Place attachment in New Zealand

Abstract

The ongoing world wide focus on communities, the building of social capital, and of residential mobility has produced a stream of research on attachment to place. Surprisingly, very little published work appears to have been conducted in New Zealand. This study draws on the results of a recent nationwide sample survey of 1001 residents throughout New Zealand designed to ascertain the relative importance of the different dimensions of place attachment. Most previous work agrees that place attachment is driven out of several different relationships to place which are expressed in terms of attitudes (e.g. sentiment and emotions) and behaviour (formal and informal social networks). Application of principal components factor analysis to 25 questions pertaining to place attachment uncovers five principal dimensions: sentimental, friends, relatives, participation, and evaluation accounting for nearly half of the total variance. Clearly in New Zealand, as the international literature has shown, attachment to place is indeed multidimensional. Of particular interest in this study is the extent to which each of these five dimensions varies according to the characteristics of both the sampled residents and the place they live in, and the extent to which there is an interaction between the two. Certain types of people in certain types of place are likely to exhibit disproportionate levels of place attachment on one or more dimensions. Multiple regression models applied to each of the five dimensions of attachment showed that different subpopulations are indeed sensitive to different dimensions, in relatively predictable ways: the elderly are more highly attached through sentimental feelings towards the community; families with children are attached through friends and participation; while respondents of Maori ethnicity are more like to be attached by family than those of New Zealand European ethnicity. Of particular interest was the way in which, even after controlling for the characteristics of respondents, different types of places evoked different dimensions of attachment. Population size is a relatively poor discriminator when it comes to attachment - notwithstanding the prevalence of the linear model of attachment in the literature. Place defined in terms of type, that is in terms of a mix of size and position on an urban-rural axis, proved more robust. Also important was the socio-economic level of the community as judged by its level of deprivation. Both the linear and systemic models of place attachment have limited applicability in the New Zealand context. A more robust model is one in which type of place is allowed to interact with different socio and demographic characteristics of residents to produce different degrees of attachment by dimension. While it is still meaningful to talk in terms of place attachment, when it comes to measuring it and assessing its importance with respect to any given outcome, especially in a policy context, its inherent multidimensionality must be recognised as well as the sensitivity of that mix by population type.