Beyond Mean and Variance: Economic Risk Versus Perceived Risk of Farming, Foraging, and Fishing Activities in Southweste
- PDF / 1,422,565 Bytes
- 15 Pages / 595.276 x 790.866 pts Page_size
- 67 Downloads / 197 Views
Beyond Mean and Variance: Economic Risk Versus Perceived Risk of Farming, Foraging, and Fishing Activities in Southwestern Madagascar Bram Tucker & Jaovola Tombo & Tsiazonera & Patricia Hajasoa & Charlotte Nagnisaha & Vorisoa Rene Lahitoka & Christian Zahatsy
Published online: 8 February 2013 # Springer Science+Business Media New York 2013
Abstract Economic analyses of risk typically employ mean and variance calculations such as the coefficient of variation (CV) and z-score model. In this study we examine to what degree these simple measures approximate people’s perceptions of the risk associated with different farming, foraging, and fishing subsistence activities in southwestern Madagascar. CV calculated using production data for 15 activities (ranked) explains about 18 % of participants’ risk ratings from 24 focus groups. CV corresponds better to the ratings of men (37 %) and farmers (40 %), but does not predict the ratings by women, foragers, or Vezo fishermen, suggesting that economic measures of risk have a male and agrarian bias. Narrative responses suggest that risk is qualitatively different for foragers and fishermen, who face hazard (danger) and the cosmological dread of dying far from home. We conclude by predicting the ethnographic and ecological contexts when researchers should go beyond mean and variance when estimating risk. Keywords Subsistence risk . Risk perception . Agriculture . Foraging . Madagascar B. Tucker (*) Department of Anthropology, University of Georgia, 250A Baldwin Hall, Jackson St, Athens, GA 30602, USA e-mail: [email protected] J. Tombo : Tsiazonera : P. Hajasoa : C. Nagnisaha : V. R. Lahitoka : C. Zahatsy Université de Toliara, and Centre de Recherche sur l’Art et Les Traditions Orales à Madagascar (CeDRATOM), BP 185, Toliara 601, Madagascar
Introduction Social and behavioral scientists working in the neoclassical economic tradition commonly define risk as unpredictable variation, or as probability of shortfall or loss. Risk is then calculated in terms of mean and variance, often relative to a utility function for gains and losses (Bernoulli 1954 [1738]; Binswanger 1980; Friedman and Savage 1948; Kahneman and Tversky 1979; Kuznar 2001; Winterhalder et al. 1999). This interpretation of risk is sometimes called pure risk. Among its appeals is that it is relatively easy to measure. If one wants to know the risk associated with planting a crop such as maize, one need only measure production in a sample of maize fields and calculate mean production and standard deviation (SD). The coefficient of variation (CV), which is the simplest risk metric, is the ratio of SD to mean. Human ecologists frequently employ pure risk measures such as this to explain and predict subsistence behavior among hunter-gatherers, fishermen, farmers, and herders (Bliege Bird and Bird 2008; Goland 1993; Hegmon 1989; Kohler and van West 1996; Tucker et al. 2010; Winterhalder 1986; Winterhalder and Goland 1997). But from the perspective of the rural subsistence decision-maker facing real life choices such as which
Data Loading...
