social psychology : bocchiaro Flashcards
Aim
To see how many people will comply with an unethical request and how many will respond by ‘whistle-blowing’ to a
higher authority. A secondary aim was to compare actual rates of disobedience and whistle-blowing to estimated rates.
method
149 students were given an unethical request, to write a statement designed to convince other students to participate in a
traumatic sensory deprivation experiment. Participants were then left alone to see what they would do. In a separate procedure 138 students were told about the scenario and asked both what they
would do, and what they thought the average student would do.
results
When questioned, most participants said that they would report the unethical experiment and that the average student
would disobey. However, 76.5% actually obeyed and only 9.4%
‘blew the whistle.’
CONTEXT
The authors were inspired by Milgram’s research into destructive obedience.
However, they noted that Milgram’s research focused on obedience and did
not tell us much at all about disobedience. One of their aims therefore was to
discover more about disobedience, specifically disobedience to unjust authority.
Procedures such as Milgram’s are generally discouraged now, on ethical grounds,
so it was not possible to replicate or extend his research. Instead a new way was
required of researching obedience in response to destructive instructions.
Aside from obeying or disobeying such instructions, there is a third option,
known as whistle-blowing, which involves reporting an unethical incident to
higher authorities. Whistle-blowing is a very challenging option because it
involves ‘taking on’ your immediate superiors, those who have authority over
you. The authors note that there is a lack of research into this option and point
out that very little is known about what whistle-blowers think or feel when they
make the decision to blow the whistle. Neither is much known about whether
whistle-blowers have particular psychological characteristics in common, such as personality.
There are various ways in which whistle-blowing could be investigated. One way
is to interview whistle-blowers after the fact. However, the authors rejected this
approach because participants may have had considerable time since the event
to draw their own conclusions about why they acted as they did. The authors
also rejected a scenario-based method of investigation, in which people are
asked how they would act in particular situations, on the basis that this has poor
validity unless the situation is a very familiar one.
Milgram did not just investigate obedient behaviour, he also researched
participants’ beliefs about how obedient they thought people would be when
given destructive orders. He found that people tended to wildly underestimate
rates of obedience. Bocchiaro et al. picked up on this idea. As well as carrying out
an experiment to see how many people obeyed, disobeyed and blew the whistle,
they also surveyed people to gather information about estimated responses.
It was predicted that participants would underestimate the tendency for
compliance with unethical instructions.
Bocchiaro et al. also picked up on the distinction between dispositional and
situational factors affecting obedience and compliance, and their participants
were assessed for dispositional characteristics, including personality and social
values.
conclusion
Although most people believe they will disobey
unethical instructions and report unethical conduct, in practice the
majority comply with unethical instructions.
AIM
The main aim of the study was to investigate the rates of obedience,
disobedience and whistle-blowing in a situation where no physical violence was
involved but where it was quite clear that the instructions were ethically wrong.
There were two additional aims:
1 To investigate the accuracy of people’s estimates of obedience, disobedience
and whistle-blowing in this situation.
2 To investigate the role of dispositional factors in obedience, disobedience
and whistle-blowing.
METHOD - pilot studies
Eight pilot studies, involving a total of 92 participants, were carried out in order
to be sure that the procedure was both credible (i.e. they didn’t realise that they
were being deceived), and ethically acceptable to participants. It was found that
from the participants’ perspective the procedure was both believable and ethical.
METHOD : Participants
All participants were undergraduate students from the VU University of
Amsterdam. Ninety-two people took part in the pilot studies, and a further 149
in the main experimental procedure (96 women and 53 men, with a mean age of
20.8 years). The sample consisted of volunteers recruited by flyers posted in the
university cafeteria. In addition, 138 different participants were surveyed about
how they believed they would respond in the experimental situation.
METHOD : design and procedure
The main study was carried out in laboratory conditions. Each participant was
paid seven euros or given course credits. They arrived alone and were met by
a stern experimenter who informed them that they were carrying out research
into sensory deprivation. They said that in their last study all participants
had panicked, and some had asked for the procedure to be stopped – which
it had not been. They aimed to carry out a similar study but were waiting for
ethical approval from the university. Each participant was instructed to write
a statement to convince other students to take part in the sensory deprivation
procedure. They were told that they had to use at least two words from the
choice of ‘exciting’, ‘incredible’, ‘great’ and ‘superb’, and not to mention the
negative effects of sensory deprivation. They were also offered regular paid work
in the future.
The participant was then left alone in a room with a computer on which to
compose their statement, a mailbox, and some ethics committee forms. If a
participant believed the study was unethical they had the option of completing
a form and putting it in the mailbox. Obedience/disobedience was assessed by
whether or not the participant composed the statement. Whistle-blowing was
assessed by whether they completed an ethics form and mailed it.
After seven minutes the experimenter returned and led the participant back to
the first room, where they were given a set of dispositional measures:
*The HEXACO-PI-R personality test, which measures six personality traits:
honesty-humility, emotionality, extraversion, agreeableness (niceness),
conscientiousness, and openness to experience.
*The Decomposed Games measure of social values, i.e. the extent to which
personal values are oriented towards benefit for all or just for the self.
*Religiosity was assessed by asking participants about their religious affiliation
(i.e. what their religion was), frequency of worship, and extent of faith.
Participants were then debriefed, with a particular emphasis on why they had
been deceived in the course of the study. They gave written consent for their data
to be used. The whole procedure took around 40 minutes.
Separately, 138 different participants were asked to estimate likely obedience
levels in this situation. They were provided with a detailed description of the
procedure and asked ‘what would you do?’ and ‘what would the average student
at your university do?’
RESULTS
There was a dramatic difference between the estimates of obedience gathered
in the second procedure, and the actual rates seen in the first, experimental
procedure. Only 3.6% of participants believed that they personally would obey,
and 64.5% believed that they would blow the whistle. The remaining 31.9%
believed that they would disobey. In relation to the average student estimates
were somewhat more accurate but obedience was still underestimated; 18.8%
were estimated to obey, 43.9% to disobey and 37.3% to blow the whistle. In fact
76.5% obeyed, 14.1% disobeyed and only 9.4% blew the whistle. Among the
whistle-blowers, the majority (6%) had obeyed by writing a message.
RESULTS : dispositional factors
None of the six personality traits assessed was associated with levels of
obedience, disobedience or whistle-blowing. Nor was social value orientation.
With regard to religiosity, neither religious affiliation nor frequency of worship
predicted behaviour. However there was a moderate relationship with depth of
faith, with those expressing strong religious faith being slightly more likely to
whistle-blow.
CONCLUSIONS
1 People are very obedient and whistle-blowing is uncommon.
2 People overestimate the tendency to blow the whistle and underestimate the
likelihood of obedience.
3 There is little or no evidence to suggest that dispositional factors affect
obedience or whistle-blowing.
4 On a theoretical level, results support the findings of previous research
showing that we tend to see ourselves as ‘special’ and rate ourselves as less
likely to follow destructive orders.
5 Results have implications for social psychology research. The inaccuracy
of estimates of behaviour in this situation suggests that all scenario-based
research lacks validity.
EVALUATION : The research method
The study was a laboratory procedure. Like Milgram, the
authors refer to it as an ‘experimental’ study, although it had
only one condition. Laboratory studies such as these have
particular strengths and limitations. Because the procedure
takes place in a highly controlled environment it is possible
to eliminate many extraneous variables and be reasonably
confident that it is the independent variable we are interested
in that is affecting the dependent variable. Laboratory
procedures are straightforward to replicate, making them
reliable. The potential weakness of laboratory studies lies in
the realism of the environment and the participants’ tasks. In
this case the procedure was very lifelike because the situation
being investigated was that of a psychologist carrying out a
study, and this is exactly what happened – there was nothing
artificial about the procedure.
EVALUATION : Qualitative and quantitative data
The data gathered in this study was quantitative, in the
form of percentages of participants displaying obedience,
disobedience and whistle-blowing behaviour. This was a
strength because the researchers were interested in making
comparisons between rates of disobedience and whistle-
blowing (by definition quantitative), and comparing these to
estimates. This requires the use of quantitative data.
EVALUATION : Ethical considerations
As a behavioural study of obedience, this study might have
raised some of the same ethical issues as Milgram’s study,
however the researchers were clever in designing the study so
as to minimise these problems. The situation was relatively
low in stress because participants were not ordered to inflict
direct harm, as Milgram’s participants were. Also, they were
left alone when deciding whether or not to obey, and were
not ‘prodded’. This is quite different from having to refuse
to obey an experimenter face-to-face while being told that
‘the experiment requires that you continue’. In addition,
extensive piloting was carried out to establish that participants
considered the procedure to be acceptable, and participants
had the opportunity to withdraw their data if they were not
satisfied with the ethical conduct of the study.
The design of the study necessarily involved deceit. This is
an ethical issue, however deceit can be acceptable in research,
provided that participants are informed of the true nature of
the study as soon as possible and that they are happy about
the study once they are aware of it. These conditions were met.
