The Knowledge Economy and Postsecondary Education: Report of a Workshop (2002)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

PART II: WORKSHOP PAPERS

student financial aid. The paper then presents a brief overview of trends within the military and the labor market that may influence participation in postsecondary education. The final section of the paper reviews the limitations of current data sources for monitoring changes in postsecondary education, and suggests areas for improvement. This paper focuses primarily on trends within the past three decades, from 1970 to 2000' although shorter time Periods are used when data are not available for all 30 years. _ ma. . , .. a. . ~ ~ . . . . .. . . . . .. . .. .. l'~roughout, the paper relies on analyses of federal data, particularly data collected by the U.S. Department of Education's National Center for Education Statistics (NCES). NCES data are a rich source of relatively objective, reliable data with which to describe postsecondary education. They do, however, have some limitations. First, the national portrait provided by these data necessarily masks differences that exist among states and regions of the country (e.g., enrollment trends among Hispanic students are likely to differ in the Southwest compared to the Midwest). These more detailed analyses were beyond the scope of this paper.) Also, for reasons discussed at the end of the paper, existing national data pnmanly describe "traditional" postsecondary institutions and the attainment of "traditional" postsecondary education credentials. Finally, because of both the reliance on national data and the broad scope of this paper, many topics could not be covered in depth, and many of the complex issues raised by these data received admittedly cursory treatment. It is hoped that these shortcomings are outweighed by the rigor and breadth of the infonnation presented. YOUNG ADULT POPULAR TRENDS Although about 40 percent of college students are over age 24, young adults aged ~ ~ to 22 are often considered the key constituency for postsecondary education. This population of young adults has fluctuated in size over the past three decades, increasing in the 1970s (as the baby boomers reached college age) and declining during the 1980s and early 1990s. The number of young adults increased from 23.7 million in 1970 to 30.2 million in ~ 98 I, then declined to a low of 24.8 million in ~ 996. Since ~ 996, the size ofthe IS- to 24-year-old cohort has increased to 26.0 million in 1999, and it is expected to continue to grow in size over the next five decades (U.S. Census Bureau, 1996, 20001. Because the federal government did not separate out Hispanics in its data collections until the mid-1970s, trend data on the racial/ethnic composition ofthe young- adult cohort are more limited. Over the roughly two decades Tom ~ 980 to ~ 999, the proportion of young adults who are White2 declined Tom 78 percent to 66 percent. At the same time, the proportion of Blacks increased slightly from 13 percent to 14 percent, the proportion of Hispanics increased from ~ percent to ~ 5 percent, and the proportion of other minorities (Asians and Native Americans) increased Tom 2 percent to 5 percent (U.S. Census Bureau, 1996, 2000~. As will be seen later, these changes in the ~ The reader interested in state-level data is referred to publications produced by the State Higher Education Executive Officers (SHEEO) and He National Center for Public Policy and Higher Education (e.g., Measuring Up 2000~. 2 Throughout this paper, "White" refers to non-Hispanic Whites and "Black" refers to non-Hispanic Blacks. The Knowledge Economy and Postsecondary Education: Report of a Workshop Chapter 1 12

1 Demographic and Attainment Trends in Postsecondary Education Lisa Godsons "More people are going to college!" is hardly an attention-grabbing headline, as it describes a long-standing trend in American educa- tion. This trend reflects continued increases in the skills required by the labor market and by society in general. Within this world of increasing skill demands, America's public and private postsecondary education institutions have firmly maintained their role and mission. In recent years, however, these postsecondary institutions have faced growing competition. In particular, the growth of alternative providers (such as for-profit institutions, "virtual" universities, and corporate universities) and alternative credentials (such as company-based certifi- cates) have called into question the efficacy of the traditional postsecondary institution and its ability to continue its dominant role as the (nearly) exclusive provider of postcompulsory education. Other chapters in this volume examine alternative postsecondary education providers and pedagogies and the ways in which tradi- tional postsecondary institutions are adapting to changing conditions. This chapter provides a context for the remainder of the volume, by providing a broad overview of trends within postsecondary education, as well as trends in the civilian labor market and the military that may affect the demand for postsecondary education. These three activities (postsecondary education, civilian work, and military service) constitute the three main career options available to those leaving high school. To put these options in perspective, among students who were eighth- ~Lisa Hudson is an education statistician at the National Center for Education Statistics in the U.S. Department of Education. The views in this paper are those of the author. No official support by the U.S. Department of Education is intended or should be inferred. 13

graders in 1988 (and thus expected to graduate in 1992), 74 percent were working for pay or looking for work in 1994, and 53 percent were in a postsecondary education program. (About 35 percent were engaged in both activities.) Only 3 percent of these former students were in the military, and 7 percent were full-time homemakers (Berktold, Gels, and Kaufman, 1998~. STRUCTURE OF THE CHAPTER Although this chapter examines all three postsecondary school activities, the main emphasis is on postsecondary education. The chapter begins by examining characteristics of the young adult population. A number of aspects of postsecondary education are then examined, including trends in postsecondary enrollment levels and rates, the composition of students in postsecondary education, the number and types of degrees awarded, and student financial aid. The chapter then presents a brief overview of trends within the military and the labor market that may influence participation in postsecondary education. The final section reviews the limitations of current data sources for monitoring changes in postsecondary education and suggests areas for improvement. This chapter focuses primarily on trends within the past three decades, from 1970 to 2000, although shorter time periods are used when data are not available for all 30 years. Throughout, the chapter relies on analyses of federal data, particularly data collected by the U.S. Department of Education's National Center for Education Statistics (NCES). NCES data are a rich source of relatively objective, reliable data with which to describe postsecondary education. They do, how- ever, have some limitations. First, the national portrait provided by these data necessarily masks differences that exist among states and regions of the country (e.g., enrollment trends among Hispanic students are likely to differ in the Southwest compared to the Midwest). These more detailed analyses were beyond the scope of this chapter. Also, for reasons discussed at the end, existing national data primarily describe "traditional" postsecondary institutions and the attainment of "tradi- tional" postsecondary education credentials. Finally, because of both the reliance on national data and the broad scope of this chapter, many topics could not be covered in depth, and many of the complex issues raised by these data received admittedly cursory treatment. YOUNG ADULT POPULATION TRENDS Although about 40 percent of college students are over age 24, young adults aged 18-22 are often considered the key constituency The reader interested in state-level data is referred to publications produced by the State Higher Education Executive Officers (SHEEO) and the National Center for Public Policy and Higher Education (e.g., Measuring Up 2000~. 14 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

for postsecondary education. This population of young adults has fluctuated in size over the past three decades, increasing in the 1970s (as the baby boomers reached college age) and declining during the 1980s and early 1990s. The number of young adults increased from 23.7 million in 1970 to 30.2 million in 1981, then declined to a low of 24.8 million in 1996. Since 1996, the size of the 18-24-year-old cohort has increased to 26.0 million in 1999, and it is expected to continue to grow in size over the next five decades (U.S. Census Bureau, 1996, 2000~. Because the federal government did not separate out Hispanics in its data collections until the mid-1970s, trend data on the racial/ethnic composition of the young adult cohort are more limited. Over the roughly two decades from 1980 to 1999, the proportion of young adults who are White2 declined from 78 percent to 66 percent. At the same time, the proportion of Blacks increased slightly from 13 percent to 14 percent, the proportion of Hispanics increased from 8 percent to 15 percent, and the proportion of other minorities (Asians and Native Americans) increased from 2 percent to 5 percent (U.S. Census Bureau, 1996, 2000~. As will be seen later, these changes in the racial/ethnic composition of the young adult population are reflected in changes in the college student population over time. Looking at a slightly older group of adults, those aged 25-29, shows that the education level of adults has increased over time, as more individuals have completed high school, enrolled in college, and earned a college degree (Figure 1-1~. From 1971 to 1999, the percentage of adults aged 25-29 who completed high school increased from 78 percent to 88 percent; the percentage who had at least some college education increased from 44 percent to 66 percent; and the percentage who had at least a bachelor' s degree increased from 22 percent to 32 percent (NCES, 2000~. The proportion of these adults complet- ing at least some college has increased faster than the proportion completing high school, suggesting that the college enrollment rate has been increasing. As will be discussed later, this rate has indeed been rising. But before students can go to college, they must leave high school. High School Dropout and Completion Rates Completing high school increases a student's chances of attend- ing college, and completing high school through a regular diploma rather than an alternative route increases a student's chances of both going on to college (Snyder, 2001) and of completing college once he or she has started (Boesel, Alsalam, and Smith, 1998~. Thus, to maxi- mize their opportunity to enter and complete college, students should 2Throughout this paper, "White" refers to non-Hispanic Whites and "Black" refers to non-Hispanic Blacks. LISA HUDSON 15

100- 80 - 60 - 40 - 20 - O - _ _ _ ~ I, D 9' ~ ~ 9' ,6 <7 ~ ~ ,6 ,~ On, ~ Hi or A, 9,4 a. a D ~ ~ ° ~ ~ ~ ~ O O ~ ~ ~~ ~ ~ o ~ ¢, o ~ ~ ,~7 ~ ,~_ ~ High school I ~ Some college _ Bachelor's degree . . . ~ 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 Year FIGURE 1-1 Percentage of 25-29 year olds who have completed at least high school, some college, or a bachelor's degree: 1971-1999. SOURCE: Data from National Center for Education Statistics (2000, pp. 154-156~. ideally graduate from high school with a regular high school diploma.3 While most high school students do this, many do not. For example, in 1999, 86 percent of 18-24 year olds who were not enrolled in high school had completed high school, 77 percent by graduating from high school and 9 percent through an alternative means such as the General Educational Development (GED) test. Thus, 23 percent of these young adults had failed to graduate from high school through the traditional path. These figures represent a decline in high school dropout rates and corresponding increase in completion rates since the 1970s. Eleven percent of 16-24 year olds were dropouts4 in 1999, down from over 14 percent in 1972 (Kaufman, Kwon, Klein, and Chapman, 2000). However, while dropout rates have declined since the early 1970s, they were fairly steady during the l990s; similarly, the high school completion rate has increased slightly since the early 1970s, but remained flat in the l990s. 3High school graduation maximizes other opportunities as well: Graduating from high school with a regular diploma is also related to lower levels of unemployment and higher wages, compared to not completing high school or completing through an alternative program (Boesel et al., 1998~. 4This measure of dropouts includes all young adults aged 16-24 who are not in school and have not earned a high school credential. This measure undercounts school dropout rates, since some of these young adults may have dropped out of high school but subsequently earned a credential. 16 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

These trends are occurring along with increased academic course taking among high school students (Levesque, Lauen, Teitelbaum, Alt, and Librera, 2000) and relatively steady or increasing academic achievement on national standardized tests (NCES, 2000; Smith, 1996~. Taken together, these findings suggest that school reform and accountability efforts in the past few decades may have improved learning outcomes for many high school students, although, at least in recent years, they seem to have had little effect on high school completion rates. It is not clear to what extent these trends have affected postsecondary education for example, it is not (yet) known whether these learning gains have reduced the need for remediation at the college level or to what extent they account for increasing enrollments at the postsecondary level. Immediate Transition to College and SAT Scores The most successful route to a college degree is to enter college immediately after high school graduation (NCES, 1997~. The percentage of high school completers who make this immediate transition remained relatively constant at about 50 percent from 1972 to 1980 but then increased to 66 percent by 1998 (NCES, 2000~. The number of students who took the SAT also increased from 1975 to 1999 (College Board, 2000c). This increase has occurred despite a declining cohort of 17 year olds, so that in 1975 the number of SAT-takers was 23 percent of the number of 17 year olds, while in 1999, SAT-takers were 31 percent of the 17-year-old population. Most of this increase occurred during the 1980s; by 1987, SAT-takers were 29 percent of the 17- year-old population (Snyder and Hoffman, 1991; College Board, 2000c; Snyder, 2001~. At the same time that more high school graduates are going directly to college and more students are taking the SAT, SAT scores have been holding steady or increasing (Snyder, 2001~. Average verbal SAT scores declined from 507 to 505 from 1986 to 1987 but have remained constant at 505 since then (up to 1999~. However, over this same time period, verbal SAT scores increased for each racial/ethnic group except Hispanics, whose scored dropped. These within-group trends suggest two reasons for the lack of an overall increase in verbal scores. The first reason is the drop in scores among Hispanics; the second reason is the increasing percentage of minorities attending college. Since all minority groups have lower verbal scores than Whites, this enrollment increase lowers the overall average score. Average math SAT scores increased from 501 to 514 from 1986 to 1999. Part of this increase may be due to increasing enrollments of Asian students, who have higher average math scores than other racial/ethnic groups, but it also reflects an increase in scores among each racial/ethnic group. These positive trends reflect a high school student body that appears to be, on average, better prepared to enter college. The next sections LISA HUDSON 17

1600Q 1200Q o o x E 800Q o 400Q All institutions f f f ~ ~ f f Public institutions _4-year institutions _2-year institutions Private institutions ~ ~ f f f ~ f O _ f ~ ~ ~ f ~ I ~ Am, f ~ , ~ I ~ take a closer look at college enrollment trends in general and the students who are enrolling in college. COLLEGE ENROLLMENTS The number of students enrolled in college has been increasing for at least the last three decades (Figure 1-2~. This growth has been fueled by increasing college enrollment rates among high-school graduates and among adults in general (Figure 1-3), rather than from increases in the number of high school graduates or college-age adults (defined as adults aged 18-24~. In fact, enrollment increases have occurred in spite of a declining cohort of college-age adults over most of the last two decades and relatively constant high school graduation rates. The increase in college enrollment was particularly steep during the 1970s, when community colleges were expanding. Interestingly, however, the college enrollment rate of 18-24 year olds was fairly constant over this period. During the 1970s, college enrollment growth appears to have resulted from enrollment rate increases among older adults (aged 25-34, see Figure 1-3) combined with a growing cohort of adults in this age category (U.S. Census Bureau, 2000~. Enrollment growth continued throughout the 1980s, fueled primarily by an increasing enrollment rate among college-age adults. Since 1992, however, enrollment appears to have leveled off, and the increase in the enrollment rate of college-age adults has slowed. The apparent leveling off of enrollment does not appear to be due to changes in the fly f ~ f ~ f i,, f <I ~ - O 1~. 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 Year FIGURE 1-2 Total fall enrollment in degree-granting postsecondary institutions, by control and level of institution: 1970-1997. SOURCE: Data from Snyder (2001, p. 203~. 18 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

~ Age 18-24 50 - 40 - ° 30- ~ 20 - 10 O - Age 25-34 Age 35+ - - - ,6 0 ~ ~ q$E ~ ~ ,0 9,, ~ O ~ ~ - ~ ~ ~ ° ~ ~ O ~ ~ ~ O ~ ~ ° ° ~ ~ l l l l l l l l l l l 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 Year FIGURE 1-3 Percentage of the population enrolled in postsecondary education, by age group: 1970-1998. SOURCES: Data from National Center for Education Statistics (2000, pp. 114-115) and U.S. Census Bureau, (2000, p. 167~. size of the college-age population. Although this cohort became smaller during the l990s, it shrank less during the l990s than in previous decades, when enrollment grew. Data on the wage premiums associ- ated with college education may provide one clue as to why growth in enrollment rates and levels may be slowing. College Wage Premiums College enrollment rates can be viewed as an indicator of labor market demand for a college education; when demand is high, the enrollment rate increases, and vice versa. Another indicator of labor market demand for a college education is the wage premium associ- ated with a college education. This measure indicates how much a college-educated worker earns compared to a worker who has only a high school education (Figures 1-4a and 1-4b). Comparing the trend in Figure 1-3 with the trends in Figures 1-4a and 1-4b shows that the enrollment rate among adults aged 18-24 began to increase a few years after the wage premium for a college education began to rise. Throughout most of the 1980s, both the relative returns to a college education and young adult enrollment rates increased, suggesting a strong labor market increase in the demand for a college education during that decade. In the l990s, however, the wage premium for a college education leveled off and college enroll- LISA HUDSON 19

a) ~ 1.00 - 0.50 - ~ Bachelor's degree or higher 0ff0009 Some college 2.00 - 1.50 - ~ Grades 9-11 JO _ - _ -_ _ _ ~ ~ ~ ~ ~ ^ 9 ~ ~ ~9 p 45, a o 9 ~ ~ o as To To 9,< no Den . _ ~ ~ ~ — .... ~ __ _ -_ ~ .,~,, A in, 9,5> o ~ 9 9 ~ ~ ~ 9 ~ a g ~ °~ ~ ~ f ~ ~ ~ L ooo 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 Year FIGURE 1-4a Ratio of median annual earnings of male wage and salary workers aged 25-34 whose highest education level was grades 9-11, some college, or a bachelor's degree or higher, compared to those whose highest education was a high school diploma or GED: 1970-1998. SOURCE: National Center for Education Statistics (2000, p. 144~. —- - - Bachelor's degree or higher 9 ~ ° ~ Some college 2.00 - 1.50 o ._ a) ~ 1.00 . _ 0.50 0.00 _ ~ am, —Grades9-11 in_ _ ,,.%, _ _ _ ~ _ ^_ ~ ~ _ 1 ~ 9 ~ is 9 ~ ~ # A ~9 ~ ~ 4'~ ~ o p of it, sap ~ ~ ~ ~ ,~ 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 Year FIGURE 1-4b Ratio of median annual earnings of female wage and salary workers aged 25-34 whose highest education level was grades 9-11, some college, or a bachelor's degree or higher, compared to those whose highest education was a high school diploma or GED: 1970-1998. SOURCE: National Center for Education Statistics (2000, p. 144~. 20 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

ment rates fluctuated, suggesting that the labor market may have (at least temporarily) met its demand for college-educated workers. Of course, many factors in addition to the wage premium can affect college enrollment rates, and some factors (such as a growing high- skill economy) may drive both measures in the same direction. None- theless, these trends seem to suggest that labor market demand for a college education was particularly strong during the 1980s and may have leveled off in the late l990s. Enrollments Among Types of Postsecondary Institutions The overall increase in college enrollments in the last three decades has occurred within public institutions, private institutions, four-year institutions, and two-year institutions (see Figure 1-2~. During the 1970s, as the community college system grew, enrollment increases were larger at public rather than private institutions and at two-year rather than four-year institutions. As a result of these changes, from 1970 to 1980, public institution enrollments increased from 75 percent to 78 percent of all postsecondary enrollments, and four-year institution enrollments decreased from 73 percent to 63 percent of all enroll- ments. Since 1980, the share of enrollments at four-year institutions has dropped only slightly to 62 percent, and the share at public insti- tutions has not changed. Thus, the last decade has been characterized by fairly stable enrollment shares across public and private institutions and across four-year and two-year institutions. Table 1-1 provides a more detailed look at enrollments in the four major types of postsecondary institutions public four-year, public two-year, private four-year, and private two-year in 1981 and 1998. (This time period was selected because the criteria NCES uses to define the two-year sector have changed over time, such that data on two-year institutions before 1981 are not comparable with data in the TABLE 1-1 Fall Enrollments in Postseconclary Institutions and Distribution of Enrollments among Institutions, by Type of Institution: ~ 98 ~ and ~ 998 1981 1998 Type of institution Enrollment % Distribution Enrollment % Distribution Public 4-year 5,166,324 41.8 5,903,837 40.6 Public 2-year 4,480,708 36.2 5,272,347 36.2 Private 4-year 2,489,137 20.1 3,128,908 21.5 Private 2-year 235,503 1.9 244,097 1.7 All institutions 12,371,672 100.0 14,549,189 100.0 SOURCE: Snyder (2001, p. 203~. NOTE: The 1998 data are for degree-granting institutions. Data in 1981 were not available for degree-granting institutions, so data in this year represent two-year and four-year institutions of higher education. In any given year, enrollment estimates for these two types of institutions differ by about 1 percent. LISA HUDSON 21

1990s.) This table shows, first of all, that enrollments at public institutions are almost evenly split between those at the four-year level and those at the two-year level, while enrollments at private institutions are heavily concentrated at the four-year level. From the other perspective, public four-year institutions enroll about twice as many students as private four-year institutions, while public two-year institutions enroll more than 10 times as many students as their private counterparts. From 1981 to 1998, there was little change within the two-year sector; public institutions' share of enrollments increased only slightly from 95 percent to 96 percent. Over the longer time period from 1970 to 1997, there was a shift within the four-year sector from public nshtubons to private ~nshtubons, resulting In a decline In public ~nshtubons share of four-year enrollments from 74 percent to 65 percent. This relative growth in the private four-year sector is somewhat surprising, since private four-year institutions have had higher tuition increases in the past three decades than have public institutions. However, the shift from public to private institutions occurred only during the 1970s and 1990s, not the 1980s. It was during the 1980s that private four- year institutions increased in tuition costs relative to public four-year institutions; their relative costs have declined since 1990-1991 (see Snyder, 2001, and Table 1-10 later in this chapter). So the public-to- private shift seems to reflect a combination of increasing interest in attending private four-year institutions combined with the mitigating influence of relative costs. In sum, except for a shift from public to private four-year institu- tions, there has been little change in the past two decades in the type of institution in which students (in general) enroll. If past trends are any guide, private four-year institutions may increase their enroll- ments relative to public four-year institutions in the future if their relative tuition costs do not increase. Who Enrolls in Colleges The modal college student in 1970 was a young (under age 24) White male who attended school full-time; today the modal college student looks much the same, except she is female. Attending school along with today's modal student is a more diverse student body. College students vary more now than they did during the 1970s in age (as the population has aged and more older students have enrolled in college), minority composition (reflecting population changes and en- rollment growth among some minorities), and attendance status (as more students have enrolled part-time). Most of these changes in the composition of the student body occurred during the 1970s, rather than during the 1980s or 1990s. 5All data in this section are student counts. Except for Table I-2, all enrollment data are from the U.S. Department of Education's Integrated Postsecondary Educa- tion Data System (IPEDS). IPEDS includes all students enrolled in for-credit courses; it excludes students enrolled only in noncredit or continuing education courses. 22 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

| ~ age 14-24 ~ age 25-34 ~ age 35 or older | 1 998 1 990 1 980 1 970 21.1 23.9 7 18 2 0 25 50 75 100 Percentage of enrollments FIGURE 1-5 Distribution of fall enrollments in degree-granting institutions, by age group: 1970, 1980, 1990, and 1998. SOURCE: Data from Snyder (2001, p. 204~. As shown in Figure 1-5, the proportion of young college students (under age 25) declined from 72 percent to 63 percent during the 1970s. This change was driven by increasing enrollment rates among older students and by the changing age distribution of the population. Although this proportion declined further (to 58 percent) during the 1980s, by 1998 the proportion of young college students had in- creased slightly to 59 percent (reflecting primarily the strong increases in enrollment rates among recent high school graduates seen earlier in Figure 1-3~. The projected increases in the size of the college-age population (discussed above as the "young adult population") and increasing enrollment rates among these college-age adults (Figure 1-3) suggest that future college enrollment growth may be concentrated among younger students rather than older students, as was the case during earlier decades. But labor market trends, discussed in a later section, may increase the enrollments of older students, assuming postsecondary institutions are seen as a viable alternative for worker training; this is a more difficult trend to predict. The 1970s also saw a relatively large shift in enrollments from male to female and from full-time to part-time (Figures 1-6 and 1-7~. Female enrollments rose from 41 percent to 51 percent of all enroll- ments during this decade, while part-time enrollments increased from 32 percent to 41 percent. Since 1980, however, the proportion of students who are female has increased only from 51 percent to 56 percent, and the proportion of students attending part-time has remained LISA HUDSON 23

1 998 1 990 1 980 1 970 o Male Female 50 Percentage of enrollments FIGURE 1-6 Distribution of fall enrollments in degree-granting institutions, by sex: SOURCE: Data from Snyder (2001, p. 204~. 1 998 1 990 1 980 1 970 75 100 1970, 1980, 1990, and 1998. o Full-time Part-time 50 Percentage of enrollments 100 FIGURE 1-7 Distribution of fall enrollments in degree-granting institutions, by attendance status: 1970, 1980, 1990, and 1998. SOURCE: Data from Snyder (2001, p. 204~. 24 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

fairly constant. Thus, in terms of age, sex, and attendance status, there has been relatively little change in the student body over the last two decades, even as enrollments have increased. The growing predominance of females in college has created con- cern in some quarters about the absence of males (e.g., Mortenson, 1999~. A closer look at the enrollment rates of males and females in postsecondary education suggests that males are increasingly less likely than females to enroll in college (Table 1-2), but it is unclear how this finding should be interpreted. Male college enrollment rates were artificially high during the late 1960s and early 1970s, when young men could receive a draft deferment by enrolling in college. The 1970s also ushered in the women's movement, with changing mores that have, over time, made it more acceptable for young women to postpone marriage and childbearing in order to further their educa- tion. So the higher enrollment rates of males relative to females in the 1970s reflect both an inflated rate for males caused by the Viet- nam War, and a deflated rate for women caused by more traditional mores. While both male and female enrollment rates have been increas- ing since 1980 (for females since 1970), the increase has been larger for females than for males, so that by 1999, females' enrollment rate was 1.4 percentage points higher than the rate for males. One might ~ _ ~ _ _ assume that "equal access" implies that male and female college par- ticipation rates should now be equal. However, to the extent that the labor market provides good-paying jobs that do not require a college education in traditionally male-dominated fields (such as construc- tion), the labor market may serve as a greater deterrent to college attendance for males than for females. Differences in pay between males and females may also encourage females to obtain more educa- tion in order to have earnings that match those of males. (In 1998, among those with income, males aged 16-24 and 25-34 earned more than females in the same age groups tU.S. Census Bureau, 20001.) In sum, it is difficult to tell whether sex differences in pay or job oppor- tunities (or job interests) lead to these differential participation rates, or whether the lower rate of males reflects a more systemic educa- tional problem facing males in today's society. These uncertainties about the causes and interpretations of these trends make it difficult to predict whether and to what extent females' participation rate will continue to outstrip that of males. TABLE 1-2 Enrollment in Two- and Four-Year Colleges of Persons Aged 14-34, by Sex: October ~ 970, ~ 975, ~ 980, ~ 990, and ~ 999 Sex 1970 1975 1980 1990 1999 Male 14.6 14.9 13.0 13.6 15.0 Female 9.2 11.6 12.8 14.4 16.4 SOURCES: Current Populations Survey, October supplement, 1999; U.S. Census Bureau, special tabulations. LISA HUDSON 25

Id a) |~ White Al Black ~ Hispanic ~11 Asian ~ Native American | 1 997 1 990 1 980 1 976 0.7 ~ 0.7 ~ o 50 Percentage of enrollments 75 100 FIGURE 1-8 Distribution of fall enrollments in degree-granting institutions, by race/ethnicity: 1976, 1980, 1990, and 1997. SOURCE: Data from Snyder (2001, p. 236~. The racial/ethnic composition of the student body has changed significantly during the last two decades (Figure 1-8).6 From 1980 to 1997, White students aged 14-34 declined from 84 percent to 73 per- cent of all enrollments; this decline was accompanied by increases among all minority groups, but particularly among Hispanics and Asians. For example, from 1980 to 1997, the proportions of students who were Black or Native American increased only slightly (from over 9 percent to 11 percent, and from 0.7 percent to 1 percent, respec- tively), while the proportions who were Hispanic or Asian each more than doubled. These changes in student-body composition partly reflect changes in the young adult population. Over the same time period, the propor- tion of 14-34 year olds who were White declined from 81 percent to 66 percent, the proportion of Blacks increased from 10 percent to 13 percent, and the proportion of Hispanics and other minorities more than doubled. The increases for Blacks and other non-Hispanic minorities also reflect increasing college participation rates over time. In fact, the proportion of Black students enrolled in postsecondary education did not increase until the 1990s, the decade in which their participa- tion rate increased (Table 1-3~. Among other non-Hispanic minorities,7 representation in postsecondary education and participation rates increased 6Since Hispanics were not separately identified in federal data collections until the mid-1970s, data by race/ethnicity cannot be provided for 1970. 7The other minority category in Table 1-3 includes Asians and Native Americans. In 1980, 72 percent of this combined group consisted of Asians; by 1999, 82 percent of the group consisted of Asians. 26 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

TABLE 1-3 Enrollment in Two- and Four-Year Colleges of Persons Aged 14-34, by Race/Ethnicity: October ~ 975, ~ 980, ~ 990, and ~ 999 Race/ethnicity 1975 1980 1990 1999 White 13.7 13.5 14.8 16.6 Black 10.7 10.4 10.7 14.6 Hispanic 9.9 8.2 6.7 9.6 Other minority 17.3 17.0 18.8 23.0 SOURCES: Current Populations Survey, October supplement, 1999; U.S. Census Bureau, special tabulations. during both the 1980s and l990s. The participation rate for Hispanics, however, did not increase during these decades, suggesting that their increased representation in postsecondary education reflects only their increased representation in the population. The stagnant participation rate of Hispanics in postsecondary education may be due partly to changes in the size and composition of the immigrant population over time. The main areas from which the United States attracts Hispanic immigrants are Mexico, Cuba, the Dominican Republic, Central America, and South America (Guzman, 2001~. The rate of immigration from these countries grew dramatically during the 1980s and l990s, from 147,880 annual immigrants in the 1970s to 405,737 annual immigrants in the l990s (U.S. Census Bureau, 1992, 2000~. As a result, the percentage of the U.S. Hispanic popula- tion that is foreign born increased from 25 percent in 1980 to 39 percent in 2000. Hispanic immigrants are also increasingly likely to come from Mexico and Central America, rather than other countries. During the 1970s, immigrants from Mexico and Central America comprised 52 percent of immigrants from the countries listed above; during the l990s (up to 1998), they comprised 72 percent of these immigrants (U.S. Census Bureau, 1992, 2000~. The net result of these racial/ethnic group trends suggests mixed progress in minority participation in postsecondary education. To sum- marize this progress, Table 1-4 presents an indicator of the extent to which each racial/ethnic group was overrepresented or underrepresented in postsecondary education in 1975 and 1999. As the table shows, Blacks have achieved a more equitable participation in postsecondary education over time, as their underrepresentation has shrunk in size, and Asians (the predominant group in the other minority category) have increased their overrepresentation in postsecondary education. Hispanics, however, not only remain underrepresented in postsecondary education but also were more underrepresented in 1999 than in 1975. Finally, Table 1-5 shows that the increase in female enrollments from 1976 to 1997 reflects increases in female participation among all racial/ethnic groups. However, the nature of the increase varies by racial/ethnic group. Among Blacks, enrollments have been predomi- LISA HUDSON 27

TABLE 1-4 Index of Degree, Those Aged 14-34, Racial/Ethnic Groups are Over- or Unclerrepresentec3 in Two- and Four-Year Colleges: ~ 975 and ~ 999 Index Racial/ethnic group 1975 1999 White 0.04 0.07 Black -0.19 -0.08 Hispanic -0.25 -0.41 Other minority 0.29 0.43 SOURCES: Current Population Survey, October supplement, 1999; U.S. Census Bureau, special tabulations. NOTE: An index value greater than 0 signifies overrepresentation in two- and four-year colleges; a value lower than 0 signifies underrepresentation. The index was calculated by dividing the proportion of college students aged 14-34 who are in a racial/ethnic group by the proportion of the population aged 14-34 in the racial/ethnic group and subtracting one from that value. TABLE 1-5 Percentage of Fall Enrollments in Degree-Granting Institutions Who Are Male, by Racial/Ethnic Group: ~ 976, ~ 980, ~ 990, and ~ 997 Race/ethnicity 1976 1980 1990 1997 White 53.0 48.5 45.3 44.3 Black 45.5 41.9 38.9 37.3 Hispanic 54.6 49.1 45.2 43.2 Asian 54.8 52.8 51.5 48.6 Native American 50.6 45.1 41.9 41.5 SOURCE: Snyder (2001, p. 237~. nantly female since at least 1976 and have become more so over time. Among Native Americans, enrollments have shifted from a fairly eq- uitable sex division to enrollments that are predominantly female. Finally, among Whites and Hispanics, enrollments have shifted from predominantly male to predominantly female, while among Asians the shift has been from predominantly male enrollments to a fairly equitable sex division. The net result is that as of 1997 females were over- represented among the enrollments of all racial/ethnic groups except Asians. While equity concerns typically focus on females, in this situation it is the continued and significant underrepresentation of Black males in postsecondary education relative to other racial/ethnic groups as well as to females that is particularly troubling. This underrepresentation seems to reflect larger problems in the lives of American Black males that are not, by this indicator, being resolved over time. (Also see the Annuals of the American Academy of Political and Social Science, 569, May 2000, a special issue devoted to the American Black male.) 28 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

100 - 80- - o ~ 60- a `0 40- c) 20 - O- . 4.9 ._ ~! 39 40 1976 1980 1990 1997 50 53 55 L 52 55 57 57 58 57 54 48 _ .... _ ~ . ~ Public 4-year Private 4-year Public 2-year Private 2-year FIGURE 1-9 Percentage of fall enrollments in degree-granting institutions who are female, by institution type: 1970, 1980, 1990, and 1997. SOURCE: Data from Snyder (2001, p. 208~. Trends Among Institutions Looking more closely at enrollments in the four basic types of postsecondary institutions helps clarify the nature of the trends noted above. This section examines patterns and trends in the gender and racial/ethnic composition of students in each type of institution. The percentage of females enrolled does not vary much among different types of institutions, although enrollments at public two- year institutions are slightly more female than are enrollments at other types of institutions (57 percent versus 54-55 percent, respec- tively; as shown in Figure 1-9~. Although the percentage of students who are female increased within each type of institution from 1970 to 1997 with the largest increases in each case occurring during the 1970s it is only among four-year institutions that female enrollments continued to increase (relative to male enrollments) during the l990s. Among two-year institutions, the proportion of female students remained constant or declined during the l990s. It is not clear why the proportion of female enrollments leveled off in two-year institutions; this level- ing off could reflect (among other things) a movement of females from two-year to four-year programs and/or a shift in offerings within two-year institutions toward courses and programs that are more appealing to males than to females. Two-year institutions, in accordance with community colleges mission of opening access to postsecondary education (e.g., through , lower tuition rates, greater geographic accessibility, and more relaxed admissions standards), enroll a higher proportion of minority students LISA HUDSON 29

35 - 30 - tn cam ~ 25- o 0, a' 1 0 20 - 15 - 5 o ~ 1 976 · 1 980 ~ 1 990 1~ 1 997 14 _ ~9 15 32 Public 4-year Private 4-year 22 20 20 Public 2-year 20 Private 2-year FIGURE 1-10 Percentage of fall enrollments in degree-granting institutions who are minority, by institution type: 1976, 1980, 1990, and 1997. SOURCE: Data from Snyder (2001, p. 236~. than do four-year institutions (31-32 percent versus 22-24 percent, respectively; as shown in Figure 1-10~. However, within the four-year and two-year sectors, public and private institutions enroll roughly equivalent percentages of minority students. This overall enrollment pattern has not changed over time, as increases in minority represen- tation have been roughly equivalent across all four types of institu- tions. Looking at specific minority groups (Figures 1-11 through 1-13) shows that within the four-year sector, public and private institutions enroll similar proportions of Black, Hispanic, and Asian students, cur- rently and over time.8 But within the two-year sector, minority group enrollments vary among public and private institutions. Private two- year institutions enroll a larger proportion of Black students than do public two-year institutions, while the opposite is true for Hispanic and Asian students. Taken together, these data show that the higher representation of minority students in the two-year sector is limited to the higher representation of Black and Hispanic students within these institutions in particular, to the relatively high representation of Blacks in private two-year institutions and Hispanics in public two-year institutions. As noted above, minority enrollments increased during both the 1980s and l990s. For Hispanics and Asians, this trend appears to result from their increased representation in all types of institutions Because of their low representation in postsecondary education, Native American students are not examined separately in this section. 30 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

35 - 30 - E 25 o a) - o a, 15- a) a, 10 20 - 5 o 1 976 1 980 1 990 1 997 16 11 Public 4-year Private 4-year 16 11 11 10 10 Public 2-year 12 Private 2-year FIGURE 1-11 Percentage of fall enrollments in degree-granting institutions who are Black, by institution type: 1976, 1980, 1990, and 1997. SOURCE: Data from Snyder (2001, p. 236~. 35 30 In ~, 25 - o cat 10 O Public 4-year Private 4-year Public 2-year 1976 1980 1990 1997 20 10 Private 2-year FIGURE 1-12 Percentage of fall enrollments in degree-granting institutions who are Hispanic, by institution type: 1976, 1980, 1990, and 1997. SOURCE: Data from Snyder (2001, p. 236~. LISA HUDSON during both decades, but with an especially large increase in the representation of Hispanics within both public and private two-year institutions (Figures 1-12 and 1-13~. Figure 1-11 also shows that the overall increase in the proportion of Black enrollments during the l990s reflects an increasing representation of Black students in four- year institutions and public two-year institutions. The proportion of 31

35 - . 30 - - o o ~ 15- Cal 25 - 20 - 10 - 4 6 6 6 1 970 1 980 1 990 1 997 Public 4-year Private 4-year Public 2-year Private 2-year FIGURE 1-13 Percentage of fall enrollments in degree-granting institutions who are Asian, by institution type: 1976, 1980, 1990, and 1997. SOURCE: Data from Snyder (2001, p. 236~. students enrolled in private two-year institutions who were Black de- clined during the l990s, suggesting that during this decade Black student enrollments may have shifted from private two-year institu- tions to public two-year and four-year institutions. In summary, over the past two to three decades, changes in college enrollments have largely reflected the changing minority composition of the American population. However, the increasing representation of women, Blacks, and Asians reflects both an increasing rate of par- ticipation among these groups and (for the two minority groups) their growing representation in the adult population. The growth of Hispanics in postsecondary education, on the other hand, reflects only their increase in the population, not an increasing rate of participation in postsecondary education. Increases also occurred at different points in time for each group. While the 1970s was the primary decade of enrollment growth for women, the 1980s and l990s were growth decades for Hispanics and Asians, and only the l990s were a decade of growth for Blacks. Regardless of their race or sex, most of these students enrolled in postsecondary education to earn a degree. Degree awards are exam- ined next. DEGREES AWARDED Bachelor's degrees made up 52 percent of all degrees awarded by postsecondary institutions in 1997-1998. The next most commonly awarded degree was the associate degree, constituting 24 percent of 32 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

all degrees, followed by the master's degree (19 percent), first pro- fessional degrees (3 percent), and doctoral degrees (2 percent). Over- all, public institutions award the majority of degrees, 67 percent of all degrees in 1997-1998. With the exception of doctoral degrees, the proportion of degrees awarded by public institutions compared to private institutions declined as the level of the degree increases. For example, in 1997-1998, public institutions awarded 81 percent of associate degrees, 66 percent of bachelor's degrees, 55 percent of master's degrees, and 40 percent of first professional degrees. As will be seen below, this is a change from past decades, when public insti- tutions awarded about two-thirds of all master's degrees. As one would expect given rising enrollments, the number of degrees awarded increased from the 1970s through the 1990s (Table 1-6), with an overall increase of 65 percent from 1970-1971 to 1997- 1998. (Over the same time period, enrollments increased 67 percent.) The marked growth in associate degrees over this period (121 percent) is the result of both the expansion of the community college system during the 1970s and the addition of for-profit private schools to TABLE 1-6 Number of Degrees Awarclec3 by Degree-Granting Postseconclary Institutions and Percentage Change in Number of Degrees Awarclecl, by Type of Institution and Level of Degree: ~ 970-197 i, ~ 982-1983, and ~ 997-1998 Number of degrees awarded in: % change in number of degrees, Institution and degree type 1970-1971 1982-1983 1997-1998 1982-1983 to 1997-1998 All institutions Associate degree 252,311 449,420 558,555 24 Bachelor's degree 839,730 969,510 1,184,406 22 Master's degree 230,509 289,921 430,164 48 Doctoral degree 32,107 32,775 46,010 40 First professional degree 37,946 73,054 78,598 8 Total degrees 1,392,603 1,814,680 2,297,733 27 Public institutions Associate degree 215,645 377,617 455,084 21 Bachelor's degree 557,996 646,317 784,296 21 Master's degree 151,603 176,246 235,922 34 Doctoral degree 20,788 21,186 29,715 40 First professional degree 16,139 29,757 31,233 5 Total degrees 962,171 1,251,123 1,536,250 23 Private institutions Associate degree 36,666 71,803 103,471 44 Bachelor' s degree 281,734 323,193 400,110 24 Master's degree 78,906 113,675 194,242 71 Doctoral degree 11,319 11,589 16,295 41 First professional degree 21,807 43,297 47,365 9 Total degrees 430,432 563,557 761,483 35 SOURCE: Snyder (2001, p. 306~. LISA HUDSON 33

NCES' postsecondary data collection system. Since the second factor artificially inflates counts of associate degrees and private institu- tions' share of these degrees, trends by degree level are examined in Table 1-6 starting in 1982-1983, the year after for-profit schools were fully incorporated into the data system. The trend data from 1982-1983 to 1997-1998 show that the number of degrees awarded over this time period increased 27 percent. Although increases occurred at all degree levels and at both public and private institutions, increases were most notable in the private sector and among master's and doctoral degrees. The greater growth in degree awards within the private sector is due primarily to rela- tively large increases in the number of associate and master's degrees awarded by these institutions. From 1982-1983 to 1997-1998, the number of associate degrees awarded by public institutions increased 21 percent while the number awarded by private institutions increased 44 percent. Similarly, the number of master's degrees awarded by public institutions increased 34 percent, compared to 71 percent in private institutions. As a result, public institutions' share of associate degrees declined over this period from 84 percent to 81 percent, and their share of master's degrees declined from 61 percent to 55 percent (Table 1-7). Because of these declines, the total percentage of degrees awarded by public institutions declined slightly from 69 percent to 67 percent from 1982-1983 to 1997-1998. As shown below, business degrees account for much of the increase in awards at the master's degree level, and employers are particularly likely to financially support students in these programs. Thus, this relatively large increase in the private sector may result from an increase in employers' willingness to pay for their employees to attain MBAs (making private institutions more affordable to individuals) or because private institutions are increasingly targeting their programs to local business needs. The relatively strong growth among private institutions in associate degrees (which are largely vocational in nature) lends support to the second hypothesis. TABLE 1-7 Percentage of All Degrees Awarclec3 by Degree-Granting Institutions That Were Awarclec3 by Public Institutions, by Level of Degree: 1982-1983 and 1997-1998 Level of degree 1982-1983 1997-1998 Associate degree 84 81 Bachelor's degree 67 66 Master's degree 61 55 Doctoral degree 65 65 First professional degree 41 40 All degrees 69 67 SOURCE: Snyder (2001, p. 306~. 34 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

TABLE 1-S Number of Associate, Bachelor's, and Master's Degrees Conferred by Degree-Granting Institutions: ~ 997-1998 Field of Study Associate degrees Bachelor' s degrees Master's degrees Agriculture/natural resources Architecture/related programs Area, ethnic, and cultural studies Biological/life sciences Business Communications Communications technologies Computer and information sciences Construction trades Education Engineering Engineering technologies English language/literature Foreign languages/literatures Health professions/related programs Home economics Law and legal studies Liberal arts/humanities/general studies Library science Mathematics Mechanics/repair Multi/interdisciplinary studies Parks/recreation studies Philosophy and religion Physical sciences and science technologies Precision production Protective services Psychology Public administration ROTC/military technologies Social sciences and history Theological studies/religious vocations Transportation/material moving Visual and performing arts Not classified by field of study Total 6,673 265 104 2,113 104,659 2,368 1,602 13,870 2,172 9,278 2,149 32,748 1,609 543 92,031 8,292 7,797 186,248 96 844 10,616 9,401 895 94 2,286 11,085 19,002 1,765 4,156 22 4,196 570 1,009 14,980 3,017 558,555 23,284 7,652 6,153 65,868 233,119 49,385 729 26,852 182 105,968 59,910 13,727 49,708 14,451 84,379 17,296 2,017 33,202 73 12,328 91 26,163 16,781 8,207 19,416 407 25,076 73,972 20,408 3 125,040 5,903 3,206 52,077 1,373 1,184,406 4,475 4,347 1,617 6,261 102,171 5,611 564 11,246 16 114,691 25,936 1,136 7,795 2,927 39,260 2,914 3,228 2,801 4,871 3,643 o 2,677 2,024 1,307 5,361 15 2,000 13,747 25,144 o 14,938 4,692 736 11,145 868 430,164 SOURCE: Snyder (2001, p. 306). FIELDS OF STUDY Table 1-8 shows that certain fields of study predominate at the three most common degree levels (associate, bachelor's, and master's).9 9Because of their relatively low frequency, first professional and doctoral degrees are not discussed in the remainder of this section. LISA HUDSON 35

At the associate degree level, the two predominant degrees are liberal arts/humanities/general studies and business; these two fields repre- sent half of all associate degrees awarded in 1997-1998. At the bachelor's degree level, three disciplines predominate: business, the catch-all category of social sciences and history, and education. These three areas include almost 40 percent of all bachelor's degrees. At the master's degree level, business and education predominate, accounting for half of all master's degrees. Most notable in these figures is the predominance of business degrees. Business is by far the most common degree awarded at the bachelor's level and is the second most common degree at the associate and master's levels. The different roles of public and private institutions are reflected in the level and types of degrees each awards. As noted above, public institutions granted 81 percent of the associate degrees awarded in 1997-1998. This relatively high proportion indicates the strong role played by community colleges at this degree level. Their role as a transfer institution is also reflected in the field of study for their degree awards. Fully 95 percent of the associate degrees awarded in liberal arts/humanities/general studies in 1997-1998 were awarded by public institutions; this field of study accounted for almost 40 percent of all the associate degrees awarded by public institutions (Snyder, 2001). Also of note is the relatively greater emphasis on education and lesser emphasis on business in public rather than private institutions. While public institutions awarded 81 percent of all associate degrees, they awarded 86 percent of associate degrees in education and 69 per- cent of associate degrees in business. Likewise, while public institu- tions awarded 66 percent of all bachelor's degrees, they awarded 74 percent of bachelor's degrees in education and 59 percent of bachelor's degrees in business. Finally, while public institutions awarded 55 percent of all master's degrees, they awarded 62 percent of education degrees at this level and 39 percent of business degrees. This difference in orientation is important because, as discussed below, the trend in the past few decades has been a shift away from education degrees toward busi- ness degrees. Trends in Fields of Study Historical data on fields of study are not available for associate degrees~° and are only available for selected bachelor's and master's degree disciplines. Even these limited data, however, provide an inter- esting view of how college majors are changing over time. From 1970-1971 to 1997-1998, increases in the number of bachelor's and master's degrees were driven in good measure by the large and fast-growing field of business (Table 1-9). Bachelor's degrees awarded lo, 1 n~ a, 1Vlhe N(~S classification system for associate degrees has changed over times so associate degree fields are not comparable across years. 36 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

TABLE 1-9 Number of Bachelor's and Master's Degrees Awarclec3 in Selected Field s of Study: ~ 970- ~ 97 ~ and ~ 997- ~ 99 Field of study Number of bachelor's degrees in: % 1970-1971 1997-1998 change Number of master's degrees in: % 1970-1971 1997-1998 change Computer/information sciences Communications Public administration Health professions Business Psychology Agriculture/natural resources Biological/life sciences Visual/performing arts ~ ~ · . ng~neer~ng Architecture/related programs Physical sciences/science technologies Social sciences/history English language/literature Foreign languages Education Mathematics Total number of degrees 2,388 10,802 5,466 25,226 114,729 38,187 12,672 35,743 30,394 50,046 5,570 21,412 155,324 64,342 19,055 176,307 24,937 839,730 26,852 1,024 50,114 364 20,408 273 84,379 234 233,119 103 73,972 94 23,284 84 65,868 84 52,077 71 73,910 48 7,652 37 19,416 -9 125,040 -20 49,708 -23 12,769 -33 105,968 -40 12,328 -51 1,184,40 41 1,588 1,770 7,785 5,749 25,977 5,717 2,457 5,728 6,675 16,309 1,705 6,367 16,539 10,686 5,217 87,666 5,695 230,509 11,246 5,611 25,144 39,260 102,171 13,747 4,475 6,261 11,145 25,936 4,347 5,361 14,938 7,795 2,927 114,691 3,643 430,164 608 217 223 583 293 140 82 9 67 59 155 -16 -10 -27 -44 31 -36 87 SOURCE: Snyder (2001, pp. 295-2961. in business more than doubled, increasing by over 100,000. At the master' s degree level, the number of business degrees almost quadrupled, increasing by over 75,000. Nonetheless, a broad range of fields increased at a faster-than-average rate over this time period, including (in addi- tion to business) computer/information sciences, communications, public administration, health professions, and psychology. Even during this period of overall growth, several degree fields declined in number. These declining fields of study include physical sciences/science technology, social sciences/history, English language/ literature, foreign languages, mathematics, and education (at the bachelor's level only). In general, the shift in the past three decades appears to be away from the humanities and hard sciences toward business, technical, and health fields. This shift in degree awards is consistent with the view that the economy is shifting toward an information services economy and with recent growth in the management, technical, and health care sectors of the labor market (see, e.g., Bureau of Labor Statistics [BLS], 2000 and "The Labor Force" section of this chapter). This shift also could be interpreted as signaling a growing vocationalism in postsecondary education. LISA HUDSON 37

THE ROLE OF TECHNOLOGY New information technologies have the potential to change postsecondary education through three main avenues: virtual univer- sities (universities that offer courses exclusively through the Internet), distance education (courses delivered to remote locations via audio, video, or computer technologies), and new instructional practices offered within traditional settings (see Bransford, Vye, and Bateman, Chapter 6 of this volume for examples). Virtual universities and new instructional practices are not well . covered or well identified in current federal data collection systems; other chapters in this publication examine these issues in more detail. NCES has, however, initiated a series of regular data collections on distance education. Currently, two special- focus surveys have been done on this topic, and a third survey is underway. The results of the first two surveys are briefly summarized here. NCES' first survey on distance education examined the nature and prevalence of distance education among higher education institutions in 1994-1995; the second examined distance education among all two-year and four-year postsecondary institutions in 1997-1998. ~ ~ The more recent survey revealed that about one-third of all postsecondary insti- tutions offered at least one distance education course in 1997-1998, and that many of the institutions that did not planned to do so in the near future. However, half of the institutions stated that they did not offer any distance education programs and did not plan to within the next three years (Lewis, Snow, Farris, and Levin, 1999~. Public institutions were more likely than private institutions to offer distance education courses; about 80 percent of all distance edu- cation courses in 1997-1998 were offered by public institutions. Smaller institutions were also less likely than larger institutions to offer distance education courses. Since private institutions are smaller on average than public institutions, private institutions may be less likely to offer distance education because their smaller size makes them less likely to have the institutional capacity to support this type of instruction- or perhaps their smaller size indicates a greater emphasis on a personal approach to education that is inconsistent with the use of distance education. From 1994-1995 to 1997-1998, there was a significant increase in the use of distance education among higher education institutions. Over this three-year period, distance education course offerings doubled, and distance education degree and certificate programs nearly doubled. Much of this increase was among higher education institutions that had already offered distance education in 1994-1995. As a result, the 1lAccording to the NCES definitions used in these surveys, institutions of higher education are a subset of postsecondary institutions. 12In fall 1997, the average enrollment at public institutions was 6,530 students; the average enrollment at private institutions was 1,343 (Snyder, 2001, pp. 202 and 286~. 38 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

percentage of higher education institutions offering distance educa- tion courses increased by one-third, and there was no increase in the number of institutions offering distance education programs. TRENDS IN POSTSECONDARY COSTS AND STUDENT FINANCIAL AID Extensive analyses have been conducted on college costs and student financial aid (e.g., Berkner, 1998; McPherson and Schapiro, 1998~. This chapter merely notes some of the key trends in these financial measures over time and some indicators of their potential effect on postsecondary students. This section focuses on the federal financial aid system, which provides about 70 percent of all student financial aid. (Private foundations, postsecondary education institutions, and state governments are the other main providers of student aid.) Concerns about the effects of college costs on postsecondary access and completion have grown in recent years, as costs escalated during the 1980s and 1990s (Table 1 - 10), outpacing both family income and increases in federal financial aid programs (College Board, 2000a, 2000b). Rising college costs not only increase the need for student financial aid programs, they also raise the costs of these programs, particularly when the demand for postsecondary education also increases as has been the case in the past two decades. One way the federal govern- ment has dealt with these rising program costs has been to shift from a grant-based system to a less costly loan-based system. This trend was exacerbated in 1992, when federal legislation increased the ceiling on federal loan amounts and loosened the needs assessment require- ments for loan eligibility. Thus, as Figure 1-14 shows, in the 1980s, average grant aid and loan aid per (full-time equivalent) student were TABLE 1-10 Average Unclergracluate Tuition and Required Fees Paid by Full-Time Equivalent Students in Degree-Granting Institutions, by Type of Institution: Selected Years from ~976-1977 to ~999-2000 Public institutions Private institutions Year Two-year Four-year Two-year Four-year 1976-1977 $ 829 $1,807 $4,662 $ 7,420 1980-1981 792 1,628 4,886 7,324 1985-1986 991 2,039 5,679 9,467 1990-1991 1,052 2,410 7,111 11,596 1995-1996 1,355 3,114 7,756 13,386 1999-2000 1,336 3,351 8,107 14,690 SOURCE: Snyder (2001, pp. 344-345). NOTE: Tuition amounts have been adjusted to constant 1999 dollars. LISA HUDSON 39

—Average total aid per FTE ~ Average grant aid per FTE 7000 ~ . 6000 - 5000 - 4000 - 3000 - 2000 - 1 000 - —Average loan aid per FTE ......................................................................................................................................................................................................................................................... - ~C _ _ _' _ _ _ ~ ~— .................. ,~s O ~ - A.................. Fib A O- l l l l l l l l l l l l l l l l l l l l l l l l l l l l l 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 Year FIGURE 1-14 Average aid per full-time equivalent (FTE) student in constant dollars, overall and by type of aid: 1970-1999. SOURCE: Data from College Board (2000a) and Snyder (2001, p. 230~. about equal, but since 1992 average loan aid has increasingly out- weighed grant aid. So while the percentage of students who receive financial aid and the average amount of aid per student have increased (Snyder, 2001; College Board, 2000a), more aid has been awarded in the form of loans than grants. In the mid-1970s, about 20 percent of student aid dollars were distributed as loans, but by 1999, about 60 percent was in the form of loans (College Board, 2000a). In addition, real-dollar funding for the major federal need-based grant program (Pell grants) has declined so that the maximum Fell grant in 1999-2000 had a lower purchasing power than it had two decades earlier (College Board, 2000a). As noted by McPherson and Schapiro (1998), this shift from grants to loans has the net effect of shifting federal aid from support directed to low-income students to support that is more broadly targeted on middle- and upper-middle-income students. This shift is also supported by the newest federal initiatives to financially support students tuition tax credits. These forms of aid are even more "needs" neutral than student loans and thus represent a further targeting of student aid away from lower-income students toward middle- and upper-income students. These changes would be understandable if increases in college costs and declines in federal aid had negatively affected the ability of students from middle- and upper-income backgrounds to attend col- lege but, as explained below, there is no evidence that this occurred. 40 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

Effect on Student Access . These changes in college costs and federal financial aid have inevitably raised concern about their potential effects on students' access to college, persistence in college, and debt burden. On the surface, the evidence suggests no negative effects college enroll- ments have been increasing, even as tuition has gone up and grant aid has gone down. Part of the reason for this seeming lack of effect on access may be that the middle- and upper-income families that have traditionally supplied the majority of college students have not been significantly affected by these changes. Econometric studies (e.g., McPherson and Schapiro, 1998; Kane, 1995) have found that the distribution of student aid and high tuition costs adversely affect the enrollment of low-income students but have little effect on the enroll- ment of middle- and upper-income students. This differential effect can be at least partially explained by differences in the degree to which cost increases have eaten into family incomes. As the incomes of families at the lowest levels have remained relatively flat, increas- ing attendance costs have taken a proportionately bigger "bite" out of the pockets of lower-income families. From 1971-1972 to 1998-1999, attendance costs increased from 42 percent to 62 percent of the income of families in the lowest income quintile, but increased only from 13 percent to 16 percent of the income of families in the middle income quintile, and remained at a fairly steady 5 to 6 percent of the income of families in the highest income quintile (College Board, 2000b). McPherson and Schapiro also found that from 1980 to 1994 a constant proportion of entering freshmen from low-income families enrolled in public two-year colleges, while middle- and upper-income students increasingly chose to attend four-year schools. Specifically, from 1980 to 1994, the percentage of entering freshmen from all income groups who entered public two-year institutions declined from 36 percent to 31 percent, while the percentage of low-income fresh- men who entered public two-year institutions remained relatively constant at 46 percent in 1980 and 47 percent in 1994. These findings suavest .. . . . . .. . . ^. . . . . . .. ... .. that trends in college costs and ilnanclal aid may be llmltlng the choice of institutions only among low-income students, as well as decreasing their likelihood of entering any postsecondary institution. Another indicator of the selective effect of college costs and financial aid policies comes from data on students' immediate transition from high school to college. As seen in Figure 1-15, between 1980 and 1986, the immediate transition rate for Black and Hispanic students declined or remained constant while the rate for White students increased. As a result, the gap in immediate enrollment rates between Whites and each of these minority groups increased from the early 1970s to the late 1990s (NCES, 1999~. The divergence in trends during the 1980s might reflect the combination of rising college costs, the shift in federal aid from grants to loans, and an American economy in recession. Periods of economic decline may make the joint effects of LISA HUDSON 41

100 - 80 - o `, 60 =p 40 20 ~ O - —White Hispanic —Black l 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 Year FIGURE 1-15 Percentage of high school completers aged 16-24 who were enrolled in college the October after completing high school, by student race/ethnicity: 1972-1998. SOURCE: Data from National Center for Education Statistics (2000, p. 149~. NOTE: The 1973-1997 data for Black and Hispanic students are 3-year rolling averages. high college costs and restricted grant availability more noticeable or widespread than they would otherwise be. Student Debt Burden Concerns have also been raised about the effects of student finan- cial aid on students who can attend college only by assuming a sig- nificant amount of student loan debt. One concern is that the prospect of assuming a large amount of debt may keep some individuals from entering college or may force some students to leave college prema- turely. Another concern is that the debt assumed by students may have detrimental effects on their lives after leaving school. Unfortu- nately, data to assess these concerns are scarce. Recent news reports suggest that the debt burden assumed by young adults is a growing problem (e.g., USA Today, 2001~. Not all of a young adults debt burden is attributable to college debt, of course, but anecdotal ev~- dence suggests that for some young adults, college debt may be an important factor in their total debt burden. How much debt do students have? Systematic data on this topic are available from a study of 1992-1993 bachelor's degree recipients (Choy and Gels, 1997~. This study has two limitations that should be kept in mind. First, the focus on bachelor's degree recipients excludes (1 ) students who left school without completing a degree, some of whom 42 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

TABLE ~ - ~ ~ Percentage Distribution of ~ 992- ~ 993 B achelor' s Degree Recipients by Borrowing and Debt Status in 1994 Status as of 1994 % Never borrowed Borrowed Borrowed and now owe: $0 $1-5,000 $5,000-9,999 $10,000-14,999 $15,000-19,999 $20,000 or more 51 49 8 14 13 8 4 3 SOURCE: Choy and Geis (1997~. may have dropped out due to excessive debt burden, and (2) indi- viduals who have foregone college enrollment because of potential debt burden. Second, the study reflects the experiences of students who received financial aid before the 1992 changes in federal student aid policy that resulted in higher levels of student borrowing. Choy and Geis ~1997) found that about half of all 1992-1993 bachelor's degree recipients had borrowed money for their education, including borrowing from friends or relatives as well as through student loan programs (Table 1-11~; the average amount borrowed was $10,200. In 1994, about one year after graduation, 17 percent of these student borrowers no longer owed money, leaving 41 percent of 1992-1993 bachelor's degree recipients in debt; these former students owed, on average, $9,068. The amount owed varied widely, with 14 percent of all bachelor's degree recipients owing less than $5,000, 15 percent owing more than $10,000, and 3 percent owing more than $20,000. Choy and Geis also examined two potential effects of this debt burden on bachelor's degree recipients effects on career choices (whether those with more debt were compelled to seek higher salary jobs) and on further postsecondary education (whether those with more debt postponed or gave up on further study). There was no evidence that bachelor's degree recipients with higher amounts of debt gave more weight to income when seeking a job than did other students. Bachelor's degree recipients with more debt, however, were less likely to apply to graduate school or to enroll in further postsecondary study within a year of receiving their bachelor's degree. In addition, 10 percent of student borrowers who had considered but not applied to graduate school stated that high amounts of student debt were a factor in their decision to delay further education. Because student loan amounts have increased substantially since the time of the Choy and Geis study, it is likely that the amount of student debt has also increased substantially, and the effects of this LISA HUDSON 43

debt burden may be more extensive now than they were in this study. The next NCES survey of bachelor's degree recipients, in 2001, can be used to address this issue. A surprising feature of these relatively dramatic trends in postseconda~y costs and student aid is their lack of dramatic effect on postsecondary enrollments in general (excluding the tentative evidence of negative effects on low-income families). For example, while tuition increases and the shift in state and federal aid from institutions to students should be making the postsecondary education market more competitive, students seem to still be attending the same types of institutions as in the past (at least within the nonprofit sector), at ever-increasing rates. THE MILITARY OPTION If college costs too much or is otherwise unappealing to a high school student, the military provides a potential alternative choice.~3 Even though the military "takes" a relatively small proportion of young adults each year (roughly 200,000 new recruits annually), it is an important alternative because of its interest in "high-quality" recruits, and because of the training and education benefits it provides. The military attempts to attract the types of young adults that many col- leges and universities would like to enroll (thus competing with postsecondary education), but it also provides service members with a range of education benefits that may encourage their participation in postsecondary education. In fact, military surveys have found that over one-third of 16-21 year olds rate "pay for education" as the single most important reason to consider joining the military. For the most part, however, the military serves primarily as an alternative to entering the labor market, rather than as an alternative to college attendance. As is true for recruitment into any sector of the labor market, the ease of military recruitment depends on the state of the economy. When the economy is strong and jobs are plentiful, as has been the case in recent years, recruitment is more difficult; conversely, when the economy is in recession and jobs are scarce, recruitment is relatively easy. Military Recruitment Military enlistment requirements include an age requirement (generally ages 17-35) and education requirements. The education requirements differ for enlisted personnel (who constitute 85 percent of the military) and for officers (who constitute the remaining 15 percent). With few |3In this chapter, military refers to active-duty personnel in the Air Force, Army, Navy, and Marines. Members of the Reserve Forces and the Coast Guard (which is operated by the Department of Transportation in peacetime and by the Department of Defense in wartime) are not counted here as military. 44 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

exceptions, enlisted personnel must have completed high school. Interestingly, this education requirement does not exist to ensure that enlisted recruits have a minimal aptitude (separate aptitude testing requirements exist for this) but to minimize attrition. Military studies have shown that recruits who have not completed high school have a relatively high attrition rate; thus, high school dropouts are not viewed as a cost-effective military investment (Office of the Assistant Secretary of Defense, 2000b). Military officers typically must have a four-year college degree. Prospective officers can receive their college education through military academies, but few do so. In 1999, for example, only 17 percent of officer accessions, or 2,859 new recruits, entered the services from military academies (Office of the Assistant Secretary of Defense, 2000a). The majority of officer accessions come from ROTC programs at participating colleges and universities or from other postbaccalaureate routes. Thus, although military academies can serve as an alternative college provider for prospective officers, the academies' relatively small size and limited training capabilities (e.g., they do not train military doctors or lawyers) mean that they are a "small-time" com- petitor for traditional postsecondary institutions. Nonetheless, military academies do tend to be highly selective, attracting students that many other colleges would also like to enroll. The military strives to enlist personnel who perform well both in training and on the job. To ensure that this goal is met, every poten- tial new recruit into the military takes the Armed Services Vocational Aptitude Battery (ASVAB). One component of the ASVAB, the Armed Forces Qualification Test (AFQT), assesses math and verbal skills. The military uses the AFQT to predict trainability and job perfor- mance; it serves as the screening device to ensure that new recruits meet the military's targets for "high-quality" enlistments. In 1981, about 20 percent of new recruits came from the lowest allowable training level (of five levels) on the AFQT. Since that time, regula- tions have been put into place to ensure that few new recruits are at this low level; as a result, fewer than 2 percent of new recruits in 1999 were at the lowest allowable training level. The percentage of new recruits at the middle two AFQT levels increased from 49 percent in 1981 to 61 percent in 1999, and the percentage of new recruits in the highest two AFQT levels increased from 30 percent to 37 percent. Partly because of its stringent enlistment requirements, the mili- tary does not draw its recruits predominantly from the lower social classes or lower ability levels, as some might believe. For example, in 1999, almost all active duty enlisted accessions had a high school diploma or equivalent, and 93 percent of these accessions held a regular high school diploma (compared to 77 percent of 18-24 year olds in general) (Office of the Assistant Secretary for Defense, 2000a). In addition, the percentage of recruits who score in the below-aver- age category on the AFQT is lower than the percentage among their civilian counterparts. LISA HUDSON 45

The military does, however, attract relatively high proportions of Blacks and other non-Hispanic minorities. For example, in 1999,22 percent of enlisted personnel and 8 percent of officers were Black, compared to 13 percent of civilians aged 18-44. Hispanics, on the other hand, are underrepresented in the military, comprising only 9 percent of the enlisted force and 4 percent of officers, compared to 13 percent of civilians aged 18-44. Women also continue to be underrepresented in the military, although their representation has increased from 2 per- cent of all enlisted personnel in 1973 to 14 percent in 1999. (All data are from Office of the Assistant Secretary for Defense, 2000a.) This increase in female participation in the military parallels increases in female participation in postsecondary education and (as discussed be- low) in the civilian labor force. Over time, the size and composition of the military are largely affected by defense policy, which in turn is sensitive to both political changes (such as the breakup of the Soviet Union) and changes in job demands within the military (e.g., as equipment becomes more tech- nologically complex). From 1973, the year in which the all-volunteer force began, the size of the military was relatively stable until the drawdown of the 1990s. From 1989 to 1999, the active-duty military declined in size from about 1.8 million to 1.1 million (Office of the Assistant Secretary of Defense, 2000a). In the foreseeable future, the size of the military is expected to remain stable at around 1.3-1.4 million, although its "quality" may continue to change in response to a growing reliance on new technologies. Military Training and the Montgomery G} Billy The U.S. military is one of the largest training institutions in the world. Virtually every member of the military receives extensive job training upon entry to the service, as well as periodically during his or her service tenure. Formal training, in addition to on-thejob training, is often an important consideration for military promotions. To meet these training needs, the military provides training opportunities at its own "school houses." In addition, the military offers a number of education benefits. Tuition assistance programs are common, and oppor- tunities for paid, full-time college study are provided on a selective basis. For purposes of this chapter, however, the most important edu- cation benefit the military provides is the Montgomery GI Bill (MGIB). Since the end of World War II, the military has offered a "GI Bill" that provides education benefits, including college financial aid, for separating service members. The original purpose of these benefits was to compensate service members for educational opportunities lost while in the conscripted service and to ease their transition to civilian |4Information in this section of the chapter is drawn from Office of the Assistant Secretary of Defense (2000b). 46 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

life. Since the initiation of the all-volunteer force in 1973, the provi- sion of education benefits has served a somewhat different purpose. First implemented in 1984, today's Montgomery GI Bill is intended to ease the transition to civilian life and to serve as a recruitment and retention incentive. The MGIB provides military personnel with an education fund (to which the service member contributes $1,200) that can be used to pay for college costs at any time from initial separa- tion up to 10 years after separation from the military. In year 2000, the benefit for those who enlisted for at least three years was $536 per month for up to 36 months, or $6,432 per year, for a total poten- tial benefit of $19,296.~5 (In comparison, the 1999 average Fell grant per recipient was $1,923 and the average federal student loan per recipient was $4,667 ECollege Board, 2000a].) New military recruits are automatically enrolled in the MGIB program unless they specifically request withdrawal from the pro- gram. This policy results in a high enrollment rate, currently around 96 percent. Not all participants use their benefits though. In 1999, 56 percent of eligible veterans had used at least some of their MGIB benefits. This translates to about 650,000 students attending college on an MGIB benefit between 1992 and 1999. Since benefits can be used over a 10-year period and all MGIB-eligible veterans are still within that 10-year timeframe, this usage rate is expected to increase over time. From an educational perspective, two important effects of the MGIB are not known. One is the extent to which the program increases college enrollments. It is not known how many of the veterans who use the MGIB would have entered postsecondary education without the benefit and how many are enrolling in school primarily because of the availability of the benefit. Second, the college completion rate of veterans who use the MGIB is as yet unknown. Nonetheless, the 56 percent usage rate does suggest that the MGIB is easing the transi- tion to college for many adults. Largely because of the MGIB, today' s military may function for some young adults not as an alternative to college but as an alternative entry route to college. From the military's perspective, the MGIB is a highly successful and valuable program. To ensure that the program continues to func- tion as an effective recruitment and retention incentive, a number of modifications have been proposed by Congress and/or the adminis- tration to enhance the MGIB program. These proposals include the following: . Allowing MGIB benefits to be used by any member of the service member's immediate family; 1SIn most branches of the military, additional college funds are available to those who enlist in selected critical or hard-to-fill occupations. These additional funds can increase the total MGIB benefit to $50,000. LISA HUDSON 47

· Expanding the program to include coverage of the costs of licensing or certification; · Expanding the program to cover the costs of "high-tech/short- term" programs offered by business, such as Novell Network Engineer certification courses; and · Increasing the monthly stipend to cover the average costs of tuition and expenses at public postsecondary institutions and indexing the stipend to changes in education costs. It is worth noting that these proposed changes suggest a strong interest in expanding the MGIB program to include coverage of education and training programs that are often provided by agencies or institu- tions other than the colleges or universities. THE LABOR FORCE Like the military, the civilian labor force serves as an alternative pursuit to college enrollment for students leaving high school. As is also true of the military, the labor market often provides incentives for college study by requiring initial or further skill development that can be obtained through postsecondary education (e.g., continuing education requirements for doctors), by offering raises or promotions for those who obtain further education, and through the provision of employer financial support for workers who go to college. Labor Force Composition and Trends In 1999, about 139.4 million adults aged 16 and older were in the U.S. civilian labor force. In comparison, 68.4 million adults were out of the labor force (U.S. Census Bureau, 2000~. Among adults aged 16 and older, this represents a labor force participation rate of 67 per- cent. Among those in the prime working years, aged 25-64, the labor force participation rate is 80 percent. The labor force participation rate is related to education level, as those with higher levels of education participate at higher rates than those with lower levels of education. For example, in 1999, 63 percent of adults aged 25-64 with no high school diploma participated in the labor force, as did 78 percent of those with only a high school diploma, 83 percent of those with only some college, and 88 percent of college graduates (U.S. Census Bureau, 2000~. Over the decades, the labor force participation rate has been rising, primarily because of women's increased participation. From 1950 (the earliest year of published data) to 2000, women's participation rate has increased while men's participation rate has decreased. Since 1965, women's participation rate has increased faster than the rate for men has decreased, resulting in a steady increase in the rate of overall labor force participation, from 59 percent in 1965 to 67 percent in 48 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

2000.~6 The declining participation rate for men appears to be due to changing policies concerning social security, disability benefits, and pension benefits, which have made it easier for men, particularly older men, to leave the labor market (Fullerton, 1999~. The Labor Force in Postsecondary Education Some adults who are in the labor force are also enrolled in postsecondary education, either because they are students who need to work to pay for their schooling or because they are workers who have decided to return to school. In 1995-1996, about 13.3 million of the 19.5 million students who were enrolled in postsecondary educa- tion (68 percent) were also in the labor force. This suggests that in 1995 about 10 percent of the total labor force was enrolled in college. About 3.9 million of these working students (20 percent of all students, and 3 percent of the labor force) defined themselves primarily as employees who were going to school ("student employees"), a group that is in many ways distinct from other students. Among undergraduates, student employees tend to be concen- trated in public two-year institutions, suggesting that workers seeking further education and training are particularly likely to attend public two-year institutions (which are primarily community colleges). In 1995-1996, 67 percent of undergraduate student employees were enrolled in public two-year institutions, 17 percent in public four-year institutions, 5 percent in private for-profit institutions, and most of the remaining 11 percent in private four-year institutions (Lee and Clery, 1999~. The Demand for Skills Historically, skill demands in the labor market have increased over time, and the recent past is no exception. For example, a 1994 national survey of employers found that 57 percent reported that skill demands were increasing for jobs in their companies while only 2 percent reported a decline in skill demands (National Center on the Educa- tional Quality of the Workforce, 1995~. Nonetheless, the majority of all current labor market jobs do not require education beyond the high school level. In 1998, 72 percent of all occupations required only work experience or on-thejob training. In comparison, 7 percent of all occupations required an associate degree or vocational training, and 22 percent required a bachelor's degree or higher (BLS, 2000~. Because these requirements are based primarily on the education composition of labor market participants, they closely match the edu- cation level of the population. In 1998, 7 percent of adults age 25 or 16From 1965 to 2000, women's participation rate rose from 35 percept to 60per- cent, while men's participation rate dropped from 81 percent to 75 percent. (These labor force participation data are from the Bureau of Labor Statistics Web site http://stats.bls.gov.) LISA HUDSON 49

older had an associate degree, and 24 percent had a bachelor's degree or higher (Snyder, 2000~. Occupational projections suggest that the largest number of new jobs in the coming decade will be in occupations that require only short-term on-thejob training, mainly because that is the education requirement for most of today's existing jobs (BLS, 2000~. Job growth, however, is more concentrated in jobs that require postsecondary edu- cation. Projections of employment growth from 1998 to 2008 show that 57 percent of new jobs will be in occupations that do not require postsecondary education, 11 percent will be in occupations that require an associate degree or vocational training, and 33 percent will be in occupations that require a bachelor's degree or higher (BLS, 2000~. The faster-than-average growth among jobs that require postsecondary education is expected to result mainly from increases in health and computer-related occupations. For example, among occupations at the associate degree/vocational training level, the fastest-growing occupations are registered nurses, computer support specialists, and licensed practical nurses. At the bachelor's degree level, the fastest-growing occupa- tions are computer systems analysts, general managers/executives, and computer engineers. These projections reflect where growth has occurred in the recent past and, as seen above, are largely consistent with recent trends in degree fields of study, where growth has been most pronounced in business, technical, and health fields. Skill demands in the labor market appear to be increasing both because of the changing nature of the labor market as a whole (i.e., the shift to higher skill jobs) and because the skills required for specific jobs are also increasing. This increase in skill demands is reflected in increasing proportions of workers Participating in work-related educa- tion activities (including college enrollment). particularly notable among workers in the trades occupations and in sales and service occupations, as opposed to professional occupations (Creighton and Hudson, 2002~. 1 1 0 .. .. These increases are Employer Financial Support Employers often provide support for the further education of their employees, including participation in postsecondary education. For example, in a 1995 survey of business enterprises with at least 50 employees, the BLS found that 61 percent of these employers offered tuition reimbursement programs in 1994 (Frazis, Gittleman, Horrigan, and Joyce, 1997~. This training practice was second only to the financing of off-site training (including conference attendance) among the edu- cation and training benefits provided by employers. Another Perspective on the role of emolovers in supporting 1 1 1 ~ 1 1 "7 . . . ~ ~ . . . mu. . postsecondary education comes from surveys of college students. lhls perspective shows that while many employers offer tuition assistance, relatively few college students receive it. Using the NCES National Postsecondary Student Aid Study, Lee and Clery (1999) found that 50 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

6 percent of all undergraduates in 1995-1996 (about 700,000 under- graduates) received financial aid from their employers. However, among undergraduates who consider themselves to be primarily employees who are going to school, a much higher percentage 25 percent- received employer aid. Employer financial aid is also more common among graduate students than among undergraduates; among graduate students who considered themselves primarily employees, 42 percent received employer aid. Lee and Clery (1999) also found that students in some fields of study were more likely than those in others to receive employer aid. At the undergraduate level, over one-third of students enrolled in business, engineering, and computer/information science programs received employer aid, compared to one-quarter of those in health programs and no more than one-fifth of those in other program areas. At the graduate level, students in business programs were more likely to receive emolover aid than were students in all other Program areas (14 percent versus no more than ~ percent). lnese ilncllngs suggest that recent growth in business and computer-related degrees may be partially the result of employer support for workers to obtain these degrees. Yet another perspective on employer support for postsecondary education comes from surveys of adults. Lee and Clery (1999) also used NCES' Adult Education Survey to examine the extent to which adults received employer support for their participation in "credential programs." Because of ambiguity in the definition of this term, credential programs may include vocational training programs and noncredit courses taken to receive continuing education requirements or other formal credentials, in addition to for-credit college enrollments. Among adults in these programs, 24 percent received employer financial support, and 33 percent received some other form of employer support (such as time off from work). About half of adults in credential programs (53 percent) received one or the other type of employer assistance. The likelihood of receiving employer financial aid for a creden- tial program varied depending on one's occupation, with workers in occupations that have higher skill demands (and workers with higher incomes) being more likely to receive employer financial support than those in occupations with lower skill demands (and lower incomes). For example, in 1995, one-half of executives, administrators, and managers who enrolled in credential programs received financial assistance from their employers. This figure compares to 10 percent of those employed in marketing and sales and 4 percent of those who were handlers, cleaners, helpers, or laborers (Lee and Clery, 1999~. These findings suggest that employer support for college education may increase in the future, as the labor market (slowly) shifts to the management and technical jobs that employers most often support. The findings summarized above demonstrate one important difference between federal student financial aid and employer aid. While the federal government provides financial support for postsecondary education LISA HUDSON 51

for the benefit of society, employers provide this support primarily for the benefit of their company. This goal means that employers tend to financially support those workers who are most likely to increase company productivity or profitability as a result of their education- i.e., managers, skilled technical workers, and other high-skill, high- demand workers who cannot be easily hired with the requisite skills or who cannot continue to function effectively without further educa- tion. Thus, while employer aid can and does support postsecondary education, it tends to do so in a way that further exacerbates differ- ences between the educational "haves" and "have note." One reason so many employers provide tuition assistance is that federal policy provides incentives to employers to do so through "Sec- tion 127" benefits. This legislation allows employers to provide their employees (as of 2000) up to $5,250 tax-free to pay for undergraduate tuition. Employers have the additional incentive of not having to pay their share of the FICA contribution on this funding (i.e., the aid does not count as earnings). Current Section 127 legislation had been scheduled to expire in December 2001, but the Economic Growth and Tax Relief Act of 2001 recently made these benefits permanent, which means that this policy will be in place for at least the next 10 years (when the new Act expires). The 2001 Act also extended these benefits to cover graduate school tuition. Both the lon.~-term provision of this benefit _ _ ~ 1 ·, , · , 1 , 1 1 r ,1 1 and its extension to graduate school may further encourage employer support for postsecondary education, and thus may further encourage the participation in postsecondary education of working adults. Postsecondary Institutions as a Provider of Worker Training In addition to postsecondary education, workers often receive other types of education and training, much of it provided by their employer. The 1995 BLS survey of employers found that 93 percent of enter- prises that have at least 50 employees provided some type of formal training for their workers and that 70 percent of workers in these enterprises received formal employer-provided training over a one- year period (Frazis et al., 1997~. Postsecondary institutions are a source for some of this employer-provided training but not the bulk of it. According to the BLS survey, only 17 percent of employers used postsecondary institutions as a training source for employer-provided training (BLS, 1996~. Another perspective on the role of postsecondary education as a provider of adult education and training comes from the 1995 NCES Adult Education Survey. This survey shows that among all adults who took courses that were not part of a credential program, postsecondary institutions were the instructional provider for 31 percent of these adult learners, second only to business and industry (36 percent) (Hudson, 1999~. 52 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

The Adult Education Survey also found that, although half of adults enrolled in postsecondary institutions were taking courses leading to a credential, half were not. This implies that adults are as likely to use postsecondary institutions for noncredential purposes as for credential purposes. This finding is inconsistent with NCES student surveys showing that most students are enrolled in degree programs.~7 This inconsistency suggests that a significant amount of continuing educa- tion and other noncredit course taking is occurring within postsecondary institutions that is not captured by NCES' regular student surveys. This limitation, in turn, implies that the student enrollment data dis- cussed earlier in this chapter show how postsecondary institutions are used by only about half of those who receive instruction from these institutions. One type of college course taking that is missing from NCES surveys is course taking designed to lead to an industry or company credential only. Not much is yet known about these activities. Adelman (2000) has recently examined credentialing in the information tech- nology (IT) industry, the largest sector of the industry credentialing movement. His data show that while IT credentials were virtually unheard of a decade ago, as of January 2000, 1.7 million credentials had been awarded by the IT industry. It would appear that the credentialing "movement" is well underway in the IT industry. A less advanced, but broader effort to encourage industry credentialing is being advanced by the National Skill Standards Board (NSSB), an organization initiated by the National Skill Standards Act of 1994. The NSSB is a coalition of leaders from business, labor, employee, education, and community organizations who are working to build "a voluntary national system of skill standards, assessments, and certification systems" to enhance workforce development. The NSSB proposes to develop skill standards in 15 industry sectors. At present, standards have been developed in the manufacturing and the sales and service industries; standards are under development in the education and training and the hospitality and tourism industries. It is not yet clear how the work of the NSSB will link to postsecondary education, but the general goal seems to be to develop a credentialing system that is industry-based, portable, and ultimately international in scope much like the existing IT credentialing system. These initiatives raise questions about the trade-off between a broad, formal education and a more narrow credentialing of skills. Most educators would argue that the acquisition of narrow skills instead of a broad education is a bad choice for individuals, reducing their labor market flexibility (as well as their general intellectual foundation). But as occupations become increasingly specialized and technical, the credentialing of skills instead of or in addition to general educa- |7For example, the 1995-1996 National Postsecondary Student Aid Study found that only 3 percent of beginning postsecondary students were not enrolled in a degree or certificate program (Kojaku and Nunez, 1998~. LISA HUDSON 53

tional credentialing is likely to grow in popularity. It remains to be seen what role postsecondary institutions should and will play in this credentialing movement. SUMMARY From the national data, postsecondary education appears to be doing quite well. Enrollment levels, enrollment rates, and degree comple- tions have been increasing, in decades when the size of the college- age cohort was shrinking, costs were rising, and student aid was uneven at best. Further, since one of the key predictors of college attendance is whether one's parents went to college (see, e.g., Kane, 1994), postsecondary education is also reinforced through a self-perpetuating process: The more adults there are who have a college education, the more children there will be in the next generation who also seek a college education. In turn, the more highly educated workers there are in society, the more high-skill jobs the economy can support, further increasing education and skill demands. Given these trends, plus projections of a growing cohort of college-age adults in the next few decades. postsecondary education would seem to be in a good position overall. But the national data also hint at some potential problems. Increasing college costs may be limiting access for some students, student loan programs may lead to undesirable debt burdens, and business and industry appear to be pushing for credentialing processes that could operate independently of the postsecondary education system. Surveys of employers and adults also show that the majority of adult course taking occurs outside of postsecondary education. Taken together, these findings suggest that the combination of a cost-restricted postsecondary education system on the supply side and a growing interest in further education on the demand side may be setting the stage for the growth of alternative education systems and providers. As the other chapters in this volume demonstrate, postsecondary institutions are increasingly adopting new missions, new education programs, and new instructional delivery strategies, while new pro- viders (e.g., virtual universities, corporate universities, industry credentialing agencies) are offering a wider range of alternative learning routes. Unfortunately, these alternative strategies and systems are not well covered in national data systems, so they do not appear in the portrait of postsecondary education created by these data. It is reasonable to ask why these alternative systems are so diffi- cult to assess within a national data collection. The basic problem is that alternative and emerging systems often do not meet the criteria necessary for cost-effective collection of systematic, reliable data. First, a national data collection depends on a clear and consistent definition of all the entities from which one intends to draw a survey sample. Thus, before one can survey postsecondary institutions, one must opera- tionally define them and then be able to identify all institutions that meet the definition. This task becomes more difficult when new institutions 54 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

open and close at a rapid rate or when institutions or systems emerge that do not fit existing definitions. Data collections also rely on the willingness of survey participants to share information and on their capacity to provide information. Both of these respondent character- istics are often compromised in new and emerging systems. For-profit postsecondary education institutions, for example, are sometimes unwilling to respond to surveys for fear that their competitors will learn too much about them. Finally, new alternatives are by definition different from the norm, so that existing survey instruments and procedures may simply be unable to capture or describe them. In short, it is always difficult for national surveys to accurately capture an emerging system or a system in flux. So national data collections are probably not the best source for finding out what is happening "at the margins." This is not to say that the current data collection system for postsecondary education cannot or should not be improved. There are a number of ways in which the current system could be adapted to better capture the full breadth of educa- tion alternatives facing adults. Three proposals are suggested here. First, until better methods are devised for capturing information directly from alternative providers, the best source of information on these providers is the adults who enroll in education programs. The NCES Adult Education Survey is our best source of information on adults, but its sample size is typically too small to allow analysis of participation in activities that may be relatively new and small scale. A larger sample of adults is needed. Second, it would be useful to regularly survey employers about existing policies and practices that may influence workers' participation in traditional postseconda~y education and other forms of learning. Previous employer surveys have been conducted (but discontinued) by the Department of Labor and the Department of Education; perhaps a joint Labor-Education survey effort should be attempted. ~ '' Nay, more focused surveys, such as the NCbb surveys on distance education, He needed to monitor emerging delivery systems that cannot be captured in existing surveys. Potential topics for these focused surveys include customized training, continuing education, the use of new technologies in traditional classrooms, and institutions' role in industry certification. To end on a positive note, existing federal data sources provide a wealth of information about the traditional postsecondary education system, only a small part of which could be included in this chapter. We know much more about postsecondary education today than we ever have in the past and even with a moving target our knowl- edge is sure to improve in the future. REFERENCES Adelman, C. (2000). A parallel universe. Change, 32(3), May/June, 20-29. Berkner, L. (1998). Student financing of undergraduate education: 1995-96 (NCES 98-076). Washington, DC: U.S. Department of Education, National Center for Education Statistics. LISA HUDSON 55

Berkner, L. (2000~. Trends in undergraduate borrowing: Federal student loans in 1989-90, 1992-93, and 1995-96 (NCES 2000- 151). Washington, DC: U.S. Department of Education, National Center for Education Statistics. Berktold, J., Gels, S., and Kaufman, P. (1998~. Subsequent educational attainment of high school dropouts (NCES 98-085~. Washington, DC: U.S. Department of Education, National Center for Education Statistics. Boesel, D., Alsalam, N., and Smith, T.M. (1998~. Educational and labor market performance of GED recipients. Washington, DC: U.S. Department of Edu- cation, Office of Educational Research and Improvement, National Library of Education. Bureau of Labor Statistics. (1996~. BLS reports on the amount offormal and informal training received by employees. BLS news release, December 16, 1996. Available: http://stats.bls.gov/news.release/sept.nws.htm [October 25, 2001]. Bureau of Labor Statistics. (2000~. Charting the projections: 1998-2008. Occupa- tional Outlook Quarterly, 43~4) Winter 1999-2000, 8-38. Choy, S.P., and Gels, S. (1997~. Early labor force experiences and debt burden (NCES 91-286~. Washington, DC: U.S. Department of Education, National Center for Education Statistics. College Board. (2000a). Trends in student aid. New York: College Entrance Exami- nation Board. College Board. (2000b). Trends in college pricing. New York: College Entrance Examination Board. College Board. (2000c). 1996-2000 profiles of college-bound seniors, SAT national profile reports. Available: http ://www.collegeboard.org/sat/cbsenior/yr 1997/ nat/cb s 1997.html; http :llwww. collegeboard. org/s at/cb senior/yr 199 81natl cb s 1998.html; http :llwww. collegeboard. org/s at/cb senior/yr 1999/NAT/ cbs l 999.html; http://www.collegeboard.org/sat/cbsenior/yr2000/nat/cbs2000.html. [December 10, 2001] Creighton, S., and Hudson, L. (2002~. Participation trends and patterns in adult education: 1991 to 1999 (NCES 2002-119~. Washington, DC: U.S. Depart- ment of Education, National Center for Education Statistics. Frazis, H., Gittleman, M., Horrigan, M., and Joyce, M. (1997~. Formal and informal training: Evidence from a matched employee-employer survey. Advances in the Study of Entrepreneurship, Innovation, and Economic Growth, 9, 47-82. Frazis, H., Gittleman, M., Horrigan, M., and Joyce, M. (1998~. Results from the 1995 survey of employer-provided training. Monthly Labor Review, 121~6), 3-13. Fullerton, H.N., Jr. (1999~. Labor force participation: 75 years of change, 1950-98 and 1998-2025. Monthly Labor Review, 122~12), 3-12. Guzman, B. (2001~. Census 2000 brief: The Hispanic population. Washington, DC: U.S. Department of Commerce, U.S. Census Bureau. Horn, L.J., and Berktold, J. (1998~. Profile of undergraduates in U.S. postsecondary education institutions: 1995-96 (NCES 98-084~. Washington, DC: U.S. Depart- ment of Education, National Center for Education Statistics. Hudson, L. (1999~. Adult participation in lifelong learning: An examination of noncredential coursetaking. Paper presented at the annual meeting of the Association for Institutional Research, Seattle, WA, May. Kane, T.J. (1994~. College entry by blacks since 1970: The role of college costs, family background, and the returns to education. Journal of Political Economy, 102~5), 878-911. Kane, T.J. (1995~. Rising public college tuition and college entry: How well do public subsidies promote access to college? NBER Working Paper No. 5164, July. Cambridge, MA: National Bureau of Economic Research. Kaufman, P., Kwon, J.Y., Klein, S., and Chapman, C.D. (2000~. Dropout rates in the United States: 1999 (NCES 2001-022~. Washington, DC: U.S. Department of Education, National Center for Education Statistics. 56 DEMOGRAPHIC AND ATTAINMENT TRENDS IN POSTSECONDARY EDUCATION

Kojaku, L.K., and Nunez, A. (1998~. Descriptive summary of 1995-96 beginning postsecondary students (NCES 1999-030~. Washington, DC: U.S. Depart- ment of Education, National Center for Education Statistics. Lee, J.B., and Clery, S.B. (1999~. Employer aid for postsecondary education (NCES 1999-181~. Washington, DC: U.S. Department of Education, National Center for Education Statistics. Levesque, K., Lauen, D., Teitelbaum, P., Alt, M., and Librera, S. (2000~. Vocational education in the United States: Toward the year 2000 (NCES 2000-029~. Washington, DC: U.S. Department of Education, National Center for Edu- cation Statistics. Lewis, L., Snow, K., Farris, E., and Levin, D. (1999~. Distance education at postsecondary education institutions: 1997-98 (NCES 2000-013~. Washington, DC: U.S. Department of Education, National Center for Education Statistics. McPherson, M.S., and Schapiro, M.O. (1998~. The student aid game. Princeton, NJ: Princeton University Press. Mortenson, T.G. (1999~. Where are the boys? The growing gender gap in higher education. The College Board Review, 188, August, 8-17. National Center for Education Statistics. (1997~. Indicator of the month, subbaccalaureate persistence and attainment. Washington, DC: U.S. Department of Education. National Center for Education Statistics. (1999~. The condition of education 1999 (NCES 1999-022~. Washington, DC: U.S. Department of Education. National Center for Education Statistics. (2000~. The condition of education 2000 (NCES 2000-062~. Washington, DC: U.S. Department of Education. National Center for Public Policy and Higher Education. (n.d.~. Measuring up 2000: The state-by-state report card for higher education. San Jose, CA: Author. National Center on the Educational Quality of the Workforce. (1995~. First findings from the EQW national employer survey. Philadelphia: University of Penn- sylvania. Office of the Assistant Secretary of Defense (Force Management Policy). (2000a). Population representation in the military services: Fiscal year 1999. Arlington, VA: U.S. Department of Defense, Defense Manpower Data Center. Office of the Assistant Secretary of Defense (Force Management Policy). (2000b). Biennial report to Congress on the Montgomery GI Bill education benefits program. Arlington, VA: U.S. Department of Defense, Defense Manpower Data Center. Smith, T.M. (1996~. The condition of education 1996 (NCES 96-304~. Washington, DC: U.S. Department of Education, National Center for Education Statistics. Snyder, T.D. (2000~. Digest of education statistics 1999 (NCES 2000-031~. Washington, DC: U.S. Department of Education, National Center for Education Statistics. Snyder, T.D. (2001~. Digest of education statistics 2000 (NCES 2001-034~. Washington, DC: U.S. Department of Education, National Center for Education Statistics. Snyder, T.D. and Hoffman, C.M. (1991~. Digest of education statistics 1991 (NCES 91-697~. Washington, DC: U.S. Department of Education, National Center for Education Statistics. USA Today (2001~. Debt smothers young Americans, Feb. 13, pp. 1-2. U.S. Census Bureau. (1992~. Statistical abstract of the United States: 1992. Washington, DC: U.S. Government Printing Office. U.S. Census Bureau. (1996~. Statistical abstract of the United States: 1996. Washington, DC: U.S. Government Printing Office. U.S. Census Bureau. (2000~. Statistical abstract of the United States: 2000. Washington, DC: U.S. Government Printing Office. LISA HUDSON 57

U.S. Census Bureau (2000~. Statistical Abstract of the United States: 2000. Washington, DC: U.S. Government Printing Office. U.S. Department of Education, National Center for Education Statistics (19971. Indicator of the Month, Subbaccalaureate Persistence and Attainment. Washington, DC: U.S. Government Pnuting Office. U.S. Department of Education, National Center for Education Statistics (1999 ). Me Condition of Education 1999 (NCES 1999-022~. Washington, DC: Author. U.S. Department of Education, National Center for Education Statistics (2000~. We Condition' of Education 2000 (NCES 2000-062~. Washington, DC: Author. U.S. Department of Labor, Bureau of Labor Statistics (20001. Occupational Outiook Quarterly, Winter 1999-2000, 8-38. . The Knowledge Economy and Postsecondary Education: Report of a Workshop Chapter 1 s8