Part 2: Scientific inference Flashcards
Inference:
An act or a process of reaching a conclusion from a set of premises, which can express, for instance, known facts or evidence.
Premise:
a statement in an argument that justifies a conclusion.
Conclusion:
A statement that follows logically from premises.
Generalization:
Inductive inference from a sample to a general conclusion.
Projection:
Inductive inference from past samples to future samples.
Inductive inference:
In an inductive inference, the premises support the conclusion but does not guarantee its truth.
Deductive inference:
In a valid deductive inference, true premises necessitate the truth of the conclusion.
Conditional claim:
A claim involving the logical operator “if”, for instance of the form “If A then B”.
Modus ponens:
A deductive inference of the form: If A then B, A, therefore B.
Modus tollens:
A deductive inference of the form: If A then B, not B, therefore not A.
Amplicative:
Inferences that go beyond what is stated in the premises – in particular, inductive inferences are amplicative.
Explicative:
Inferences that do not go beyond what is stated (implicitly) in the premises – in particular, deductive inferences are explicative.
Truth preservation:
The conclusion must be true if the premises are true, see deductive inference.
Fallible:
The conclusion can be false even if premises are true.
Infinite regress:
A never ending chain of propositions being justified by other propositions which in turn are justified by other propositions and so on.
Foundationalism:
Propositions are justified by being inferred from foundational premises which do not need additional justification, for instance necessarily true premises.
Coherentism:
Propositions are justified by being compatible with a coherent set of propositions, where each proposition in the set is compatible with every other proposition in the set.
Falsification:
Rejecting a hypothesis as a result of an empirical test.
Confirmation:
Increasing the confidence in a hypothesis as a result of an empirical test.
Hypothesis:
A proposition that can be true or false but is not necessarily true or false, and that preferably either has some generality or is about something not directly observable.
Tautology:
A proposition which is necessarily true or false.
The 5 steps of the Hypothetico-Deductive method?
- Formulate a hypothesis H
- Deduce observable consequences {Ci} from H.
- Test whether {Ci} is true or not.
- If {Ci} is false, infer that H is false.
- If {Ci} is true, increase confidence in H
Direct observation:
Observations of objects and properties that are accessible through the use of human senses.
Operationalization:
A way to measure something which cannot be directly observed or that cannot be observed directly with sufficient precision, by connecting this feature with something causally connected to something that can be observed directly.
Falsification in the context of hypothesis testing:
To prove a hypothesis false.
Falsification in the context of ethics:
Unethical behaviour of changing data to suit your opinion or target.
Asymmetry between falsification and confirmation:
No amount of confirming observations can deductively confirm the hypothesis, but one falsifying observation can deductively falsify the hypothesis.
Falsificationism:
The view that science should proceed only through valid falsification, and never use confirmation.
Falsifiable:
A hypothesis is falsifiable if it is possible to show that it is false, even if it has not yet been shown to be false.
Demarcation of science:
Distinguishing science from non-science by providing criteria for counting something as science.
Corroboration:
A hypothesis is corroborated if it has withstood multiple falsification attempts.
Auxiliary hypothesis:
A hypothesis used to test another hypothesis, but which one does not intend to test, for instance background assumptions necessary to infer the empirical conclusion.
Conjunction:
Two propositions joined by the logical operator AND. The conjunction is true if and only if both propositions are true
Duhem-Quine thesis:
No hypothesis can be tested without the use of auxiliary hypotheses.
Ad hoc:
The modification of a claim is ad hoc if (i) the claim has previously been falsified, (ii) the modification saves the claim from this falsification and (iii) it makes the claim less falsifiable – i.e. it does not allow deriving any new testable consequences.
Frequentism:
Probabilities are frequencies of repeatable observable events.
Severe test:
A hypothesis test is a severe test if the probability to observe a consequence would be low if the hypothesis were false.
Under-determination:
An inference is underdetermined if there are multiple conclusion that would be equally supported by the premises.
