Kilham and mann obedience study 1974

Discussion 1 Conformity

Article

Would You Deliver an Electric Shock in 2015? Obedience in the Experimental Paradigm Developed by Stanley Milgram in the 50 Years Following the Original Studies

Dariusz Doliński 1 , Tomasz Grzyb

1 , Michał Folwarczny

1 , Patrycja Grzybała

1 ,

Karolina Krzyszycha 1 , Karolina Martynowska

1 , and Jakub Trojanowski

1

Abstract

In spite of the over 50 years which have passed since the original experiments conducted by Stanley Milgram on obedience, these experiments are still considered a turning point in our thinking about the role of the situation in human behavior. While ethical considerations prevent a full replication of the experiments from being prepared, a certain picture of the level of obedience of participants can be drawn using the procedure proposed by Burger. In our experiment, we have expanded it by controlling for the sex of participants and of the learner. The results achieved show a level of participants’ obedience toward instructions similarly high to that of the original Milgram studies. Results regarding the influence of the sex of participants and of the “learner,” as well as of personality characteristics, do not allow us to unequivocally accept or reject the hypotheses offered.

Keywords

conformity, obedience, social influence

Experiments conducted by Milgram (1963, 1965), in which the

study participant is encouraged by the experimenter to admin-

ister an electric shock to another person, are generally consid-

ered to be one of the most important (if not the most important)

in the field of social psychology (e.g., Benjamin & Simpson,

2009; Blass, 2004). The entire series of experiments carried out

by Milgram (1974) demonstrated that under conditions of pres-

sure from an authority, the majority of people will carry out his

commands even when they are informed at the beginning that

they have the right to end their participation in the experiment

at any time, while the information placed on the device used in

emitting electric shocks states unequivocally that it can damage

the health of the “learner,” or even kill him.

Following the publishing of Milgram’s work (1963, 1965),

there were discussions in the psychological literature concern-

ing the ethical aspect of such experiments (e.g., Fischer, 1968;

Kaufmann, 1967). While a few replication experiments were

carried out in the 1970s in various countries (e.g., Kilham &

Mann, 1974; Shanab & Yahya, 1978), further work within this

paradigm was then halted.

Naturally, an attempt was made at finding various alterna-

tives to direct replications of the original Milgram studies. For

example, Slater et al. (2006) conducted an experiment in which

the “electric shock” was administered not to a living human but

rather a computer-generated avatar. Participants in this experi-

ment were seated in front of a screen displaying a picture of a

woman (“the learner”) reacting in real time to electric shocks.

Another idea for creating an ethically acceptable procedure to

examine obedience was to assign unpleasant descriptors to rel-

atively pleasant images (Haslam, Reicher, & Birney, 2014).

The researchers prepared a series of 30 pictures sorted on the

basis of their attractiveness (beginning from the least pleasant

to the most pleasant). The participants’ task consisted in

selecting from among four negative adjectives the one which

best described a given image. It should be noted that while the

pictures became increasingly attractive as the study contin-

ued, the adjectives remained negative, which led to increasing

discomfort on the part of the participants. In the opinion of the

experiment’s designers, this procedure was to evoke a dis-

comfort similar to that experienced by participants in the orig-

inal Milgram studies. It should be noted that we may have

serious doubts regarding the extent to which this procedure

really reflects the realism of the Milgram experiments and

whether the impact of authority on obedience is what has

1 Faculty of Psychology in Wrocław, SWPS University of Social Sciences and

Humanities, Wrocław, Poland

Corresponding Author:

Tomasz Grzyb, Faculty of Psychology in Wrocław, SWPS University of Social

Sciences and Humanities, Ostrowskiego 30b, 53-238 Wrocław, Poland.

Email: tgrzyb@swps.edu.pl

Social Psychological and Personality Science 2017, Vol. 8(8) 927-933 ª The Author(s) 2017 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1948550617693060 journals.sagepub.com/home/spp

https://us.sagepub.com/en-us/journals-permissions
https://doi.org/10.1177/1948550617693060
http://journals.sagepub.com/home/spp
http://crossmark.crossref.org/dialog/?doi=10.1177%2F1948550617693060&domain=pdf&date_stamp=2017-03-14
essentially been tested here. One thing is certain: Since the

original experiments by Stanley Milgram, we have yet to find

a successful way of reconciling realism with care for the

well-being of study participants.

A few years ago, however, Burger (2009) noted that in the

original studies by Milgram, a decisive majority of people who

pressed the 10th button (33 people of 40) could then be con-

vinced to press all of the remaining ones (26 people of 33).

He thus arrived at the conclusion that conducting an experi-

ment in which participants would be encouraged only to press

10 successive (and not 30 successive) buttons would, on the

one hand, significantly reduce their level of discomfort, while

on the other it would allow for a direct comparison of obedi-

ence in pressing the 10th button and, indirectly, that is through

performing an appropriate estimate and comparison of total

obedience. Burger asked himself the question of what level

of obedience would be recorded in the United States almost a

half-century after the Milgram experiments. He thus replicated

Experiment No. 5 in which the alleged learner reports heart

problems at the beginning of the experiment, and before the

10th shock is administered he demands the halting of the

experiment, again reminding those present of his health prob-

lems. It turned out that 70% of Americans could be induced to press the 10th button, which led Burger (2009) to the conclu-

sion that “average Americans react to this laboratory situation

today much the way they did 45 years ago” (p. 9).

In our study, we decided to apply the empirical scheme of

Milgram (1974), Experiment 2 with Burger’s (2009) idea of

using only 10 buttons. It is worth emphasizing that Milgram

noted almost identical reactions by the participants in Experi-

ments No. 2 and No. 5 (the 10th button was pressed by 34 of

40 in Experiment 2, and 33 of 40 in Experiment 5, and while

button 30 was pressed by 25 participants in Experiment 2, and

26 in Experiment 5). Experiment 5 has been more frequently

replicated around the world than Experiment 2, but for the sole

reason that it is more spectacular and its results are more shock-

ing. From the perspective of estimating obedience levels, both

paradigms are, however, equally valid, while ethical considera-

tions argue for the choice of Experiment 2 in which people are

not encouraged to administer an electric shock to an individual

suffering from heart problems and who demands that his partic-

ipation in the experiment be concluded.

Our objective was to examine how high a level of obedience

we would encounter among residents of Poland. It should be

emphasized that tests in the Milgram paradigm have never been

conducted in Central Europe. The unique history of the coun-

tries in the region made the issue of obedience toward authority

seems exceptionally interesting to us. After World War II,

which began with Germany’s attack on Poland in 1939 and

concluded in 1945, the countries located in Eastern Europe

were made dependent on the Soviet Union, while the commu-

nist system was forced on them. One of the foundations of that

system was significant curbs on democracy and the demand of

strict obedience to authority. The official press used censorship

to develop an impression of the authorities’ infallibility and

moral legitimacy to ruling through the use of orders and

decrees. The primary and secondary school curricula also mar-

ginalized the role of such ideas as individual freedom and the

right to decide about one’s own affairs (Hodos, 1999; Naimark

& Gibianskii, 1997). However, the year 1989 marked a sea

change for the entire region. The understanding reached by the

communist authorities and anticommunist opposition initiated

a rapid series of changes across all of Eastern Europe. Free

press, democratic elections, and free speech became the norm

(Petersen, 2001; Rothschild and Wingfield, 2007). However,

in recent years we have observed a surge in the popularity of

the political party named “Law and Justice” [Pol.: “Prawo i

Sprawiedliwość”], which won an absolute majority in the last

parliamentary elections. In both the verbal arena and in its

actions, this party values governing with a strong hand rather

than freedom and democracy. Its efforts to limit the role of

democratic institutions and eliminate pluralism in the media

have met with extensive social approval (public opinion polls

show this party with support remaining stable at over 30%). This all means that both the historical experiences of Poles

and the current political situation may have a complicated

and opaque impact on obedience levels. By the same toke,

we felt it would be interesting to replicate the Milgram

experiment in this country.

Besides, we also took advantage of introducing a factor into

the experimental design that had previously never been tested

to a satisfactory extent.

Discussion of Milgram’s experiments in the psychological

literature generally oscillates around consternation at the uni-

versal nature of people’s pliability. For example, emphasis is

placed on the fact that the sex of participants in experiments

on obedience is not a factor that differentiates their reactions

(see Blass, 1991, for review). When considering the role of sex

in experiments carried out in the Milgram paradigm, we turned

our attention to something entirely different. However, before

we say what that was, let us take a look at three typical descrip-

tions of Milgram’s experiment that can be found in the psycho-

logical literature.

1. “Participants sat in front of an imposing shock genera-

tor and were instructed to administer an electric shocks

to the learner for each incorrect answer” (Burger, 2009,

p. 1).

2. “Who among us was not surprised and sobered to learn

that 65% of his subjects delivered the full series of painful and escalating shocks to an innocent partner?”

(Gilbert, 1981, p. 690).

3. “First, of course, is the unexpected enormity of the basic

findings themselves—that 65% of a sample of average American adult men were willing to punish another per-

son with increasingly higher voltages of electric shock”

(Blass, 1991, p. 398).

We have no doubt that the intention of these and other

authors writing about experiments on obedience toward author-

ity is not to present a false picture of reality, but it is worth

noting that the confederate is defined here using words which

928 Social Psychological and Personality Science 8(8)

are devoid of biological sex (learner, “partner,” and “person”);

meanwhile, in nearly all studies on obedience carried out under

the Milgram paradigm the learner who was allegedly being

electrified was a man.

Existing empirical data thus demonstrates that study partici-

pants are inclined to administer a shock to a man sitting behind

a wall. However, it is not clear whether the behavior of partici-

pants would change in conditions in which the experimenter

instructed them to give the shock to a woman. Why do we think

that the sex of the learner in experiments performed in the Mil-

gram paradigm may be significant?

Because women are physically weaker and more susceptible

to physical violence than men, in accordance with cultural

norms they should be treated more favorably and gently than

men (Anderson, 2000; Muller-Funk, 2012). This assumption

is supported in the results of meta-analyses of experiments

regarding altruism, which show that women receive assistance

from others more often than men (Eagly & Crowley, 1986;

Piliavin & Unger, 1985), as well as meta-analyses of experi-

ments concerning aggression, which show that it is more preva-

lent in conditions where its target is a man than when

aggression should be directed at a woman (Eagly & Steffen,

1986). It could be assumed that results will be similar in the

case of experiments carried out in the Milgram paradigm.

Administering an electric shock is an obvious violation of the

norm to refrain from harming an innocent person. Shocking a

woman with electricity, however, is also an infringement of the

norm to treat people with greater leniency who belong to vul-

nerable groups. It is therefore a more urgent violation of cul-

tural norms than shocking a man with electricity. We also

think that the sex of the learner will be of particular signifi-

cance when the participants are males. Traditional European

and North American norms (collectively “Western”) assume

that men are obliged to behave nobly toward women, and thus

to avoid causing them harm, both in word and in deed (Genov-

ese, 2000; Girouard, 1981).

On the other hand, in some milieus there are cultural

norms which hold that men should treat women as their

inferiors and require obedience and pliancy (Crowell &

Burges, 1996; Fontes & McCloskey, 2011). This, in turn,

would mean that male participants in an experiment con-

ducted within the Milgram paradigm would not have any

problem with administering an electric shock to a woman

who made a mistake in answering.

The issue of the learner’s sex as a determinant of obedience

is complex and warrants empirical study. Meanwhile, we are

aware of only three experiments in which the learner was a

woman. In the first one (Constanzo, 1976), sex of the learner

was manipulated. No evidence was obtained for the influence

of this factor on the level of obedience. However, this study

was never published, the experiment was conducted four

decades ago, the procedure employed was significantly modi-

fied from the original one as applied by Milgram, and the

cohort of participants consisted exclusively of university stu-

dents, which may also have known one another. In the study

by Shanab and Yahya (1977), the participant and the learner

were always of the same sex, which did not allow for a deter-

mination of whether the sex of the person to be zapped with

electricity influences the obedience of participants. In the study

by Slater et al. (2006), participants were asked to put on 3-D

stereo glasses and then instructed to administer a shock to a vir-

tual woman appearing on the screen. The absence in the study

design of conditions in which the learner was a virtual man also

prevented evaluation of the role played by the sex of the person

inflicting pain (albeit virtual) in the degree of obedience.

The issue of the role that the student’s sex may play remains

therefore an open question, requiring empirical exploration. In

our experiment, we decided to include 80 participants (40 of

each sex). For an experiment performed within the Milgram

paradigm, this is an exceptionally large number. However,

we are aware that it may also be too small for a definitive

understanding of the role played in obedience toward authority

by sex of the participant and the sex of the learner.

Procedure

Participants were offered Polish złoty (PLN) 50 (equivalent to around US$15) for about an hour’s time participating in psy-

chological research “dedicated to memory and learning.” They

were recruited in one of two ways. Some of them were

approached on the street, near the university. Others were

acquired with the help of students of the university, who

recruited participants from among their acquaintances who

were not students of that institution. Those eliminated from

the selection procedure were individuals who had taken a psy-

chology course as students, as well as those who responded to

a question about familiarity with psychological experiments

in a manner indicating they may have come across a descrip-

tion of the Milgram studies. People who had ever sought the

assistance of a psychiatrist or psychologist, who had experi-

enced some trauma, and those who had episodes of alcohol

or drug abuse in their history were all eliminated. The age

of participants ranged from 18 to 69 with M ¼ 27.36 (standard deviation ¼ 11.07).

After arriving to the psychological laboratory of the univer-

sity, participants and the confederate (the latter pretending to

be a participant) completed two or three questionnaires. 1

Sub-

sequently, the experimenter explained that the study would

address the impact of punishments on learning and memory

processes and required a division into the roles of learner and

teacher. He gave the participants PLN 50 for their participation

in the experiment, and then requested that they draw lots for

their role by selecting one of two pieces of paper. Each of them

contained the word “teacher,” but the confederate (a man or a

woman depending on the experimental conditions) announced

that he or she had selected the paper with the word learner. The

experimenter asked the participants to sign an informed con-

sent form for participation in the experiment. The form stated

that the participant could interrupt participation in the study

at any moment. The experimenter additionally emphasized that

any decision to do so would not require the return of the com-

pensation paid for participation. The trio then proceeded to the

Doliński et al. 929

laboratory, where the experimenter showed the generator

(identical in appearance to the one used by Milgram) and

explained that the learner’s role would consist in learning by

heart associations between certain syllables, and then handed

the learner a piece of paper containing eight pairs of syllables

while leading him or her into a neighboring room (because

we did not have access to the original list of words used by Mil-

gram, we decided that the best approach would be to use sylla-

bles, as they are neutral in content). He then returned, handed

the teacher a list of 45 pairs of syllables written in a different

order than those on the paper received by the learner, and then

explained that the teacher’s task was to read one syllable and

wait for the learner’s response. If the response was correct, the

next syllable was to be read. If incorrect, the teacher was to

wait for the experimenter’s instructions. The experimenter also

demonstrated the device’s functioning, showing the teacher

that pressing particular levers would activate a shock of elec-

tricity measuring from 15 V to 450 V (a picture of the replica

of Milgram’s device as used in the experiment is located in the

Online Appendix). To demonstrate the functioning of the

device, the experimenter administered a shock of 45 V. Next,

the experimenter attached electrodes to the learner’s wrist in

the presence of the participant. The experimenter asked the

study participant to sit at the generator, and then himself sat

approximately 3 m from the participant. He then instructed the

participant to read the first syllable. The learner responded cor-

rectly, and the participant then read the following syllable. The

learner did not make a mistake in this and in the following

attempts, until the seventh syllable, when he erred. The experi-

menter instructed the teacher to press the first lever. Next, the

learner gave a few proper responses but then made a mistake

with the 10th syllable. The experimenter instructed the teacher

to press Lever No. 2. The next mistake occurred with Syllable

No. 13 and led to the instruction to press the third lever. Suc-

cessive impulses of electricity (following successive mistakes

that occurred at numbers 15, 18, 22, 24, 27, 31, and 34) resulted

in screams of increasing pain from the learner. These screams

were recorded and played back at appropriate moments. If the

teacher vacillated, the experimenter exhorted him or her using

similar prompts to those applied by Milgram (1974): “Please

continue,” “The experiment requires that you continue,” “It is

absolutely essential that you continue,” and “You have no other

choice, you must go on.” A note was made of the moment when

the participant refused further participation in the experiment, as

well as whether any doubts were expressed that required the

experimenter to deploy one of the aforementioned messages.

Just after the participant pressed the 10th button (or refused

to continue the experiment), the experimenter asked the ques-

tion “Do you think it hurts?” This was a way of making sure

that the participants were aware of the real pain being adminis-

tered to the person in the neighboring room. All the partici-

pants, save for one man who expressed doubt as to whether

the person sitting behind the wall was being shocked by elec-

tricity, responded in the affirmative way. The individual who

spoke of doubts was removed from the pool of results and

replaced by another man.

The experiment was conducted by one of two experimenters

(either a male in his mid-30s or one in his mid-50s). The role of

the learner was assigned to one of two women in their mid-20s

or one of two men in their mid-20s.

The experiment was conducted after receiving approval

from the departmental ethics commission (Decision No.

2014/E/02/2), which, following a thorough analysis and con-

sideration of the benefits that could result from the study, gave

its approval and oversaw how it was conducted. An important

element in the procedure was the detailed and painstaking

debriefing held individually with each participant following the

conclusion of the experiment. During this debriefing, partici-

pants were told of the details of the procedure, apologized for

being deceived at the start of the experiment as to its objectives

and course, and they received an explanation of why it was

done in that way. Each conversation was conducted by a qual-

ified clinical psychologist and lasted from several to several

dozen minutes. Participants were also informed that if they had

any further questions or concerns about the course of the study,

particularly if they felt any discomfort about their own partic-

ipation, that they should get in contact using a special telephone

number provided to them.

Results

Because initial analyses demonstrated that neither the manner

in which participants were recruited nor the person of the

experimenter, the female learner, or the male learner had any

impact on the structure of results, these factors were not taken

into account in further analysis. Dominant majority of the par-

ticipants pressed the 10th (the last in this variant of Milgram

experiment) lever. Exact number of participants who finished

on particular levers is shown in Figure 1.

The overall sample size is 80, and the observed proportion

of participants who pressed the 10th button is 90% (this is also

Figure 1. Number of participants who finished withdrawal from experiment on particular levers.

930 Social Psychological and Personality Science 8(8)

the effect size). The 95% confidence interval (CI) is from 83.43% to 96.57%.

We also examined the impact of the learner’s sex on obedi-

ence. Results are displayed in Figure 2. It is worth remarking

that although the number of people refusing to carry out the

commands of the experimenter was 3 times greater when the

student was a woman, the small sample size does not allow

us to draw excessively far-reaching conclusions. (This result

was not statistically significant, Wald w2 ¼ .341, df ¼ 1, p ¼ .559, Cohen’s d ¼ .13.)

Because of the very low percentage of people resigning

from further participation in the study, we decided to also ana-

lyze the doubts raised by participants during the course of the

experiment. In Table 1, we have correlated information about

sex, age, and the moment of withdrawal (or expression of

doubt) of each person who did not demonstrate total obedience

toward the experimenter’s instructions.

Discussion

It is exceptionally interesting that in spite of the many years

which have passed since the original Milgram experiments, the

proportion of people submitting themselves to the authority of

the experimenter remains very high. The result of 90% obedi- ence which we have achieved, 95% CI [83.43%, 96.57%], is very close to the number of people pressing the 10th button

in the original Milgram studies. For example, in Milgram’s

(1974) Experiment No. 2, replicated in our study, 34 of 40 peo-

ple pressed Button No. 10 (85% of participants, the 95% CI extends from 70.54% to 93.32%).

In the Milgram procedure, participant is issued with unam-

biguous orders from a person who is an authority, who leaves

no room for freedom of decision, does not suggest taking time

to think about reactions, or to select from among the options

available. In our experiment, participants demonstrated such

total obedience that we achieved a ceiling effect, making it

exceptionally difficult to demonstrate the influence of any

moderators of the dependent variable. From a certain perspec-

tive, it is worth drawing attention to the interesting proportion

of refusals to continue the experiment in the case of differences

in the learner’s sex. When it was a woman being “zapped,” par-

ticipants were 3 times more likely to withdraw from the experi-

ment (regardless of their own sex). However, the fact that only

10% of our participants failed to perform all of the experimen- ter’s commands means that this difference is far from statisti-

cally significant.

Our results can thus not serve as grounds for definitive con-

clusions on the role of learner sex in the experiment—with all

certainty the results allow for the declaration neither that such

an impact is present nor that it is not present. However, in our

view the results are worth noting and may provide inspiration

for further studies in the paradigm.

That said, we are forced to admit that we did not confirm the

hypothesis that the sex of the person being shocked with elec-

tricity would influence the level of obedience displayed by par-

ticipants. Our search for factors differentiating the behaviors of

participants in the Milgram paradigm is consistent with the

long tradition of such studies (some of which have been

described in earlier fragments of this article). Searches have

also been conducted for the sources of obedience (apart from

“agentic state”) in the experimental situation itself (e.g., Col-

lins & Brief, 1995; Gilbert, 1981; Lutsky, 1995). However, it

should be remarked that the search for such mediating variables

generally concludes with the admission that the original

explanations proposed by Milgram are difficult to refute,

Figure 2. Sex of the “learner” and obedience.

Table 1. Information About Participants Expressing Doubts in the Course of the Study.

Sex Age Number of Prompts Switch Number Decision

Male 35 4 5 Stop Female 58 4 5 Stop Female 21 4 6 Stop Male 26 4 7 Stop Female 24 4 7 Stop Female 26 4 8 Stop Female 44 4 9 Stop Female 25 4 9 Stop Male 19 1 8 Continue Male 35 1 6 Continue Male 26 1 9 Continue Male 23 2 5, 9 Continue Male 20 2 6 Continue Female 25 1 9 Continue Female 26 3 5 Continue Female 33 1 8 Continue Female 23 1 6 Continue Female 24 1 9 Continue Female 21 2 9 Continue Female 20 1 6 Continue Female 23 2 8 Continue

Doliński et al. 931

and—significantly—relatively stable over time. An exception

to this rule can be found in the studies of Reicher, Haslam, and

Miller (2014), indicating that participants in studies on

obedience can be motivated rather by appeals to science than

by orders. This is, however, only a more precise labeling of

the reason why participants carry out the commands of the

experimenter–scientist. In other words, we may expect that

contemporary replication experiments on obedience will also

refer in their explanations to agentic state as the primary

mechanism for explaining the behavior of study participants.

It would seem that the results of our experiment also provide

indirect support for this explanation.

In summary, it can be said that such a high level of

obedience among participants, very similar to that attained in

the 1960s in the original Milgram studies, is exceptionally

fascinating. Elms (1995) wrote that Milgram told his students

to ask important research questions and to gather data which

would be interesting even after 100 years had passed. Over

50 years have passed since the original Milgram experiments,

and it seems today we are headed in the right direction to con-

tinue in the next half-century seeking the sources of obedience

and compliance among study participants.

Acknowledgments

This research is supported by the BST research Grant No. 25/16/2015.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to

the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, author-

ship, and/or publication of this article.

Supplemental Material

The supplemental material is available in the online version of

the article.

Note

1. While conducting this study, we also planned to examine the role

played by individual factors. We took the following variables into

account: (1) Rotter’s (1966) locus of control because previous

study results on the role of that factor as a determinant of obedience

in the Milgram paradigm are inconsistent and inconclusive (see

Blass, 1991), (2) the role of empathy, which, while from the theo-

retical perspective would seem a rather obvious “candidate” for the

role of moderator of obedience, has only been directly examined—

to the best of our knowledge—by Burger (2009) who demonstrated

that empathy influences the verbal expression of doubt by partici-

pants during the experiment but did not show any link with the

actual level of obedience, and (3) only in respect of men—accep-

tance of the norms of the culture of honor (Cohen, Nisbett, Bowdle,

& Schwartz, 1996; Nisbett & Cohen, 1996). We expected that men

who particularly strongly accept the rules of the culture of honor

would demonstrate very low rates of compliance in conditions

where the experimenter instructs them to zap a woman with

electricity. Because the results we achieved were inconclusive, and

this issue is not of fundamental importance to the main subject of

the article, we present both the scales applied and results on the

links between those personality characteristics and obedience in the

Online Appendix.

References

Anderson, M. L. (2000). Thinking about women: Sociological

perspectives in sex and gender. Boston, MA: Allyn and Bacon.

Benjamin, L. T., & Simpson, J. A. (2009). The power of the situation:

The impact of Milgram’s obedience studies on personality and

social psychology. American Psychologist, 64, 12–19. doi:10.

1037/a0014077

Blass, T. (1991). Understanding behavior in the Milgram obedience

experiment: The role of personality, situations, and their interac-

tions. Journal of Personality and Social Psychology, 60, 398–413.

Blass, T. (2004). The man who shocked the world: The life and legacy

of Stanley Milgram. New York, NY: Basic Books.

Burger, J. (2009). Replicating Milgram: Would people still obey

today? American Psychologist, 64, 1–11. doi:10.1037/a0010932

Cohen, D., Nisbett, R. E., Bowdle, B. F., & Schwartz, N. (1996).

Insult, aggression, and southern culture of honor: An

“experimental ethnography.” Journal of Personality and Social

Psychology, 70, 945–960.