Walking is the most popular physical activity reported by adults, and the benefits of walking are well documented in the literature (Fox & Hillsdon, 2007). Walkability refers to the friendliness of neighbourhood characteristics the enables pedestrian walking (Grant, Edwards, Sveistrup, Andrew, & Egan, 2010). A well-known and validated perceived walkability scale, the “Neighborhood Environment Walkability Scale” (NEWS; Saelens, Sallis, Black, et al., 2003) was developed in San Diego, California. Several adapted versions of the NEWS have been previously constructed for continued use in warm climates; however, adaptations of the NEWS for climates with winter walkability have not been developed. The main objective of this dissertation was to adapt the NEWS for colder climates and to assess the reliability and validity of the adapted measure, the NEWS-North, by conducting three studies. The first study consisted of two phases. For the first phase, three focus groups reviewed the NEWS questionnaire in a Canadian context and provided feedback on how to adapt the NEWS for use in Canada and other northern climates. The inclusion of a winter walkability section as well as the removal of American spellings and measurements were among some of the suggestions from the focus groups. An initial draft of the NEWS-North was created based on the results from the focus group. For the second phase, this initial draft was sent to a panel of walkability experts to assess content validation. All of the nine subscales were evaluated as highly important during the content validation. In addition, the experts renamed several section titles to improve precision.
In the second study and third study, data from an online pilot test in Ottawa, Canada of the NEWS-North was used. The validity based on the internal structure of the questionnaire (i.e., factorial validity) and reliability (i.e., and internal consistency and test-retest reliability) of the NEWS-North was examined in Study Two. In this study, only subscales with the same Likert scale were examined (i.e., seven subscales) following the instructions of the adaptation protocol. Our results suggested a 7-factor model structure based on a multilevel exploratory factor analyses (MEFA). In addition, the 7-factor MEFA model had good test-retest reliability (ICC = 0.85 to 0.93) as well as adequate internal consistency (α = 0.53 to 0.75). In Study Three, the same seven subscales identified by the MEFA were included. In this study, a two-level Hierarchical Linear Modeling (HLM) with spatial dependency analyses was used to examine the convergent validity in the NEWS-North across 64 Ottawa neighbourhoods. The HLM analyses examined each subscale of the NEWS-North with and without any control variables. More specifically, they determined how each subscale of the NEWS-North predicted walking behaviour in the last two weeks at each level. The individual level (Level One) was comprised of the seven subscales of the NEWS-North from self-reported data along with three control variables (i.e., dog ownership, season, and gender). The neighbourhood level (Level Two) was comprised of the six equivalent subscales measured objectively with one control variable. At the time of this study, unfortunately, an equivalent winter subscale was not available at the neighbourhood level. At the individual level without any control variables, only the sidewalk subscale was a marginally significant predictor of walking. However, the sidewalk subscale reached significance while controlling for dog ownership, season, and gender. At the neighbourhood level without controlling for resident median income after tax, two NEWS-North equivalent subscales were found to predict walking behaviours. The sidewalk subscale was marginally significant and the land-use access subscale (via Franks walkability index Frank et al., 2010) was significant. While controlling for neighbourhood resident median income after- tax, both the land-use access subscale and sidewalk subscale were significant predictors of walking over the past two weeks. Overall, this dissertation supports the development for a perceived walkability questionnaire for seasonal climates. In addition, the preliminary results overall provide initial evidence for the reliability and validity of the NEWS-North. Overall implications, limitations, and future applied uses of the NEWS-North are discussed.
| Identifer | oai:union.ndltd.org:uottawa.ca/oai:ruor.uottawa.ca:10393/37055 |
| Date | January 2017 |
| Creators | Takacs, Trista Anne |
| Contributors | Kristjansson, Elizabeth Anne |
| Publisher | Université d'Ottawa / University of Ottawa |
| Source Sets | Université d’Ottawa |
| Language | English |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.002 seconds