Compulsory education is more than one hundred years old in the United States and recent information clearly indicates that as a nation we are having serious problems with absenteeism and truancy. Nearly ten percent of children between the ages of 7 and 18 are not enrolled in school. Dropout rates vary immensely but reach as high as twenty percent and absenteeism ranges from 12 percent in some middle schools to about 20 percent in high schools (Neon and Seltzer, 1989; Sommer, 1985a). Indeed, many of the students who leave school are "pushouts" from a system that has not remedied the problems faced by many of its diverse young people (Ziesmer, 1984). Fully 25 to 30 percent of adolescents do not graduate from high school (Bureau of the Census, 1985). Truancy is an old problem highly deserving of innovative attempts to make inroads into a national tragedy.
Truancy is a multifaceted issue in the lives of students, the schools, the Juvenile court, and the economic base of our communities (Levine, 1984). The research problem in this paper is threefold: 1) to assess the value of the skills and training of liberal arts and science college students as truancy prevention facilitators in a high school setting, and how this experience affects the education and career of the college students; 2) to descriptively look at the value of a truancy prevention program to seven traditional high schools and relevant community agencies in a middle-sized urban community; and 3) to evaluate a program which uses intervention strategies to improve attendance among students having large numbers of unexcused absences.
The significance of this research lays In the fact that a grant ( written by the senior author) to hire, train, and supervise seven cooperative education students with liberal arts backgrounds became the centerpiece in a community's response to a growing problem. A 1987 state law mandated accurate reporting of high absence rates to the district attorney's office in each judicial district in the state.
Many states are debating and passing similar statutes that they hope will prevent high rates of absenteeism and the negative consequences that follow. Documentation of the consequences of truancy is considerable. School failure and dropout (Elliot and Voss, 1974), delinquency (Sommer, 1985a), illiteracy and unemployment (Kozol, 1985; Doss and Holly, 1985), and poverty (Dearman and Pliske, 1983) are but a few of the potential problems that may await the habitual truant. Any program that can provide an alternative to the children who face such consequences and manage itself inside traditional agencies like the schools and courts, demonstrates its value repeatedly. The research reported here analyzes data from the first year of a three year truancy prevention program.
Programs designed to change student truancy and absenteeism have been discussed by Fiordaliso et al. ( 1977) and several show high rates of success at the junior high level simply by reporting to parents when students were absent and continuing to report on improved attendance patterns. The parents were called and sent letters as "feedback" about their student's attendance. Parental involvement and awareness were effective in reducing truancy in about 70 percent of the experimental groups' subjects. The control group showed only 40 percent improving in attendance (Fiordaliso et al., 1977, p. 191). Herbst and Sontheimer (1987) reported the effective use of college students as tutors in an "alternative education program" that dramatically altered the dropout pattern for two small counties in Ohio.
The use of liberal arts and science students in social service programs is not new (Newman and Watts, 1989). As part of a community psychology cooperative education class, students were asked to design social intervention programs relevant to some agencies in their community. Wilson's (1987) study documents the correlates of positive attitudes toward cooperative education ... noting that resistance is common. While similar resistance is apparent at the authors' university, the cooperative education program places about 110 liberal arts majors each semester ... only seven of whom are included in this research. Ideas for evaluation models of cooperative education programs (Armstrong, 1989, Wilson, 1989) are helpful in evaluating specific programs, as the one reported here, as well. Indeed, some of the ideas are used in this evaluation of a truancy prevention program and have helped to modify the three year project as it proceeds (Armstrong, 1989; p.53).
Numerous studies on truancy highlight some of the important variables which influence absenteeism and the prevention of it. First, age is related to truancy and high rates of absence in that the older the students get while in secondary schools, the higher their nonattendance (Sommer, 1985b; Weitzman et al. 1985; and Barber and Kagey, 1977). Second, the lack of success most schools have in changing patterns of high absenteeism is clearly related to the pumt1ve approaches (detention, suspensions, etc.) used in response to truancy (Duckworth and deJung, 1989; Hyman et al., 1982; and Safer, 1986). Finally, Compher (1982) suggests that the student, parents, and school may be "triangulated" in ways that encourage rather than discourage absences. Summarily, high rates of truancy among 14 and 15 year olds, coupled with the ineffective means used by schools to handle excessive absences, can create a situation where the students and their families no longer care about attendance.
Literature relevant to truancy and related problems suggests that traditional approaches in solving the problem, usually punitive, are less effective than alternatives that reach into the home or try to work directly to reinvest the student in the school environment. Working with students who are in the critical moments of deciding about their academic future is delicate and risky. In many cases there is little chance to reach them, but in some cases truancy programs can make a difference through "feedback systems," tutoring, intervening for the student with a teacher, or simply showing an interest in the student who is cutting class. Finding these mechanisms for change seems a worthy goal.
The independent and dependent variables in this study are simply defined. Truancy, the dependent variable, is defined as ten to fifteen hours of unexcused absence during the preceding few weeks. Each truancy prevention facilitator (TPF) was allowed to make such determinations based on daily computer monitoring of students showing excessive absences. The official definition in the state, which is binding on all school districts to report to the district attorney, is 35 unexcused hours of absence ... or about 5 total days. Since the idea was to intervene early in the pattern of attendance and "prevent" a referral to the district attorney, most of the TPF used 10 to 15 hours of absence as a rule of thumb to initiate their concern and eventual response. Since the students were tracked before, during, and after their contact with the cooperative education students or TPFs, two groups of truants were eventually identified as "successes" or "failures" based on whether their absence pattern improved or worsened. Intervention strategies, the independent variables in this study, amounted to such things as "number of informal contacts," "number of calls to the home to talk with parents," "registered letter sent to home at 35 absences," and "number of weeks actively working with the truant by the TPF". Several other intervention strategies were used including: 1) referred to principal; 2) individual counseling with the high school student or truant; and 3) behavioral contracts made with the truant. In many respects, the TPF's were encouraged to be creative in responding to the students, and sometimes take a truant to lunch as a reward for better attendance. A TPF also got one truant a job in the school office. Defining all the specific interventions is impossible, but the major ones for which data are available have been listed.
The first sample of interest is the seven truancy prevention facilitators placed in seven traditional high schools by the cooperative education coordinator for the Liberal Arts and Sciences. All the college student TPFs were undergraduate majors in sociology, psychology, social work, and administration of justice. This particular sample of co-op students was composed of five females and two males. One of the TPFs was black. The overall GPA for the sample had a mean of 3.12. Four of the students had GPAs over 3.2. Only six TPFs managed to collect and return the research data requested by the coordinator.
From the six TPFs data were collected on the second sample of 660 freshman and sophomore truants in six high schools. Each of these students worked with high school students presenting attendance problems. Some worked with as few as 50 during the year, while other TPFs worked with nearly 170 truants. TPFs were asked to split their 15 hour employment week equally between the demands of required paperwork and working directly with truants. All the truants were ages 14 and 15. The data set on this large number of high school truants was incomplete at times and sketchy. Descriptive results will be presented for this sample.
A final sample of 99 truants for whom there was a complete data set on the independent and dependent variables. This amounted to being able to compute the number of excused and unexcused absences for pretest and post-test periods... no longer than nine weeks each. Intervention strategies, the independent variables, were used to test the effectiveness of the program in reducing truancy.
Since there are three related research questions in this paper, several different analysis techniques are used to determine the effectiveness of the TPFs and the overall value of the program to the schools and the community-wide effort to impact truancy. Predominantly descriptive statistics are employed in this effort. The actual program evaluation employed a stepwise discriminant function analysis (SPSS, Inc., 1988) to determine which of the intervention strategies used by the TPFs showed predictive value in discriminating between groups of truants whose attendance improved or worsened across the study period.
Other variables that might predict changes in the dependent variable, truancy, have not been included in this analysis because they do not constitute a part of the specific program evaluation. Variables such as sex, age, race, social class, previous standardized test scores, and previous grades might help to explain changes in truancy behavior. Family and school variables also have a bearing on school attendance, and such variables will be included in the longitudinal study planned for the second and third years of the project.
The first two research questions in this paper can be addressed with the descriptive data collected. These two research questions are: 1) Could college students majoring in liberal arts and sciences become effective change agents in the high school as truancy prevention facilitators, and how did this experience affect the co-op student's education and career?; and 2) Could a small program hiring seven college students be of value to meeting the community's compliance with a new law obligating referrals to the district attorney upon 35 unexcused absences?
One important benefit to students who co-op during college is the "career testing experience." In this sample of seven social science majors who operated as truancy prevention facilitators, there is evidence that working with truant youth in the high school setting provided similar benefits. Two of the students continued their studies in the previously chosen major field and are pursuing the baccalaureate degree. Two of the students went into social service work closely related to their TPF experience. Two of the students are in graduate school ... one pursuing an MA in Family Counseling and the other an MBA. This latter student was a psychology undergraduate major and clearly stated, that as a result of his work in the high schools, he "didn't want to work with sick people." He applied to an MBA program in another state and was accepted. Even in cases where the experience leads to changing one's mind about career paths, this seems to be a benefit of "career testing. " We have no followup information on one student.
For most of the students, continued schooling in their chosen fields and continued employment with similar populations were the norm. As hard as they worked, they all spoke positively about their experiences and the part those experiences played in their decisions about school and career. One finding that was uniform across this group of seven was the frustration they felt in not having enough time to do all the work with students who needed the attention. It is our observation that this is a common experience for first semester social service placements.
As established by the national data on truancy and absenteeism, young people are gone from school in very large numbers. The high schools in the study's home community are no exception. The average daily attendance ranges from 90.4 percent as a high to a low of 84.8 percent (USO 259, 1991 ). This means that in the high school with the least amount of absenteeism there is still one in ten students missing on any given day. In the high school with the greatest amount of absenteeism, one in six students is missing each day. In a traditional high school population of approximately 11,300, there will be about 1500 students absent each day; about half of the absences are excused. The cooperative education students placed as truancy prevention facilitators worked only with freshmen and sophomores ... those aged 14 and 15 and those with the highest rates of truancy. At fifteen hours of employment per week, their total impact on this massive nonattendance problem would be minimal, but there are many ways to assess the value of these students and the work they did for the schools and the community.
In the six high schools on which data were available from the TPFs, 660 high school students were contacted during the year. Four hundred and twenty registered letters were sent to the home informing the parents that a response was necessary or the student would be referred to the district attorney's office for truancy. It was an enormous task to collect all the information to put into the DA's form and send to the home. Of these 420 letters sent, 185 referrals were made to the DA. Again, the information sent to the DA was extensive and laborious to collect. Had not the TPFs been there, the task of data collection would have fallen to the assistant principals or not been done at all. It was clear from the principals and school personnel that these administrative services were crucial.
These data indicate that the TPFs were able to work with over 70 percent of the truants in the high school setting without making a referral to the district attorney. Thirty-three percent of the students showing attendance problems were intervened with effectively before 35 absences were reached. Such interventions amounted to informal contacts in the halls, calls to the student's homes, regular counseling with the TPF, meetings with parents, referrals to the principal, etc. system-wide, the Truancy Prevention Program, underwritten by a grant to the cooperative education office on the university campus, provided 105 hours of weekly support to the unified school district (at no cost to them).
Several indicators that support the first two research questions can be found in some of the descriptive results. First, twenty-six percent ( 171 students) were contacted informally with no further action being taken. This statistic indicates the preventive power of early intervention with a subset of the population who may respond to some of the less punitive methods used by the TPFs. Second, 128 calls were made to the homes by TPFs, and 88 calls were made from the home to the TPFs. Third, the TPFs reported counseling with sixteen percent (108 students) about matters relating to their absences, while twelve percent (76 students) were "met with regularly." Fourth, the TPFs were encouraged to be creative with their use of skills in dealing with the truants: eight percent (51 students) agreed to behavioral contracts and four percent (28 students) were put into discussion groups. It seems apparent from these descriptive data that the TPFs not only used their social science skills derived from psychology, sociology, and social work, but they also were able to communicate effectively with the high-school-age population.
Local and statewide significance of the Truancy Prevention Program is also documentable. The senior author on the paper was invited to a statewide meeting to present her program to a conference of associated college students. She and her program were chosen as a model for effective use of moneys spent from an Educational Opportunity Fund ... the source of the grant moneys.
Research question three, a more specific program evaluation research problem, states: What intervention strategies employed by the truancy prevention facilitators in the prevention program might work to decrease truancy? Some preliminary descriptive results about the sample of 99 will be reviewed before the multivariate test of the research question is presented.
The range of pretest absences per week was from two to 35. (Thirty-five absences is the total number of absences one can have in five days at seven periods per day.) The range of posttest absences per week was from one to thirty-five. The mean number of pretest absences was 5.0 per week, and the mean number of posttest absences was 4.5 per week. As can be seen, this is a population that has a highly variable attendance pattern, but it is also a population that has some severe attendance problems. In all cases, the numbers of excused and unexcused absences are about equal ... 50 percent excused and 50 percent unexcused. This pattern altered slightly from the pre- to the posttest period in that there was a decrease in the number of unexcused absences and an increase in the number of excused absences.
A simple ratio of posttest absences per week divided by pretest absences per week was used to generate a set of scores allowing dichotomization of truants into a "success" group and a "failure" group. The "successes" had fewer posttest absences and the "failures" had more posttest absences. Stepwise discriminate function analysis, a multivariate technique appropriate to these nominal data was used. This analysis simply takes each independent variable in order of its ability to discriminate between the truancy two groups. Table 1 shows that four of the intervention strategies used by the TPFs were significant predictors of group membership for the "successes" and "failures."
| Step | Variable Entered | Corr. w/Fn | Wilks Lambda | Signif. |
|---|---|---|---|---|
| 1 | V24 CALLTOHM | +0.613 | .95516 | .04 |
| 2 | V33 LETTER35 | -0.501 | .93873 | .05 |
| 3 | V26 INFCONT | +0.287 | .92391 | .05 |
| 4 | V37 WKSWORK | -0.100 | .88965 | .03 |
Another way of seeing this relationship is to say that four of the strategies were able to predict changes in truancy between the pre- and posttest periods. Six independent variables or intervention strategies were put into the model, and four were significant discriminators between the truancy subgroups.
The four intervention strategies and their relationship to the function are presented in Table 1, as 1) call to home, 2) letter sent to home at 35 absences, 3) informal contacts, and 4) weeks worked with the TPF. All four variables significantly impact changes in truancy. Calling to the home and informal contacts correlate positively with the dependent variable and the letters sent at 35 absences as well as the weeks the TPFs worked with truants correlate negatively. The relative size and directions of these correlations were expected. For instance, sending a letter indicates a very high rate of absenteeism is already established, and working longer with the truant is probably an indication that there is diminishing hope the student will improve his/her attendance pattern. These four intervention strategies suggest that the prevention program was able to significantly intervene and alter patterns of poor attendance.
The three related research questions investigated in this study are supported by the findings. There is descriptive evidence available to indicate that liberal arts and science students do work effectively as truancy prevention facilitators in the high school setting, and there is demonstrated importance to the school and community in complying with the state statute on truancy. Finally, the evidence for the research question evaluating the program and the significance of intervention strategies in affecting attendance is positive and statistically significant.
The findings demonstrate that the three research questions were supported. Not only were the liberal arts students effective as administrative personnel and as change agents with the truants, but the specific strategies that accounted for changes in attendance behavior were identified. Traditional high schools, with typically-punitive measures to fight truancy, have not worked well in recent times. Cnaan and Seltzer ( 1988, p. 176) argue that "overly restrictive rules or unfair application of rules in school" puts excessive stress on students. They go on to say that "Use of suspension and expulsion as punishment also may enhance truancy." Support for this position can also be found in the work of Hyman, et al. (1982). Most students who skip school are apparently unaffected by threat or punishment.
When a mandatory reporting law was passed in the investigators' home state, new approaches to affecting absenteeism were tried. It seemed important that the population with the highest truancy and dropout rate be served, and it seemed just as important to "prevent" rather than "punish." Such was the justification for the funded research grant that has underwritten the Truancy Prevention Program for three years.
Liberal arts and science cooperative education students from a local university were placed in high school settings to become the change agents. These truancy prevention facilitators were close to the age of the high school population they were serving. The TPFs could be viewed in a less-threatening light than principals, and their social science training had given them ideas and insights about adolescents who might have special needs. As change agents, they were to use their skills to create informal, positive relations with many of the truants and work to decrease the hostility between the truant, school, and the truant's family. Through early detection and monitoring of the absence pattern, the TPFs were able to contact the student and the home with comments and move toward a collective decision about the attendance problem.
The TPFs were sometimes able to work through the in-school problems and make out-of-school referrals when indicated. This early detection and tracking allowed intervention to occur before long strings of absences and school failure might set in and permanently sever the relationship between the school and truant. It is clear from the findings that the intervention strategies employed by the TPFs worked to accomplish these goals.
A second and equally important justification for the prevention program was its administrative value to the school and the community. Large amounts of work were done by the TPFs as they looked at absences, completed important legal paper work, and attended to the less-serious cases coming to the principal's office. The school district, district attorney's office, and juvenile court have all become dependent on these administrative activities. "Tracking of students, inadequate reporting, recording, and follow-up to prevent truancy are additional factors of school characteristics that may contribute to truancy" ( Cnaan and Seltzer, 1988, p. 176). As has been demonstrated here and by Herbst and Sontheimer (1987), college students are effective change agents with a high school population.
Truancy is a multifaceted problem. Finding ways to counteract the devastating effects of truancy will require the attention of many, many people and many community agencies. Neither passing a law nor applying many of the other traditional approaches to stop truancy will by themselves be effective. It is obvious from the research reported here that the causes of truancy can be personal, familial, school-based, or relate to other factors such as friends.
Effectively intervening in such situations argues strongly for a complex model of understanding the causes of truancy. The "ecosystems perspective" of Cnaan and Seltzer ( 1988) is one example of such a theory.
"Thus, based on an ecosystems perspective, practitioners in schools are expected to broaden their pursuit of causes of truancy to include a complex set of personal and environmental causes and their interactions." ( Cnaan and Seltzer, 1988, p. 174.)
The cooperative education program at the university has used research funds to develop student positions which allowed the program to effectively participate in a community-wide effort to decrease truancy. The findings in this paper strongly support the value of cooperation between community agencies, the university, and liberal arts and science majors. The results of this project are beginning to be seen and requests for more cooperative education involvement at different places in the school system and related agencies are occurring.
The project was not without its limitations, however. With the TPF's focused on other very important matters, data collection was frequently left as a much lower priority. Incomplete data sets from the schools resulted in a much smaller sample size for analysis than we had hoped. Better training and monitoring of the college students is indicated by this limitation. Another limitation of the project centered on the politics relating to schools, courts, and community agencies ... of which the university is one. Any community-wide program that intends to decrease truancy on a large scale will encounter personal, professional, and bureaucratic agendas for self-aggrandizement. Those interested in a similar project could benefit from an understanding of the political ramifications of such involvement and by finding an early base of support. Further scientific investigations of truancy might best focus on some of the simple interventions having to do with contacting parents by phone and making more informal contacts with students very early in their non-attendance. Since this Truancy Prevention Program is ongoing, refinements in procedures anticipate two more years of improved data collection and results.