<p> This research aimed to determine which factors are significant for modeling deer-vehicle collisions, regarding the white-tailed deer, for a city in Illinois. In past research, very detailed variables were used with precise measurements. To simply the model, generalized variables were used: six land use classifications, traffic volume, road width, speed limit, three reproductive seasons of the white-tailed deer, and six classes for time of the day. While all the tested variables proved to be significant, the model rejected the time of the day variable. The binary logistic regression model used to predict the odds of a collision included a sample dataset of occurrences of collisions within the city and random sample points where a collision has not occurred. The model was able to explain 73% of the random and report collision sites, using the variables of significance to the model. Using GIS also aided in determining where the highest number of collisions occurs within the city and is helpful for evaluating roads for mitigation. To conclude, this study can be used in future research to advocate the use of a binary logistic regression model or to support the use of the significant variables of the model.</p>
| Identifer | oai:union.ndltd.org:PROQUEST/oai:pqdtoai.proquest.com:1545177 |
| Date | 24 October 2013 |
| Creators | Lograsso, Kamiliah L. |
| Publisher | Southern Illinois University at Edwardsville |
| Source Sets | ProQuest.com |
| Language | English |
| Detected Language | English |
| Type | thesis |
Page generated in 0.0014 seconds