The relationship between attitudes and behaviours towards environmental conservation amongst farmers and urban dwellers in Western Australia

Hatwell, Diane (2000) The relationship between attitudes and behaviours towards environmental conservation amongst farmers and urban dwellers in Western Australia. PhD thesis, Murdoch University.

Abstract

The original work of Thurstone ( 1928) on the measurement of attitude has been continued by many researchers. Although Thurstone immediately cautioned against the prediction of behaviour from attitude, and argued that the measurement of attitude in its own right was important, many researchers have studied the relationship between attitude and behaviour. Their general conclusion is that, except in specific situations where the behaviour and attitude are linked in particular ways, attitude and behaviour are not strongly correlated, and therefore, have suggested that they are different din1ensions. Duncan (1985) presents the hypothesis that attitude and behaviour, rather than representing two different dimensions of some construct, can be conceptualised as manifestations of the same underlying disposition but at different levels of intensity. Duncan further suggests that attitude is "easier" than behaviour.

This study was designed to focus on the relationship between attitude and behaviour towards land degradation of farmers in the Central Wheatbelt of the State of Western Australia, and between attitude and behaviour towards urban pollution in Perth, the capital of this state. Land degradation and urban pollution are significant problems for the Central Wheatbelt and the metropolitan area of Western Australia, respectively. Farmers and urban dwellers were invited to complete written surveys on these issues. The key responses were agreement or disagreement to statements reflecting attitudes and reported behaviour. Demographic information and information regarding the level of commitment of the respondents to the attitude was also gathered.

In the study of attitudes through responses to statements, there are two main response mechanisms. In one, there is an ideal direction and it is expected that the more positive the attitude, the more likely a particular responses. For example, the responses may be Disagree (D) or Agree (A) which are scored O and 1 respectively. Then, the greater the probability of the Agree response (scored 1), and the higher the score across more than one statement, the more positive the inferred attitude. The models that are used for analysing such responses have a monotonic form and are termed cumulative. In the second, there is an ideal point, and the closer a statement is to the person's ideal point, the more likely it is that the person will choose the Agree response, and the further away the statement is from the ideal point in either direction (more positive or more negative), the more likely a Disagree response. In this case, the total score across statements cannot be used directly to infer attitude. The models that are used for analysing such responses are single-peaked and are termed unfolding. In both kinds of analyses, persons and statements are located on an attitude continuum. Duncan's hypothesis, that attitude and behaviour may be located on the same continuum. suggests that in some studies, at least, the fact that they are found not to be related may be a methodological artefact.

In the present study, the statements in the scale reporting behaviour with respect to environmental issues were written explicitly according to the cumulative response mechanism, and the statements in the scale reflecting attitude were written according to the unfolding response. However, in part to explore the matter of methodological artefacts, both scales were analysed according to both the cumulative and unfolding models, after the data were configured to suit each analysis.

The study found that, as Duncan bad suggested, attitude and behaviour could be placed on the same continuum, as different manifestations of the same construct. However, the further contention that attitude is "easier" than behaviour appeared too simple according to these data. The data appear to support the contention that attitude alone is not sufficient to explain behaviour, and that attitude and level of involvement may be more effective. The demographic information collected explained some but not all assumptions about the relationship between attitude and behaviour. One particularly noteworthy result was that the correlation between attitude and behaviour was very low (order of 0.1) in the urban sample while moderate (order of 0.5) in the rural sample. It is suggested that part of this difference between rural and urban relationships between attitude and behaviour is explained by the differing situations of the two groups - farmers' attitudes and behaviour to land degradation relate directly to their livelihood. This has implications, not only for the seriousness with which the environmental degradation is viewed, but also for their responses which are an integral part of their working conditions. In contrast, for urban dwellers most behaviours relating to dealing with urban pollution are not directly related to their livelihood and must be carried out in their leisure time. To the degree that this observation explains the difference in the relationship between attitude and behaviour between the two groups, to that degree it shows that the relationship between attitude and behaviour is moderated by other related factors. Studies that test this relationship between attitude and behaviour towards the environment are issues for further research.

Item Type:	Thesis (PhD)
Murdoch Affiliation(s):	Division of Social Sciences, Humanities and Education
Notes:	Note to the author: If you would like to make your thesis openly available on Murdoch University Library's Research Repository, please contact: repository@murdoch.edu.au. Thank you.
Supervisor(s):	UNSPECIFIED
URI:	http://researchrepository.murdoch.edu.au/id/eprint/50385

Item Control Page