Neb. Researchers Develop New Methods to Collect Dropout Data
Speaking as if they are on the verge of a revolution, researchers from the University of Nebraska say they have a way of making the painstaking process of collecting data on public school dropouts ancient history.
The researchers recently released two independent studies: One expands the common characteristics of a dropout, and the other offers an alternate method of measuring dropouts and tracking those at high risk of dropping out by computer throughout their school careers.
Most schools, if they collect dropout data at all, use sketchy information that often varies from school to school. And they record their dropout information on paper, a process that invites error, the researchers say.
"Schools have been doing this for a hundred years,'' says Jack Frymier, a professor of education at the University of Nebraska at Omaha and a senior research fellow at Phi Delta Kappa, the professional education fraternity, who co-authored both studies. "It's a dumb way to [collect data], and that's what we are trying to attack head on.''
The quest by Mr. Frymier and his colleagues, Jack McKay and Gary Hartzell, could have ramifications nationwide. The second national education goal pledges that, by 2000, the high school graduation rate will increase to at least 90 percent.
But while the National Education Goals Panel has chosen two methods of measuring progress toward that goal--the percentage of 19- and 20-year-olds and the percentage of 23- and 24-year-olds with high school credentials--individual schools currently lack both a common definition of what a dropout is and effective ways they can boost their own graduation rates.
'In the School Hall'
According to the U.S. Department of Education, a dropout is an individual who has:
- Been enrolled in grades 7-12 at any time during the previous school year;
- Was not enrolled at the beginning of the current school year; and
- Had not graduated from high school or completed a state- or district-approved education program.
But in practice, that definition has proved difficult to apply, researchers say. In some places, students who have transferred, been suspended, or died are routinely included in state dropout counts. In others, states count truants as dropouts, or fail to factor in transfer students, or have no consistent measure from year to year.
But even with a common statistical definition, schools lack a method of determining which of their students are dropping out. Without such a method, educators say, they can do little to keep them in school.
That is where the Nebraska studies come in.
"Whatever policies are made by government, the dropout problem is going to be solved in the school hall where the teachers interact with the actual kids,'' says Mr. Hartzell, a professor of educational administration at the University of Nebraska at Omaha.
In the first study, Mr. Frymier reviewed 114 state and local studies on dropouts and identified specific factors that "caused students to dropout of school, or life.''
He then asked teachers in 41 states to survey 22,000 students on roughly 45 items, including their academic failures, personal tragedies, drug and alcohol use, parents' education, and medical history.
The result is an expanded definition of the characteristics of a dropout that includes such variables as economic status, criminal history, and low birthweight, along with school attendance.
'Holding Power Index'
Working with the characteristics identified in the first study, the authors of the second study attempted to find a way to use technology to standardize and simplify the data-collection process.
Following a cohort of 1,669 students in four Omaha-area high schools from 1987 to 1991, a team of seven University of Nebraska researchers collected data on, among other factors, students' gender, birth date and weight, program of study, and race.
The researchers plugged the information into a software program and came up with what they dubbed the "Holding Power Index,'' which calculates the ability of a school to retain students who may be at risk of dropping out.
The index, which expresses each school's "holding power'' as a percentage of its enrollment, represents a more positive way of presenting data on dropouts by showing how often schools succeed, rather than how often they fail, the researchers say.
At the same time, they say, it is superior to conventional data in that it can show how successful schools are in retaining groups of students.
Mr. Hartzell, one of the study's authors, remembers using the traditional method--subtracting the number of students who graduated from the number who matriculated.
"When I was a high school principal in Southern California, I heard, 'The dropout rate was 25 percent,''' Mr. Hartzell says. "But that didn't tell me anything about my school.''
"This program,'' he continues, "can tell me, for example, that I am doing well with academically oriented white girls, but I'm not doing well with Hispanic vocational-track boys.''
Moreover, he says, school personnel can use the program to find out how certain situations--such as changes in the student-teacher ratio, the extent of parental involvement, and the "commitment to diversity''--affect the dropout rates in their school.
In addition to helping individual schools, the Holding Power Index can also feed into state and national data that are being compiled by the National Center for Educational Statistics. The 1992-93 N.C.E.S. report on dropout rates, scheduled to be released in November, will be the first to include data that are comparable in all states, according to Lee M. Hoffman, the chief of the general surveys and analysis branch at the N.C.E.S..
Ms. Hoffman says she sees the Holding Power Index as "potentially very useful.''
"The biggest need in the field is a fairly nonburdensome way for schools to keep track of students, and programs like this would be good for that purpose,'' she says.
High Marks From Educators
Other educators have also given the program high marks.
The Council of the Great City Schools has endorsed it, according to Mr. Hartzell, and the National Association of Secondary School Principals is so enthusiastic about it that the group arranged a partnership with the authors to advertise and to help distribute the software package in exchange for a share of the proceeds.
Phi Delta Kappa, the educational honorary fraternity based in Bloomington, Ind., which sponsored Mr. Frymier's original survey through a $25,000 grant from the John D. and Catherine T. MacArthur Foundation, has pledged technical support for the software package.
Some school administrators, however, have expressed concerns about the $395 cost of the software package. School funds, they argue, could be better spent on individualized instruction, counseling, and mentoring programs.
But others contend that the Holding Power program is a worthy investment because it helps educators channel their funds more efficiently.
"The way it is now, so many things go into dropout formulas that you're comparing apples and oranges,'' says Sue Evanich, an assistant principal at Westside Middle School in Omaha. "Up to now, it's been a guessing game as to whether a program has had an impact.''
The system could also help identify potential dropouts before they leave school, which is vital to stemming the problem, Ms. Hoffman of the N.C.E.S. notes.
"You don't just wake up one morning and say you are going to drop out of school,'' she says. "It's a gradual disengagement.''
For the studies' authors, though, the most worthwhile aspect of the program is its ability to help schools keep students in school, rather than simply bemoan the dropout problem.
"What we were searching for was a way to do something to focus on
something besides the negative,'' Mr. Frymier says. "Nobody wants to
just focus on the problems; they want to know what strengths we can