Does a high level of human activity and site type affect the level of coyote activity?

Statistical Assumptions/Tests

Figure 7 shows the distribution of coyote relative activity data in the natural, rural and urban site types. Each site shows a normal distribution of data with no skewing, and thus no need for transformations. These distributions allowed me to proceed with ANOVA tests.

                                     Figure 7. Distribution of coyote relative activity data in natural, rural and urban site types.

Figure 8 shows the residual plot of the coyote relative activity data for the natural, rural and urban site types. The plot shows that each site’s data has equal variance and gives no indication of non-normal distributions, unequal variances or outliers. Again, no transformations were required, and it was acceptable to proceed with ANOVA.

                                   Figure 8. Residual plot of coyote relative activity data in natural, rural and urban site types.

Limitations

The data from this study does not take population density and seasonal variations in behaviour into account. Both human and coyote population densities in an area likely impact relative activity levels, and coyote activity patterns (and like human activity patterns) also vary depending on the season. As a result, this data currently only applies to the study sites during the season the data was collected. The data can be applied to the overall population when population densities and seasonal behaviour variations have been incorporated.

---