The 2001 report by Indiana University's Metcalf compares four groups of students, from the time they entered first grade in 1998-99 through the end of second grade. There are two groups of voucher students: a 3-year voucher group that entered the voucher program at the start of kindergarten, and a 2-year voucher group that entered the program at the start of first grade. The study also included two groups of public school students: those whose families had applied for a voucher and either not received or not used one (called "applicant/non-recipients"), and their classmates, whose families had not applied for a voucher (called "non-applicants").53
Metcalf found that both public school groups in the study demonstrated greater learning gains in the subjects tested (language, reading and math) over the two years of the study than either of the voucher student groups. While voucher students had higher total test scores entering first grade, this advantage quickly began to erode. In fact, one group of public school students students - the applicant/non-rrecipients - surpassed the 2-year voucher students (with a score of 576) and had completely caught up with the 3-year voucher students by the end of second grade (scores of 583 each). And the non-applicant public school students (577) pulled slightly ahead of the 2-year voucher students and closed what had been an 11-point gap between themselves and the 3-year voucher students to only six points.54
Some pro-voucher groups have tried to frame the 2001 data as proof that vouchers are superior to public schools by focusing on the test scores of the 3-year voucher students.55 However, even Metcalf notes that other variables such as more school experience or added parental involvement could have affected his sample population, favoring 3-year voucher students and giving them an added advantage.56
Due to budget cuts, full-day kindergarten in public schools was eliminated in 1995.57 While full-day kindergarten was in fact restored by the 1997-98 school year, families would have had to have enrolled their child in the voucher program before the decision restoring funding for public school kindergarten was made. It is thus entirely possible that some Cleveland families actually chose the voucher program for 1997-98 specifically because it gave them access to kindergarten opportunities otherwise unavailable.
Second, despite the advantage that students in the 3-year voucher group had entering first grade, both groups of public school students made greater gains over the course of the two years that were studied - and one public school group actually caught up with the 3-year voucher students.
Third, and most importantly, Metcalf concluded that the analysis of student test results from voucher schools and public schools “presented no clear or consistent pattern tha[t] can be attributable to [voucher] program participation.>/i>"58 Echoing this view, officials at the Ohio Department of Education summed up the study in distinctly lukewarm terms, noting that voucher students "perform at a similar academic level as public school students."59
One issue that emerges from the latest Cleveland evaluation - released in early December - is that the voucher student population is not representative of the city’s school population from which it is drawn. Metcalf found that the voucher students were disproportionately white, and that this phenomenon increased over the course of the study. Although only 16 percent of the public school non-applicant sample were white, nearly 30 percent of the voucher students were white at the start of the study. This imbalance was further accentuated when 59 percent of those entering the voucher program in second grade were white. By this point, the voucher group was also notably less poor than the public school students. This is consistent with concerns that vouchers are likely to exacerbate racial segregation and other inequalities in U.S. schools.60
Two 1999 reports by voucher proponents purported to show that vouchers will, in fact, integrate, rather than segregate, private schools. However, available evidence does not corroborate this view. For example, a study cited by the voucher proponents claims that voucher students are more likely than public school students to attend a school whose racial composition is representative of the community-at-large. This claim was based on an apples-to-oranges comparison of Cleveland metropolitan data—which includes suburban schools—with the Cleveland school district. However, suburban schools are not participating in the voucher program. By using the relevant apples-to-apples comparison of Cleveland city school data and utilizing the guidelines set up under the desegregation order that applies to the city’s public schools, Cleveland public school students are about four times more likely to attend integrated schools than voucher students.61