Dutch psychologists Meeus and Raaijmakers (1995) ran a series of studies into 'administrative violence' - the kind of harmful behaviour that might actually take place in an office or school. Like Milgram's study, Meeus and Raaijmakers used a confederate (or 'stooge') participant, who was here given the role of a job applicant, taking a test as part of his application. The real participant was given a list of 15 insults about the applicant's personality or performance, and encouraged to make these remarks by an experimenter who used 4 verbal 'prods' like those in the Milgram study. Participants were also led to believe that the applicant could be stressed out by the insults to the extent that he would fail the test, and if this happened he would not get the job, and would be unemployed.
Despite complaining that the remarks were unfair and being fully aware of the harmful consequences, 91% of participants made all 15 negative remarks. This is highly comparable to Milgram's findings - it appears that in a real-life context too, people will obey an authority figure and do harm to others for no good reason, and against their own conscience.
Participants later said that the giving of insults was the fault of the experimenter rather than themselves - according to Milgram's theory, they had entered an agentic state.
A key finding was that although the participants found the task very stressful, the stress was not enough to make them stop - and they also attempted to hide their stress from the victim.
References
Meeus, W.H.J. and Raaijmakers, Q.A.W. (1995). Obedience in modern society: The Utrecht studies. Journal of Social Issues, 51(3), 155-175.
Milgram, S. (1963). Behavioural study of obedience.
