RESUMEN
Previous studies do not agree on the strengths and directions of the effects between risk appraisal, nonprotective, and protective responses in private flood mitigation. This may be due to the widespread use of cross-sectional survey designs, which infer causality from theoretical considerations alone. The present longitudinal study, in contrast, builds on the logic that cause precedes effect to confirm the direction of effects. Drawing on two-wave survey data from 554 flood-prone households in Austria, cross-lagged autoregressive models analyze pairwise combinations between risk perception, fear, five nonprotective responses (fatalism, denial, wishful thinking, reliance on social support, reliance on public protection), and seven specific protective responses (ranging from coordination with neighbors to structural modifications of the building). These factors show substantial temporal stability, in particular for nonprotective responses and fear. Only in very few instances can effects over time be confirmed statistically. Nonprotective responses emerge as the major drivers; foremost, denial, and reliance on public protection limit private flood mitigation. This overall null finding on causality may trace back to the 1.5 years' time span and the absence of any policy intervention or flood disaster between survey waves, and the high stability of protection motivation theory components. This finding puts into question the theoretically assumed causal relationships and the effects found in cross-sectional studies. The high trait-like stability requires perseverance in risk management efforts to change attitudes and capabilities. Finding nonprotective responses as key determinants in an overall picture of stability suggests that this factor merits a stronger role in future risk research.
RESUMEN
Self-efficacy is one of the strongest and most consistent drivers of private flood mitigation behavior; however, the factors influencing self-efficacy in the context of flooding remain unclear. The present study examines three potential antecedents of self-efficacy: personal and vicarious experiences of floods or building-related events, social norms for private flood preparedness, and personal competencies such as technical abilities and social skills. While controlling for other drivers in a protection motivation theory (PMT) framework, these antecedents are tested as precursors of self-efficacy and intentions to improve flood resilience. Structural equation modeling is applied to conduct mediation analyses with survey data of 381 flood-prone households in Austria. Contrary to theoretical expectations, personal and vicarious experiences do not predict self-efficacy, presumably because rare flood events and changing hazard characteristics do not facilitate generalizable performance accomplishments. Social norms strongly and consistently influence self-efficacy, especially for actions observable by others, and also directly influence protective responses. Personal competencies increase self-efficacy and support protective action, particularly with regard to preventive and structural measures. The strength and direction of the antecedents of self-efficacy as well as of other PMT determinants vary between general and specific protective responses. This study provides important insights for risk managers, suggesting that interventions involving social norms and personal competencies can be effective in stimulating self-efficacy and, in turn, private flood mitigation. Interventions and research should clearly differentiate between general intention and the implementation of specific measures, and should address cumulative, synergistic, or tradeoff interrelations between multiple measures.
RESUMEN
The recent shift to individualisation of flood risk calls for a stronger involvement of private actors. Bottom-up citizen initiatives (BUIs) may bring together governmental bodies with people at risk. Drawing on a screening of existing BUIs in Europe, North America, and Australia and an in-depth analysis of three study sites, this paper maps BUI activities to stages in the risk management cycle and discusses the institutional, relational and social proximity between BUIs and other stakeholders. Flood BUIs often take over roles that the authorities are not willing or able to fulfil. BUIs emerge out of frustration with current risk policies, after a catastrophic flood event, government-initiated engagement projects or targeted funding opportunities. BUIs can take different forms, ranging from oppositional pressure groups, self-help movements for disaster response and recovery, to initiatives formally installed by law. While self-organised BUIs benefit from high proximity to their home communities, formalised BUIs are deeper embedded in existing institutional structures. In order to gain a stronger voice in the risk debate, BUIs need to expand from the local level to catchment areas and exchange expertise and resources in nationwide or cross-border networks. However, BUIs may create parallel political structures that are not democratically legitimised.
RESUMEN
Flood preparedness of private households is regarded an essential building block of integrated flood risk management. In the past decade, numerous empirical studies have employed the protection motivation theory (PMT) to explain flood mitigation behavior at the household level. However, much of this research has produced mixed results and could not consistently confirm the strength and direction of the relationships between the PMT components. Based on a survey of 2,007 households in flood-prone areas, this study revisits the model structure of the PMT by means of structural equation modeling. Compared to the methods used in previous studies, this modeling technique allows us to capture the PMT components in greater detail and to comprehensively test their hypothesized interrelations. Our results point to two separate routes leading to two different response types: A protective route from coping appraisal to protective behavior, and a non-protective route from threat appraisal to non-protective responses. Risk perception is not found to be part of the protective route, neither are non-protective responses confirmed to undermine protection motivation. The two separate routes are observed consistently across all combinations of the six protective and four non-protective responses assessed in this study. In the light of encouraging private flood adaptation, risk communication measures should specifically target the protective route and avoid (accidentally) providing incentives that fall within the non-protective route. This cross-sectional study, however, cannot establish how the two routes interrelate over time. More experimental and longitudinal research is required to address potential feedback effects and the role of decision stages.