Title
Subscribe Today
Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13
Topics
Discussion
Announcements
 

Who Drops Out of High School and Why? Findings from a National Study


by Ruth B. Ekstrom, Margaret E. Goertz, Judith M. Pollack & Donald A. Rock - 1986

Using data from High School and Beyond, a national longitudinal study of American high school students, the authors investigated who drops out, why one student drops out but not another, what dropouts do while peers remain in school, and what impact dropping out has on gains in tested achievement. (Source: ERIC)

The research on which this paper is based was supported under Contract No. 300-83-0247 for the U.S. Department of Education, National Center for Education Statistics (NCES). The opinions and findings expressed here do not necessarily reflect the position or policy of NCES and no official endorsement should be inferred. The authors contributed jointly and equally to this paper: the order of listing is alphabetical.


Using the most comprehensive data set on school dropouts that we have to date, the High School and Beyond study, Ruth Ekstrom, Margaret Goertz, Judith Pollack, and Donald Rock provide an analysis of the salient characteristics of the dropout population.


In 1983, the National Center for Education Statistics (NCES) contracted with Educational Testing Service (ETS) to conduct a study using NCESs High School and Beyond (HS&B) data base. This study, which was part of NCESs Study of Excellence in High School Education, included a longitudinal analysis relating growth and development of 1980 high school sophomores to their school experience over the period 1980-1982. Since these data provide information on administrative practices and school policy, on curriculum and requirements, and on student outcomes, a thorough investigation of models of the process of schooling and of casual relationships among school and student characteristics was possible.


One of the issues examined in the longitudinal study was: How did the cognitive achievement and attitudes of high school dropouts differ from those of teenagers who chose to stay in high school? The research focused on four questions:


Who drops out?

Why does one student and not another drop out?

What happens to dropouts during the time that their peers remain in school?

What is the impact of dropping out on gains in tested achievement?


This article summarizes the answers to these questions.1


Previous research has indicated that high school attrition is related to background, achievement and attitudes, and individual behaviors.


The two background characteristics that are most strongly related to dropping out are socioeconomic status (SES) and race/ethnicity. Students of lower socioeconomic status have been consistently shown to have higher dropout rates than high socioeconomic status students.2 Dropout occurs more often among Hispanics than among blacks, and more often among blacks than whites.3 Other background factors associated with dropout include coming from a single-parent family,4 coming from a large family,5 and living in the South or in a large city.6


Low academic achievement, as indicated by low test scores and low grades, has also been consistently associated with high school attrition. Low scores on standardized tests have been found to be good predictors of dropout.7 Academic failure, as indicated by low grades, is also consistently related to dropout.8 Students who become dropouts have been shown to be dissatisfied with school and to have lower self-esteem.9 Students with no plans for postsecondary education have also been shown more likely to become dropouts.10


Student behaviors that have been found to be associated with dropout include enrollment in a nonacademic (vocational or general) curriculum11 and problem behaviors such as delinquency and truancy.12 Other researchers have pointed out the role that employment during high school13 and pregnancy14 play in dropout.

METHODOLOGY


The analyses were conducted using data from High School and Beyond, a national longitudinal study of American high school students sponsored by NCES. The data in HS&B are drawn from a highly stratified national probability sample of about thirty thousand high school sophomores who attended about one thousand public and private high schools in 1980. Students were administered base-year survey and achievement tests in vocabulary, reading, mathematics, science, writing, and civics. A follow-up survey collected data from and retested over twenty-two thousand of these students who were seniors in 1982 and over two thousand of the individuals who had dropped out of school by 1982.


The findings reported in this article are based on three different kinds of analysis. First, descriptive analysis was used to describe who stayed in school and who dropped out between the sophomore and senior years. Students who stayed in school (stayers) were compared with those who did not complete school (dropouts) on a number of dimensions: race/ethnicity, socioeconomic status, family structure, home educational support system, ability and attitudes, and school behaviors. Second, path analysis was used to explain why some students and not others drop out of school, The robustness of some of the estimates in the path-analysis results was further verified by comparing these estimates with those of a propensity analysis.15 Third, a value-added analysis was conducted to estimate the relative impact of staying in or dropping out of school on gains in tested achievement.

WHO DROPS OUT OF SCHOOL?


Students who later became dropouts differed significantly in their sophomore year from those who chose to remain in school. These differences include background, educational achievement and other school-related behaviors, out-of-school activities, educational aspirations, and attitudes toward self and society. Thirty percent of the dropouts reported leaving school during or before the end of tenth grade, 44 percent during or before the end of eleventh grade, and 26 percent during twelfth grade.

BACKGROUND


As shown in Figure 1, dropouts are disproportionately from low SES families and racial/ethnic minority groups. While 15 percent of students who were sophomores in 1980 did not complete high school two years later, nearly 25 percent of black students dropped out. Dropouts were also more likely to be older, to be males rather than females, and to attend public schools in urban areas in the South or West.


Dropouts tended to come from homes with a weaker educational support system. Compared with stayers, dropouts: (1) had fewer study aids present in their homes, (2) had less opportunity for non-school related learning, (3) were less likely to have both natural parents living at home, (4) had mothers with lower levels of formal education, (5) had mothers with lower educational expectations for their offspring, (6) had mothers who were more likely to be working, and (7) had parents who were less likely to be interested in or to monitor both in-school and out-of-school activities.

EDUCATIONAL ACHIEVEMENT AND OTHER SCHOOL-RELATED BEHAVIORS AND ATTITUDES


Dropouts exhibited different school behaviors. They had lower school grades and lower test scores, did less homework, and reported more disciplinary problems in school.


It appears that the gap between stayers and dropouts is greater in the area of school performance (as measured by reported school grades) than it is in tested achievement. The typical sophomore who remained in school reported a grade average of B, while those who dropped out reported grades of mostly Cs, a difference of about one standard deviation. The typical dropouts grades were at approximately the sixteenth percentile of the school stayers.


[39_688.htm_g/00001.jpg]


The dropouts had lower sophomore-year scores on all of the HS&B achievement tests than the stayers. The mean score differences were smallest in science and largest in mathematics. The dropouts science test scores placed them at about the twenty-eighth percentile of school stayers while their mathematics scores placed them at about the twenty-third percentile of the stayers.


Not surprisingly, the dropouts reported doing less homework as sophomores than did the school stayers (an average of 2.2 hours of homework a week compared with 3.4 hours a week reported by stayers).


[39_688.htm_g/00002.jpg]


The dropouts were also more likely to report having behavior problems while in school. As shown in Table 1, dropouts were more likely than stayers to have cut classes, to have had disciplinary problems, to have been suspended from school, or to have had trouble with the police. The dropouts also reported higher rates of absenteeism and tardiness.


The dropouts appear to feel alienated from school life. They report lower levels of participation in most extracurricular activities, especially in athletics. They are less likely to feel satisfied with the way their education is going, to be interested in school, or to like working hard in school (see Table 2). They are also less likely to feel that they are popular with other students, to feel that other students see them as good students, as athletes, or as important, and more likely to feel that other students see them as troublemakers.


The dropouts appear to have chosen friends who are also more alienated from school than the friends of the stayers (see Table 3). The largest dropout-stayer differences in close friends involve plans to attend college and being interested in school.


[39_688.htm_g/00003.jpg]


[39_688.htm_g/00004.jpg]

OUT-OF-SCHOOL BEHAVIOR


Students who later became dropouts reported spending more time in their sophomore year riding around or going on dates than students who stayed in school (see Table 4). Dropouts were less likely than stayers to discuss their experiences with their parents and, as indicated earlier, parents of dropouts are reported doing less monitoring of the students activities both in school and out. Dropouts also reported spending less time reading than did stayers.


Dropouts were slightly more likely to be working for pay during their sophomore year (47 percent) than were stayers (42 percent). The future dropouts reported working more hours per week than the stayers and receiving a higher hourly wage. The dropouts were more likely to report finding their current or most recent job more enjoyable than school (66 percent) than the stayers (54 percent). The dropouts also reported that their job was more important to them than school more often than did the stayers (23 percent vs. 10 percent).


[39_688.htm_g/00005.jpg]

EDUCATIONAL EXPECTATIONS


As sophomores the future dropouts expected to attain less education than did the stayers. The typical stayer thought he or she would complete between two and four years of college, while the typical dropout thought he or she would finish high school and take some junior college training.

ATTITUDES TOWARD SELF AND SOCIETY


The HS&B questionnaires included scales to measure self-esteem, locus of control, gender role attitudes, and life values. Dropouts differed from stayers on most of these scales.


Self-esteem items that focused on whether the students had a positive attitude toward themselves or felt of equal worth compared with others showed no practical or significant differences between dropouts and stayers. However, when asked if they were satisfied with themselves or if they had much to be proud of, dropouts were significantly more likely than stayers to show lower self-concept.


On most of the locus-of-control items, dropouts responded with a significantly more externalized sense of control, indicating that they are more likely than stayers to feel that their destiny is out of their hands.


The gender-role-attitudes scale showed that the females who became dropouts were significantly more likely to agree with items such as Most women are happiest when making a home and It is usually better if the man is the achiever and the woman takes care of the home than were female stayers.


The contrast in life values of stayers and dropouts is shown in Table 5. Dropouts are more likely than stayers to give importance to getting away from this part of the country and to having lots of money.

WHY DOES ONE STUDENT AND NOT ANOTHER DROP OUT OF SCHOOL?


The preceding section provides a descriptive summary of how the populations of stayers and dropouts differed in their sophomore year in high school. In this section, the question of why one student rather than another drops out will be examined by analyzing: (1) their self-reported reasons for dropping out of school and (2) the results of a path analysis.


[39_688.htm_g/00006.jpg]

SELF-REPORT OF REASONS FOR DROPPING OUT OF HIGH SCHOOL


In 1982, students who dropped out of school were asked their reason(s) for leaving. The students could check as many reasons as they felt relevant. The major reasons, chosen by 10 percent or more of the dropouts, are shown in Table 6, both for the group as a whole and separately for males and females. Other minor reasons for dropping out included travel (7 percent), and inability to get into desired program, inability to get along with other students, and illness (all 6 percent).


The most frequently reported reasons for leaving school for the total group were poor grades and not liking school. This suggests that about one-third of all dropouts leave high school because they do not achieve in school and/or because they are alienated from school. Males are somewhat more likely to leave school for these reasons than are females.


[39_688.htm_g/00007.jpg]


Males are also more than twice as likely as females to report leaving high school because of behavior problems, including not being able to get along with teachers (21 percent of male dropouts) and being expelled or suspended (13 percent of male dropouts). Males were also more likely than females to leave high school because of economic-related issues. Fourteen percent of the males, as contrasted with 8 percent of the females, said they left school because they had to help support the family; 27 percent of the males and 11 percent of the females said they left school because they were offered a job. Females, in contrast, are more likely than males to leave high school for personal/family formation reasons. Nearly a third (31 percent) of female dropouts reported that they left high school to marry and nearly a quarter (23 percent) stated that they left school because of pregnancy.


The diversity of the reasons given for dropping out of high school, encompassing academic, behavioral, economic, and personal factors, suggested that there is no single, simple cause underlying this problem and led to the development of a complex path-analysis model that was tested in the next phase of this study.

PATH ANALYSIS


Student self-reports provide a list of reasons for leaving school, but they do not yield much insight into the causal factors that led a student to drop out of school. A path model was developed that relates demographics, family educational support, behavior, sophomore-year ability and attitudes, and student school behaviors to the students decision to stay in or drop out of school. The model, summarized in Figure 2, contrasts whites with blacks, and whites with Mexican-Americans and Puerto Ricans, to identify possible level differences in home educational support systems and student behaviors. In addition, the model was run separately within racial/ethnic groups to determine whether the educational process works the same way for minority as for white students.


The direct effects of the explanatory variables that had the largest standardized path coefficients, for total group or for the racial/ethnic subgroups, are shown in Table 7.


The demographic variables related to the decision to stay in or drop out of high school were, in approximate order of importance:


Intact family: White and Hispanic, but not black, students who came from an intact, two-parent family were less likely to drop out of school.


SES: Students of higher socioeconomic status were less likely to drop out.


Race/ethnicity: Other things being equal, whites and Hispanics were more likely to drop out of school than blacks. The critical control variables here are sophomore-year grades and achievement test scores.


Region: Whites in the South were more likely to drop out than whites in other regions, assuming all other variables were held constant. Blacks in the South were less likely to drop out than blacks in other regions.


Sex: White and Hispanic males were more likely to drop out than females; black females were more likely to drop out than black males.


[39_688.htm_g/00008.jpg]


[39_688.htm_g/00009.jpg]


The only family-education support variable related to dropping out was study aids in the home. The more study aids available, the less likely white students were to drop out.


Three sophomore-year ability and attitude variables were related to staying in or dropping out of school. They were:


Grades: Students with low grades were more likely to drop out. Grades appear to be more important in the dropping out decision for whites and Hispanics than for blacks.


Mathematics test score: Poor mathematics skills, as measured by the sophomore-year achievement test, were related to dropout, especially for whites.


Locus of control: An externalized locus of control, or the feeling that one can do little to control ones destiny, was negatively related to dropout, especially among minority students.


One school behavior variable, having behavioral problems, was also related to dropout. Students who cut classes, had disciplinary problems, had been suspended, and/or had trouble with the police were much more likely to drop out.


It is clear that having behavior problems and having low grades are the major determinants of dropout. What factors affect these behaviors? The same path analysis showed that several demographic and family variables were related to behavioral problems and to grades.

Behavior Problems


Students exhibiting problem behaviors, such as cutting classes and having disciplinary problems, during their sophomore year tended to be males with low verbal ability (as measured by the vocabulary test scores) and with a sense that they had little control over their lives (externalized locus of control). They tended to come from homes that failed to provide a supportive educational environment. The mothers had low educational aspirations for these students and the parents were not involved in helping the student select a high school curriculum. The major direct effects of the demographic, family, and ability and attitude variables on sophomore-year problem behavior are shown in Table 8.

Grades


Self-reported grades thus far in high school, as of the sophomore year, were highest for students with high verbal ability (as measured by the vocabulary test) who did not engage in problem. behavior and who spent more time doing homework. The typical student with high grades was a female whose family provided strong educational support (as indicated by the mothers educational aspirations for the student and by parental involvement in the selection of the students high school curriculum). These sophomores also tended to be enrolled in the academic curriculum, to do more homework, to have fewer behavior problems, to be involved in extracurricular activities, and to have an internalized locus of control. Other things being equal, grades in the South and in rural schools tended to be higher. The major direct effects of the demographic, family, ability and attitude, and school-behavior variables on grades are shown in Table 9.

WHAT HAPPENED TO DROPOUTS BETWEEN 1980 AND 1982?


HS&B included a follow-up survey of dropouts that collected information on their activities, attitudes, and behaviors between the time they dropped out of school and 1982.


At the time of the follow-up survey, 47 percent of the dropouts were working full-time or part-time, 10 percent were taking courses or participating in job training programs, 16 percent were homemakers, 3 percent were in military service, and 29 percent were looking for work. These percentages varied, however, by gender and by race/ethnicity. For example, more whites and males reported working for pay than did blacks and females.


[39_688.htm_g/00010.jpg]


Dropouts reduced their educational expectations or plans between 1980 and 1982. As sophomores, 40 percent reported they would be disappointed if they did not graduate from college. Two years later, this figure was 26 percent. However, 58 percent of the dropouts reported in 1982 that they planned to complete high school eventually. While dropouts had lowered their educational aspirations, they had an improved self-concept and, as indicated by the locus of control scale, more sense of control over life in 1982 than in 1980.


During the 1980 to 1982 period, 21 percent of the dropouts reported that they had participated in a job training program and/or educational activities other than formal educational course work. Seventeen percent had enrolled in an educational institution, and by 1982, 14 percent reported they had obtained a General Educational Development (GED) high school equivalency certificate.


[39_688.htm_g/00011.jpg]

THE IMPACT OF DROPPING OUT OF SCHOOL ON GAINS IN TESTED ACHIEVEMENT


A value-added analysis was carried out to estimate the relative impact of early dropout (before the end of the junior year) on achievement gains as contrasted with later dropouts and with stayers in each curriculum. The groups contrasted with the baseline early-dropout group included:


Late dropouts and/or early dropouts who subsequently received additional education or training, such as formal tutoring or GED work

School stayers in the general curriculum

School stayers in the academic curriculum

School stayers in the vocational curriculum


Table 10 presents the standardized adjusted gains averaged across achievement areas by various subgroups and curriculum classifications. The entries in the table indicate the gains in pretest standard deviation units by curriculum type and by racial/ethnic and gender groups.


[39_688.htm_g/00012.jpg]


The results clearly show that staying in school positively impacts ones gains in achievement, and that staying in school in the academic or, to a lesser extent, in the general curriculum leads to larger overall gains than staying in the vocational curriculum.


The results also show that females, and to a lesser extent minorities, are relatively bigger losers when they drop out of school. Blacks and females fall the furthest behind in the language development areas of vocabulary, reading, and writing when they leave school early. Because females and minorities tend to take fewer high school courses in science and mathematics than do males and whites, the impact of dropout is less for them in these areas.

SUMMARY


Identifying who drops out of school and why and assessing the impact of this decision on future values, behaviors, and achievement are difficult tasks. Educators and policymakers do not share a common definition of dropout. Students drop out of school for a variety of personal reasons, and the impact of leaving school is affected by when an individual drops out, what he or she does after dropping out, and the outcome measures employed.


The analyses of the 1980/1982 HS&B data reported in this paper shed some light on the dropout problem. First, the critical variables related to dropping out are school performance, as measured by grades, and extent of problem behavior. These variables are more important in explaining dropout behavior than sophomore ability, as measured by test scores.


Second, problem behavior and grades appear to be determined in part by the home educational support system. The mothers educational aspirations for the student, the number of study aids in the home, parental involvement in curriculum choice, and the provision of opportunities for nonschool learning all affect school academic performance and/or deportment.


Third, regardless of ethnicity, gender, or curriculum choice, staying in school increases achievement gains in all tested areas. Students in the academic curriculum gained most, followed by students in the general and then the vocational curriculum. Females and minorities suffered the greatest with respect to unrealized achievement gains if they dropped out of school. These unrealized achievement gains for women and minorities were largest in the language-development areas of vocabulary, reading, and writing.


These findings have significant implications for the development of policies that deal with the dropout problem. No single program or policy can meet the needs of the diverse dropout population. Three major types of programs are needed: (1) programs to help pregnant teenagers remain in school; (2) programs to help youth with economic needs combine work and education; and (3) programs directed toward students who perform poorly because they are dissatisfied with the school environment.


The study also showed that the students home environment has a critical, although indirect, impact on the decision to leave school. Policies should be developed to help parents increase their interest in and monitoring of their childrens school progress. It is also important to identify potential dropouts prior to the high school years and to begin interventions when the first behavioral signs (e.g., disciplinary problems, poor grades, poor attendance) are noted.



Cite This Article as: Teachers College Record Volume 87 Number 3, 1986, p. 356-373
https://www.tcrecord.org ID Number: 688, Date Accessed: 3/9/2022 11:05:48 AM

Purchase Reprint Rights for this article or review
 
 
Member Center
In Print
This Month's Issue

Submit
EMAIL

Twitter

RSS