Any field study showing convincing evidence of group selection would be a significant contribution to the field of evolutionary biology. Pruitt and Goodnight1 claim to provide such evidence in a 14–18-month field experiment on spiders. However, we contend that apparent flaws in their predictions, assumptions, methods and interpretations undermine this claim. We believe that the data presented are unreliable and are equally consistent with both group selection and individual-level selection; thus, we question the conclusion of Pruitt and Goodnight1 that group selection has produced the observed patterns. There is a Reply to this Brief Communication Arising by Pruitt, J. N. & Goodnight, C. J. Nature 524, http://dx.doi.org/10.1038/nature14597 (2015).