Better Late than Never? Examining late high school graduates
For most of us, thinking of high school graduation brings memories of walking across a stage to "Pomp and Circumstance" after four years of work. But for a persistent and overlooked group of students, the picture is different. At each graduation, there are students who take longer than four years to receive their diploma.
While we admire their persistence,&is it worth the extra effort? Keeping these students in school, and getting them to graduation day, takes a lot of time and effort on everyone's part, and a lot of extra resources from the school district. Little1&is known about&late high school graduates'&eventual outcomes. If they ultimately receive no benefits from the extra time, a lot of school districts might want to rethink their policies.
So is it better to graduate late than never? To answer this question, we compared late graduates to:
GED recipients&students who earned a General Educational Development credential or equivalent.
Dropouts&students who left high school before 2000 without receiving any type of high school credential.
On-time graduates&students who graduated in four years or less with a standard high school diploma.
The quick answer: Late graduates do better
In every aspect of life&not just academic, but work, civic life, and even health&on-time graduates had the best outcomes, so school districts should make on-time graduation for all students their first priority.
But the extra work that late graduates and their high schools put into earning a diploma pays off&not only in late graduates& academic outcomes, but in jobs, involvement in civic life, and commitment to healthy lifestyles. Some examples you&ll see in this&guide include:
Late graduates are much more likely to go on to obtain either a bachelors' or associates' degree.
Late graduates are more likely to be employed, and more likely to have full-time employment than GED recipients and dropouts. Late graduates are also significantly better off in terms of job benefits.
Late graduates are significantly more likely to have voted in a recent election.
Late graduates are more likely to make better health choices&they are more likely to be non-smokers than GED recipients and dropouts and they exercise more.
Essentially, a student facing late graduation has two options: Stay in school and earn a regular diploma late, or leave school and earn either nothing or a GED. Late graduates fare better compared to students who leave school and those who earn a GED.
In making these comparisons, we asked the following questions:
Who are late graduates?
How do late graduates fare after high school?
Why are late graduates late?
How do late graduates compare to similar on-time graduates?
What should school board members do about late graduates?
Why should policymakers give credit to schools for late graduates?
For the first three comparisons we determined2&whether there are statistically significant differences3&between late graduates and GED recipients&late graduates' next option&and whether there are statistically significant differences between late graduates and dropouts.
In order to determine whether the differences are actually associated with students& graduation status, we then compared late graduates to on-time graduates from similar socioeconomic backgrounds and similar initial achievement levels4.
Who are late graduates?
Of the 2.9 million eighth graders in 1988, an estimated 4.6 percent, or approximately 135,000, were late graduates (Table 1). While this may seem like a small group to focus on, keep in mind that some high schools in your district might have a much higher late graduate percentage than the nationwide average.
Table 1: High School Graduation Status of the Class of 1992 eight years late
Table figures are in percentages&
On-time Graduates
Late Graduates
GED Recipients
Minority students (black, Hispanic, or American Indian) are significantly more likely to graduate late (10 percent) than their white peers&of whom just 3 percent graduate late (Table 2).
Table 2: High School Graduation Status of the Class of 1992 eight years later by race
Table figures are in percentages&
On-time Graduates
Late Graduates
GED Recipients
Native Amer.
Students from minority groups aren't the only ones more likely to graduate late, so are students from poorer families. Seven percent of students from families in the bottom 25 percent of socioeconomic status (SES) graduated late, but just 2.4 percent of students from families in the top 25 percent did (Table 3). Students from urban districts (7.1 percent) were also more likely to graduate late than students from suburban (4.2 percent) or rural (3.0 percent) districts (Table 4).
Table 3: High School Graduation Status of the Class of 1992 eight years later&
by socioeconomic status&
Table figures are in percentages
Ontime Graduates
Late Graduates
GED Recipients
Highest-Q4
Table 4: Urbanicity of the High School's Community by High School Graduation Status
Table figures are in percentages
Ontime Graduates
Late Graduates
GED Recipients
Eighth-graders whose family typically spoke a language other than English at home (language minority students) were almost twice as likely to graduate late as their non-language minority classmates (8.1 percent to 4.2 percent respectively) (Table 5).
Table 5: High School Graduation Status of the Class of 1992 eight years later by Language Minority Status
Table figures are in percentages
Ontime Graduates
Late Graduates
GED Recipients
Language Minority
Non-Language Minority
Finally, eighth graders with two or more of the six5&risk factors identified by NCES for dropping out were more likely to graduate late (9.7 percent) than students with no risk factors (2.7 percent) or just one (4.7 percent) (Table 6).
Table 6: High School Graduation Status of the Class of 1992 eight years later by risk factors of dropping out
Table figures are in percentages
Ontime Graduates
Late Graduates
GED Recipients
No risk factors
One risk factor
Two or more risk factors
How do late graduates fare after high school?
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
Do late graduates start or complete college?
Late graduates distinguish themselves not so much by enrolling in college, but in completion. While they are no more likely (59 percent) than GED recipients (51 percent) to enroll in college, they are much more likely to go on to obtain either an Associates or Bachelors degree. Late graduates were nearly four times as likely to earn at least an Associates degree (12 percent) than GED recipients (3.2 percent) (Chart 7).
(Since there are only a small number6&of dropouts with complete postsecondary transcripts, we will not include them in the rest of our discussion on college performance.)
How did late graduates do while attending college?
Graduating with a degree is a matter of accumulating enough credits with passing grades. Late graduates were more likely to complete a degree program, even though they did not earn grades as high as GED recipients. (Keep in mind, however, that we do not know the academic rigor of these courses.)
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
Credits. For instance, late graduates earned, on average, forty-two undergraduate credits, while GED recipients earned thirty-one credits (Chart 8). Although the difference seems to be large, for various technical reasons, it is not statistically significant7. The results are basically the same when examining the number of credits earned at only four-year institutions (Chart 9) or community colleges (Chart 10).
Remediation. However, the credit data does not include remedial courses, because such courses don&t count towards college credit. The majority&72 percent&of late graduates enrolled in at least one remedial course (Chart 11). This is significantly more than the GED recipients who enrolled in at least one remedial course (49 percent). At first glance this looks like bad news for late graduates, but it may not be. The higher rate of late graduates
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
actually enrolled in remedial courses may indicate that late graduates are taking the courses they need to be successful in credit baring courses later on, more so than GED recipients. A study by the Community College Research Center (2008) found that nearly four out of ten community college students referred to remedial math or reading courses did not even enroll in them. These students were less likely to earn any college credits than students who went on to complete remedial course sequences (Bailey, Jeong,&and Cho 2008).
Looking at math and reading remediation rates specifically shows that a greater percentage of late graduates than GED recipients enrolled in remediation in either subject (Charts&12and&13). (We do not have data for the number of students recommended for remediation.)
Grades. Even though late graduates were more likely to earn a college degree than GED recipients, they sometimes did not earn higher grades. Overall, late graduates, on average, earned lower8&GPAs (2.1) than GED recipients (2.4) (Chart 14). However, when comparing GPAs earned at four-year institutions and at community colleges separately, late graduates earned similar GPAs as GED recipients (Charts&15&and&16). Keep in mind, however, that late graduates likely took a greater number of higher-level courses since they earned more credits and were more likely to earn a degree.
Even though late graduates earned lower grades in college than GED recipients, they persevered to eventually earn more degrees. It does show that students who persevere and graduate late go on to a better future.
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
Do late graduates get and keep good jobs?
High schools prepare students for work as well as college. Since a large number of late graduates go straight into the workforce, evaluating their experience means seeing if they are better able to get and keep good jobs than either GED recipients or dropouts. (All employment data&unless otherwise noted&is from 2000, when most of these students were twenty-six years old.)
Our first indicator for late graduates is whether they are actually employed. Eighty-five percent of late graduates are employed, compared to&just 77 percent of GED recipients* and 81 percent of dropouts (Chart 17). Furthermore, late graduates are more likely to be employed full-time than GED recipients* (Chart&18). Additionally, research (Wald and Martinez 2003) has shown that high school graduates are not just more l when unemployed, they are so for shorter lengths of time than GED recipients or dropouts.
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
Not all jobs are created equal, so it is also important to look at job quality. Wages are one measure of quality. On average, late graduates& wages were essentially the same as GED recipients and dropouts in 1999. Although there was some variation between the groups, the differences are not statistically significant (Chart&19). Keep in mind, however, that because late graduates are more likely to earn at least an Associates degree, as they get older their incomes are likely to grow faster than GED recipients and dropouts (Day and Newburger 2002).
While there are no significant differences in annual income between groups, on average, there are differences at the bottom of the income scale. Late graduates are less likely than GED recipients* and dropouts* to earn incomes that fall within the lowest 20 percent.& Nearly one in three (29 percent) GED recipients and 39 percent of dropouts earned such low incomes, while just 19 percent of late graduates did so (Chart 20). This is probably due to the fact that fewer GED recipients and dropouts were employed full time.
Benefits, which are as crucial as wages, provide a clearer picture of the differences in job quality. Late graduates are significantly better off. Of the late graduates who were employed any time after 1994, close to two-thirds (63 percent) held a job that offered retirement benefits compared to just over half (53 percent) of GED recipients and less than half of dropouts* (45 percent) (Chart 21).
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
With health insurance, too, late graduates fare better than GED recipients and dropouts. Seventy-six percent of late graduates have coverage, compared to 66* and 61 percent of GED recipients and dropouts, respectively (Chart 22). When looking specifically at who received health insurance through their employers, half of late graduates did, while 40 percent of GED recipients* and 42 percent of dropouts did so (Chart 23).
Interestingly, graduating late does not make much of a difference in job satisfaction. Overall, late graduates are no more satisfied with their jobs than GED recipients and dropouts (Chart 24). More specifically, late graduates are slightly less satisfied with their pay than GED recipients. They are, however, just as satisfied with their benefits, promotion opportunities and job security. Perhaps this is because with their education they want a better job and are just not satisfied with holding an entry-level position in their mid-twenties.
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
Although late graduates do not earn significantly more money and are no more satisfied with their jobs than GED recipients, late graduates have higher quality jobs. They are more likely to have a job, and one that is more likely to offer retirement and health benefits. Furthermore, the beneficial impact of additional years of education on annual income increases as time goes on, especially after age twenty-six (Day and Newburger 2002), so late graduates may see more of an economic benefit as time goes on.
Are late graduates good citizens?
Though often overlooked, public education is also intended to prepare students to be good citizens and lead healthy lives. While this quality is hard to quantify, we can look at two distinct actions of many good citizens&voting and community service&and three indictors of a healthy lifestyle.
Voting. Where the real difference in voting practices really occurs is in who actually votes. Although late graduates are no more likely to be registered to vote
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
than GED recipients (Chart 25), late graduates were significantly more likely to have voted in a recent election (40 percent versus 29 percent) (Chart 26). (Charts 25 and 26 also show that late graduates were more likely both to be registered and to vote than dropouts.)
Community service.& High school graduation status has little effect on volunteer work. Even though late graduates were more likely than dropouts to volunteer in their communities, they were no more likely than GED recipients (Chart 27).
Health. Here We examined three indicators: Smoking, drinking, and exercising. Late graduates were more likely to be non-smokers than GED recipients and dropouts (Chart 28). There was no difference in drinking habits among the groups (Chart 29). Late gra 82 percent of late graduates exercised at least&one or&two days a week, compared to 75 percent of GED recipients and 55 percent of dropouts (Chart 30).
Just as with college and work outcomes, the data demonstrates once again that students are better off graduating late than never. Late graduates were
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
more likely to take a greater role in their communities than GED recipients and dropouts by voting in recent elections. Late graduates also tend to live a healthier lifestyle by smoking less and exercising more than GED recipients and dropouts. Such habits likely will lead to longer, healthier lives.
Better late than never
Overall, late graduates fare much better after high school than their classmates who went on to earn a GED or dropped out. Late graduates are more likely than both groups to earn a college degree, hold a quality job, vote in elections, and live a healthier lifestyle. As a matter of fact, the data is fairly consistent in showing that students& with higher high school graduation statuses are more successful after high school.
Why are late graduates late?
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
Why do late graduates need more time? Here we look at the differences in the courses students took in eighth grade and high school, and their grades, to determine how academic preparation affects graduation status.
Math. When it comes to the rigor of eighth grade math courses, late graduates&are no more prepared to move on to high school math than GED recipients or dropouts. They had not taken more rigorous courses than GED recipients or dropouts. At the top of the achievement spectrum, about one-third (34 percent) of late graduates took algebra in eighth grade, which is not significantly more than GED recipients (27 percent), though it is significantly more than dropouts (19 percent) (Chart 31). Late graduates (12 percent) also took remedial math courses at about the same rate as GED recipients and dropouts (14 percent and 16 percent) (Chart 32).
In respect to achievement late graduates perform similarly to GED recipients and dropouts. Late graduates are no more likely than GED recipients and dropouts
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
to receive As and Bs in their sixth-eighth grade courses, and scored similar to GED recipients on standardized tests (Charts 33&and&34). The math course-taking and achievement data show that late graduates are as ill prepared to go on to high school math as GED recipients and dropouts. And, we will show later that late graduates are less academically prepared than on-time graduates.
English. In English, the picture is slightly different. Late graduates (19 percent) are no more likely to have&taken remedial English in eighth grade than GED recipients (14 percent) and dropouts (16 percent) (Chart 35).
As for achievement, when compared to GED recipients, late graduates are more likely to earn high grades and just as likely to have scored low on standardized tests. When compared to dropouts, however, late graduates are more likely to receive high grades but less likely to score low on standardized tests (Charts&36&and&37).
When course-taking, grades, and test scores are viewed together, the data show that late graduates are not substantially more prepared for high school
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
than GED recipients or dropouts. Since late graduates start high school equally ill-prepared, their high school journey may provide some insight into why these students finish.
High school performance. At the eighth grade level, GED recipients, dropouts, and late graduates look similiar. But in high school, late graduates& performance begins to look different. (Keep in mind that the information in this section was collected in the summer of 1992, immediately after the on-time graduation date. It doesn&t include the courses late graduates took to finish their high school requirements.)
Overall, late graduates not surprisingly earned more credits in their first four years of high school in both math and English than GED recipients and dropouts (Charts&38and39). Of course, most GED recipients and dropouts leave high school within the first year or two.
In math, just under half (43 percent) of late graduates fall behind in math early in their high school careers by taking non-academic math courses, if any math at all, in&ninth grade (Chart 40). This is similar to the 42 percent of GED recipients but less than
Source: National Center for Education Statistics. NELS:88/2000 and PETS:2000
the 62 percent of dropouts. So at the start all three groups fall behind their classmates in high school requirements.&Districts should ensure that all ninth graders have adequate math courses to start their high school career and stay on track to graduate on time.
Not only do late graduates earn more credits in high school, they also perform better. For instance, on average, they earn higher GPAs than GED recipients and dropouts. While in high school, late graduates earn a GPA of 1.8, which is significantly higher than the GPAs of GED recipients (1.6) and of dropouts (1.3) (Chart 41).&
The data does not explain, however, why some students persevere through high school to graduate, while others leave school to get a GED or drop out entirely. Perhaps&their higher achievement levels in high school help them persevere both in high school and later in college.
How do late graduates compare to similar on-time graduates?
Here we compare late graduates and on-time graduates who have similar SES and initial achievement9. On-time graduates, as a group, come from a broad range of socioeconomic backgrounds and achievement levels, while late graduates tend to come from poorer socioeconomic backgrounds and are initially lower achievers. Although on-time graduates& results are better overall in every category, comparing similar students ensures that the findings are associated only with students& graduating on-time or late and not SES background or initial achievement levels.
Our focus here is on college, job, civic, and health outcomes. It is important, however, to keep in mind that in eighth grade and high school on-time graduates, on average, earn higher grades and take more rigorous courses than late graduates. Furthermore, even after high school, the average on-time graduate is typically more successful than the average late graduate (when comparing similar students).
College. Late graduates are less likely10&than similar&on-time graduates to receive an Associates or Bachelors degree. Furthermore, on-time graduates earn 34 more11&credits on average and obtain higher grades than late graduates.
Work. Late graduates were as likely12&as similar on-time graduates to be employed in 2000 than. However, employed on-time graduates earned significantly13&more income&about $2,400 more in 1999&than late graduates of similar background and skills. This would amount to nearly $100,000 more upon retirement in 40 years if the gap remained constant. Interestingly, late graduates were just as likely14&as on-time graduates to have held a job that offered retirement benefits. However, on-time graduates were more likely15&to be covered by health insurance than late graduates. Similar to college, on-time graduates were only slightly more successful in their work than late graduates.
Voting. Late graduates are as likely as similar on-time graduates to have voted16&in a recent election.
Community service. There was no significant17&difference in volunteerism and the data shows that late graduates were just as involved in their communities.
Health. Similar on-time graduates were more likely18&to be non-smokers. Late graduates were just as likely19&as similar on-time graduates to exercise at least one or two days a week.
When we compare all late graduates to all on-time graduates, it is clear on-time graduates are much more successful after high school. However, when we compare students of similar SES and and initial acheivement, the picture is far more mixed. In a perfect world students are better off graduating on-time. But if students do fall behind, they are much better off graduating late than never.
The path of a late graduate
Little is known about late graduates, and more research needs to be conducted. What we can surmise about the path of the late graduate from this data, however, reveals that they start out in the same group as those who will eventually drop out or receive a GED. We also see that& minority students or non-native English speaking students are more likely to graduate late. Students who live in poorer households, attend high schools in urban districts, and have two or more of the risk factors for dropping our are also more like to graduate late.
Moreover, late graduates' preparation for high school is close to that of dropouts and GED recipients. In the eighth grade, they were no more prepared to go on to high school math or English. And they fall further behind their on-time classmates in ninth grade, where they mainly take non-academic math courses, if any math course at all.
But in high school, the picture shifts slightly. Late graduates start making better grades than those who will eventually drop out or receive a GED. These results align with those Nobel Laureate economist James Heckman found in his research on GED recipients: That GED recipients were more likely to drop out of college (Cameron and Heckman 1993). Heckman also found that GED recipients had cognitive skills similar to high school dropouts. But, compared to students who graduated, GED recipients lacked non-cognitive skills such as thinking ahead, persisting at a task, and adapting to an environment (Heckman and Rubinstein 2001). Finally&perhaps simply due to perseverance&late graduates accumulate significantly more credits than GED recipients or dropouts.
The results of late graduates' persistence are striking. As stated earlier, late graduates do markedly better in all arenas than GED recipients or dropouts. And, controlled for SES and achievement level, they come close to on-time graduates' achievement. It's clear once again: The effort to keep students in school pays off.
What should school boards do about late graduates?
Until recently, NCLB did not give credit to schools for students who earned a high school diploma beyond the traditional four years. (States can now have late graduates counted, however, they must apply.) But given the benefits we've illustrated, shouldn&t high schools be recognized for the time and resources used ensuring that students receive the help they need to earn a high school diploma, whether or not the process takes longer than four years?
Accountability systems that only credit schools graduating students on time hold a narrow view of defining success after high school, especially for lower-achieving students. This reticence to recognize late graduates may be due to concern that late graduates aren&t as likely to earn a four-year college degree, and there is some truth in that. However, late graduates do have similar-quality jobs as on-time graduates from similar backgrounds and achievement levels and they display similar citizenship patterns.
We have seen that late graduates are more likely to earn&a college&degree than GED recipients. Throughout high school and into college and the workplace students with higher high school graduation statuses are more and more successful. Of course this means it's best for students to graduate on time. But it also clearly shows that students are better off graduating late than never.
What does this mean for school board members?
Schools should continue to help students graduate on time. Nonetheless, it is worth the district&s time and resources to graduate students who fall behind, even if it takes longer than four years.
The investment in these students could be reduced if a curriculum&were in place to ensure that students are ready for high school work when they leave the eighth grade.
All students should take an academic math course in ninth grade to remain in the mathematics pipeline.
Identify possible dropouts in middle school and establish effective dropout programs. [See &&]
Provide support in high school for low-achieving students to develop the qualities Heckman talks about&thinking ahead, persistence, and adapting to an environment. Teaching work habits and giving help with homework and other supports could make the difference between a student graduating or dropping out.
Why should policymakers give credit to schools for late graduates?
Although on-time graduation is best, late graduates are still better situated after high school than students who dropped out or received a GED.
Schools should be encouraged, through accountability systems, to keep all students in school until they graduate&whether or not it takes longer than four years. It takes time and resources for schools to continue working with late graduates, but these efforts should be encouraged not discouraged.
Accountability systems should give schools credit for all students who graduate late, not just those from certain groups such as special education students or English language learners. Students from all backgrounds fall behind their peers for a variety of reasons&many outside the school's control&so schools should get credit for graduating these students with a regular high school diploma.
&Note:&* Indicates difference is at the 90 percent interval instead of 95 percent.
1&There are, however, studies by&Kienzl, G. and Kena, G. (2006) and Sass, T. R. that provide some insight inot the eventual outcomes of late high&&school graduates. Neither is comprehensive.&&&
2&Statistical significant differences are determined using t-tests within the AM statistical software www.am.air.org.
3&Unless noted otherwise, statistical differences are evaluated at the 95 percent confidence interval, meaning there is only a 5 percent chance the difference happened by chance. But it does not necessarily mean the differences are educationally significant. For more information about statistical significance, see our piece "."
4&We did this by controlling for students& socioeconomic status (SES) in the&eighth grade and their initial achievement as measured by their combined scores on the&eighth grade math and reading assessments. SES and&eighth grade test scores were included as control variables in AM&s logit and regression functions depending on whether the dependent variable was dichotomous or continuous.
5&Students are considered at risk if: (1) they come from a& low-income family, (2) they come from a single parent family, (3) their parents have low educational levels, (4)& they are not proficient in English, (5) they spend&three or more hours at home alone a day, and (6) they had a sibling who had dropped out.&
6&PSE transcripts are only available for 48 dropouts (unweighted). PSE transcripts were available for 8020 students overall.&
8&Difference at the 90% level not 95% level.&
9&NELS variables BYSES and BY2XCOMP are control variables in AM&s Logit and Regression functions depending on whether the dependent variable was dichotomous or continuous, respectively.
10&Comparing similar on-time and late graduates using logit regression model
11&Comparing similar on-time and late graduates using linear regression model&
12&Comparing similar on-time and late graduates using logit regression model
13&Statistically significant at the 90% level when comparing incomes from similar students using linear regression model.&
14&Comparing similar on-time and late graduates using logit regression model&
15&Comparing similar on-time and late graduates using logit regression model
16&Comparing similar on-time and late graduates using logit regression model
17&Comparing similar on-time and late graduates using logit regression model
18&Comparing similar on-time and late graduates using logit regression model
19&Comparing similar on-time and late graduates using logit regression model
This guide was prepared by Jim Hull, the Center for Public Education's policy analyst. Special thanks to Cliff Adleman, author of&Toolbox Revisited, for his insights on calculating high school graduations using the NELS/PETS datasets. Differences and errors, however, in calculating graduation rates are solely those of the author. Special thanks also to Larry Ogle, formerly of the National Center for Education Statistics, and Terrance Moore, of the Ohio Department of Education, for their insights and suggestions. Errors and opinions found within this guide, however, are solely those of the author.
&Posted:&February 11, 2009
&2009 Center for Public Education
Display name as (required):
Comments (max 2000 characters):
Pre-KindergartenInvesting in high-quality pre-kindergarten education yields benefits for kids, school, and communities.Data FirstHow good are your schools? How can they improve? Data can provide answers.