Seminar 3 Flashcards
What is an ad hoc hypothesis? Mark all correct options.
a) An ad hoc hypothesis does not allow deriving any new testable consequences.
b) An ad hoc hypothesis makes it more difficult to falsify the main hypothesis.
c) An ad hoc hypothesis explains data that contradicts the main hypothesis. In this way the ad hoc hypothesis blocks the main hypothesis from being falsified.
d) An ad hoc hypothesis does not affect how hard it is to falsify the main hypothesis.
e) When you get a contradicting result you should propose an ad hoc hypothesis.
f) Since science aims to propose testable hypotheses, to say that a hypothesis is “ad hoc” is to say that it is a bad hypothesis.
Answer: a), b), c), f)
Which of the following statements are correct about the hypotheses in Neyman-Pearson hypothesis testing? Mark all correct options.
a) Test participants are always blinded about what the original hypothesis is.
b) If the p-value is lower than the significance level, we say that the test has rejected the original hypothesis.
c) The original hypothesis is always complemented by an alternative hypothesis.
Answer: b), c)
What is true regarding the p-value? Mark all correct options.
a) The p-value is the probability of observing a result at least as extreme as the one observed, given that the hypothesis is false.
b) The p-value is the probability of observing a result at least as extreme as the one observed, given that the hypothesis is true.
c) The p-value can exceed 1.
d) The p-value can be lower than the significance level.
Answer: b), d)
What is true regarding the use of statistical tools in experiments? Mark all correct options.
a) Using treatment and control groups is one way to get information on the sampling distribution.
b) One can perform meaningful experiments without making use of statistical tools.
c) The p-value of the data from an experiment is calculated by comparing the experimental result to the significance level.
d) One must always make use of statistical tools in order to produce meaningful results.
e) The p-value of the data from an experiment is calculated by comparing the experimental result to the sampling distribution.
Answer: a), b), e)
The Deductive-Nomological (DN) model of explanation does not (necessarily) take causal relations into account. Why might this make the DN model insufficient as an account of explanation? Mark all correct options.
a) All good scientific hypotheses are causal claims. Thus, since some DN explanations don’t account for the causal relations, the DN model is insufficient for scientific explanations.
b) Some generalizations express spurious correlations, but are nevertheless true. Therefore, using the DN-account, one can sometimes ‘explain’ phenomena by referencing completely irrelevant facts in conjunction with such generalizations.
c) Intuitively, we can explain a particular effect by identifying its cause in conjunction with true law-like generalizations, e.g. “The water in the pot is boiling because it was heated to 100 °C and water boils at 100 °C at sea level”. Given the symmetry of the DN account, the account also allows us to explain a cause with reference to an effect of the cause in conjunction with true law-like generalizations. This, however, seems too permissive.
d) The DN account erroneously predicts the following to be a genuine explanation: (1) Dan takes birth-control pills. (2) Dan is a male. (3: Law-like generalization): No male who takes birth-control pills becomes pregnant. (4: Explanandum): Dan has not become pregnant.
Answer: b), c), d)
Consider the following claim: “Negative stress has been found to be linked to extensive mobile phone use.” Which of the alternatives below are compatible with this claim? Mark all correct options.
a) Extensive phone use can be hypothesized to cause negative stress, or negative stress can be thought to bring about an increase in phone use.
b) This is a correlation. From this we cannot infer that any particular causal relation holds between the variables.
c) The cause of extensive mobile phone use is the extensive mobile phone use itself.
d) Negative stress and phone use have a common cause: the psychological phenomenon “fear of missing out”.
Answer: a), b), d)
What is true about the Duhem-Quine thesis? Mark all correct options.
a) According to the Duhem-Quine thesis, a seemingly falsifying observation might appear because our underlying scientific theory is incorrect.
b) The Duhem-Quine thesis is a way of distinguishing science from pseudo-science.
c) A hypothesis on the form ”our measurement instruments work as intended” is a common auxiliary hypothesis.
d) The Duhem-Quine thesis only targets scientists who use bad methodology.
Answer: a), c)
Feedback:
a) The Duhem-Quine thesis suggests that when an observation appears to contradict a scientific theory, it may not necessarily mean the theory is incorrect. Instead, it may be due to other factors, such as auxiliary hypotheses or the experimental setup.
c) In the context of the Duhem-Quine thesis, auxiliary hypotheses are often used to protect the core scientific theory from being easily falsified. The hypothesis that “our measurement instruments work as intended” is an example of a common auxiliary hypothesis.
