Transcript Ethics in Experimental Research

Showing concern for the welfare of
human subjects






Tuskegee syphilis study of 1932
Stanley Milgram’s conformity research
of 1963
commercially funded, “for profit”
research (2001): conflicts of interest
and the “file drawer” problem
Pharmaceutical manufacturers
“outsourcing” drug trials to poor
countries with lax standards (2004)
U.S. energy Dept. and radiation tests
on civilians (1970’s)
LSD and the CIA’s MK Ultra program
(1973)



More than 100 boys living
in an orphanage were fed
Quaker Oats with
radioactive iron and
calcium in the 1950's.
The diet was part of an
experiment to prove that
the nutrients in Quaker
oatmeal travel throughout
the body.
A class action settlement
for $1.85 million was
reached in 1998



During and after
WWII, American
soldiers were forced
to observe nuclear
blasts within 50 miles
of ground zero.
Thousands of these
soldiers later died of
leukemia and other
rare forms of cancer.
Their families were
barred from suing the
federal government





“The Monster Study”
In 1938, Wendell Johnson and Mary
Tudor trained orphans to be more
conscious of small speech errors.
Johnson’s theory was that punishing
fluency errors made them worse.
All five stutterers in the test group
showed increased stuttering; five out of
six of the normal children exhibited
worse fluency.
The experiment, referred to by some as
the “Monster Experiment” turned
some of the children into lifelong
stutterers despite later efforts to
reverse the damage.


2005: South Korean
researcher, Woo Suk
Hwang, fabricated
evidence that he had
successfully cloned
human embryos.
The journal Science,
retracted two studies
he had published.




The price of bringing a new
drug to market is about $1
million per day
Much of that cost is devoted
to human clinical trials
western drug makers are
outsourcing safety and
efficacy studies to
developing countries, a large
proportion of them to India
and Russia.
There are currently some 400
clinical trials underway in
India
Good ends
Bad ends
Good means
Ethical
research
Bad means
Machiavellian
research
Subjective
ethic
(backfires)
Unethical
research

Autonomy:



Beneficence:


Free-choice, no pressure to participate,
consideration of “at risk” groups or individuals
persons with diminished autonomy
“do no harm,” ensure the well-being of participants
Justice:
fair distribution of risks and benefits of research
subject recruitment, selection
 subject compensation




“More than 1 percent of scientists report
direct knowledge of an instance of
misconduct.” Elliot (2000). How prevalent
is fraud? That’s a multi-million dollar
question. Science, 290, pp. 1662-1663
Motivations include
 tenure and promotion
 pressure to “publish or perish”
 lucrative grants, patents
 fame, notoriety, prestige

Intentional fraud


Cyril Burt’s research on
monozygotic twins—
fudging the results of IQ
tests
William Summerlin’s
cancer research—faked
results of tumor shrinkage

Unintentional fraud


Weitzman’s research on
women's and men's
incomes following
divorce—blamed
“computer error” for
erroneous results
Pons and Fleischmann's
research on “cold”
fusion—couldn’t be
replicated by other
researchers

minimizing
psychological risks


minimizing physical
risks


Example: simulations that
accentuate racist, sexist, or
homophobic attitudes
Example: behavioral
psychologists’ penchant
for shocking subjects in the
60’s and 70’s
showing concern for the
welfare of participants

Example: Stanley
Milgram’s conformity
research

Before conducting any research using
human participants, a participant’s
voluntary informed consent must first be
obtained:



Voluntary: the subject willingly agrees to participate in the
study, and is free to withdraw at any time without penalty
Informed: the subject is aware of any risks (physical or
psychological) associated with participating
Consent: the subject’s consent is unambiguous, e.g., a
signed permission form (no such things as “implied
consent”)
Low-risk anonymous surve
Observations gathered in
public places
Information in the public
domain


Kinch’s study on the “Pygmalion effect”
Problems associated with using freshman in
experimental research



Can students under 18 legally give their consent?
Should participation in experiments be a course
requirement?
Ethics of participant-observation

Going “under cover” to study groups may violate
their rate to privacy

Elms (1982) recommends the following strictures for
the use of deception in experimental research:
 As a last resort: When there is no other feasible way
to obtain the desired information
 example: studies on student cheating

When the benefits substantially outweigh the risks
 example: controlled double-blind studies on drug efficacy



When subjects are given the option to withdraw at
any time, without penalty
When any physical or psychological harm is
temporary
When subjects are debriefed and the research
procedures are made available for public review



Humphrey’s (1970) “tea
room” trade research
Personnel action—
failure to ensure
anonymity
Incest case—failure to
remove identifying
information


Anonymity: no one
including the
experimenter can
match the data to
specific individuals
Confidentiality: the
experimenter knows
participants’ identities
but takes steps to
protect participant’s
privacy.



at the department level
 Human Subjects Committees
at the university level:
 Institutional Review Boards (IRBs)
professional associations
 American Psychological Association’s
“Ethical Guidelines”
 Code of Ethics” of the American Speech
Hearing and Language Association

Dehoaxing:


Desensitizing:



undoing the cover story and revealing the true
purpose of the investigation
addressing any lingering psychological or emotional
concerns associated with participating in the
investigation
Explaining the benefits of participation to
subjects
Thanking subjects and providing for future
contact if necessary



the “subjects” versus
“participants” controversy
avoiding “isms” in research;
sexism, racism, ethnocentrism,
ageism, etc.
ethics of withholding
treatment from control groups