Thursday, July 23, 2015

Wednesday, July 8, 2015

Abductive Arguments (Inference to the Best Explanation)

An abductive argument (also known as an inference to the best explanation) is an argument in which a hypothesis is inferred from some data on the grounds that it offers the best available explanation of that data.1 Though it may appear as a special type of induction, many philosophers view it as a separate type of inference.

The following example is useful in drawing the distinction between deduction, induction and abduction:

Deductive Reasoning: Suppose a bag contains only red marbles, and you take one out. You may infer by deductive reasoning that the marble is red.

Inductive Reasoning: Suppose you do not know the color of the marbles in the bag, and you take out a handful and they are all red. You may infer by inductive reasoning that all the marbles in the bag are red.

Abductive Reasoning: Suppose you find a red marble in the vicinity of a bag of red marbles. You may infer by abductive reasoning that the marble is from the bag.

Hence we can say that with a deductively valid inference, it is impossible for the premises to be true and the conclusion false. With an inductively strong inference, it is improbable for the premises to be true and the conclusion false. In an abductively weighty inference, it is implausible for the premises to be true and the conclusion false.

Abduction is essentially a kind of guessing by forming the most plausible explanation for a given set of facts or data. It's inference comprises of three steps. First, it begins with the observation of the data, evidence, facts, etc. Second, it forms various explanations that can be given to explain the observations in the first step. Third, it selects the best explanation and draws the conclusion that the selected explanation is acceptable as a hypothesis. Here is the process in standard form:

P1. D exists.
P2. H1 would explain D. 
P3. H1 would offer the best (available) explanation of D. 
C. Therefore, probably, 4. H1

Abductive arguments are commonly used in many areas including law, archaeology, history, science and medical diagnosis. A medical example would include when a doctor examines a patient with certain symptoms and tries to reason from those symptoms to a disease or condition that would explain them. A legal example would be when a police detective gathers evidence then forms a hypothesis as to who committed a crime.

Evaluating Abductive Arguments
The strength of an abductive argument depends of several factors.
1. how decisively H surpasses the alternatives.
2. how good H is by itself, independently of considering the alternatives (we should be cautious about accepting a hypothesis, even if it is clearly the best one we have, if it is not sufficiently plausible in itself)
3. judgments of the reliability of the data
4. how much confidence there is that all plausible explanations have been considered (how thorough was the search for alternative explanations)

Additional factors to consider are:
1. pragmatic considerations, including the costs of being wrong, and the benefits of being right 
2. how strong the need is to come to a conclusion at all, especially considering the possibility of seeking further evidence before deciding.

1. A Practical Study of Argument

2. Abductive, presumptive and plausible arguments

Tuesday, July 7, 2015

Bradford Hill Criteria for Causation (epidemiology)

The Bradford Hill criteria for causation are a group of criteria or guidelines used to help determine if an observed association is potentially causal. They were established in 1965 by the English epidemiologist Sir Austin Bradford Hill.

Research to determine the cause of disease is a principal aim of epidemiology. As most epidemiological studies are observational rather than experimental, a number of possible explanations for an observed association must be considered before a cause-effect relationship can be inferred. In his 1965 paper The environment and disease: association or causation, Hill proposed the following nine guidelines to help assess if a causal relationship exists:

1. Strength: (effect size): A small association does not mean that there is not a causal effect, though the larger the association, the more likely that it is causal.

2. Consistency: (reproducibility): Consistent findings observed by different persons in different places with different samples strengthens the likelihood of an effect.
3. Specificity: Causation is likely if a very specific population at a specific site and disease with no other likely explanation. The more specific an association between a factor and an effect is, the bigger the probability of a causal relationship.

4. Temporality: The effect has to occur after the cause (and if there is an expected delay between the cause and expected effect, then the effect must occur after that delay).

5. Biological gradient: Greater exposure should generally lead to greater incidence of the effect. However, in some cases, the mere presence of the factor can trigger the effect. In other cases, an inverse proportion is observed: greater exposure leads to lower incidence.

6. Plausibility: A plausible mechanism between cause and effect is helpful (but Hill noted that knowledge of the mechanism is limited by current knowledge).

7. Coherence: Coherence between epidemiological and laboratory findings increases the likelihood of an effect. However, Hill noted that "... lack of such [laboratory] evidence cannot nullify the epidemiological effect on associations".

8. Experiment: "Occasionally it is possible to appeal to experimental evidence".

9. Analogy: The effect of similar factors may be considered.