The influence of economic affluence and environmental conditions on an individual's concern for the environment: a Greek case study (2005–2007)

Abstract

This paper tests the impact of economic affluence and objective environmental conditions on an individual's concern for the environment across Greek regions and over time. Our data originate from three nation-wide surveys conducted in Greece between 2005 and 2007 (N = 3592 individuals). By employing a multilevel hierarchical modelling technique, we are able to check for the influence of micro-level and macro-level predictor variables simultaneously. We find that an individual's environmental concern is largely influenced by micro-level variables and, in particular, positively influenced by family income. Macro-level variables, such as community affluence and objective environmental conditions, turned out to be statistically non-significant. Furthermore, we find that the effects of the various predictors on an individual's concern for the environment vary significantly over time and across regions.

Keywords:

• environmental concern
• Greece
• affluence
• environmental conditions
• hierarchical linear modelling

Acknowledgements

The authors thank Dr Christoforos Vernardakis for granting permission to use the data collected in the various V-PRC surveys.

Notes

Thus, shortly after the period covered in this study, the percentage of Greeks answering that “the protection of the environment is ‘very important’ to me personally” stood at 88% (third highest and compared with a 64% for the EU27) (Eurobarometer 2008, p. 9). More relevant to the research questions addressed in this paper, following a couple of years of economic recession, this percentage was down to 70% in 2011 (EU 27–54%) (Eurobarometer 2011, p. 9).

Of equal importance is the role of socio-psychological variables (e.g. values, norms, etc.). Yet, since our data set does not include information on any of these variables, we will not discuss the relevant literature concerning their influence on environmental concern.

This is a tentative hypothesis since the available research is evenly split between those reporting a positive influence and those reporting a negative one. We decided to opt for the “positive” hypothesis since this is in line with the original suggestion made by Inglehart (1995).

At the time of the surveys, there were 51 prefectures in Greece. Following the 2011 overhaul of the Greek administrative system, they have been roughly replaced by “peripheral units” (74 of them).

Since the opinion surveys were carried out in early March of each year, in our models, we used macro-level data for the year preceding the survey: it is only logical to assume that, for example, the respondents of March 2006 might have been influenced by the car-related pollution occurring in their area in 2005 (i.e. the year before the survey) –in other words, they might be influenced by factors already in place and operating and not by future (and unforeseen/yet-to-materialise) developments.

In our case, data collected within the same prefecture can be assumed to be correlated, thus forming a specific cluster of measurements.