Skip Navigation
Digest of Education Statistics: 2004
Digest of Education Statistics: 2004

NCES 2006-005
October 2005


In the fall of 2004, about 71.7 million persons were enrolled in American schools and colleges (table 1). About 4.3 million persons were employed as elementary and secondary school teachers and as college faculty. Other professional, administrative, and support staff at educational institutions numbered 5.0 million. Thus, about 81.0 million people were involved, directly or indirectly, in providing or receiving formal education. All data for 2004 in this Introduction are projected. Some data for other years are projected or estimated as noted.

Elementary/Secondary Education


Enrollment in public elementary and secondary schools rose 22 percent between 1985 and 2004 (table 2). The fastest public school growth occurred in the elementary grades (prekindergarten through grade 8), where enrollment rose 25 percent over this period, from 27.0 million to 33.7 million. Private school enrollment grew more slowly than public school enrollment from 1985 to 2004, rising 14 percent, from 5.6 million to 6.3 million. As a result, the proportion of students enrolled in private schools declined slightly, from 12.4 percent in 1985 to 11.6 percent in 2004. Since the enrollment rates of kindergarten and elementary school-age children have not changed much in recent years (table 7), increases in public and private elementary school enrollment have been driven primarily by increases in the number of children in this age group. Public secondary school enrollment declined 8 percent from 1985 to 1990, but then rose 29 percent from 1990 to 2004, for a net increase of 18 percent.

The National Center for Education Statistics (NCES) forecasts record levels of total elementary and secondary enrollment through at least 2014, as the school-age population continues to rise. The projected fall 2005 public school enrollment marks a new record, and new records are expected every year through 2014, the last year for which NCES enrollment projections have been developed (table 3). Public elementary school enrollment (prekindergarten through grade 8) is projected to show a slight decline between 2002 and 2005, and then increase, so that the fall 2014 projected enrollment is 6 percent higher than the 2004 projected enrollment. Public secondary school enrollment (grades 9 through 12) is expected to show a net decline of 2 percent between 2004 and 2014.


A projected 3.5 million elementary and secondary school teachers were engaged in classroom instruction in the fall of 2004 (table 4). This number has risen in recent years, up 27 percent since 1990. The 2004 projected number of teachers includes 3.1 million public school teachers and 0.4 million private school teachers.

The number of public school teachers has risen faster than the number of students over the past 10 years, resulting in declines in the pupil/teacher ratio (table 64). In the fall of 2004, there were a projected 15.6 public school pupils per teacher, compared with 17.3 public school pupils per teacher 10 years earlier.

The salaries of public school teachers, which lost purchasing power to inflation during the 1970s, rose faster than the inflation rate in the 1980s (table 77). Since 1990–91, salaries for teachers have generally maintained pace with inflation. The average salary for teachers in 2002–03 was $45,822, about 2 percent higher than in 1992–93, after adjustment for inflation.

Student Performance

Most of the student performance data in the Digest are drawn from the National Assessment of Educational Progress (NAEP). NAEP conducts assessments using three basic designs: main NAEP, state NAEP, and long-term trend NAEP. These three basic designs are described in the paragraphs that follow. The main NAEP assessments provide current information for the nation and specific geographic regions. They include students drawn from both public and private schools and report results for student achievement at grades 4, 8, and 12. The main NAEP assessments follow the frameworks developed by the National Assessment Governing Board and use the latest advances in assessment methodology. The NAEP frameworks are designed to reflect changes in educational objectives and curricula.

Since 1990, NAEP assessments have also been conducted at the state level. States that choose to participate receive assessment results that report on the performance of students in that state. In its content, the state assessment is identical to the assessment conducted nationally. However, because the national NAEP samples prior to 2002 were not designed to support the reporting of accurate and representative state-level results, separate representative samples of students were selected for each participating jurisdiction/state. Since 2002, a combined sample of public schools has been selected for both state and national NAEP. The national sample is a subset of the combined sample of students assessed in each participating state, plus an additional sample from the states that did not participate in the state assessment.

NAEP also conducts long-term trend assessments, which provide information on changes in the basic achievement of America's youth since the early 1970s. They are administered nationally and report student performance at ages 9, 13, and 17 in reading and mathematics. Measuring trends of student achievement or change over time requires the precise replication of past procedures. Therefore, the long-term trend instrument does not evolve based on changes in curricula or in educational practices.


Overall achievement scores on the NAEP long-term trend reading assessment for the country's 9-, 13-, and 17-year-old students are mixed. The average reading scores at ages 9 and 13 were higher in 2004 than in 1971 (table 110). The average score for 17-year-olds in 2004 was similar to that in 1971.

For Black 9-, 13-, and 17-year-olds, average reading scores in 2004 were higher than in 1971. At age 9, Black students scored higher on average in 2004 than in any previous administration year. For ages 13 and 17, scores increased between 1971 and 2004. For White students, the average scores for 9- and 13-year-olds were also higher in 2004 than in 1971. Separate data for Hispanics were not gathered in 1971, but as with the other racial/ethnic groups, the average reading score for Hispanic students at age 9 was higher in 2004 than in any other assessment year. The average score for Hispanic students at age 13 shows an increase between 1975 and 2004. The scores for 17-year-old Hispanic students also increased between 1975 and 2004, but no measurable changes were seen between 1999 and 2004.

The 2003 main NAEP reading assessment of states found that reading proficiency varied widely among fourth-graders in the 53 participating jurisdictions (50 states, Department of Defense overseas and domestic schools, and the District of Columbia) (table 114). The U.S. average score was 216. The scores for the participating jurisdictions ranged from 188 in the District of Columbia and 203 in New Mexico to 228 in New Hampshire, Massachusetts, and Connecticut.


Results from NAEP long-term trend assessments of mathematics proficiency indicate that the scores of 9- and 13-year-old students were higher in 2004 than in 1973 (table 120). For White, Black, and Hispanic 9-, 13-, and 17-year-olds, average mathematics scale scores were higher in 2004 than in 1973.

The 2003 main NAEP assessment of states found that mathematics proficiency varied widely among public school eighth-graders in the 53 participating jurisdictions (50 states, Department of Defense overseas and domestic schools, and the District of Columbia) (table 124). Overall, 67 percent of these eighth-grade students performed at or above the Basic level in mathematics, and 27 percent performed at or above the Proficient level. Mississippi and the District of Columbia had significantly fewer than 50 percent of students performing at least at the Basic level in math.

International Comparisons

In 2003, the performance of U.S. 15-year-olds, as measured by the Program for International Student Assessment (PISA), in mathematics literacy and problem solving was lower than the average performance for most Organization for Economic Cooperation and Development (OECD) countries (table 399). Along with the scale scores, PISA also used six proficiency levels (levels 1 through 6, with level 6 being the highest level of proficiency) to describe student performance in mathematics literacy (table 400). In mathematics literacy, the United States had greater percentages of students below level 1 and at levels 1 and 2 than the OECD average percentages. The United States also had a lower percentage of students at levels 4, 5, and 6 than the OECD average percentages.

Educational Technology

There has been widespread introduction of computers into the schools in recent years. In 2003, the average public school contained 136 instructional computers (table 424). One important technological advance that has come to classrooms following the introduction of computers has been connections to the Internet. The proportion of instructional rooms with Internet access increased from 51 percent in 1998 to 93 percent in 2003 (figure 30). Nearly all schools had access to the Internet in 2003 (table 424).


Postsecondary Education

College Enrollment

College enrollment hit a record level of 16.6 million in fall 2002. Another record of 17.1 million is anticipated for fall 2004 (table 3). Enrollment is expected to increase by an additional 14 percent between 2004 and 2014. Despite decreases in the traditional college age population during the late 1980s and early 1990s, total enrollment increased during the late 1980s and early 1990s (tables 7, 15, 172, and 183). Between 1992 and 2002, the number of full time students increased by 22 percent compared to a 5 percent increase in part-time students (table 171). During the same time period, the number of men enrolled rose 10 percent, while the number of women enrolled increased by 18 percent.

Faculty and Staff

In the fall of 2003, there were 1.2 million faculty members in degree-granting institutions (table 224), including 0.6 million full-time and 0.5 part-time faculty. In 1998, full-time instructional faculty and staff generally taught more hours and more students than part-time instructors (tables 229 and 230). About 21 percent of full-time faculty taught 15 or more hours per week, compared with 9 percent of part-time faculty. About 13 percent of full-time faculty taught 150 or more students, compared with 4 percent of part-time faculty.

Graduates, Degrees, and Attainment

The estimated number of high school graduates in 2003–04 was 3.1 million (table 102), including 2.8 million public school graduates and 0.3 million private school graduates. The number of high school graduates is lower than at its peak in 1976–77, when 3.2 million students earned diplomas. The number of General Educational Development (GED) credentials issued rose from 332,000 in 1977 to 648,000 in 2001, before falling to 311,000 in 2002 (table 105). The percent of dropouts among 16- to 24-year-olds declined over this period, from 14 percent in 1977 to 10 percent in 2003 (table 107). The projections of the number of postsecondary degrees conferred during the 2003–04 school year by degree level show 667,000 associate’s degrees; 1,401,000 bachelor's degrees; 531,000 master's degrees; 82,200 first-professional degrees; and 45,700 doctor's degrees (table 247).

The U.S. Census Bureau collects annual statistics on the educational attainment of the population. Between 1990 and 2004, the proportion of the adult population 25 years of age and over who had completed high school rose from 78 percent to 85 percent, and the proportion of adults with a bachelor’s degree increased from 21 percent to 28 percent (table 8). Over the same period, the proportion of young adults (25- to 29-year-olds) completing bachelor's degrees rose from 23 percent to 29 percent. The 87 percent of 25- to 29-year-olds who had completed high school in 2004 was about the same as it was in 1990.

Education Expenditures

Expenditures for public and private education, from kindergarten through graduate school (excluding postsecondary schools not awarding associate’s or higher degrees), are estimated at $866 billion for 2003–04 (table 29). Expenditures of elementary and secondary schools are expected to total $514 billion, while those of colleges and universities are expected to total $351 billion. Total expenditures for education are expected to amount to 7.9 percent of the gross domestic product in 2003–04, about 0.7 percentage points higher than in 1993–94 (table 29).

Interpreting Statistics

Readers should be aware of the limitations of statistics. These limitations vary with the exact nature of a particular survey. For example, estimates based on a sample of institutions will differ somewhat from the figures that would have been obtained if a complete census had been taken using the same survey procedures. Standard errors are available for sample survey data appearing in this report. In many cases, standard errors for all items appear in the printed table. In other cases, only standard errors for key items appear in the printed table. For some tables, standard errors are available from NCES upon request. Although some of the surveys conducted by NCES are census type surveys, all surveys are subject to design, reporting, and processing errors and errors due to nonresponse. Differences in sampling, data collection procedures, coverage of target population, timing, phrasing of questions, scope of nonresponse, interviewer training, data processing, coding, and so forth mean that the results from the different sources may not be strictly comparable. More information on survey methodologies can be found in the Guide to Sources in the appendix. Unless noted otherwise, all financial data in this report are in current dollars, not adjusted for changes in the purchasing power of the dollar. Price indexes for inflation adjustments can be found in table 35.

Common data elements are collected in different ways in different surveys. Since the Digest relies on a number of data sources, there are discrepancies in definitions and data across tables in the volume. For example, several different surveys collect data on public school enrollment, and while similar, the estimates are not identical. The definitions of racial/ethnic groups also differ across surveys, particularly with respect to whether Hispanic origin is considered an ethnic group regardless of race, or counted separately as a racial/ethnic group. Individual tables note the definitions used in the given studies.

All statements cited in the text about differences between two or more groups or changes over time were tested for statistical significance and are statistically significant at the .05 level. Two test procedures were used primarily, depending on the nature of the statement tested. The most commonly used test procedures were t tests, equivalence tests, and linear trend tests. Equivalence tests were used to determine whether two statistics are substantively equivalent or substantively different. This was accomplished by using a hypothesis test to determine whether the confidence interval of the difference between sample estimates is substantively significant, i.e., greater or less than a preset substantively important difference. In most cases involving percentages, a difference of 3.0 was used to determine substantive equivalence or difference. In some comparisons involving only very small percentages, a lower difference was used. In cases involving only relatively large values, a larger difference was used, such as $1,000 in the case of annual salaries.