computer science classes, and the use of computers to support other subjects, cannot happen in schools that have antiquated wiring. The evidence is conclusive that the numerous City school buildings with these deficiencies impede learning. Plaintiffs also offered probative evidence that the totality of conditions in crumbling facilities can have a pernicious effect [187 [187 Misc.2d Misc.2d 47] 47] on student achievement. As former SED Commissioner Thomas Sobol testified: If you ask the children to attend school in conditions where plaster is crumbling, the roof is leaking and classes are being held in unlikely places because of overcrowded conditions, that says something to the child about how you diminish the value of the activity and of the child's participation in it and perhaps of the child himself. If, on the other hand, you send a child to a school in well-appointed or [adequate facilities] that sends the opposite message. That says this counts. You count. Do well. The court finds that this evidence is credible as it is based on the experience and intuition of knowledgeable educators. However, this evidence does not attempt to gauge the magnitude of the effect of school facility condition upon student performance. Unlike plaintiffs, defendants attempted to measure empirically the effect (if any) that school facilities have on student performance. Defendants offered statistical analyses performed by their expert, Dr. Hanushek, which purport to show no link between a school's disrepair and the test score performance of its students. Dr. Hanushek first examined whether school building conditions, as measured by repair costs per square foot, in high and moderate poverty schools were related to student scores on standardized reading and math tests. This analysis tended to show that higher-performing students in elementary and middle schools are clustered in schools with high per square foot repair costs, i.e., those in worse condition. Dr. Hanushek thus concluded that facility repair needs do not cause performance differences among students. Dr. Hanushek further analyzed all New York City public elementary schools using, among other variables, per square foot repair costs and something called "facility scores" allegedly derived from the BCAS rating system. In these regression analyses Dr. Hanushek attempted to control for socioeconomic deficits of a school's students by accounting for various factors in each of the schools including the median income of parents, the limited English proficiency rate for students, and rate of special education participation. He again reached the conclusion that facility disrepair is not causing negative student performance. Dr. Hanushek's statistical analyses on this question are of limited probative value. First, the underlying data regarding [187 [187 Misc.2d Misc.2d 48] 48] per square foot repair costs and facility scores is questionable. Dr. Hanushek testified that he did not compile this data, and the court admitted these analyses subject to the establishment of a foundation by another witness. Defendants' attempts to do so were both convoluted and incomplete. Defendants at least arguably established that the per square foot repair costs were calculated at the direction of Mr. O'Toole and then provided to Dr. Hanushek through defendants' counsel and another defense expert. However, no witness laid a foundation for the facility scores data. Accordingly, to the extent that they rely on facility scores Dr. Hanushek's regression analyses have no probative value. It appears that facility scores are compilations of BCAS scores (1-5) for building components, without any weighting assigned to the various components. Therefore, for the reasons discussed in subsections F (1) (c) and (d) above, even if the court were to consider them, facility scores would not adequately measure facility conditions. The per square foot repair costs used by Dr. Hanushek are compromised to an unknowable degree by Mr. O'Toole's use of cost estimates included in the BCAS, rather than on more accurate refinements to those cost estimates later generated by BOE for inclusion in its 2000-2004 five-year plan. Less important, but still a source of doubt, is defendants' failure to set forth how they calculated per square foot costs. This calculation appears to have been a simple division of a building's square footage by its BCAS repair cost estimates, but defendants failed to establish the validity of this calculation. Dr. Hanushek's regression analysis is hampered by its time frame, which encompassed a single school year. As stated above, education is a cumulative undertaking. A student's performance cannot be measured by the resources he was provided in a single year, but rather should be measured by the resources provided over a number of years. Accordingly, longitudinal studies of student performance over a period of years are more probative than one-year snapshots. Perhaps it could be argued that because Dr. Hanushek's analyses concerning school facilities encompassed the performance of all the students in a given school, it thereby provided a kind of compressed longitudinal study wherein the performance of students in different grades in a given school stands in for a study of a cohort of students as they progress through grades over multiple years. However, defendants did not make this argument and the court finds no support for it in the record. [187 [187 Misc.2d Misc.2d 49] 49] For the reasons stated, the physical condition of New York City's schools has a negative effect upon the academic performance of the City's public school students. However, the magnitude of that effect is unclear from the evidence at trial. G . O v e r c r ow d i n g a n d C l a s s S i z e The poor physical state of New York City public school facilities coupled with an influx of new students into the system in the late 1980's and the first half of the 1990's has resulted in severe overcrowding in many of its schools. Overcrowding, which exists at every level, is most severe in elementary and high schools. Overcrowding has a negative effect on student achievement. Public school enrollment grew rapidly in the 1950's and 1960's and peaked in 1971 at approximately 1,150,000 students. Between 1971 and 1982, enrollment shrank by approximately 220,000, though some districts did experience expansion during this period. New York City's fiscal crisis in the mid-1970's caused the City to shed a number of school buildings, decrease the money spent on maintaining school facilities, and slow the creation of new schools. Thus the City was ill-prepared when enrollment began to increase in 1983 and exploded in 1989. In the five years from October 1988 to October 1993, enrollment grew by close to 80,000 students and reached a total of 1,016,000 students. Rapid growth continued until 1996, when it slowed somewhat. Enrollment in the 1999-2000 school year was 1,102,000. The credible evidence at trial indicates that steady growth of the system will continue. While projections generated by the Grier Partnership, consultants to BOE, indicate that enrollment will level off and begin to drop in the middle of the current decade, recent data on birth rates and immigration that postdates the Grier Partnership report cast doubt on its predictions. In any event, the evidence