JVER v27n3 - Attrition, Completion, and Graduation Rates in Georgia Technical Colleges, Before and After the Initiation of the HOPE Grant
Attrition, Completion, and Graduation Rates in Georgia Technical Colleges Before and After the Initiation of the HOPE Grant
Teresa Resch Helen C. Hall Coosa Valley Technical College University of Georgia Abstract
Two groups of students enrolled in Georgia technical college diploma programs were studied: one group matriculated in fall 1992 (n=9,463) and a second group matriculated in fall 1997 (n=12,467). Z-scores and logistic regression were used to determine differences and relationships in attrition, completion, and graduation rates before and after the initiation of the HOPE (Helping Outstanding Students Educationally) scholarship program. Attrition and completion rates were greater after the initiation of the HOPE grant, while the graduation rate was less. The attrition and completion rates were also higher for students who received the HOPE grant compared to students who received need-based financial aid. Logistic regression analysis illustrated that students were less likely to leave school in 1992 than in 1997, and part-time students were more likely to be completers than full-time students. Students were less likely to be completers in 1992, but more likely to graduate in 1992 than in 1997. When controlling statistically for all independent variables, 50% of the change in attrition rates and 37% of the change in decreasing graduation rates occurred after the initiation of HOPE in 1997.The study of student attrition in postsecondary education takes on much importance as colleges work to retain students. Attrition is defined as students who quit attending school prior to completing requirements for graduation from a diploma program. In studies of attrition at the 2-year postsecondary level, distinctions are made between program completion and graduation. Students completing at least 50% of a recognized program of study and then gaining employment in the field of study are considered completers, although they have not graduated from any particular program ( Council on Occupational Education, 2000 ).
Historically, approximately one-half of all traditional freshmen entering college ultimately graduate. Conversely, the attrition rate for nontraditional students in two-year colleges is close to 60% ( Lombard, 1992 ). In fact, of the three million students who enrolled in two-year postsecondary institutions in 1995-96, after three years, 36% did not earn a degree or certificate and were no longer enrolled in school, 6% did not earn a degree but were still enrolled in school, and 58% had attained a degree or certificate after three years ( Berkner, Carroll, Clune, & Horn, 2000 ). Kerka (1995) speculated that attrition rates have increased because students in colleges and universities are increasing at more widely varying stages of the life cycle compared to the traditional 18- to 22- year old cohort.
While attrition is a problem, colleges have struggled in their attempts to gather good information on attrition, and without such data are hampered in efforts to launch successful retention programs. According to Tinto (1987) , most college students leave voluntarily, and their decisions to withdraw stem most often from personal, social, or financial problems. Barton (1997) suggested that to make higher education obtainable there are several critical issues to be examined including the formulations for financial aid, non-completion rates at postsecondary institutions, and improving graduation rates in high schools. College Board Online (1996) described four distinct factors influencing student attrition: student experience factors, finances, cost and external factors, and institutional variations. Financial aid lowers the net cost of attendance and increases the probability of persistence. Roslund (1998) completed a study of 600 non-returning students from Davenport College Career Center and found that financial aid problems were the number one reason for not returning. It appears there is a direct relationship between financial aid concerns and student retention.
According to DeSalvatore and Hughes (2000) , for the third year in a row in 1999-2000 year, Georgia ranked number one for students receiving state financial aid to attend postsecondary education. In fact, their national survey found that 77.9% of Georgia's undergraduate students received state-financed grants and scholarships to attend Georgia public and private colleges and universities during the 1998-1999 academic year. Georgia's high ranking is attributed to the Helping Outstanding Pupils Educationally (HOPE) Scholarship program funded by the Georgia Lottery for Education. HOPE became available in September 1993 to all qualified Georgia citizens. For the first seven years, 556,030 students received more than one billion dollars in scholarships to pay for tuition, fees, and books ( Georgia Student Finance Commission, 2000a ). Among Georgia's 33 public technical colleges, 243,000 students received HOPE scholarships with total awards in excess of $21 million between September 1, 1993, and December 9, 2000.
Of the 64,539 students in 2-year postsecondary technical colleges who received financial aid during the fall 2000 academic quarter, 46,532 received HOPE Scholarship or Grant funds, 16,639 received Pell Grant funds, 1,673 received funds through Veterans Administration, and 1,008 received funds through the Job Training Partnership Act. During this same quarter, 5,847 students were enrolled in developmental studies for English/reading and 6,706 in developmental studies for math ( Georgia Department of Technical and Adult Education, 2000a ).
According to Georgia's HOPE Scholarship Program Regulations for the 2000-2001 academic year ( Georgia Student Finance Commission, 2000b ), non-traditional students, GED recipients, recent high school graduates, and home-study students are eligible to receive a HOPE grant to cover tuition, HOPE-approved mandatory fees, and a book allowance to seek a technical diploma or certificate at a public 2-year postsecondary institution in Georgia if specific requirements are met. All individuals who have been legal residents of Georgia for at least one year, regardless of grade average or high school graduation date, may be eligible for a HOPE Grant. Selective service registration is required for all males over the age of 18. Students must not be in default of a student loan and must be free of drug convictions for 90 days in order to qualify for HOPE Grant funds. In order to retain the HOPE grant funds at technical colleges, students must make satisfactory progress toward earning a diploma or certificate. According to the Georgia Student Finance Commission (2000b) , graduates from 91 high schools had a 40% or better renewal rate for the HOPE scholarship. This means that approximately 60% of the high school graduates eligible in one academic year were no longer eligible for HOPE funds the following year.
Variables of interest in attrition studies include full-time and part-time status, entry-level education, financial aid plans, age, gender, ethnicity, and types of programs of study. Tinto (1982) suggested developing group-specific models of student disengagement to include gender, race, age, and social status backgrounds. Metzner and Bean (1987) proposed that dropout decisions for nontraditional students are based both singularly and interactively on six constructs which include background and defining variables, academic variables, environmental variables, psychological outcomes, academic outcomes, and intent to leave. Catt (1998) found that the obstacles most likely to inhibit student persistence were loneliness, financial issues, housing problems, security concerns, and the inability to commit to the college or local community. Horton (1998) included prerequisite requirements for courses, student age and enrollment status at the time the courses were taken, ACT subscores and composite scores, type of high school diploma, type of high school attended, and gender as variables in his study of student attrition. Pardee (1992) conducted a study at a medium-sized California community college and concluded that (a) the typical returning student was a white female between the ages of 28 and 32, taking less than six units during the evening and working in excess of 40 hours per week, (b) 30% of the students had been out of school for 5 years or longer, 23.7% for one year, and 10% for two or three years, (c) desire to learn was the most important influence to return to college for both men and women and for all ethnicities, except black students, (d) other significant influences were improved earning potential, increased value on education, improved emotional outlook, occupation requires, and dissatisfaction with job, (e) the six top-ranked influences corresponded closely to the top-ranked trigger influences that were identified before a student drops out of school, and (f) 73% of students were returning to the college they had left originally. Nippert (2000) concluded that women are somewhat more likely to complete their degrees than men and academic activities, college GPA, and choosing to re-enroll had a positive effect on educational attainment.
Purpose of Study
Problem and Purpose Statement
Financial aid issues are a major problem for students and the most common reasons students give when dropping out of school. A large percentage (72% in fall 2000) of students at Georgia's technical colleges utilize HOPE scholarship and Grant funds ( Georgia Department of Technical and Adult Education, 2000a ). HOPE funds are being spent with little or no documentation of the impact they have on educational attainment. Therefore, this study investigated the relationship of the HOPE Grant to student attrition, completion, and graduation from diploma programs at Georgia technical colleges.
The purpose of this causal-comparative study was to determine the attrition, completion, and graduation rates of students in Georgia technical college diploma programs (less than 90 quarter credits) before and after the initiation of the HOPE grant in 1993 and to explain the relationship between selected dependent and independent variables. The HOPE grant was initiated in September 1993, therefore two groups of students were included: one group matriculated in 1992, the year before the initiation of the HOPE grant program; a second group matriculated in 1997, five years after the initiation of the HOPE grant program. We compared the dependent (response) variables-attrition, completion, and graduation rates of students-based on the independent (explanatory) variables-(a) full-time and part-time enrollment status, (b) age, (c) gender, (d) ethnicity, (e) program divisions, and (f) need-based financial aid. Attrition, completion, and graduation rates of students who received Pell Grant and/or JTPA funds (need-based financial aid) were compared with students who received only HOPE grant funds.
Research Questions
The following research questions were addressed:
- Is there a significant difference in attrition, completion, and graduation rates in Georgia technical colleges before and after the initiation of the HOPE grant based on (a) full-time and part-time enrollment status, (b) age, (c) gender, (d) ethnicity, or (e) program division?
- Is there a significant difference in attrition, completion, and graduation rates in Georgia technical colleges between those students who received the HOPE grant and those students who received need-based financial aid?
- What are the strengths of relationships between the dependent variables-attrition, completion, and graduation-and the independent variables-full-time and part-time enrollment status, age, gender, ethnicity, and program division?
- Controlling for all independent variables-full-time and part-time enrollment status, age, gender, ethnicity, and program division-what is the relationship of the initiation of the HOPE grant and attrition, completion, and graduation rates?
Theoretical Framework
The framework for this study was based on theories of student attrition from St. John (1991, 1992) , Tinto (1993) , and Bean and Metzner (1985) . St. John (1991) reported that evidence existed from econometric studies concluding that student financial aid was an effective means of promoting equal opportunity and in promoting persistence in higher education. St. John reported that some studies found that financial aid was effective, while others concluded it had no significance. Because of conflicting findings, St. John (1992) recommended two models for evaluating the effects of financial aid, which he referred to as the Basic Attendance Model and Workable Persistence Model. These models used existing institutional data sources. The Basic Attendance Model includes social background (gender, age, ethnicity, dependency status, financial need), academic preparation (test scores, high school, some college), student aid (any aid, grants, loans, loans and work, grants, work, all other types of aid, amounts), and attendance. The Workable Persistence Model includes all of the parts of the Basic Attendance Model plus academic experience (grades and programs of study) and college experiences (special programs and extracurricular activities).
According to Tinto (1987) , most traditional college students leave voluntarily and their decisions to withdraw stem most often from personal, social, and financial problems. Tinto (1982) suggested developing group-specific models of student disengagement to include gender, race, age, and social status background. Models of attrition that include descriptions/levels of social and academic interactions without including gender, race, age, and social status tend to underestimate and even distort the characteristics of dropouts among various groups of students, especially those from disadvantaged backgrounds. Tinto (1988) suggested that students are more likely to be successful in college if they go through rites of passage that include separation from past associations, transition that begins when the person begins to interact with members of the new group, and the last phase, incorporation. Incorporation is the taking on of new patterns of interaction with members of the new group and establishing competent membership in that group as a participant member.
Tinto (1993) revisited his theories on student attrition, particularly as they related to traditional and nontraditional students at two-year and four-year, public and private institutions. His major emphasis was that student attrition is most affected by a lack of social and academic integration with the community. The community is described as the school, faculty, and students.
Bean and Metzner (1985) developed a model of student attrition that states that older students (nontraditional) drop out of school because of one or more of the following variables: (a) academic performance, (b) intent to leave, (c) previous performance and educational goals, and (d) environmental variables. Environmental variables (e.g., finances, hours of employment, outside encouragement, family responsibilities, opportunity to transfer) have a greater impact on decisions of adult students to leave than academic variables (study habits, academic advising, absenteeism, major certainty, course availability). The Bean-Metzner model suggests that making environmental factors conducive to completion could compensate for weak academic support. Metzner and Bean (1987) proposed that dropout decisions for nontraditional students are based both singularly and interactively on six constructs which include background and defining variables, academic variables, environmental variables, psychological outcomes, academic outcomes, and intent to leave. In contrast with Tinto's expectations, the social integration variable was not found to have significant effect on nontraditional student attrition. The Bean and Metzner model indicated that the most significant variables influencing dropout decisions for nontraditional students are academic performance, intent to leave, background and defining variables, high school performance, educational goals, and environmental variables.
Method
Population and Sample
The population for this study was students who were enrolled in diploma programs (less than 90 quarter credits) at Georgia's technical colleges during the fall 1992 and fall 1997 academic quarters. The 33 technical colleges and 17 satellite campuses in the state of Georgia are units of the Georgia Department of Technical and Adult Education (GDTAE). In fiscal year 1992, 53,302 students were enrolled in credit courses at Georgia technical institutes; during fall quarter, 19,018 were full-time students and 12,845 were part-time students. In fiscal year 1997, 76,300 students were enrolled in credit courses at Georgia technical institutes; during fall quarter, 21,715 were full-time students and 25,889 were part-time students ( Department Technical and Adult Education, 2000 ). Student data were limited to those enrolled in diploma programs requiring less than 90 quarter credits for the program of study. Full-time students normally complete diploma programs requiring between 60-90 quarter credits of course work in four quarters. During fall 1992, 12,486 students were enrolled in diploma programs with less than 90 quarter credits in the program of study. During fall 1997, 15,840 students were enrolled in diploma programs with less than 90 quarter credits in the program of study. These two groups of students served as the population for the study. If a student did not have an exit status such as graduate, completer, or leaver recorded in their file, they were excluded. The remaining number of students in fall 1992 was 9,593 and fall 1997 was 12,734. The data files were further reviewed with incomplete records excluded. The number of eligible students to be selected in the sample for research questions 1, 2, and 3 was 9,463 students in fall 1992 and 12,467 students in fall 1997. For research questions, 4, 5, and 6 only the 12,467 students in fall 1997 were included. Of the 12,467 students, 4,667 students received need-based financial aid, 5,879 students received HOPE funding, and 1,921 students received no financial aid. For research questions 7, 8, 9, and 10, the number of eligible students in the sample included all 21,930 students.
Diploma programs were categorized into six divisions by the GDTAE: (a) agricultural/natural resource technologies, (b) business technologies, (c) engineering science technologies, (d) health technologies, (e) industrial technologies, and (f) personal/public service technologies. No students from the engineering science technologies division were included because programs in this division contain more than 90 quarter credits in the program of study.
Data was acquired from information available in BANNER, a computer software program used as a student management system by all public Georgia technical colleges. Biannually, the director of student services at each college reviews and updates students' records in the BANNER system to denote whether a student is a leaver, completer, or a graduate from a program. Students who quit attending school prior to completing requirements for graduation from a diploma program are considered a leaver. Those who return to school after being in non-attendance for one quarter are not counted as leavers. Students completing at least 50% of the program of study and then gaining employment in the field of study ( Council on Occupational Education, 2000 ) are designated as completers. Only students who completed all courses in the diploma program and met all other graduation requirements were considered graduates.
Full-time and part-time students enrolled in fall 1992 were tracked until fall 1994, while students enrolled in fall 1997 were tracked until fall 1999 to allow sufficient time for students to complete their programs of study. Full-time students were enrolled in 12 credits of course work or more, part-time students were enrolled in fewer than 12 credits or course work. The query from BANNER provided data to establish exit status (leaver, completer, or graduate) for fall 1994 for each student enrolled in diploma programs in fall 1992 and for fall 1999 for each student enrolled in diploma programs in fall 1997. The data were used to determine if a significant difference existed in attrition, completion and graduation rates before and after the initiation of the HOPE grant based on full-time and part-time enrollment status. For remaining BANNER queries, full-time and part-time students were considered as one group. The next query provided data to establish exit status based on age, gender, ethnicity, and program divisions for each student in fall 1994 for the students enrolled in diploma programs in fall 1992 and for each student in fall 1999 for students enrolled in diploma programs in fall 1997. For this study, need-based financial aid was determined by whether students had received a Pell Grant or JTPA funds. The next BANNER query provided data that designated exit status (leaver, completer, or graduate) for students who had received Pell Grant or JTPA and HOPE Grant funds for each student in fall 1999 for students enrolled in diploma programs (with less than 90 quarter credits) in fall 1997. Students who received Pell Grant or JTPA funds were compared with students who received only HOPE Grant funds. The data was input into SAS for analysis.
Data Analysis
Analyses using z-scores and logistic regression were used to determine differences in attrition, completion, and graduation rates before and after the initiation of the HOPE grant. A z-score is a standard score frequently used in educational research that is derived from standard deviation units. Also, z-scores are continuous and have equality units. Thus, a person's relative standing on two or more measurements can be compared by converting the raw scores to z-scores ( Gall, Borg, & Gall, 1996 ; Huck, 2000 ). The z-distribution is used when samples are large and is used to determine the level of statistical significance of an observed difference between the groups. Logistic regression deals with the relationship among variables where one variable is the dependent variable, while the other(s) is/are independent variables. The independent variable can be continuous or categorical. In this study, all independent variables are categorical.
The purpose of logistic regression can be either prediction or explanation ( Huck, 2000 ). Logistic regression revolves around a core concept called the odds ratio. The odds ratio measures the strength of association between an independent variable and dependent variable. A subset of independent variables in a typical logistic regression is referred to as control variables. Such variables are included to assess the relationship between dependent and independent variables. The primary focus is on non-control independent variables, with the goal being to identify the extent to which each one plays a role in explaining why changes exist with the dependent variable. Most researchers utilize logistic regression so they can discuss the explanatory power of each independent variable using the concept of odds. By using the estimates in logistic regression, researchers also try to find a good set or model of independent variables that can help predict or explain group membership on the dependent variable.
The dependent variables in this study were attrition, completion, and graduation rates of students at Georgia technical colleges before and after the initiation of the HOPE grant. The nominal independent variables were: (a) full-time and part-time enrollment status, (b) age, (c) gender, (d) ethnicity, (e) program divisions, and (f) financial need to describe each student enrolled in diploma programs in fall 1992 and fall 1997. Age categories were consistent with those used in BANNER data collection from GDTAE. Ethnic categories from the initial query included American Indian, Asian, Black, Hispanic, White, non-resident alien, and multi-racial. For analytic purposes, I recoded the ethnic designation into White ( n =6406 for fall 1992, n =7428 for fall 1997), Black ( n =2809 for fall 1992, n =4552 for fall 1997), and others ( n =248 for fall 1992, n =487 for fall 1997) groups. The other group included American Indian, Asian, Hispanic, non-resident alien, and multi-racial students.
Findings
Conclusions
The sample for this study was 9,463 students in fall 1992 and 12,467 students in fall 1997. Overall, attrition and completion rates were greater after the HOPE grant than before, while the graduation rate was greater before the HOPE grant by nearly 10%. The z-scores in all categories for attrition rates were statistically significant except in the age categories, 31-35 years and 36-40 years; ethnicity category, other; and the agriculture/natural program and business division. z-scores in all categories for completion rates were statistically significant except in the ethnicity category, other. Completion rates were lowest in the health and personal/public divisions. The z-scores in all categories for graduation rates were statistically significant except in the ethnicity category, other; and the agriculture/natural program.
Of the total sample of 12,467 students from fall 1997, 4,667 students received need-based financial aid, 5,879 students received HOPE funding, and 1,921 students received no financial aid. The z-score for attrition rates between students who received need-based financial aid and those that received the HOPE grant was statistically significant, z(3591)=-11.75, p .0001. There was a statistically significant difference in the completion rate of students who received need-based financial aid and those that received the HOPE grant, z(2041)=-7.51, p <.0001. The completion rate increased by nearly 5% for students who received need-based financial aid compared to students who received only the HOPE grant in 1997. There was a statistically significant z-score for the graduation rate of students who received need-based financial aid when compared to those who received the HOPE grant, z(4914)=17.10, p <.0001. The graduation rates decreased by nearly 17% for those students who received need-based financial aid compared to students who received only the HOPE grant in 1997.
In the type III analysis of effects using logistic regression, a statistically significant relationship existed between the dependent variable, attrition, and all independent variables. A student was less likely to be a leaver in 1992 than in 1997. A statistically significant relationship existed between the dependent variable, completion, and all independent variables. Full-time students were 50% less likely to be completers compared to part-time students. A student was less likely to be a completer in 1992 than in 1997. A statistically significant association also existed between the dependent variable, graduation, and all independent variables. Agricultural students were one third less likely to be graduates compared to students in personal/public programs. A student was more likely to be a graduate in 1992 than in 1997. Controlling for all independent variables, 50% of the total change in attrition rates was attributed to after the initiation of HOPE in 1997. Controlling for all independent variables, 37% of the total change in graduation rates was attributed to after the initiation of the HOPE in 1997. These results provide evidence of the association of the HOPE grant to the dependent variables, attrition, completion, and graduation, and how the independent variables illustrate this association relationship.
Summary
Attrition rates increased from 1992 to 1997 for all independent variables except for the ethnicity, others category, and the agricultural/natural program division. Completion rates increased from 1992 to 1997 in all independent variables except ethnicity, other. Graduation rates decreased from 1992 to 1997 for all independent variables except ethnicity, others, and the agricultural/natural program division. Attrition rates were higher for students who received the HOPE grant compared to students who received need-based financial aid for students in 1997. Completion rates in 1997 were higher for students who received the HOPE grant compared to students who received need-based financial aid. Graduation rates were lower in 1997 for students who received the HOPE grant compared to students who received need-based financial aid. All independent variables-full-time and part-time enrollment status, age, gender, ethnicity, and program divisions-had an effect on attrition, completion and graduation rates. The parameters that best predicted a leaver included being female, Black, between 16-20 years of age, part-time, and enrolled in a business program in 1997. The parameters that best predicted a completer were male, White, between 26-30 years of age, part-time, and in an agricultural program in 1997. The parameters that best predicted a graduate included being female, White, between the ages of 36-40, full-time, and in a health program in 1992. Controlling for all independent variables the percentage of leavers changed from 27.5% in 1992 to 30.6% in 1997. Controlling for all independent variables the percentage of completers changed from 17.8% in 1992 to 18.3% in 1997. Controlling for all independent variables the percentage of graduates changed from 54.7% in 1992 to 51% in 1997. Controlling for all independent variables in the study 50% of the total change in attrition rates is contributed after the initiation of the HOPE. Controlling for all independent variables in the study 37% of the total change in graduation rates is contributed after the initiation of the HOPE.
In conclusion, overall attrition and completion rates were greater after the HOPE grant than before, while the graduation rate was greater before the HOPE grant by nearly 10%. Attrition rates were less and graduation rates were higher for students receiving need-based financial aid compared to students that received the HOPE grant. The program divisions, health and personal public services, have the highest graduation rates, 58% and 61%, respectively. Results of this study should be compared with other studies of attrition, completion, and graduation rates to note if similar changes reported in this study were apparent in other schools.
Discussion
The study of student attrition in postsecondary education is an endeavor that takes on much importance as colleges work to retain students. Historically, approximately one-half of all traditional freshmen entering college ultimately graduate, conversely the attrition rate for nontraditional students is close to 60% ( Lombard, 1992) . In fact, of roughly three million students who first enrolled in two-year postsecondary institutions in 1995-96, 36% did not earn a degree or certificate and were no longer enrolled in school, 6% did not earn a degree but were still in enrolled in school, and 58% had attained a degree or certificate after three years ( Berkner, Carroll, Clune, & Horn, 2000 ).
According to Tinto (1987) , student's decisions to withdraw stem most often from personal, social, and financial problems. Nippert (2000) concluded that women are somewhat more likely to complete their degree than men and that academic activities, college GPA, and choosing to re-enroll have a positive effect on educational attainment. Bean and Metzner (1985) concluded that older students (nontraditional) drop out of school because of one or more of the following variables academic performance, intent to leave, previous performance and educational goals, and environmental variables.
In this study 9,463 students from fall 1992 were included with 2,907 designated as leavers a 30.7% attrition rate. In the comparison cohort from fall 1997, 12,467 students were studied with 4,442 students designated as leavers for an attrition rate of 35.6%. If the attrition rate were the same in 1997 as in 1992, an additional 612 students would have been completers or graduates in the group of students studied from 1997. The attrition rate of 35.6% is a 5% increase from 1992 to 1997 and is similar to that stated in the study of more than three million students in two-year postsecondary institutions. Tom (1999) reported the following reasons could be interpreted as contributing to attrition: 27% reported that loss of income was a major reason, 30% cited conflict of job and school as a major reason, and 21% mentioned the untimeliness of course offerings. Tom's findings suggest that finances and conflict of job and school are related to the attrition of the student. K. Breeden (personal communication on December 19, 2000) suggested that typically, when unemployment rates decline, school enrollment and retention decreases because it is easier to find employment. An expanding economy will likely produce greater attrition ( Walleri, 1981 ). However, DTAE has set new enrollment records every quarter for more than 35 consecutive quarters ( Georgia Department of Technical and Adult Education, 2000 ). According to the DTAE Statistical Information FY 2000, 46,076 students were enrolled in credit programs in fiscal year 1990 with 6,227 graduates; and 101,194 students were enrolled in credit programs in fiscal year 2000 with 15,304 graduates. According to the Bureau of Labor Statistics Data (2000), the unemployment rate for Georgia in November 1990 was 6.1% and the unemployment rate was 3.0% for November 2000.
It is difficult to compare attrition, completion, and graduation rates in Georgia's technical colleges from year to year because there is limited data published from the Department of Technical and Adult Education. However, there have been other sources to discuss attrition, completion and graduation rates. According to the American Medical Association (2000) , data collected from 4,365 programs and 203,838 students indicated that attrition rates range from a low 2% to a high of 33.3% and an average 11.8% attrition in various medical programs in the United States. Attrition rates in the health division at Georgia's technical colleges were 28% in 1992 and 34% in 1997. The attrition rates are higher than the national average. Completion rates were lowest in the health and personal/public divisions. This may be because students in health and personal/public divisions must complete the entire program of study to qualify to sit for state and national certification exams. The program divisions, health and personal/public services, have the highest graduation rates, 58% and 61%, respectively. Most programs in these divisions have measurable outcomes for the graduates such as certification exams. In the other program divisions, the programs have very few measurable outcomes, such as certification exams, that may attribute to their lower graduation rates and higher completion rates as compared to the health and personal/public divisions. Completion rates were higher for part-time students than for full-time students. This could be due to the fact that students who attend school on a part-time basis are likely to be employed. Eighty-five percent of students enrolled in Georgia's technical colleges are employed in the labor force ( Georgia Department of Technical and Adult Education, 2000 ). In fact, much attrition in vocational-technical education can be explained simply by students leaving school due to job opportunities, especially where and when there is a shortage of skilled laborers ( Walleri, 1981 ). Bean and Metzner (1985) indicated that environmental variables (finances, hours or employment, outside encouragement, family responsibilities, and opportunity to transfer) have a greater impact on decisions of adult students to leave than academic variables (study habits, academic advising, absenteeism, major certainty, and course availability).
Attrition and completion rates were lower and graduation rates were higher for students who received need-based financial aid compared to students who received the HOPE grant. Attrition rates increased by 5% after the initiation of the HOPE grant, while the graduation rate was greater before HOPE by 10%. Roslund (1998) completed a study of 600 non-returning students from Davenport College Career Center and found that financial aid problems were the number one reason for not returning.
I found that parameters that best predict a leaver include being female, Black, between the ages of 16-20, part-time, and enrolled in a business program in 1997. Parameters that best predict a completer includes being male, White, between the ages of 26-30, part-time, and in an agricultural program in 1997. The parameters that best predict a graduate includes being female, White, between the ages of 36-40, full-time, and in a health program in 1992. Knowing these predictors is valuable to the management of all technical colleges. By recognizing characteristics of students that can predict a student leaving prior to graduation, interventions can be put into place to help the student succeed in school. Students can be identified who need additional assistance once the characteristics of students who have a tendency to not succeed are recognized.
Overall attrition rates increased by 6% and completion rates increased by 5% after the HOPE grant, while the graduation rate was greater before the HOPE grant by nearly 10%. All of these rate changes were statistically significant. The sample size for the study was 9,463 student in 1992 and 12,467 in 1997. To compare students who received HOPE grant funds to students who received need-based financial aid the sample size was 5,879 students who received HOPE funding and 4,667 students who received need-based financial aid. Large sample sizes can produce statistically significant result even though there is limited practical significance associated with the findings ( Huck, 2000 ). All of the independent variables had an affect on attrition, completion, and graduation rates. In a practical sense, controlling for all of the independent variables in the study, 50% of the total change in attrition rates is attributed to after the initiation of the HOPE in 1997 and controlling for all of the independent variables in the study 37% of the total change in graduation rates is attributed to after the initiation of the HOPE in 1997. Changes in attrition, completion, and graduation rates reported in this study, should be further researched to confirm the results in this study and recommendations put into place to increase graduation rates and decrease attrition rates.
References
American Medical Association. (1999) . Number of programs and enrollments, attrition, and graduation by occupation academic year 1998-1999 [Brochure]. Author.
Barton, P . (1997, November 14). Toward inequality: Disturbing trends in higher education. Policy Information Center . Retrieved June 12, 2000 from
http://www.ets.org/research/pic/towine.htmlBean, J. P., & Metzner, B. S. (1985) . A conceptual model of nontraditional undergraduate student attrition. Review of Educational Research , 55, 485-540.
Berkner, L., Carroll, D., Clune, M., & Horn, L. (2000, April) . Beginning postsecondary students longitudinal study: Descriptive summary of 1995-96 beginning postsecondary students: Three years later (NCES 2000-154) Washington, DC: National Center for Education Statistics.
Bureau of Labor Statistics. (2000, May 17). College enrollment and work activity of 1999 high school graduates . Retrieved December 22, 2000 from
http://stats.bls.gov/newsrels.htmBureau of Labor Statistics Data. (2000, December 22). Local area unemployment statistics . Retrieved December 22, 2000 from
http://146.142.4.24/cgi-bin/ SeriesHist?lasst13000003Catt, S. R. (1998) . Adjustment problems of freshman attending a distant, non-residential community college. (Doctoral dissertation, University of Pittsburgh, 1998).
College Board Online. (1996, Summer) . Organizational issues in enrollment management. Enrollment Management Review . Retrieved March 30, 2000 from
http://www.collegeboard.com/aes/emr/emrsu96/html/organize.htmlCouncil on Occupational Education. (2000) . Handbook of accreditation 2000 edition . Atlanta: Author.
Department of Technical and Adult Education . (2000, December 1). Statistical information FY 2000 . Atlanta: Author.
DeSalvatore, K., & Hughes, L. (2000) . 30th annual survey report 1998-99 academic year . New York: National Association of State Student Grant and Aid Programs, New York State Higher Education Services Corporation.
Gall, M., Borg, W., & Gall, J. (1996) . Educational research an introduction (6 th ed.). White Plains, NY: Longman.
Georgia Department of Technical and Adult Education [DTAE] . (1999). 1999 annual report . Atlanta: DTAE Communications Office.
Georgia Department of Technical and Adult Education [DTAE]. (2000a) . Knowledge management system [Enrollment reports]. Retrieved January 1, 2001 from
http://kms.dtae.org/Georgia Department of Technical and Adult Education [DTAE]. (2000b). Annual report 2000 . Atlanta: DTAE Communications Office.
Georgia Department of Technical and Adult Education [DTAE]. (2000c). Policy manual of the state board of technical and adult education . Retrieved May 3, 2001 from
http://www.dtae.org/policy/policy.htmlGeorgia Department of Technical and Adult Education. (2000d). Results. Atlanta: Office of Economic Development Programs.
Georgia Student Finance Commission. (1999). Celebrating achievement annual report 1999 . Atlanta: Author.
Georgia Student Finance Commission. (2000a) . HOPE scholarship program . Retrieved December 15, 2000 from
http://www.gsfc.org/gsfc/html_summary_grant_all_cov_H.htmGeorgia Student Finance Commission. (2000b) . Georgia's HOPE scholarship program: Regulations for the 2000-2001 academic year . Atlanta: Author.
Georgia Student Finance Commission. (2000c). Georgia's HOPE scholarship program: Program summary . Retrieved September 19, 2000
http://www.gsfc.org/press_release/hopefaq.cfmGeorgia Student Finance Commission. (2000d). High school academic requirements . Retrieved December 15, 2000 from http://www.gsfc.org/hope/hop2000.htm
Horton, S. G. (1998) . Using academic and demographic variables to predict success in the general education curriculum. (Doctoral dissertation, Louisiana State University, 1998).
Huck, S. (2000) . Reading statistics and research (3rd ed). New York: Longman.
Kerka, S. (1995) . Adult learner retention revisited . Retrieved March 6, 1999 from http://ericacve.org/docs/dig166.html
Lombard, R. (1992) . Persistence of nontraditional students in a two-year college: A test of the Bean and Metzner model. Dissertation Abstracts International , 53. (University Microfilms No. 9226453).
Metzner, B.S., & Bean, J.P. (1987) . The estimation of a conceptual model of nontraditional undergraduate student attrition. Research in Higher Education , 27, 15-38.
Nippert, K. (2000) . Influences on the educational degree attainment of two-year college students. Journal of College Student Retention: Research, Theory & Practice , 2(1), 29-40.
Pardee, R. L. (1992, March) . A descriptive profile of returning students, and the influences affecting the re-enrollment in college . Paper presented at the annual research conference of the Community College League, CA: (ERIC Document Reproduction Service No. ED 342 436)
Roslund, S. (1998) . Post-secondary retention of the non-traditional student at the Davenport college career center. (Master's thesis, Grand Valley State University, 1998).
St. John, E. P. (1991) . The impact of student financial aid: A review of recent research. Journal of Student Financial Aid , 21(1), 18-32.
St. John, E. P. (1992). Workable models for institutional research on the impact of student financial aid. Journal of Student Financial Aid , 22(3), 13-26.
Tinto, V. (1982) . Limits of theory and practice in student attrition. Journal of Higher Education , 53, 687-670.
Tinto, V. (1987) . Leaving college: Rethinking the causes and cures of student attrition . Chicago: University of Chicago Press.
Tinto, V. (1988) . Stages of student departure. Journal of Higher Education , 59, 438-454.
Tinto, V. (1993) . Leaving college: Rethinking the causes and cures of student attrition (2nd ed.). Chicago: University of Chicago Press.
Tom, G. (1999) . A post-mortem study of student attrition at the college of business administration. Journal of College Student Retention: Research, Theory & Practice , 1(3), 267-287.
Walleri, D. R. (1981, April 6) . Student retention and attrition in the community college: A review and research design . (ERIC Document Reproduction Service No 210 064)
TERESA RESCH is Director of Instructional Services, Coosa Valley Technical College, One Maurice Culbertson Drive, Rome, Georgia 30161. e-mail: tresch@cvtcollege.org
HELEN C. HALL is Professor, Department of Occupational Studies, University of Georgia, 203 River's Crossing, Athens, Georgia 30602. e-mail: hchall@uga.edu