| Reducing students' absenteeism is one of the broad educational policy goals identified in Section 1116 of the No Child Left Behind Act of 2002. Educational leaders and public policy practitioners need to understand factors that impact school attendance. This study examined the association between air pollution and daily elementary school attendance in 13 schools and six school districts located in an environmental justice community in the south Bronx and Northern Manhattan between September 2002 and June 2005. The percentage of students in attendance during 546 school days was correlated and regressed on one spatial variable (distance) and 11 air pollutant variables extracted from the Aerometric Information Retrieval System (AIRS) monitoring network sites in the study area. The daily average attendance rate of 97.062% (SD 0.954) implied an average absenteeism rate of 2.938% and displayed a skewed range from 1.4% to 9.9%. The results of the stepwise linear regression indicated that seven variables: CO_8-hour, NO2_1-hour, O3_8-hour, PM_1-hour, SO2_1-hour, SO2_24-hour, and, SO2_3-hour, explained 2.2% (R2 = 0.022) of the variance in attendance at the school level, while at the district level, 5.3% ( R2 = 0.053) of the attendance was predicted by five variables: CO_1-hour, CO_8-hour, NO2_1-hour, O 3_8-hour, and PM_1-hour. The fact that 97.8% and 94.7% of the variation in attendance at the school level and at the district level, respectively, was unaccounted for raises questions about drawing causal conclusions regarding the roles of the specific environmental variables examined in this study. Given that the data was non-normal, further analysis using non-parametric Kruskal-Wallis analysis of variance (ANOVA) indicated that the spatial variable, distance, exhibited an initial inverse relationship with attendance and then sharply exhibited a positive relationship whereby attendance seems to increase with distance. These counter-intuitive results are explained from the perspectives of associations between socioeconomic factors (specifically poverty), classroom attendance, and educational attainment. Some suggestions are made regarding potential additional strategies and variables that could account for the remaining unexplained variation in school attendance. Implications for educational leadership and public policy are highlighted for the effort to ensure that schools remain venues for incrementally sustained learning and development, rather than an environment where children's health deteriorates one breath at a time. |
