Social Science vs. the ACLU

Eugene Volokh raises the question of whether the ACLU has an anti-Christian bias -- specifically whether, as claimed by Clayton Cramer, they fail to defend free speech by Christians on the same terms that it defends non-Christian speech. Volokh makes two main arguments, both of which are expanded by others in the comments: 1) there are many reasons other than bias against the speech that an organization might decline to get involved in a free speech case, and 2) the ACLU has fought for Christians' free speech rights in a number of cases. In the comments, I suggested that a bit of social science would be necessary to really resolve the dispute:

I don't think listing cases is a very helpful way of looking at this -- it's just argument by anecdote. We need to determine the total number of free speech cases in each category that were filed, as well as the proportion of them that the ACLU participated in [update: Richard Aubrey points out that we should count as negative any cases where the ACLU is involved on the anti-speech side]. In other words, whether, over some specified time frame, the following is true: (number of cases involving non-Christian speech that the ACLU got involved in)/(total number of cases involving non-Christian speech) = (number of cases involving Christian speech that the ACLU got involved in)/(total number of cases involving Christian speech) Should there turn out to be a statistically significant difference between the two sides of the equation, that would constitute evidence of pro- or anti-Christian bias on the part of the ACLU (though one could [update: definitely should] then go back and refine the equation by factoring in other issues like the legal strength of the cases or the other resources available to the speaker's side).

I'd love to be able to give even some rough numbers as to the answer here, but as a non-lawyer I lack the resources and skills to even calculate my bare-bones equation, much less to incorporate the effect of all the mitigating factors that would be needed for a reliable answer. All I can do is mention my own suspcion, which is that the results would vindicate Volokh and the ACLU.