The study found that school disorder--as measured by major crime, minor crime, and disorder incidents--was linked to student academic performance. School disorder also influenced school attendance, which in turn was linked to student academic performance. The student background variable of poverty was indirectly related to student academic performance. The poverty level of students was related to student attendance rates and school disorder, which were more directly related to student academic performance. The author suggests ways in which schools and communities can address and improve the variables that negatively influence student academic performance. Study data were obtained from the New York City Department of Education. In order to measure the study variables and their interaction, data were obtained for the school years 2002-03 and 2003-04. Data on the variables of interest in this study were obtained for 581 of New York City's 613 elementary schools. Student academic achievement was measured by school mean scores on grade four English Language Arts and grade four Math. School size was measured by the total student enrollment at each school on October 31, 2002, and student attendance was measured by the percentage of days that students attended schools in a school year. The student background variables measured were poverty (percentage of students eligible for a free lunch) and ethnicity (percentage of White students). 3 tables and 40 references