CPS questions (1-6)
Intrinsic Difficulties
1-2. These questions ask for the
course theme(s) best illustrated by the given statement. Do not assume any
more than what is explicitly given in the question. MTF
|
(A) Rare events are difficult
to quantify |
(B) Time lags slow progress |
(C) Complexity (interactions) |
(D) Humans make difficult
experimental subjects |
(E) None |
1. (4pts) The
daughters of women who took DES during pregnancy acquired cancer more often
than did the daughters of women who did not take this drug during pregnancy.
However, these cancer cases generally did not appear until these offspring were
twenty to thirty years old. Consequently, DES -- a harmful drug -- was used for
several decades before the harmful effect was discovered.
(A) (B) (C) (D) (E)
2. (4pts) Lifespan shows numerous correlations with
diet and health, but it is difficult to identify cause and effect. First, it is difficult or impossible to do
the necessary experiments with humans because of cost and compliance
issues. Second, it may take decades to
observe the effect of someone’s diet today.
And there are potentially hundreds of problems caused by the fact that
the effect of one item in a diet (e.g., fat content) depends on what else is in
the diet. Thus, one should not hope for
a speedy understanding of what kinds of foods provide the greatest longevity.
(A) (B) (C) (D) (E)
3. (6 pts) Which of the following
options describe(s) an outcome that stems from a "tragedy of the
commons" conflict or correctly identifies a tragedy of the commons pattern? You may assume that what is
stated in each option is true; the question is whether it describes a tragedy
of the commons phenomenon. (MTF)
(A) Three unrelated people buy a ranch
together. They calculate that, with the
current market prices, their best plan is to graze more than the recommended
number of cattle on their land, sell the beef for a profit, and invest the
money in bonds. This plan fails because
the price of beef falls before their beef get sold. In future years, their
overgrazed land only supports half the number of livestock that it would have
if they had not overgrazed.
(B) Physicians often face a conflict between what is
the best treatment for the patient and what treatment will most benefit the
physician. This conflict arises because
drug companies entice physicians to sell company products, even when those
products are not the best treatments available.
(C) Some diseases for which we have vaccines, such
as influenza and measles, are caused by viruses that are transmitted from
person to person. The number of people
exposed to these viruses depends on how many people get vaccinated. As more people get vaccinated, the rate at
which unvaccinated people get the disease declines – herd immunity. People who, for their own personal reasons,
refuse to get vaccinated for these diseases increase the risk that others in
the population (who are not vaccinated) will get the diseases.
(D) Vaccines cause us to develop immunity against
particular toxins, viruses, or bacteria.
Some diseases for which we have vaccines, such as tetanus, are caused by
bacteria that come from the soil and are not transmitted from person to
person. The number of people exposed to
the tetanus bacterium thus is not affected by how many people get vaccinated. People who, for their own personal reasons,
refuse to get vaccinated for tetanus thus pose no increased risk of tetanus for
others in the population.
4. (5 pts) Mark all statements that
correctly summarize the information shown in the second part of Prisoners of
Silence video. MTF
(A) The Syracuse University institute dedicated to
facilitated communication would be harmed (perhaps dissolved) if facilitated
communication were shown to have no factual basis. Douglas Bicklen's reluctance
to accept the results of the tests of FC (his bias against accepting those
results) can be understood from the fact that he would lose a prestigious
administrative appointment/position if he agreed with the refutation of FC.
(B) Douglas Bicklen responded to the data showing
that facilitated communication does not work by stating that many autistic
individuals exhibit severe word-finding difficulties, and that this is the
reason they did not perform well in the experimental tests. This is an example
of the use of post hoc arguments (as a form of bias).
5. (5 pts) At what stages can a study or its conclusions be deliberately
biased? (MTF)
(A) Before
the experiment is designed (e.g., in deciding what models to test).
(B)
In designing the study.
(C)
In the evaluation stage.
6. (6 pts) Which of the following constitute possible bias in experimental design
rather than at other stages of the process?
MTF
(A) Searching for a
statistical test to support the desired conclusion
(B) pseudo blind
(C) assay for a narrow
spectrum of unlikely results
(D) Appeal to authority
(E) Refusal to admit error
(F) small samples
(G) change design in
mid-course
(H) Throw out unwanted
results
(I) character assassination
of opponent
Intrinsic Difficulties
7-10. These questions ask for the
course theme(s) illustrated by the given statement. Do not assume any more
than what is explicitly given in the question. That is, address only the
difficulties specifically mentioned.
There are fewer options here than on some practice questions. (One or
two answers)
|
(A) Rare events are difficult
to quantify |
(B) Time lags slow progress |
(C) Complexity (interactions) |
(D) Humans make difficult
experimental subjects |
(E) None |
7 (4 pts).
The
following is a verbatim quote from a FRONTLINE press release for the show
“Dangerous Prescription” aired 13 November, 2003 (http://www.pbs.org/wgbh/pages/frontline/press/2204.html). Which, if any, intrinsic difficulty is described in this quote?
"When a drug goes on the market, only about 3,000 patients have ever been given that drug," says Woosley, who directs a national center that studies drug side effects. "We will never know all the toxicity that can occur, especially the one [patient] in 10,000 or one in 20,000 that could be seriously harmed. Our detection will only happen after the drug is on the market and exposed to a huge number of patients."
(A) (B) (C) (D) (E)
8. (3pts) It was mentioned in class that, in the last
decade, a phase I clinical trial testing a hepatitis drug led to 5 patients
dead, a large fraction of those in the trial.
While it may have been unavoidable that at least one patient died before
the problem with the drug was discovered, the reason that so many died was that
the drug caused irreversible liver damage, the fatal effect of which was not
manifested until months after a lethal dose had been taken. Others in the trial were alive but were some
were beyond saving when the first death occurred. This trial also created a longer-term problem for government
approval of other drugs: the
unexpectedly high death rate in this trial slowed the FDA approval process for
other clinical trials, to avoid similar outcomes. This latter problem is obviously something that would not impede
progress if the experiments had been conducted with lab mice or bacteria. Which problems are indicated in the cause of
so many deaths and in the impact of that trail on the larger realm of drug
approval?
(A) (B) (C) (D) (E)
9. (3pts) One of the main themes for
this segment of the semester is that different people often have different
goals. In one specific example, we saw
in the Prisoners of Silence video that Doug Biklen had a strong incentive to
deny the scientific evidence against the validity of FC. This problem of conflicting goals underlies
many other cases in which scientific evidence is denied as well.
(A)
(B) (C) (D) (E)
10. (3pts) In the last 20 years,
experimental physicists created materials that have superconducting electrical
properties – at very cold temperatures, the materials have virtually no
resistance to electricity. The
discovery of superconductors was largely empirical (not based on prediction,
but rather by trial and error) because the superconducting materials consist of
specific combinations of unusual metals and other elements, whose collective
properties of superconductivity could not be anticipated from the properties of
the elements alone. Which difficulties
(if any) underlie this reason that it took so long for scientists to develop
superconductors?
(A)
(B) (C) (D) (E)
11. (4pts) People
often have goals other than finding the truth.
Which of the following options about conflicting goals are correct?
(MTF)
(A)
The different types of conflict discussed in class
were classified as different forms of “Tragedy of the commons”
(B)
From lecture,
the different factors affecting goals included emotional factors, material
gain, and ideology.
(C)
Physician’s rejection of Semmelweis’s discovery of
the cause of high mortality rates in maternity wards was considered an example
of emotional factors overriding a search for truth.
12. (4 pts) Which of the following
options applies to “tragedy of the commons” conflicts? (MTF)
(A)
They
are characterized by time lags, rare events, and/or interactions (complexity)
(B)
They
must include group ownership of a common resource or some other kind of group
benefit
(C)
Selfish
interests of the group members are in conflict with the common good of the
group
(D)
They
stem from biases at either the design or evaluation stage of a study.
(E)
The
conflicts that led to disagreement about the interpretation of FC experiments
in the 2nd part of “Prisoners of Silence” video were types of
tragedy of the commons conflicts.
13. (4 pts). The money game.
Which of the following options apply to the "money game "
demonstration that used strips of paper? (MTF)
(A) A
conflict existed between individual and group (class) benefit.
(B) The conflict depended on the rules of the
game. A different set of rules could
have eliminated the conflict.
(C)The outcome (this year and in previous years) was
that the self interests of many students prevented anyone from receiving any
money.
(D) The choice that every individual could make
which allowed group benefit was the $1 option.
14. (3 pts) The Notes (and lecture)
suggested that there was a fundamental difference between science and the legal
system over the way that (scientific) evidence was evaluated. What difference
was that? (One answer)
(A) Most scientists have the common goal of finding
the truth, so there is little conflict among scientists; in contrast, a trial
has conflict built into it by the nature of opposing goals between prosecution
and defense.
(B) Scientists accept ambiguity/uncertainty as part
of the nature of evidence, whereas a trial does not (for the most part) allow
ambiguity/uncertainty as an outcome -- a decision has to be made about guilt or
innocence, regardless of whether the evidence is ambiguous.
(C) Scientific data are always obtained outside of
the courtroom, so there is an inherently unsympathetic audience for data in the
legal system.
(D) None of the three major participants in a trial
-- judge, prosecuting attorney, or defense attorney -- has been trained in
scientific reasoning.
Deliberate Bias
15. (3 pts) Mark all statements that correctly
summarize the information shown in the second part of Prisoners of Silence
video (MTF)
(A) Douglas
Biklen (the guy with the beard) defended FC by saying that it didn’t matter how
many times studies failed to support it, that the important goal was to keep
testing to find evidence that it worked.
(B) Claiming that the number of tests failing to
support a model is unimportant in deciding whether to accept that model is an
example of “appeal to authority”
(C) From the viewpoints presented in the video (and
as touched on in lecture), it seems likely that the disagreement/conflict over
the validity of FC would disappear if more-conclusive, double-blind tests were
conducted, and if those tests also showed that FC was not real.
16. (6 pts) Mark all of the following
statements that are correct about deliberate bias, as covered in lecture or the
Notes (MTF):
(A)
Much
of scientific progress comes from independent thinkers. If a model that goes against the established
views/models turns out to be right, it can have a large impact. Thus, new ideas that go against the
established views should be considered more likely to be correct than ideas
that fit within established views.
(B)
It
is possible to bias the presentation of a study (e.g., to a jury) even if the
design, conduct, and analysis were carried out in an unbiased fashion.
(C)
A
major step in identifying or anticipating deliberate bias is to identify
conflicts of interest.
(D)
The published protocol for a study provides
the essential details for how the study was conducted and gives sufficient
information to detect whatever bias might be present in the design or
evaluation stages.
(E)
All
scientific models are false.
Consequently, it is a sure indicator of bias if someone claims that
their model is supported by a full analysis of the data.
(F)
A short article by E. Loftus was read to the class about her
experiences in researching a case of repressed memory (her name is unimportant;
the content of the article is). The point
of that article was that researchers can be harassed and prevented from
undertaking research by litigation, even though there is no foundation for the
litigation.
17.
(5 pts). Consider the demonstration conducted during
the last lecture (Monday), in which the class generated data on the probability
of an outcome under different experimental protocols. What was the purpose of this demonstration, what parallels apply,
and what were the details of this demonstration? Note that although this
question is the same as on the 2002 exam, the options (and what constitutes a
correct answer) are based on the demonstration in class this year (MTF)
(A)
The
demonstration used coin flips done by students.
(B)
The
demonstration illustrated how a drug company could bias the evaluation of
product safety/efficacy specifically by terminating early those trials in which
the initial results look bad and publishing only the results of completed
studies.
(C)
The
demonstration illustrated that the choice of the null model is not
important to the conclusion reached.
(D)
The
demonstration relied on the existence of sampling error. If no sampling error was present, the
desired effect could not have been demonstrated.
(E)
The
demonstration illustrated a type of bias that would usually occur at the end of
the study, after all data had been obtained.
18. (5 pts). Which apply to the biasing
of a scientific process by "controlling the null model?" MTF
(A)
A
null model is accepted until evidence actually contradicts it. The null model – and whatever goal it favors
– thus receives the “benefit of doubt” for a wide range of results that don’t
necessarily support it, simply because those results are not strong enough to
refute it.
(B)
The
step of controlling the null model can precede the study design and evaluation.
(C)
The
bias in controlling the null model arises because of the way the control is
chosen.
(D) The bias in controlling the null model can operate only if some other aspect of design or evaluation is also biased.
19-21. For each of the following,
mark all bogus experimental designs illustrated. Choose the best one answer
per question
|
(A) change design in mid-course |
(C) (true) protocol concealed |
(E) assay for a narrow spectrum
of unlikely results |
|
(B) non-random assignments |
(D) small samples |
(F) None |
19. (4 pts.) Drug companies have sponsored Continuing Medical Education classes, but invited as speakers mostly those doctors who advocate using drugs manufactured by the sponsoring company.
(A) (B) (C) (D) (E) (F)
20. (4 pts.) Pharmaceutical manufacturers have been known to fund research by non-company scientists, but stop funding these research projects when preliminary results indicate that the final experimental results will not be favorable to the pharmaceutical manufacturer.
(A) (B) (C) (D) (E) (F)
21. (4 pts.) Although the protocol of a study is written as “blind,” participants are quietly told that the real medication is bitter but that the placebo is sweet. The participants are thus aware of whether they get the real medicine or placebo.
(A) (B) (C) (D) (E) (F)
22-24
(x pts each). Mark the type of nonscientific argument (blatant bias)
illustrated in each of the following examples. Base your answer on the
information given in the question. One answer per question.
|
(A) Claim that all alternatives have not been
disproved |
(D) Build causation from
correlation |
(G) No bias is indicated |
|
(B) Identify trivial flaws
in opponent’s model |
(E)
Refusal to admit error |
|
|
(C) Draw on anecdotes or
post hoc observations |
(F)
Either-or arguments |
|
22. (4 pts) Employment at Home
Depot requires that the applicant pass
a drug test. After reading the
employment conditions, Ralph Wrench submitted to a drug test and filled out the
application form. The test result was
positive for at least one illegal substance, and he was therefore not
hired. Which type(s) of nonscientific
argument underlie(s) Home Depot’s refusal to hire him?
(A) (B) (C) (D) (E) (F) (G)
23. (4 pts) In defending the safety of their product, tobacco companies have pointed out that many non-smokers get lung cancer, so the fact that some smokers also get lung cancer does not implicate smoking as the cause of their lung cancer. What type of bias (non-scientific argument) is indicated by this defense that some non-smokers get lung cancer too?
(A) (B) (C) (D) (E) (F) (G)
24. (4 pts) Lawyers, when questioning an expert for the opposing side in a court case, often look for minor weakness in the expert's arguments (if they can't find major ones). These weaknesses might be nothing more than the standard "errors" that underlie all data or the falsity inherent in all models. The lawyer then inflates the significance of these weaknesses and blows them out of proportion in an attempt to discredit the testimony. What type of blatant bias is indicated by this inflation of minor weaknesses?
(A) (B) (C) (D) (E) (F) (G)
25. (4pts) Key code, pad #, and name. Fill in (A B) to indicate your key for this version of the exam. Be sure your name and pad number are correctly bubbled in on the scantron and that you have signed this exam form.