50 U.S. states, the unemployment rates for university graduates were lower than for those with only the equivalent of a high school education.

The relationship between education and earnings can be illustrated by calculating the mean annual earnings for a particular level of educational attainment as a percentage of the mean annual earnings of workers who completed just upper secondary education. For example, in 46 states and 7 of 12 countries university-educated males had mean earnings percentages of 150 or greater on this measure; that is, they received a 50 percent premium in earnings compared to their counterparts who only completed upper secondary education. The strength of the earnings and education relationship is indicated by the difference between the earnings premium of being a university graduate to the earnings disadvantage of completing, at most, lower secondary education. In general, the relationship between earnings and educational attainment was stronger in the U.S. states than in many other countries.

Almost without exception, higher levels of educational attainment were associated with lower rates of unemployment. Switzerland was an exception. Although their unemployment rates were generally very low, they were somewhat higher among university graduates than among those with lower educational credentials. (Indicator 26)
In the United States in 1990, the unemployment rate for people who had not completed high school (10.4 percent) was more than double that for those who had completed high school but not gone on to college (5.1 percent). A large difference in unemployment rates between those two education levels (lower and upper secondary) also existed in Canada (5 percentage points), but was not quite as large in France, Germany, or the United Kingdom (each 4 percentage points). (Indicator 26)
In all countries and all states in the early 1990s, higher levels of education were associated with higher mean annual earnings. (Indicator 27)
For university-educated females, 45 states and 9 of 12 countries had earnings ratios of 150 or greater. Similarly, for university-educated males, 46 states had ratios of 150 or greater, as did 7 of 12 countries. (Indicator 27).
In all the countries represented here, not having completed an upper secondary education resulted in the lowest earnings ratio. In 1991, Portugal had the lowest earnings ratio among the countries for the lowest level of educational attainment: below 70, for both males and females. Not having finished high school by 1990 resulted in earnings ratios that low for males in California, Louisiana, and Texas, as well as for females in those three states and also Colorado, Delaware, and Virginia. (Indicator 27)

How well have women fared relative to their male counterparts in earnings in the states and in the nations?
As Indicator 22 illustrated, not only did women still constitute a smaller portion than men of those having attained a university level of education in states and nations, but earnings within that attainment population were also unequally distributed when broken down by gender. U.S. women seem to have fared better than women in other industrialized countries relative to their male counterparts in attaining upper secondary and university levels of education. But, they were generally paid less than women in other industrialized countries relative to their male counterparts at these levels. (Indicator 28) Included in the age range 25 to 64, however, are many people who grew up in an era when occupational opportunities for women were less available than they are today. Thus, even if selection for jobs is made equitably from this point forward, the disparity in earnings would take some time to dissipate.

In all countries and states, the average annual earnings for females aged 25 to 64 was less than that of males of the same age cohort and level of educational attainment.
Half of the other countries included here reported ratios of mean annual earnings of women to men of 64 or more in 1991. All of the U.S. states had lower ratios in 1990. A similar pattern held for three of the four levels of educational attainment: half the countries had ratios of mean annual earnings of women to men higher than the ratio of the U.S. state with the highest ratio.

Do more students in the United States pursue careers as scientists and engineers than in other countries?
At first glance, it would appear that the U.S. education system puts more emphasis on science and engineering training in its higher education system than do the education systems in other countries. Science and engineering graduates generally comprise a larger proportion of their age group (at a typical graduation age - 22 years old) in the United States than they do in other countries. (Indicator 29) But, then, as was mentioned previously, the U.S. graduates more persons in the typical age group in general, regardless of the type of degree. When the number of science and engineering degrees in a nation or state are counted as a proportion of all degrees, the U.S. proportion is much lower than that in most countries.

In 1991, the number of U.S. university students who graduated with science or engineering degrees amounted to about 5 percent of the population of 22-year-olds. Among the G-7 countries in various years between 1988 and 1991, only Japan and Canada produced higher percentages of science and engineering degrees. Germany's percentage was about the same as the United States.
Four out of 30 other countries (Finland, Bulgaria, Japan, and South Korea) had percentages of science and engineering degrees among 22-year-olds of 6 or above. Twenty of the states had percentages that high.

In summary, educational attainment exhibited a strong correlation with labor market outcomes as measured by unemployment and earnings. Educational attainment was positively associated with annual earnings and negatively associated with unemployment rates in all states and all countries, except Switzerland.

Gender differences in earnings indicate that women, in general, earn less than men. The ratio of mean annual earnings of women to men varied across states and countries, but in all cases, women earned less than men having the same educational attainment. In the United States, the ratio of earnings of women to men was lower at every level of educational attainment than that of most of the other industrialized countries reporting data.

Section 6: Finance

This section includes the following indicators of education finance:

(30) Current public expenditure on education as a percentage of GDP/GSP;
(31) Current public expenditure on education as a percentage of total public expenditures;
(32) Current public expenditure per student;
(33) Current public expenditure per student as a percentage of GDP/GSP per capita;
(34) Distribution of current public expenditure on education;
(35) Teacher salaries;
(36) Sources of funds for primary and secondary education; and
(37) Sources of funds for higher education.

Through most of this section, the focus is on expenditure from public sources, rather than on total investment in education, which would include money from private sources. In some cases, expenditure from private sources amounts to a substantial portion of total educational expenditure. However, financial data on private education are not available from some countries.(11)

Which countries and states provide the strongest financial support to education?
Financial support for education can be viewed from several different angles, each of which focuses on certain factors and not on others. For example, total expenditure on education is useful for determining who spends the largest sum of money on education, but may be misleading when comparing small countries or states to larger ones, for a small country may spend less in the aggregate but may spend more per-student. Likewise, a poorer country may spend as much per student as a richer country, seeming to make a greater effort to educate its citizens; however, that would not be apparent by looking only at aggregate spending or per-student spending.

Because there is no universally superior measure of public financial support for education, several indicators are presented here. The first, current public expenditure per student (Indicator 32), presents the amount of public financial support for one student's education in each country or state.

At the primary through secondary level, the United States spent more public money per student ($4,605), and at the higher education level, the United Kingdom ($10,228) and Canada ($8,555) spent more per student, than the other G-7 countries.
For the primary through secondary level, Sweden ($5,825) had the highest level of per- student public expenditure among the countries for which data are available; and Alaska, Connecticut, New Jersey, and New York had the highest levels among the states (all above $6,400). Japan, Australia, Spain, and Hungary all spent about the same or less than Mississippi, the lowest spending state ($2,648).
At the higher education level, public expenditure per students varied greatly across both the countries and the U.S. states. The United Kingdom had the highest level of per- student expenditure among the countries ($10,228), although Alaska and Hawaii spent more. Spain and Japan both spent less public money per student on higher education than New Hampshire, the lowest spending state ($3,624).

An advantage of using per-student expenditure as an indicator of a nation's or state's financial effort to support education is that it takes into account the size of the student population. On the other hand, one disadvantage is that much of the variation between states and countries may in fact be caused by the relative wealth of that state or nation. The second finance indicator, current public education expenditure as a percentage of GDP/GSP (Indicator 30), is a measure of what states and nations spend on education in terms of the economic resources available to them.

Of the G-7 countries, only Canada had a higher level of current public expenditure as a percentage of GDP (6.1 percent) than did the United States and France (both 4.6 percent). Canada's proportion was almost twice that of Japan's (3.1 percent).
The distribution of levels of expenditure across states and countries was quite similar. Montana, Canada, West Virginia, Vermont, and New Mexico had the highest levels of educational expenditure as a percentage of GDP/GSP (6.0 percent or above). The lowest levels were found in Japan, Nevada, West Germany, and Delaware (3.3 percent or less).

Another disadvantage of the simple per-student expenditure measure is that much of the variation between states and countries may in fact reflect the relative size of the public sector in a nation or state. The third finance indicator, current public education expenditure as a percentage of total public expenditure (Indicator 31), attempts to show what states and nations spend on education in terms of the size of their public sectors generally.

Finland, Canada, and the United States had the highest level of education expenditure as a percentage of total public spending among the countries represented here; West Germany and Italy, the lowest.
The U.S. states figures on this measure generally exceeded those of the countries represented here. Two-thirds of the countries reported levels of current public education spending as a percentage of all public spending to be lower than that of Virginia, the state with the lowest level.

The second and third finance indicators provide measures of a nation's or state's spending on education in relation to its available resources or in relation to its total public spending, but education spending is also highly influenced by the size of the student population. All other factors being equal, a country or state with a relatively small student population is likely to spend a smaller portion of its GDP/GSP or of its total public spending on education than a country with a large student population. Thus, the fourth finance indicator, current public education expenditure as a percentage of GDP/GSP per capita (Indicator 33), provides a measure of fiscal effort to support education that takes into account both a country's or state's available financial resources and the size of the student population. It is calculated by dividing the first finance indicator, public expenditure per student, by a nation's or state's per- capita gross product.

On this measure, some states and countries with higher per-student expenditure (Indicator 32) appeared to be not so high when their available resources were taken into account (Indicator 33).

For example, of the 4 states - New Jersey, New York, Alaska, and Connecticut - with the highest per-student expenditure at the primary through secondary level, New Jersey, New York, and Connecticut remained among the states with the highest ratios of per-student expenditure to per-capita GSP. Alaska, however, fell below 43 other states, moving from the highest on the first measure to near the bottom on the second.
On the other hand, among countries for which data were available, those with the highest per- student expenditure at the primary through secondary level - sweden, Denmark, the United States, Norway, and Canada - remained the highest ranking countries even when available resources were taken into consideration. However, the United States fell lower when education expenditure was divided by gross product per capita.

Do states and countries differ in the relative proportion of public expenditure devoted to different levels of education?
Many factors affect this "balance," including the relative size of student populations and system- wide education goals and strategies. For example, some countries or states may choose to invest heavily in higher education in order to increase the number of professionals and managers, while others may feel a more pressing need to focus on basic education for the larger populace by providing more primary and secondary schools. It is important to note, however, that this indicator does not give a complete picture of the distribution of total resources between the two levels, since some countries (such as the United States, West Germany, and Japan) had considerable private funds going to education (see tables 3 through 6 in the supplemental Notes for examples of the relative size of private expenditures across countries).

Regarding the balance of expenditure between levels of education (Indicator 34), the United States expenditure on the primary through secondary level as a percentage of all current public education expenditure lay in the bottom half of the range among all the nations represented here. Of thes nations, Japan, Italy, and France devoted a larger share of current public expenditure to this level. If West Germany's large "undistributed" proportions were allocated entirely to the primary-secondary level, its primary- secondary shares might exceed those of the United States as well. Hungary, Spain, and Sweden had the highest percentages of current expenditure at the primary through secondary level (without counting the undistributed proportion). New Jersey, New Hampshire, and Vermont, the highest- spending U.S. states, spent a slightly larger share at that level of education. At the higher education level, Australia, Canada, Utah, North Dakota, New Mexico, and Hawaii reported relatively high proportions of spending.

Where does the funding of education originate in each nation or state? What is the balance between public and private financing or among the levels of government?
Two more finance indicators trace the path of all education expenditures back to their origin among the levels of government and between public and private sectors. The initial source of money for education sometimes differs from the ultimate spender. For example, though local school districts in the United States generally operate and fund the local public schools, much of the financing arrives in the form of transfers from state governments. Some of the state money, in turn, arrives in the form of transfers from the Federal government. The initial sources of those transferred funds, then, are state and Federal governments. Likewise, the initial source of funds spent on public schools can be either public or private.student tuition and fees are one example of a private source of public expenditure. Funding by private firms of youth apprenticeship programs in Germany and Austria is another example. Moreover, the initial source of funds spent on private schools can be either public or private. Unlike the United States, most other OECD countries maintain large numbers of privately-operated schools that are mostly or entirely publicly funded.

Tracking funds to their initial source illuminates where responsibility is actually assumed in a nation or state for financing education, either at the primary through secondary level (Indicator 36) or at the higher education level (Indicator 37).

Of the 11 other countries reporting public elementary and secondary expenditure data by level of government, only Canada raised less money for education at the national level than did Mississippi, the U.S. state that relied the most on the Federal government for funds.
In the United States, local government provided a portion of public higher education funding higher than that in any of the 11 other countries reporting data (6 percent). Conversely, the percentage of funds derived initially from the central government was lowest in the United States among all the nations. The United States and Belgium were the only 2 nations in which the share of public funding of institutions of higher education from the regional, or state, level exceeded 50 percent.

How much are teachers paid across nations and states?
Teacher salaries are an important indicator of both the level of investment in and the quality of a nation's or state's education system. Without exception across nations and states, teacher salaries constitute the greatest portion of education expenditure. The amount of money paid to teachers is a primary factor in attracting and retaining top-quality candidates to pursue careers as educators. Therefore, salaries influence the level of quality and experience with which students are instructed. This indicator (Indicator 35) presents data on average salaries for teachers for the United States and its states and for secondary school teachers with approximately 15 years of experience in other countries. The ratio of teacher salary to country or state per capita gross product is also included.

The average teacher salary in the United States for the school year 1991 to 1992 was about $34,000. That was the median among thes countries for mid-career secondary school teachers. The mid-career salaries in former West Germany, France, and Canada were highest (almost $40,000 in former West Germany). The mid-career salaries in England, Japan, and Scotland (representing the United Kingdom), and Italy were lowest (less than $22,000 in Italy).
The range of mid-career secondary school teacher salaries was slightly wider across countries than the range of average salaries for teachers across states. Teachers in Connecticut, the state with the highest salaries, received twice the income of their counterparts in south Dakota.secondary school teachers in switzerland, the country with the highest-paid teachers, received almost two-and-a-half times the salary of Italian secondary school teachers.
The ratio of a teacher's average salary to per capita gross domestic product was about 1.5 in the United States. That was higher than Italy's ratio for secondary school teachers (1.23) but lower than the ratios for others countries (England and Scotland as proxies for the United Kingdom). The ratios for France, former West Germany, England, and Scotland were about one-third higher than that of the United States.

In summary, a comparison of 1991 public education expenditures across countries finds that the United States spent more public funds per student at the primary through secondary level than did any of the others countries. At the higher education level, the United States spent more public money per student than the other G-7 countries except Canada and the United Kingdom. When public education expenditures are measured as a percentage of gross product, only Canada's ratio, among all thes countries, exceeded that of the United States, whereas France's was about the same. Finally, combining two of the previous measures into a single measure of fiscal effort - current public education expenditure per capita divided by per capita gross product - finds Canada on top again, ahead of Italy, France, and then the United States among the G-7 countries.

Comparing the U.S. states to all the countries represented here (rather than just the G-7), sometimes presents a different picture of the relative level of public education spending in the United States. Particularly because some smaller northern European countries spent at higher levels, the distribution among states was more uniform than that among countries.

The proportional allocation of public education funds from among different levels of government varies widely across nations and states. The United States relied more on both state and local governments than did other countries.

Other related NCES projects

This second edition of Education in States and Nations continues a series of occasional reports comparing the education systems of different states and countries. This series, however, is just one part of an overall NCES international effort. NCES serves as the representative for the United States in the OECD's INES project mentioned earlier. In connection with the INES project, NCES commissioned two reports to improve the comparability of education finance data across countries: The International Expenditure Comparability Study and Improving the Comparability of International Expenditure Data. These'studies have reviewed ten countries statistical reports and interviewed their officials in order to identify differences in the content and categorization of expenditures, both in national finance statistics and in data submitted to the OECD and UNESCO. The studies have developed revised estimates of countries education expenditures that adjust for deviations from an international standard. These reports should be available soon.

NCES has also sponsored another project to clarify the content of indicators published in international comparisons. Education Indicators: An International Perspective presents a set of indicators for the United States and other countries, along with additional information about the education systems in those countries. The various structures of the education systems and other contextual factors help to explain the structure of the indicators, and help U.S. readers understand the indicators in all their complexity.

These projects and others comprise a major ongoing effort to not only compare education systems across states and countries, but also to improve the comparability of data and to deepen understanding of the context of the data.

In addition to these indicators and research projects, NCES continues to work in cooperation with its counterparts in other countries to administer international assessments and collect and analyze their data. These projects include: the International Association for the Evaluation of Educational Achievement (IEA) Reading Literacy Study, conducted from 1989 to 1992; the IEA's Third International Mathematics and science study (TIMSS), being conducted now; the pilot testing of the OECD's Cross-Curricular Competency Test in 1995; and the International Adult Literacy survey, conducted in 1994. The International Adult Literacy survey (IALS) was a collaborative effort by seven governments and three intergovernmental organizations (UNESCO, Eurostat and the OECD) to fill the information gap on literacy in industrialized countries.

Notes

(1) ... Many observers attribute the origins of the current wave of education reform in the United States to the 1983 publication of A Nation at Risk. Other observers trace the origins to the late 1970s, when the first of many states passed student minimum competency requirements. The National Commission on Excellence in Education, which wrote A Nation at Risk, and many others, however, would distinguish the minimum competency movement as an earlier, separate, and failed effort to reform education (see, for example, pages 19 to 21 of A Nation at Risk). . . . return to section

(2) ... The explicit mission of the commission that wrote A Nation at Risk was to study the quality of learning and teaching in our nation's schools. since then, education reformers have often employed the language and methods of the historically parallel quality management movement. Indicators are needed in order to monitor processes and measure progress toward goals. Outcome measures are as important as input measures. Goals and standards should be universally accepted by stakeholders, clear enough to serve as a common focus, measurable, and challenging. standards, or benchmarks, from outside one's own organization serve to ground plans in a reality not defined by vested interests. . . . return to section

(3) ... It should be recognized that, in this publication, the meaning of the wordstate is the U.S. version, a sub-national, regional jurisdiction. National jurisdictions are called countries or nations throughout. . . . return to section

(3a) ... It should be recognized that, in this publication, the meaning of the wordstate is the U.S. version, a sub-national, regional jurisdiction. National jurisdictions are called countries or nations throughout. . . . return to section

(4) ... The other original National Education Goals were: 1) All children will start school ready to learn. 2) The high school graduation rate will increase to at least 90 percent. 3) students will demonstrate subject area competency at grades 4, 8, and 12 and be prepared for good citizenship, further learning, and productive employment. 7) Every school will be free of drugs and violence and offer a safe, disciplined environment conducive to learning. The two National Education Goals added in 1994 are: 4) Teachers will have access to programs to improve their skills. 8) schools will promote parental involvement. . . . return to section

(5) ... Since 1991, the National Education Goals Panel has developed education indicators that pertain to progress toward the National Goals, which are published in the annual National Education Goals Report. Other organizations making similar national efforts include the Council of Chief State School Officers, the National Science Board, and the Education Commission of the states. . . . return to section

(6) ... The increased demand for information on education and the need for improved knowledge on the functioning of education systems raised many questions not only for data collection but also the organization, reporting and interpretation of the data. These questions led authorities in the member countries of the OECD to consider new ways of comparing their education systems. Agreement was reached on the feasibility and utility of developing an international set of indicators that would present in statistical form the key features of the education systems of the member countries.

The Centre for Educational Research and Innovation responded to this demand for comparative information by initiating the Indicators of Education systems Project (INES). This project grew out of two preparatory conferences: one hosted by the government of the United states in November 1987, and the second by the French authorities in March 1988. A meeting to review progress and discuss the plan of work was subsequently convened in Austria in September 1989. The results achieved during the initial phases of the project were presented at an international conference in Lugano,Switzerland in September 1991. . . . return to section

(7) ... The nations of the OECD include Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan, Luxembourg, the Netherlands, New Zealand, Norway, Portugal,Spain, Sweden, Switzerland, Turkey, the United Kingdom, and the United States. Because Greece and Iceland did not participate in the OECD's Indicators of Education systems (INES) project, data on these countries are not included in this report. Data for several OECD observer countries, such as Hungary and the former Czechoslovakia, are included as data are available. . . . return to section

(8) ... Education at a Glance was the product of a collective effort to improve the gathering and reporting of comparative information on education in the OECD countries. In the process of developing the indicators, CERI established an international consultative mechanism for exchanging viewpoints and creating a common understanding of issues related to the definition, measurement, and organization of the indicators. Education at a Glance thus represents the combined effort of several networks and technical groups composed of policy-makers, administrators, and researchers.

The indicators were influenced by the concerns of the different parties that were involved in their development. Three principles guided the work. The first was that the indicators be targeted to a broad audience. second, total coverage through a large and complex set of measures was not the aim; rather, the indicators were selective and intended to be policy-relevant, providing information useful for decision-making and evaluation. Third, in addition to being reliable and valid at the national level, the indicators were standardized in a way that makes them comparable among the OECD countries. . . . return to section

(9) ... Several other OECD countries have federal systems like the United states in which a major responsibility for education rests with regional (provincial or state) governments. These countries are Australia, Belgium, Canada, Germany, Switzerland, and the United Kingdom. . . . return to section

(10) ... Again, international comparisons based on levels of education can sometimes cause confusion because the levels do not always have the same entrance requirements or the same duration across countries. To aid in understanding such comparisons, an explanatory note is included in the supplemental notes. . . . return to section

(11) ... See supplemental note on private higher education expenditure in Japan and the United States . . . return to section

Notes on International Comparisons

Indicator 1