Public opinions on biological diversity in Norway: Politics, science, or culture?

Abstract

Loss of biological diversity is among the most pressing environmental problems facing modern societies. Whereas today much is known about peoples' opinions on climate change, there is still a lack of knowledge regarding how people asses the issue of biological diversity. The article represents a step towards a better understanding of public opinions on the issue of biodiversity, and hence also of the prerequisites and possibilities for a more or less vigorous or legitimate policy on this issue. The analyses show that social background matters for concern about biodiversity, but only to a certain extent: women are less negative towards biodiversity protection than men, the older segments of the population, at least in Norway, are less concerned with biodiversity, and education has only a marginal effect. Hence, to a large extent, political attitudes matter and indicate that concern for biodiversity seems to work by various types of available heuristics whereby a given case is linked to political understandings of other and similar and/or familiar issues. The inclusion of other political variables (local decision-making) and what the authors label ‘cultural variables’ (e.g. view of nature and trust in science) seem also to matter decisively for attitudes towards biodiversity.

Keywords:

• biodiversity
• politics
• public opinion
• science
• values

Notes

1. For the intentions of the governmental coalition consisting of the Labour Party, Socialist Party, and Centre Party, see: http://www.regjeringen.no/upload/SMK/Vedlegg/2005/regjeringsplatform_SoriaMoria.pdf (accessed 2 October 2012)

2. TNS web portal: http://www.tns-gallup.no (accessed June 2011)

3. Latent class analysis (LCA) is often presented as the categorical counterpart to factor analysis, which is normally used for continuous variables (Hagenaars & Allan 2002). LCA is based on the premise that the covariation between a set of observed (manifest) variables results from the relationship of each of these variables to a latent variable, which then ascribes the cases in data into classes.

4. M.S. Marino, N.E. Bjørge, T. Ericson, P.A. Garnåsjordet, H. Karlsen, J. Randers & D. Rees ‘Public opinion on climate policy’. Paper in preparation in 2012.

5. The values in Fig. 5 have been estimated by taking the inverse logit (invlogit = exp(x)/1+ exp(x)) of the results from the ordered logistic regression analysis. For a first set of (lowest) values: invlogit (‘Cutpoint 1’ − (model)). For the next (second) set of values: invlogit (‘Cutpoint 2’ − (model)) − invlogit (‘Cutpoint1’ − (model)). By ‘model’, we mean the sum of regression coefficients multiplied by a chosen set of values on the independent variables (in our case, mean values except for the variable with changing values in the figure). Regressions, estimations, and figures were all computed and constructed in R, library MASS (Venables & Ripley 2002).