In Asch's (1951) classic conformity experiment an observer had to say which of three lines was equal in line to a standard. The lines were very easy to distinguish and if tested on their own observers made few if any errors. When they were tested in a group who gave their judgements publicly and all the rest of the group gave the wrong answer, three-quarters of the naive observers conformed at least once by responding incorrectly. Overall about one-third of the overall responses were conforming ones. As Eiser (1987) says 'for Asch, the important finding was that there was any conformity at all'. This study become a classic and is to be found in all texts on psychology. However, studies carried out from 1980 onwards have undermined this classic status - or at least preferred alternative explanations for the amount of conformity found.
Perrin and Spencer (1980) reproduced the Asch experiment with groups of 6 stooges and one genuine subject. But they found that the genuine subjects only once conformed out of 396 trials.
The Asch experiment is so well known toady that it is difficult to find 'naive' subjects so they used engineering, maths and chemistry students and, after the experiment, discarded the results of anyone who had heard of the original Asch study. This makes direct comparison with the Asch experiments difficult and yields to possible explanations:
Interpretation 1
'Science' students may be particularly inclined to see accurate line measurement as important and be less willing to be swayed by the group opinion. When Perrin and Spencer used young offenders on probation with people who were normally in authority over them acting as experimenters the results were similar to Asch.
Interpretation 2
During the 1950s conformity was viewed as sensible if not desirable and there were stronger pressures on students to conform than there are now. The high level of conformity found by Asch might be a 'child of its time'.
So far no one else has been able to separate these two variables and we do not know which interpretation is correct.