Data Analysis

Cross-group Comparisons

We conducted group comparisons on several issues associated with urban forests. We compared mean responses for females and males, for homeowners and renters, and across the different cities included in the study. For city comparisons, we combined Eugene and Springfield given their proximity.

Problems Associated with Urban Forests

We presented the following set of questions to respondents to assess attitudes about problems commonly associated with urban forests:

People sometimes feel that trees and nature in and around cities create problems. We would like to learn about how big of a problem you feel the following urban forest issues are in your city right now.

Directions: Please rate how severe you think a problem is in your city, by circling one number on the scale from 6 (Severe problem) to 0 (Not a problem at all).


The following figures display the results, starting with the entire sample. Following the outcome for the entire sample, figures depict outcomes for comparisons between females and males, homeowners and renters, and across the three city areas for problems associated with the presence of urban forests. The question asked how severe, on a scale from 6 ("A severe problem") to 0 ("Not a problem at all") respondents felt the problems were. The order (highest to lowest) in which the problems appear is based upon the average of the scores across the groups for each problem item.




The following tables show the statistical comparisons between groups. These tables show where statistically significant differences occur, and whether those differences could be considered minimal, typical, or substantial. Again, according to Vaske (2008), the differences, or effect sizes, indicate whether there is a relationship among variables that has practical significance. A "small" or "minimal" effect size indicates that while the relationship between the variables is statistically significant, it is nonetheless minor or of relatively small consequence. A "typical" relationship, or typical effect size, indicates a relationship that one would expect to find in most behavioral science studies. That is, it is typical of the kinds of relationships that are commonly found in this type of research. Finally, a substantial or large effect size should indicate to researchers and practitioners that the association between the variables is the result of a considerable difference and the relationship is noteworthy.