What about this old suggestion?

FROM CHUCK TUCKER 941011

          The exchanges between Bruce Abbott (last one 941011.1020)
          Tom and Rick reminded me of something that I had noted
          in a earlier post. Doesn't this come quite close to the
          "proposal" that Bruce is making? What does anyone think
          of it?
                     Regards, Chuck

          STUDIES USING THE TEST (FR0M 940929)

          I believe there are more studies using THE TEST than
          we realize. I think the Milgram studies (including his
          "crowd" studies), the studies by Sherif (including the
          autokinetic one), a study by McPhail where the students
          "walked-out" of his class in the middle of a lecture and
          he continued to lecture to two student sitting in the
          front of the room, and the Asch study (with derivations)
          all used THE TEST. The problem is that they did not
          collect the data on a person by person basis and thus it
          is impossible to find out the results of THE TEST. I think
          most conventional experimental studies use THE TEST when
          they set up "experimental" and "control" groups (sic)
          but we don't which "treatment" is the disturbance. But
          if individual data were available in any of these studies
          my bet is that they could provide some data for PCT.

          I am collecting the data from a set of studies encouraged
          by Harold Garfinkel (who Rick might have known at UCLA)
          called "breeching" studies wherein disturbances were
          purposively set up (like Candid Camera) to find out about
          a person's "taken for granted world". Erving Goffman also
          mentioned some of these studies in his works and, in fact,
          he wrote a number of essays on disturbances (one was on
          bloopers and what they mean to self-control). I hope
          that these studies can be used for replications or at
          least a way to design other studies to use THE TEST at
          higher levels. (BTW Danny Miller mentioned the Goffman
          studies in his paper at the 1993 CSG meetings)