274 Limitations There are limitations associated with survey research that may have affected this study. Non-response bias complicates the pro- cess of making inferences about populations at the neighborhood, district, or city scale. Non-response factors may have included lack of internet access or language barriers within recruited households (Dillman et al. 2009), particularly in districts that have a more diverse cultural composition or lower median household income. Analysis did not assess whether neighborhood response represented local demographics, thus the comparability of cul- tural or socioeconomic diversity within neighborhood respons- es is unknown. Also, there was a variable rate of undeliverable recruitment letters across neighborhoods. There were also vari- able rates of response to successful recruitments, with higher rates associated with more affluent districts of the city. In addition, a possible response influence is that people feeling both a work- family conflict and time pressure are more likely non-respondents (Vercruyssen et al. 2011), and may be the people who are less active in planning or maintaining their personal landscape. Given the City of Seattle’s dedication to residential tree plant- ing, the decision was made to invest up front in high quality sur- vey outreach. The tailored, multi-contact approach was used as it has been shown to generate higher response rates (Dillman et al. 2009). The use of incentives has a long history in mail sur- veys (Singer 2002), and a meta-analysis of survey studies found that prepaid incentives that are delivered with the survey (as was done in this study) yield significantly higher response rates whereas contingent (promised) incentives do not (Church 1993). Despite these efforts, a moderately acceptable response rate was achieved. Non-response bias could be addressed post hoc (as summarized by Groves 2006) by 1) comparing initial and late respondents, as late returns may be similar to non-responders; 2) comparing survey results to known population parameters, which was done on a very general demographical level; 3) con- tacting and surveying a sample of nonrespondents after the sur- vey was completed; and 4) inspecting variables known to indi- cate bias. Due to the City of Seattle’s satisfaction with the final response rate, not all of these post hoc comparisons were done. Future Research Many studies have assessed public perceptions, preferences, and satisfaction for trees in cities and within residential settings. This research differs in that it was one of the few studies to assess how homeowners interact with the trees in their yard, thereby influ- encing the canopy cover and associated benefits of urban forests. Additional studies could further assist urban forest managers in developing effective tools to communicate with and engage urban residents in tree planting and care. Similar research could be done in other cities to compare homeowner attitudes and behaviors on a national scale and identify geographic patterns of behavior in other areas. Of particular interest are the situations of where a choice must be made between two conflicting sets of benefits, such as the choice between shade trees and solar panels or plant- ing small versus large trees. Additional research themes follow. Attitudes and behavior of other private property owners This research focused solely on single-family homeowners. Addi- tional research could examine the behaviors and attitudes of rent- ers, rental property owners and landlords, multi-family homeown- ©2013 International Society of Arboriculture Dilley and Wolf: Homeowner Interactions with Residential Trees in Urban Areas ers, and commercial property owners toward trees. These groups may act in significantly different ways than single-family home- owners, meaning that urban forest managers may need to construct different programs to address canopy cover across such properties. Improved survey response rates Declining response rates are an ongoing concern in survey re- search. The Pew Center for Research found that from the 1990s to the 2000s the household response rate to standard surveys declined from 36% to 27% (Kohut et al. 2004). One firm’s meta-analysis of 199 general population surveys indicated a median survey re- sponse rate of 26% (Hamilton 2009). Kaplowitz et al. (2004) test- ed five questionnaire distribution treatments using combinations of mail and e-mail contact, finding that response rates ranged from 21% to 32%. This project’s response rate compares favorably to trends in general population surveys, but a higher rate would gen- erate more confidence in results as the basis of city programs. If future surveys of homeowners or other property owners are to be done, then new tactics should be attempted and evaluated for more effective data collection. Strategies to boost response could include developing a survey that is shorter in length (Ed- wards et al. 2002), and has higher topic interest (Martin 1994) or involvement (Van Kenhove et al. 2002). Vegetation map- ping could be used to tailor the study region and sample design to focus outreach effort on specific districts of interest, rather than sampling the entire city, and would allow survey resources to be directed into higher response approaches, such as tele- phone, door-to-door interview, or drop-off and pick-up surveys. Barriers to large tree planting Further research could be conducted on how to encourage the planting of tree species that generate the greatest environmental benefits, particularly large evergreens. Managers would benefit from a better understanding of the behavioral barriers that dis- courage large tree planting by homeowners in order to design spe- cific outreach programs. Possible barriers to large tree planting identified in pre-survey interviews included perceived tree fail- ure risk, perceived legal liability, limited capacity for households to take on large tree management tasks, and space constraints. CONCLUSIONS Urban forest planning and management involves complex human dimensions of attitudes, perceptions, and behavior. As urban for- est managers work to build sustainable urban forests, it can be helpful to understand the ways in which the public, and in par- ticular homeowners, interact with and influence urban greenery. Social scientists can contribute to understanding the social and community diversity embedded within resource systems. This study, for instance, pursued data that could be used to under- stand homeowner tree practices (such as large tree and season choice for planting) and reconcile them with the public appre- ciations for trees found in the literature review, as attitude and action are not always consistent (do Paço and Raposo 2009). Scale plays a role in both data collection and resource manage- ment. Not all cities use neighborhoods as the administrative unit of planning or program delivery, but urban forest managers can designate units within their cities to help organize and prioritize urban forest activity. Some cities have discovered that trees and parks have not been equitably provided to all of their communi-
November 2013
| Title Name |
Pages |
Delete |
Url |
| Empty |
Search Text Block
Page #page_num
#doc_title
Hi $receivername|$receiveremail,
$sendername|$senderemail wrote these comments for you:
$message
$sendername|$senderemail would like for you to view the following digital edition.
Please click on the page below to be directed to the digital edition:
$thumbnail$pagenum
$link$pagenum
Your form submission was a success. You will be contacted by Washington Gas with follow-up information regarding your request.
This process might take longer please wait
