Is religious affiliation a plausible grouping factor