Plots of mean normal factor bias and error in estimates of the intermediate unstable equilibrium (IUE) in small populations with a weak Allee effect, large populations with a weak Allee effect, small populations with a strong Allee effect, and large populations with a strong Allee effect.

<p>The magnitude of bias (<a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0110049#pone.0110049.e005" target="_blank">eq. 4</a>) can be seen in the distance of the estimate from the 0.0 line, with a bias >0.0 indicating that the magnitude of the parameter was underestimated and a bias >0.0, that it was overestimated, with the vertical bars representing error (<a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0110049#pone.0110049.e006" target="_blank">eq. 5</a>). Environmental stochasticity increases along the x-axis. Although in large populations bias (overestimation) was greatest with weak Allee effects, considering that the bias metric is normalized by its true value, in absolute numbers, overestimation could represent fewer individuals in the small Allee effect case than in the strong Allee effect one (see text for a numerical example and potential ramifications).