The process of selecting out parks from the original shape file was somewhat subjective and therefore has the potential for error. First there is the issue of what is considered a park and what is not. A polygon that is labelled as a park is not necessarily a park that is appropriate for children. There also could be areas of the city that are not called a park but could still be an outdoor green space appropriate for children to play. Furthermore, when polygons were manually selected and removed, there is the potential that certain polygons were missed that should have been removed but were not, or vice-versa, polygons that were removed that should not have been. The definition of a park is somewhat arbitrary, and therefore it is not easy to determine exactly the qualifications for a park in this study.
The grouping analysis also is a potential misrepresentation of information. The grouping analysis only includes park area that is within the CT boundary. However, there could be park area very close to the boundary of the CT that would be easily accessible to people living within the CT, but this is excluded from the analysis. This is an example of a larger issue in using GIS that draws boundaries that are not present in the real world.