Scientific Inferences Flashcards
There are two main forms of inferences: deductive and inductive. Mark all correct descriptions that apply to inductive inferences.
a) Inductive inferences are such that if they are valid, the conclusion must be true if the premises are true.
b) Inductive inferences are typically based on unfounded assumptions.
c) Inductive inferences rearrange current knowledge in such a way that they merely explicate what we already know.
d) Inductive inferences go beyond information we already have, thus they amplify our knowledge when they are used.
Answer: d)
There are two main forms of inferences: deductive and inductive. Mark all correct descriptions that apply to deductive inferences.
a) Deductive inferences go beyond information we already have, thus they amplify our knowledge when they are used.
b) In valid deductive inferences, the conclusion must be true if the premises are true.
c) Deductive inferences rearrange current knowledge in such a way that they explicate what we already know, without adding new knowledge.
d) Deductive inferences are typically based on unfounded assumptions.
Answer: b), c)
“Hume’s problem of induction” is an argument saying that inductive inferences cannot be justified. What are its premises? Mark all correct options.
a) That a particular inductive inference rule cannot be justified using deduction.
b) Inductive inference rules also require justification, and some kind of inference is needed to do this.
c) That inductive inferences are unscientific and should be avoided.
d) That there are many kinds of inferences and that inductive arguments are justified by currently unknown inference types.
e) That a particular inference rule must be justified using induction and another inductive inference rule, which in turn also needs inductive justification and so on.
f) That every inference is either inductive or deductive.
Answer: a), b), e), f)
Why is Hume’s problem of induction problematic for the justification of inductive inferences, and thus an important challenge for science? Mark all correct options.
a) Because if Hume is correct, inductive inference rules would be unreliable and science would have to rely on deductive inference rules, which are often not as valid as inductive inference rules.
b) Because if Hume is correct, all inductive inferences give necessarily true conclusions which means they are really deductive. Deductive inferences cannot be used in science.
c) Because if Hume is correct, the use of inductive inferences rules can only be justified by using other inductive inference rules, which is circular. Science uses a lot of induction and would therefore often not be justified.
Answer: c)
Mark all true claims about the Hypothetico-Deductive Method (HDM).
a) HDM is a scientific ideal that is rarely used in practice.
b) HDM is a method for testing hypotheses.
c ) HDM uses both deduction and induction.
d) HDM is based on deduction and not on induction. Therefore, it circumvents Hume’s problem of induction.
Answer: b), c)
Which of these are main components of the Hypothetico-Deductive Method (HDM)? Mark all correct options.
a) Deduce empirical consequences from the hypothesis.
b) If the empirical consequences appear then the hypothesis can be rejected using deduction.
c) Test if the empirical consequences appear.
d) If the empirical consequences appear then there is some deductive support for the hypothesis being true.
e) Formulate a research question.
f) If the empirical consequences appear then there is some inductive support for the hypothesis being true.
g) Formulate a hypothesis.
h) If the empirical consequences do not appear then the hypothesis can be rejected using deduction.
Answers: a), c), f), g), h)
Which of the following are requirements for a statement S to be used as a hypothesis? Mark all correct options.
a) S mustn’t be necessarily true or necessarily false.
b) S must be formulated in such a way that it is easily understandable for laypersons.
c) If S is to be counted as a good hypothesis, it should either be necessarily true or have the form of a question.
d) S must be on a form such that it can be true or false.
e) If S is to be counted as a good hypothesis, it should either be somewhat general or concern something that is not directly observable.
Answer: a), d), e)
Falsification and falsifiability are not the same thing. Which of the following claims are true? Mark all correct options.
a) Falsifiability is a property of a hypothesis, it has to do with the number of observable consequences that could show that it is false.
b) Falsification is an event, the observation that an implication of a hypothesis is not true, which entails that the hypothesis itself is false.
c) Falsifiability is a property of a hypothesis, it has to do with the number of observable consequences that actually have demonstrated that it is false.
d) Falsification is an event, the observation that an implication of a hypothesis is not true, which means that the test procedure went wrong (and should thus be modified).
Answer: a), b)
Mark all true claims about Popper’s falsificationism.
a) Falsificationism infers the truth of a hypothesis if all (so far) observable consequences have been observed.
b) Falsificationism is in part an attempt to avoid Hume’s problem of induction.
c) According to falsificationism one can infer the falsity of a hypothesis from even a single false observable consequence.
d) Falsificationism is similar in its structure to the falsifying part of the hypothetico-deductive method.
Answer: b), c), d)
What is the Duhem/Quine thesis? Mark all correct options.
a) It is a thesis that is used by Popper as an argument against Hume’s problem of induction.
b) It is a thesis that states that hypotheses have to be tested using empirical equipment and cannot be conclusively falsified using theory.
c) It is a thesis that states that all hypotheses are tested using auxiliary hypotheses.
d) The thesis implies that no hypothesis can, strictly speaking, be conclusively falsified.
e) It is a thesis that introduces a problem to Popper’s answer to Hume’s problem of induction.
Answer; c), d), e)
What is true regarding hypothesis confirmation? Mark all correct options.
a) Degrees of confirmation can be described qualitatively or quantitatively.
b) Degrees of confirmation can only be described quanitatively.
c) A problem for confirmation is under-determination: that there are many competing hypotheses that are compatible and supported by an empirical observation.
d) Frequentism states that one cannot meaningfully assign probabilities to hypotheses.
Answer: a), c), d)
Which of the following scenarios are examples of choices that can be made to contravene the problem of under-determination? Mark all correct options.
a) If this crop has been affected by toxins in the ground, then the observable consequences will be discoloration and underdevelopment of the seeds. However, we know from previous studies that some common types of vermin cause the same kind of discoloration. Hence, we concentrate our testing on underdevelopment of seeds.
b) The collected data on the population change over time in the area could be described by either a linear or a quadratic function. Both fit the data equally well, so we choose the linear function.
c) If our hypothesis is true it will have three separate observable effects. However, none of them give deductive proof of the hypothesis, so we have to keep looking for observable effects that do.
Answer: a), b)
What is true about the concept severe test? Mark all correct options.
a) Re-using observational data poses a problem to severe testing.
b) When setting up a severe test for a hypothesis we exclude observable consequences that are irrelevant even if they are compatible with the hypothesis.
c) Curve-fitting is always considered a type of severe testing.
d) Severe tests should be relevant, in the sense that the observations we make would be very unlikely if the hypothesis were false.
e) A problem with severe testing is that it requires a lot of background knowledge about possible alternative hypotheses.
Answer: a), b), d), e)
