Abstract
<p>Aker et al. concluded that birth experiences, measured by self-rated degree of traumatic child birth and birth satisfaction, had an impact on postpartum depressive symptoms and that this effect was mediated by rumination. Here, we fitted alternative models to the same covariance structure and these models suggested, instead, that associations could be due to a negative effect of a latent general vulnerability factor on birth satisfaction. This is an example of a common situation where competing conclusions are underdetermined by the available data. We recommend researchers to fit alternative models to their data, compare findings, and to think more deeply about what conclusions the data really can be claimed to support.</p>