Jan 22: "Political Analysis will no longer report p values in regression tables or elsewhere. There are many reasons for this change—most notably that a p value alone does not give evidence in support of a given model or the associated hypotheses. "

Feb 5: "Important clarification to PA's recently updated p-value policy: in the case of design based experiments and related procedures (permutation tests, asymptotic approximations, etc.) p-values are appropriate and may be supplied."

Feb 8: "Continuing on the subject of p-value publication, it's worth clarifying that we certainly welcome studies where a compelling case is made that p-values are the correct evidence to present."

Are the new team out of their depth?