Using school-wide data to advocate for student success.

The past three decades have seen a growing interest in improving the quality of U.S. public education, especially related to increasing the high school graduation rate. Despite the attention, the most recent national data show limited gains over the past quarter of a century. Although more than 85% of 19- to 24-year-olds nationally have earned a high school diploma or its equivalent (U.S. Department of Commerce [USDOC], 2000), approximately 5 of every 100 students over the age of 16 who were enrolled in high school in any of the past 10 years dropped out prior to completing a high school program (U.S. Department of Education [USDOE], 1999). Moreover, more than two thirds (67.3 percent) of all dropouts were 16 to 18 years old.

Critically, reform efforts have done little to improve the rate at which students graduate from a regular high school program by the typical age of 18 years. And, as we discuss, wide disparities in data collection methods have made it impossible to accurately describe the academic trajectories of students and develop responsive policies based on meaningful data. As a guiding metaphor for the present study, therefore, we imagined a busload of students bound for graduation. What we wanted to know was how many of the students who were expected to get on the bus (i.e., those eighth graders who were presently enrolled in the school district and assigned to enroll in the high school in the fall) actually got on the bus (i.e., enrolled) and then how many of these students actually got off at graduation 4 years later. Along the way, we hoped to capture an accurate picture of how many of this cohort got off prior to graduation, when and why they got off, and where they went once they got off the bus (i.e., withdrew, transferred, or were retained in grade). In addition, we were interested in identifying when and how many students got on the bus along the way, either as transfers from other schools or as members of the cohort returning to the school.

THE HIGH COST OF DROPPING OUT

Recognizing that a 12-year curriculum is the norm for virtually every U.S. public school today, the failure to receive a high school diploma "on time" places millions of young Americans at risk each year. Unable to meet the minimum requirement for advanced education or entry into the workforce, dropouts experience higher unemployment rates and lower earnings than other workers (Coley, 1995; Sherraden, 1986; USDOE, 1999). Among males age 25-34, for example, the 1998 employment rate was 87% for those who received a high school diploma or were granted a General Education Diploma (GED) versus 78.5% among those who dropped out (USDOE, 2000a). Although in 1998, females were slightly more likely than males to have completed high school by age 24 (87.1% versus 84.6%; USDOE, 1999), females with a high school diploma or GED were employed far more often than those who had dropped out of school (69.5% versus 47.3%, respectively; USDOE, 1999). Significantly, failing to complete high school presents several major social consequences that include (a) forgone national income, (b) forgone tax revenues for the support of government services, (c) increased demand for social services, (d) increased crime and antisocial behavior, (e) reduced political participation, (f) reduced intergenerational mobility, and (g) poorer levels of health (Coley; Jaffe, 1998; Rumberger, 1987; Tidwell, 1988; USDOE, 1999).

As the demographic face of America changes, it is instructive to view the dropout statistics by race and socioeconomic status. National statistics by race show that Hispanic students (7.8%) were more likely than Black (6.5%), White (4.0%), and Asian (5.0%) students to leave school prior to graduation (USDOE, 2000a). In addition, more young adults living in low-income families (11.0%) dropped out versus middle (5.0%) and high-income (2.1%) families (USDOE, 1999). These disparities highlight the fact that the dropout problem is most concentrated in large urban areas where poor and minority students tend to live (Coley, 1995). It is not surprising, therefore, that dropouts have been identified as individuals who usually experience alienation (i.e., rootlessness, hopelessness, and estrangement) from their school, home, neighborhood, and society in general (Coley; Tidwell, 1988). Further, Tidwell noted that dropouts are more likely to experience racial discrimination; receive disapproval from parents, friends, and society; and devalue themselves because of their decision to leave school.

Clearly, a global society and the status of the United States as a world leader are increasingly dependent on the development and better use of all human resources (Elam, 1993). This reality points to the importance of getting students into the right curriculum and supporting them once there if all students are to participate unconditionally in the 21st century economy (Fields & Hines, 2000; House & Martin, 1998; Schneider & Stevenson, 1999).

COUNSELORS AS ADVOCATES FOR DATA-DRIVEN POLICY AND PRACTICE

In this era of educational reform, greater emphasis is being placed upon making school personnel accountable for bringing all students to high levels of academic performance (Eriksen, 1997; Fields & Hines, 2000). Ensuring that every student has the supports necessary to remain in school until graduation has become a national priority (American School Counselor Association, 1996). All school personnel and educational policy makers are responsible for establishing responsive policies and initiating new strategies to prevent students from leaving school prior to graduation. In addition to teachers and school administrators, ensuring the success of every student falls upon school counselors, psychologists, social workers, staff, students, parents, business people, and the community at large (House & Hayes, 2002).

The significant role played by school counselors for ensuring student success places a special burden on them to advocate for responsive programs that lead to graduation for all students on time. Student files, grade reports, school-wide policies, and referrals for human services that cross their desks every day give school counselors access to critical data about student placements, academic success, course-taking patterns, and faculty as well as student performance that are critical to developing responsive programs (House & Hayes, 2000). Because they have a school-wide perspective on serving the needs of every student, school counselors are in an ideal position to assess the school for systemic barriers to academic success.

DEFINING SCHOOL DROPOUTS

Despite the importance attached to earning a high school diploma, charting the change in graduation rates remains a contentious issue. In particular, meaningful changes in educational policy must be based on meaningful data. To be truly useful, such data must allow for direct comparison from one year to the next of all students in a given age cohort, must account for the academic trajectory of every student, and must permit comparison across cohorts, and across schools, districts, and states, by race, gender, and socioeconomic class. The demand for such data notwithstanding, current practices make such comparisons difficult, if not impossible.

Typically, dropouts are described as individuals who leave school and do not graduate with their class. As Dusek (1996) explained, using this definition to measure school failure provides a cohort rate, which is based on repeated measures of a cohort of students with shared experiences. Cohort rates measure what happens to a group of students over a period of time and reveal how many students starting in a specific grade drop out over time (USDOE, 1999). Problems arise in obtaining accurate information of cohort dropout rates because students move, change schools for other reasons (e.g., entering a private school, disciplinary action, or participation in a special program), or are retained in a grade.

To correct for these problems in obtaining an accurate accounting of all students, the event dropout rate (USDOE, 2000a) was developed as a measure of the proportion of students who drop out of school in a given year, regardless of their cohort. The event rate reflects the proportion of students who leave school prior to the end of the school year without completing the high school program, whether they returned the next year or not. Although this procedure improves upon the problem of keeping accurate records on all students in a cohort, it does not account for students who complete high school requirements by returning to school or obtaining a GED or for students who transfer into a cohort. Because laws set lower limits on the age at which one may drop out, the event rate is usually calculated for students in grades 10 through 12. As a result, event dropout rates reflect changes in school population in a given year but do not give an accurate picture of what has happened to a group of students over time. Moreover, event rates ignore those students who do not attend school for significant periods of time prior to their 16th birthdays, when they become "eligible" to drop out.

As an alternative to cohort and event rates, the status dropout rate provides cumulative data on dropouts among all young adults within a specific age range (USDOE, 2000a). Because an individual may enroll in school at any age, status dropout rates usually are calculated for an age range, such as 16 to 24 years (Dusek, 1996). Although status rates accurately portray the proportion of persons who complete a high school program or its equivalent, these rates tell little about within-group differences because they do not account for members of a cohort over time. And, as with event rates, status rates typically neglect students who have not reached age 16.

COMPARING DROPOUT RATES

Differences in data collection policies and practices further complicate the comparison of dropout rates from one school district to another. Although more young people finish high school today than at any time previously, no one knows what the high school dropout rate really is in the United States. Despite efforts to standardize the data and its collection, there is no consensus definition of a high school dropout, nor is there a standard method for collecting data and computing the dropout rate.

Common sense suggests that a dropout is someone who has not graduated from, or is not currently enrolled in, a full-time, state-approved education program (Rumberger, 1987). The definition used by the U.S. Census Bureau, for example, excludes persons who have a regular or equivalent high school certificate or who are still attending school (USDOC, 2000). The National Center for Education Statistics (USDOE, 2000c), however, uses a definition based on comparing snapshot counts of students at the beginning of each school year that ignores ill and suspended students (Jaffe, 1998). Because states disagree on the definition of a dropout, they use different reporting practices that typically fail to account for students who either transfer in or who drop out prior to beginning the next school year. At the national level, the two most widely cited dropout statistics are computed from U.S. Census data, while the high school attrition rate is computed from state-level enrollment data. As a result, these two sources of information portray widely different dropout rates and probably represent lower and upper limits to the true rate.

Examining the USDOE (2000a) data reveals that the status dropout rate among 16 to 24 year-olds (which counts GED recipients and those who complete special programs as graduates) has improved only slightly over the past quarter century and has been relatively stable since the 1990s. The event dropout rate in 1999 for youth ages 15 to 24 in grades 10 to 12 indicates that 5 out of every 100 young adults enrolled in high school in October 1998 dropped out before October 1999 without successfully completing a high school program (USDOE, 2000a). When one considers that this rate is annual and accounts for students through age 24, one can see how misleading such a statistic can become when trying to develop responsive policies that rely on predictions of how many eighth graders are likely to graduate from a typical high school four years later. When comparing rates, therefore, it is important to understand what each statistic represents. In particular, developing responsive policies demands that one consider whether or not the target population is still in school and about to drop out (prevention) or already out of school and in need of special services (remediation).

The need to disaggregate data in developing responsive policies becomes especially meaningful when comparing status and event rates by race. In reference to status dropouts, for example, the rates for youth ages 16 to 24 who are out of school and who have not earned a high school credential was 11.2% in 1999 (USDOE, 2000a). Broken down by race, however, the status dropout rates for each group tell a very different story, with Whites (7.3%) and Asians (4.3%) well below the average and Blacks (12.6%) and Hispanics (28.6%) above the mean. As more young people overall are finishing high school, the decrease in the dropout rate for Blacks is especially dramatic. In 1995, about the same proportion (87%) of White and Black students graduated from high school (Jaffe, 1998; USDOC, 1996). Nonetheless, the same data revealed that among students in grades 10 through 12, the percentage of dropouts among Black students declined from 11.6% in 1974 to 6.3% in 1996 compared to 6.0% and 4.5% for White students over the same period. Disaggregating the data reveals that among Black males within these same years, there was a decline from 10.8% to 4.6%, whereas the difference among White males was 7.0% and 4.8%. Among females, however, the dropout rate decreased among Black students from 12.2% to 7.8% and among White students from 5.1% to 4.1%. Despite the fact that the gaps between dropout rates comparing Blacks to Whites or males to females are closing, the rate for Black males is now actually lower than for White males, while the rate for Black females is nearly double the rate for White females. National policy directed at reducing the dropout rate necessarily will have to account for differences in both race and gender.

Because the national data exclude students below tenth grade or those under age 16, the actual dropout rate is likely to be much greater than the percentages reported here. Because educational policies will be more effective when they support preventive strategies, school officials need data that accurately describe students' academic trajectories before they drop out.

CHARTING THE COURSE: A CASE STUDY

This study was predicated on the assumption that the goal for the participating high school was to graduate students on time (i.e., within 4 years of first enrolling in the high school as a ninth grader). Thus it was focused on student retention to graduation. Annual attempts by counselors to report graduation rates accurately to the administration, however, led to conflicting reports and the recurring demand to know exactly how many students in a given class were graduating on time. An antiquated accounting system made it difficult to obtain accurate data on transfers within the system (i.e., between the high school and a second high school or the Alternative School for severe disciplinary cases), or between the high school and other school systems. Moreover, accurate records on exactly when and the reasons why a particular student dropped out were difficult to maintain. Quite simply, dropouts do not typically stop by the counselor's office to notify officials that they will not be coming back to school.

Because school officials were interested in knowing who drops out, when, and why, they were especially interested in accounting for every student who enrolled in and failed to graduate from their school over a period of 4 years. Because many local and national studies underestimate the dropout problem by using cross-age samples, extending the age at high school completion to 24 years, or using 10th grade enrollment as the population statistic, school officials were interested in obtaining a more accurate statistic. A review of the relative merits of using the various statistics noted in the discussion above supported the use of a cohort model as the best choice to track individual students from eighth to twelfth grade. In this way, school officials hoped that they would be able to determine the academic trajectory of a cohort of students over the period of 4 years to their scheduled graduation date.

METHOD

Participants

The data used in this study were drawn from the school and district records of a cohort of eighth grade students, who were scheduled to enter a large, urban high school located in a university town in the southeastern United States in August 1993 as part of the graduating class of 1997. As one of two high schools in the district, the school in this study had a total school population of approximately 1,400 students distributed over grades 9 through 12. Of the 451 students in the cohort: 211 were female, 240 were male; 247 were Black, 196 were White, 6 were Hispanic, and 2 were Asian; 179 resided in one zip code within the district, 165 resided in another, and the remaining students resided in an additional 11 different zip code areas, ranging from a high of 42 students to a low of 1 in each of 6 different areas.

Because we were interested in identifying the academic trajectory of every student through graduation, a dropout was defined as a student who withdrew (or was withdrawn), transferred from the high school, was retained in a grade, and/or did not graduate on time with the cohort. Because many students who ultimately fail to graduate on time drop back in the system (i.e., they are retained in a grade) before they drop out, these individuals were accounted for by the use of a cohort rate. Accordingly, "drop backs" were defined as students who failed to graduate on time, but were still enrolled in school at the time of graduation.

Procedure

A research team of doctoral and master's students was recruited by the first author and coordinated with the assistance of the second and third authors. Altogether, more than 10 university faculty and graduate students worked on the team over a period of 2 years. Relying initially upon enrollment data supplied by the school district, the academic trajectories of every student scheduled to enroll in the high school after eighth grade were identified by name and cross-tabulated to identify changes in the student's status (i.e., enrolled in grade level, dropped back, withdrew, transferred) over time.

Team members reviewed quarterly printouts of student enrollment data over the 4-year period at the district level and matched these results against reports available from the school. Working in consultation with school district computing personnel, the school registrar, and the school counselors, confusing codes and changes in reporting procedures were clarified until team members had produced as reliable a listing of each student's status by quarter for 4 years as was possible using the existing data. With approximately 90 percent (n = 1,305) of the students fully accounted for, discrepancies and omissions for each of the remaining students were identified and prepared for a review with the school counselors. Meeting with them as a group, the authors (in two different meetings separated by several months) checked the team's master list against the counselors' personal records and their recollections of events related to each student's status.

Given the population of students for whom the data were missing, it was not surprising to find that the counselors were often unfamiliar with these students or reported conflicting circumstances surrounding the school behavior of specific students. Lacking a clear confirmation of a student's status, potential referral sources were identified among the faculty, administrators, staff, students, and community agencies that might provide additional information. Following up on the referrals (which often lead to further referrals), the authors interviewed 24 different referrals separately and in small groups to obtain an accurate picture of the status of each of these students on the date the cohort graduated. This process was continued until the records of all but 10 of the 451 students were located, and their status on the day of graduation and the dates of any transfer or withdrawal were confirmed.

The team then analyzed and plotted the academic trajectory for every student by quarter to determine when each graduated, withdrew, transferred, or was retained in a grade. When and where the students transferred and the reported reasons for withdrawing were also noted in the records for each student. In addition, demographic information such as gender, ethnicity, residential areas, and type of diploma were analyzed for any recurring patterns.

RESULTS

The analysis of school and district records of 451 eighth-grade students scheduled to enter high school in August 1993 indicate that 192 students graduated as scheduled with their cohort (Hayes, 2002). Because this study was intended to identify the academic trajectories of every student in the cohort, results are reported in the actual number of cases and percentages of the whole cohort unless otherwise stated. An additional 7 students had graduated previously and another 5 graduated within several weeks following graduation day. In all, 204 members of the initial cohort of 451 students (45.2%) were considered to have "graduated" from the high school. Of the graduates, 91 (44.6%) were male and 113 (55.4%) were female, representing 20.1% and 25.1% of the cohort, respectively. Of the graduates, 96 (47.1%) were Black, 102 (50.0%) were White, 5 (7.3%) were Hispanic, and 1 was Asian (.5%), representing 23.7%, 22.6%, 1.1%, and .2% of the cohort, respectively.

By contrast, 126 students had been withdrawn, another 65 had transferred, and 32 never enrolled in the ninth grade. The circumstances of 3 students for whom there are records remain unknown. Add to these numbers the 21 students who were enrolled but had been retained in grade (dropped back) at the time of graduation and the total number of students who did not receive a diploma from the high school as scheduled soars to 247 (54.8%). Of these students, 98 (39.7%) were female and 149 (60.3%) were male; 152 (61.5%) were Black, 93 (37.7%) were White, 1 (.4%) was Hispanic, and 1 (.4%) was Asian.

Of course, it can be anticipated that some students who withdrew or transferred graduated subsequently from other schools. Indeed, 17 members of this cohort graduated from the other district high school and an additional 13 members were currently enrolled in that school. In addition, 9 members had transferred to adult or post-secondary education, and 36 had transferred to other school districts, home schooling, or a private school. Further, some of those who failed to enroll may have enrolled in other schools from which they may have graduated subsequently. As of graduation day, however, none of the members of the cohort had received a GED according to state records. If one accepts (and these are extraordinary claims) that every student who transferred out of the district (n = 36) or failed to enroll (n = 32) subsequently graduated from another school, and these students were added to the total of all members of the cohort known to have received a diploma from the school district (n = 221), the most optimistic graduation figure would be 289 for a rate of 64.1%.

Clearly, the school was far from meeting its goal of preparing every student to graduate in 4 years. Because the purpose of this study was to understand the academic trajectories of students retained within the school, it was important to know which, when, and under what circumstances students left the school. It should be noted, however, that once calculated for one school, it was an easy matter to account for the same overall performance at the other high school in the district, which turned out to be a comparable 48%.

Transfer and Withdrawal

Transfer and withdrawal represent two very different sets of circumstances. For the most part, students are withdrawn by the enactment of attendance policies that remove the student from the rolls after 30 days of nonattendance. A review of the results shows that of the 126 students who were withdrawn, 85 were due to a lack of attendance. Add to this total the 10 students for whom there was no record and the 27 students for whom no reason was given in the records, and the reasons remain unknown for which 96.8% of those who were withdrawn from school failed to attend. In effect, the results reveal how but not why a student was withdrawn from school. When these 122 withdrawals are added to the 32 students who never enrolled and the 3 students whose whereabouts remain unknown, the reasons for which more than one third (34.8%; n = 157) of all the students in the cohort left school are unknown.

Transfer, on the other hand, requires that the students or their guardians initiate some action, largely due to the need to forward academic records to the receiving school in compliance with state laws. A review of the data reveals that 29 students transferred to another school or program in the district and were enrolled there on the day of graduation. Similarly, 36 students transferred to other school districts, but there was no way to know from the records on hand if they were still enrolled in that or any other school. In all, 65 members of the cohort (14.4%) transferred to another school by graduation day. Of those who transferred to other schools or programs, 33 (50.7%) were Black, 31 (47.7%) were White, and 1 (1.6%) was Hispanic; 41 (63.1%) and 24 (36.9%) of those who transferred were males and females, respectively.

One of the many things expected from this study was the ability to confirm or deny long-held assumptions operating within the school community. An examination of the transfer data reveals that the widely held assumption that students were leaving the school as part of "White flight" to a neighboring school district was not upheld. In fact, the data show that more than half of those transferring remained in the district and, of these leaving the district (n = 36), nearly one third (n = 11) were Black and more than half (n = 19) actually transferred out of the state.

Data were also analyzed according to residential address for each student to test the assumption that students from identified low-income neighborhoods were more likely to drop out of school. Results reveal that of those who graduated on time, approximately 23% (n = 47) came from low to middle SES areas and approximately 77% (n = 157) came from middle to upper SES areas. Findings also revealed that a larger proportion of students who transferred and withdrew resided in low to middle SES areas, with one area accounting for 46.6% (n = 89) of all transfers and withdrawals. Moreover, this one area accounted for 58.8% (n = 50) of all withdrawals that were due to lack of attendance.

DROPPING OUT

As noted above, 32 of the students failed to transition from the eighth to the ninth grade as scheduled. Results also revealed that 47 other students were withdrawn or transferred from the school before the end of the ninth grade. Therefore, 16.2% of the cohort dropped out by the end of the first year of high school. Another 8.2% (n = 37) of the cohort left the school during the fall of the second year, with a total of 141 (29.9%) of the cohort having dropped out by the end of the second year. During the third year, another 9.3% of the cohort (n = 42) dropped out. As many as 33 students (7.3%) left during their fourth year, 12 of them within months before graduation day. In total, during the 4-year period, August 1993 to June 1997, 216 (47.9%) of the students in the cohort left the school by either transferring to another school or being withdrawn from school.

Another assumption operating within the school was that a few students were creating most of the paperwork related to transfers and withdrawals by initiating these actions repeatedly. An analysis of the data by quarter revealed that only 13 of the 191 students (1.9%) who withdrew and/or transferred from the school did so more than twice. None transferred/withdrew more than six times in total and, in nearly every case, as many as four of these actions took place in a single quarter. Clearly, a few students were creating a lot of paperwork. Among the 240 students who graduated on time, only 1 withdrew or transferred at some point during the 4 years. Significantly, these results show that, with this one exception, every student who sought a transfer or who withdrew from school, whether or not the student subsequently dropped out of school, did not graduate on time.

Finally, many school personnel assumed that the dropout problem was largely a function of students dropping out prior to functionally beginning their studies in high school (i.e., prior to the ninth week of the first semester of the freshman year) and, therefore, beyond the control of the high school personnel. Others believed that the dropout problem began with an unresponsive but demanding high school curriculum for which students were under-prepared. An analysis of the academic trajectories revealed that 12.4% (n = 56) of the cohort had dropped out by the end of the first quarter of the ninth grade and that an additional 12.0% (n = 54) of the students dropped out by the same point the following year. Clearly both groups were correct. Further, of these students, 61 were Black and 48 were White, 62 were male and 48 were females, thereby dispelling the myth that the problem was largely confined to Black males. Indeed, all of the students who were retained in grade (i.e., dropped back) were Black (n = 21), two thirds of whom were males. As one teacher noted, "we can still work with these students." And for those who believed that the dropout problem goes away after the freshman year, the data show that another 13.8% did not pass the tenth grade, and another 9.3% did not pass the eleventh grade, leaving only 59.4% (n = 268) of the cohort to begin the senior year.

DISCUSSION

Although no one really knows the specific factors that contribute to students dropping out of school, this study identifies the critical junctures in the decision-making process that lead to dropping out. Although many believe the primary reason for dropping out of high school is economic (Fuhrmann, 1986), Zimiles and Lee (1991) argued that only a small percentage leave school for financial reasons. Clearly, the data in the present study support that dropping out is more prevalent among adolescents from poorer backgrounds; nonetheless, only one student reported financial hardship as the reason for withdrawing from school.

Students drop out of school at various times during the year, especially during transitional periods. Because they have a school-wide perspective on serving the needs of every student, school counselors are in an ideal position to assess the school for systemic barriers to academic success in the ninth or tenth grades, when declining academic performance and absences become more prevalent (Fuhrmann, 1986). In this study as well as nationally, the first 3 months of the calendar year reflect the greatest dropout activity (Eccles, Lord, & Buchanan, 1996; Eccles, Lord, Roeser, Barber, & Jozefowicz, 1997; Tidwell, 1988). Although the present study did not investigate the possible causes for the high dropout rate during the transition to high school, the data clearly support that this time is critical time for students, and one that demands a collaborative response across schools to support students in making a successful transition from middle to high school.

Although this study focused on high school students, interventions need to begin early, even at the elementary school level (Vitaro, Brendgen, & Tremblay, 1999). Counseling groups and interventions should focus on the importance of obtaining a high school education, a commitment to learning, and a sense of belonging to the school. Specific interventions should address remedial education, cultural differences, peer relationships, student-teacher relationships, and school-parent relationships. To address these areas, school counselors can consult with school administrators and staff, students, parents, and community organizations about ways to collect accurate data.

Getting Good Data is Good Practice

The collection of accurate data and the calculation of high school dropout rates are daunting, but important tasks. As indicated by varying statistics on national and local dropout rates, there are different methods of calculating the figures and different factors may be taken into account. Many reports fail to include students in the ninth grade or below. In the present study, such a practice would make it possible to accurately (if not honestly) report a graduation rate for sophomores of 65.8% that inflates the overall cohort rate well beyond the 45.2% recorded here. In addition, many studies do not report cohort dropout rates, but instead report event or status dropout rates. In the present study, the event rates for each year of the first 3 years of the study were 7.1%, 11.2%, and 16.7%, suggesting to the unaware that the graduation rate each year must be quite high. Most misleading is the accurate (if not honest) statistic that 90.7% of the senior class graduated on time. Another tool used to measure student enrollment and completion rates is a comparison of the number of ninth graders versus the number of students who graduate 4 years later. Recognizing that 7.1% of the class of 1997 never enrolled, and that 56 of the 72 students who enrolled in the school over the next 4 years to graduate with the class of 1997 did so in the freshman year, inflates the graduation rate to 58.1%.

In light of the pressures on school administrators to use "hard data" to support their success in meeting district goals, it seems likely that many graduation rates grossly overestimate the school's success in retaining students to graduation. By contrast, this study reveals an honest and courageous attempt by counselors and school administrators to obtain both accurate and useful data on every student as a first step in developing responsive policies to reduce the dropout rate and improve instruction.

From Good Practice to Good Policy

Failing to remain in school until graduation presents important considerations for future research and implications for school counselors and administrators. Having inaccurate or incomplete dropout data is misleading and seriously undermines the ability of school personnel to make informed policy decisions or develop responsive programs. In their role as student advocates, the school counselors in this school promoted an honest examination of the facts of dropping out, assisted in the collection and refinement of student data, and helped the school to serve all students better. Having accurate data assisted in the development of effective school policies and procedures that were designed to prevent behaviors that lead to dropping out, to help improve the transition from middle school to high school, to assist in the development and evaluation of dropout prevention programs targeted to specific subgroups identified in the study, and to help provide support services and care needed by troubled students, especially those self-identifying for transfer to another school or program. Based on the data reported here, the school principal was confident in his effort to implement a freshman advisory program to restructure classes for sophomores repeating failed courses and to initiate a collaborative student mentoring program for middle school students transitioning to the high school.

As the results of this study reveal, students who transferred or withdrew from the school were almost certain not to graduate on time with their cohort. Clearly, policies and procedures needed to be developed to monitor transfer requests. Despite the prevailing logic that a failing or troubled student might be more successful with a fresh start at another school, the data from this study tell a very different story. Without guided intervention to examine the reasons for the request or without a substantive program of support at the receiving school (if the student actually gets there), little improvement is likely (McNeal, 1997). Instead of sending the problem elsewhere, school personnel began to explore interventions to target the root causes (both personal and systemic) that prompted the transfer request and that signal a call for school personnel, parents, and relevant agencies to come together with the student for an honest appraisal of the situation.

In addition, school counselors can develop responsive programs that target students who exhibit dropping out behaviors such as excessive or prolonged absenteeism, withdrawing from or transferring in and out of the school, lack of involvement in school activities and academic programs, and poor academic performance (Bearden, Spencer, & Moracco, 1989; Davalos, Chavez, & Guardiola, 1999; Fine, 1991). As noted above, school personnel increased the emphasis on developing an orientation program for incoming freshmen and transfer students, given the high dropout rates among these student populations.

School counselors can also assist in the development of a positive school climate that meets the needs of the increasingly multicultural student populations that now exist in many public schools (Cummins, 1986; Reyes, Wagstaff, & Fusarelli, 1999; USDOE, 2000b). As the data support, school can be alienating for many students, especially those who are socially or economically disadvantaged. Programs need to target students who may not feel a sense of connection to the school, especially ethnic minority and language minority students (Davalos et al., 1999). Previous studies of teacher and student empowerment in the same school (Hayes & Lunsford, 1994; Hayes, Paisley, Phelps, Pearson, & Salter, 1997; Jones, 1997) had already shown substantive changes in school climate for teachers. Few lasting effects were seen in student reports of an improved school climate, however. This finding suggests that engaging the commitment of students who come and go over 4 or fewer years is a more difficult task than can be accomplished with a more stable adult population of professionals.

Dropping out of high school is a serious problem, not only for the individual, the school system, and the community, but also for U.S. society. Systemic change in the education of all students will not occur without the sustained involvement of all the critical players in the school setting, especially school counselors. School counselors who serve as advocates can help to ensure that every child receives the academic preparation necessary to choose from a wide array of post-secondary options. In particular, school counselors can lobby for improvements in accounting for students' daily attendance and academic performance, help school officials to identify and evaluate promising programs, support intervention programs targeted at dropout prevention, and lobby at the district, state, and national level to employ proper data collection procedures to ensure an accurate accounting of student attendance. In so doing, poor and minority students, who have not been served well in the past, will have a chance to acquire the skills and knowledge necessary to be full participants in the communities of the 21st century.

Portions of this report are based on a presentation by the first author to the Japan-United States Conference on Juvenile Problems and Violence in a Changing Society, Tokyo, Japan (1999) and papers submitted to the faculty of The University of Georgia by the second and third authors. This study was supported in part by grants from the College of Education at The University of Georgia and the Wallace-Reader's Digest Foundation. The authors acknowledge the assistance of members of the School Research Group, including Deryl Bailey, Melonie Bell, Lorie Blackman, Carolyn Brennan, Bernadine Campbell-Burden, Kim Davis, Lisa DiGeronimo, Marc Grimmett, Marni Mrazik, Pam Paisley, and Robert Socherman in the design and analysis of this study.

References

American School Counselor Association. (1996). Advocacy, alliances and partnerships: The role of the school counselor. Alexandria, VA: Author.

Bearden, L. J., Spencer, W. A., & Moracco, J. C. (1989). A study of high school dropouts. The School Counselor, 37, 113-120.

Coley, R. J. (1995). Dreams deferred: High school dropouts in the United States. Princeton, NJ: Educational Testing Service, Policy Information Center.

Cummins, J. (1986). Empowering minority students: A framework for intervention. Harvard Educational Review, 56(1), 18-36.

Davalos, D. B., Chavez, E. L., & Guardiola, R. J. (1999). The effects of extracurricular activity, ethnic identification, and perception of school on student dropout rates. Hispanic Journal of Behavioral Sciences, 21, 61-77.

Dusek, J. B. (1996). Adolescent development and behavior (3rd ed.). Upper Saddle River, NJ: Prentice Hall.

Eccles, J. S., Lord, S., & Buchanan, C. M. (1996). School transitions in early adolescents: What are we doing to our young people? In J. A. Graber, J. Brooks-Gunn, & A. C. Petersen (Eds.), Transitions through adolescence: Interpersonal domains and context (pp. 251-284). Mahwah, NJ: Erlbaum.

Eccles, J. S., Lord, S., Roeser, R.W., Barber, B. L., & Jozefowicz, D. M. H. (1997). The association of school transitions in early adolescence with developmental trajectories through high school. In J. Schulenberg, J. L. Maggs, & K. Hurrelmann (Eds.), Health risks and developmental transitions during adolescence (pp. 283-320). Cambridge, UK: Cambridge University.

Elam. S. (Ed.). (1993). The state of the nations public schools. Bloomington, IN: Phi Delta Kappa.

Eriksen, K. (1997). Making an impact: A handbook on counselor advocacy. Washington, DC: Taylor & Francis.

Fields, T. H., & Hines, P. L. (2000). School counselor's role in raising student achievement. In G. Duhon & T. Manson (Eds.), Preparation, collaboration, and emphasis on the family in school counseling for the new millennium (pp. 135-162). Lewiston, NY: The Edwin Mellen Press.

Fine, M. (1991). Framing dropouts: Notes on the politics of an urban high school. New York: State University of New York.

Fuhrmann, B. S. (1986). Adolescence, adolescents. Boston: Little, Brown.

Hayes, R. L. (2002). [Class of 1997 student retention data]. Unpublished raw data. (Available from the author at Department of Counseling and Human Development Services, 402 Aderhold Hall, The University of Georgia, Athens, 30602. Email: [e-mail])

Hayes, R. L., & Lunsford, B. (1994). Elements of empowerment: Enhancing efforts for school renewal. People and Education, 2(1), 83-100.

Hayes, R. L., Paisley, P. O., Phelps, R. E., Pearson, G., & Salter, R. (1997). Integrating theory and practice: Counselor educator-school counselor collaborative. Professional School Counseling, 1(1), 9-12.

House, R. M., & Hayes, R. L. (2002). School counselors: Becoming key players in education reform. Professional School Counseling, 5, 249-256.

House, R. M., & Martin, P. J. (1998). Advocating for better futures for all students: A new vision for school counselors. Education, 119, 284-291.

Jaffe, M. L. (1998). Adolescence. New York: John Wiley.

Jones, M. E. B. (1997, May). A strategic planning approach to improve school climate: A case study in school restructuring. Unpublished doctoral dissertation, The University of Georgia, Athens.

McNeal, R. B. (1997). High school dropouts: A closer examination of school effects. Social Science Quarterly, 78, 209-221.

Reyes, R, Wagstaff, L. H., & Fusarelli, L. D. (1999). Delta forces: The changing fabric of American society and education. In J. Murphy & K. Seashore Louis (Eds.), Handbook of research on educational administration (2nd ed., pp. 183-201). San Francisco: Jossey-Bass.

Rumberger, R.W. (1987). High school dropouts: A review of issues and evidence. Review of Educational Research, 57(2), 101-121.

Schneider, B., & Stevenson, D. (1999). The ambitious generation: America's teenagers motivated but directionless. New Haven, CT: Yale University.

Sherraden, M.W. (1986). School dropouts in perspective. The Educational Forum, 51(1), 15-31.

Tidwell, R. (1988). Dropouts speak out: Qualitative data on early school departures. Adolescence, 23, 939-954.

U.S. Department of Commerce, Bureau of the Census. Current Population Survey. (1996, March). Educational attainment in the United States (P20-493) [Electronic version]. Retrieved April 15, 2002, from http://www.census.gov/ prod/2/pop/p20/p20-493.pdf

U.S. Department of Commerce, Bureau of the Census. Current Population Survey. (2000, March). Educational attainment in the United States (Update)(P20-536) [Electronic version]. Retrieved April 15, 2002, from http://www.census.gov/population/socdemo/education/p20-536/p20-536.pdf

U.S. Department of Education, Office of Educational Research and Improvement, National Center for Education Statistics. (1999). The condition of education: 1999 (NCES 1999022) [Electronic version]. Retrieved April 1,2002, from http://www.nces.ed.gov/pubs99/condition99

U.S. Department of Education, Office of Educational Research and Improvement, National Center for Education Statistics. (2000a). Dropout rates in the United States: 1999 (NCES 2001022) [Electronic version]. Retrieved April 15, 2002, from http://www.nces.ed.gov/pubs2001/ 2001022.pdf

U.S. Department of Education, Office of Educational Research and Improvement, National Center for Education Statistics. (2000b). Public school student, staff, and graduate counts by state, school year 1998-99 (NCES 2000330) [Electronic version]. Retrieved April 15, 2002, from http://www.nces.ed.gov/pubs2000/2000330.pdf

U.S. Department of Education, Office of Educational Research and Improvement, National Center for Education Statistics. (2000c). A recommended approach to providing high school dropout and completion rates at the state level (NCES 2000305) [Electronic version]. Retrieved April 15, 2002, from http://www.nces.ed.gov/pubs2000/ 2000305.pdf

Vitaro, F., Brendgen, M., & Tremblay, R. E. (1999). Prevention of school dropout through the reduction of disruptive behaviors and school failure in elementary school. Journal of School Psychology, 37, 205-226.

Zimiles, H., & Lee, V. (1991). Adolescent family structure and educational progress. Developmental Psychology, 27, 314-320.

Richard L. Hayes, Ed.D., is a professor Department of Counseling and Human Development Services, The University of Georgia, Athens. E-mail: [e-mail]. Former doctoral candidates at the Department of Counseling and Human Development Services, The University of Georgia are Judi-Lee Nelson, Ph.D., who is now a postdoctoral Psychology Fellow at the Atlanta Center for Eating Disorders, and Melissa Tabin, Ph.D., who is a now an independent consultant in San Antonio, TX. George Pearson is lead counselor, Department of Counseling and Student Development, and Charles Worthy is principal; both are with Cedar Shoals High School, Clarke County School District, Athens, GA.