P. Nelson C. Sprecher, “Are Sensitive Technologies Enablers of Civil Nuclear Power? An Empirical Study”, Atoms for Peace, 3, 2 (2010).
In order to better understand, quantitatively and objectively, the factors that have been associated with the extent to which a given state relies on nuclear energy to generate electricity (termed ‘nuclear reliance’ (NR)), regression analysis was applied to a set of 14 hypothesised independent variables having associated measures constructed from a database assembled for this purpose. That process led to a linear model with five independent variables that collectively predict NR with high confidence (p < 0.05, for all predictors) and acceptable goodness of fit (R² = 0.53). This basic linear model was then employed as a tool to analyse several more-or-less current topics related to proliferation. These include the historical effectiveness of the nonproliferation regime, as regards the spread of sensitive fuel-cycle technologies; the premise underlying (fuel) assurance programmes, as intended to ensure access to (insensitive) nuclear materials and technology, in return for forgoing development of sensitive technologies; and the persistent lack of recipient states willing to accept the bargain underlying assurance programmes.