Module 9A Flashcards
Define Non-Deductive Argument:
- Most beliefs are formed through non-deductive arguments, whereSUPPORT IS LESS THAN COMPLETE but often SUFFICIENT TO WARRANT OUR BELIEFS.
- Most of what we believe on the basis of arguments relies on non-deductive reasoning
If all good arguments were deductively valid, most of our beliefs would be based on poor arguments… EXPLAIN
– Deductive validity alone does not account for the majority of our beliefs;
- non-deductive arguments play a crucial role in supporting our convictions.
Non-deductive arguments form the basis for much of our belief system.
These arguments may not offer complete support, but they provide enough justification for our beliefs.
Inductive Arguments
Definition:
In inductive inferences, conclusions covering unobserved things are drawn from:
- experiences or observations,
- expanding knowledge based on perception,
- observation,
- testimony, or
- authority.
Basis of Inductive Inference:
Inductive arguments involve REASONING FROM what we have EXPERIENCED OR OBSERVE TO CONCLUSIONS THAT ENCOMPASS UNOBSERVED ASPECTS, —CONTRIBUTING TO KNOWLEDGE EXPANSION.
Justification of Reliance on Experts
If we ….BASE KNOWLEDGE ON EXPERT OPINIONS, the RELIANCE ON EXPERTS IS JUSTIFIED THROUGH INDUCTIVE ARGUMENT.
Type of Inductive Arguments:
Inductive Analogies:
Examination of a type of inductive argument involving analogies.
Major Types of Inductive Arguments:
(i) Singular Inductions
(ii) Inductive Generalizations
(iii) Explanatory Inductions
Explanation: These are three major types of inductive arguments.
– ALSO DON’T FORGET INDUCTIVE ANALOGIES
Common Characteristics of Inductive Arguments:
- Premises Content: (INFERENCE BASIS)
– Reports on /Drawing conclusions on observations or experiences. - Conclusion Type:
– Factual statement about past, present, or future events. - Nature of Extrapolation:
—Involves extrapolating from observed to unobserved. - Probability Aspect:
— The conclusion is made probable by the premises.
Arguments in which premises are observational or experiential reports, leading to a factual conclusion about past, present, or future events. The conclusion extrapolates from observed to unobserved, and it is made probable by the premises.
Type of Inductive Argument:
Singular Inductions
Characteristics:
- Argument from EXPERIENCE of similar situations to a conclusion about a SINGLE OCCASION
Example: Dining at Zorba’s and predicting Zorba will dance on the table during the next visit.
Type of Inductive Argument:
Inductive Generalizations
Characteristics:
- Conclusions about all members of a class based on information about a sample.
- Example: Waiters at UWA cafés, where most observed waiters are students at UWA.
Type of Inductive Argument:
Explanatory Inductions
Characteristics:
- Arguments from OBSERVED DATA TO PROBABLE CAUSE..
- Example: Receipt of a postcard with Bangkok scene and Thailand postmark as evidence of safe arrival in Thailand.
Causal explanation involves the observed data (postcard) and the probable cause (being in Thailand).
Concept: Inference to the Best Explanation:
In inductive arguments, the conclusion doesn’t necessarily follow from the premises.
In cases where multiple explanations are possible, we make an inference to the best explanation.
—- It is possible that you went to India instead and asked a friend to post the card. Or perhaps you
didn’t leave W.A. at all, wishing to save money and for me to think that you were having a good time.
— In cases such as this, what we make is what is called anINFERENCE TO THE BEST EXPLANATION.
Good explanatory inductions are ones in which the explanation offered is the
BEST OF THE POSSIBLE ALTERNATIVES
Good Explanatory Inductions
In explanatory inductions, the quality of the explanation offered is crucial.
— Good explanatory inductions provide the best among possible alternatives.
Example of Inference to the Best Explanation:
Premises:
(a) Alain is out of his office every Thursday between 2 p.m and 3.30 p.m.
(b) I see him rushing off before 2 each week.
(c) He has a gleam in his eye when he returns.
(d) When he returns, he is showered, red in the face, . . .
(g) He is looking sharper, fitter, . . .
Conclusion: Alain has weekly workouts at the gym.
————–Complex Form of the Argument:————————-
There is a range of facts (a), (b), (c), . . . (g).
Good explanation: Alain goes to the gym for weekly workouts.
No other rival hypothesis provides a better explanation.
Therefore, Alain has weekly workouts at the gym.
Different Types of Facts:
Some observed facts form part of the causal explanation (e.g., (d) and (g)), while others are expected to be true if the conclusion is true (e.g., (b)).
This distinction is crucial in EVALUATING STRENGTH OF INFERENCE TO THE BEST EXPLANATION. .
Illustrative Example: Lee Harvey Oswald and Motive
– In some cases, observed facts are expected to be true if the conclusion is true.
– For example, if Lee Harvey Oswald killed President John Kennedy, we would expect to find it true that Oswald had a motive, even though the killing did not cause him to have the motive.
Evaluating Inductive Arguments
Evaluation involves assessing ACCEPTABILITY and SUPPORT in inductive arguments.
Challenge in Inductive Arguments:
EVEN IN GOOD inductive arguments, ALL PREMISES CAN BE ACCEPTABLE AND TRUE, YET THE CONCLUSION MAY FAIL TO BE TRUE WHEN IT GOES BEYOND THE EVIDENCE.
Evaluation Criteria for Inferences to the Best Explanations:
Consideration:
– To assess inferences to the best explanations, the ABILITY TO THINK OF RIVAL EXPLANATIONS IS CRUCIAL.
Identifying FACTS LIKELY -‘TRUE’ UNDER ONE EXPLANATION BUT UNLIKELY UNDER RIVALS HELPS IN IN THE ASSESSMENT.
Reliability of Observers in Inductive Arguments:
Factor:
Inductive arguments rely on observations; thus, the RELIABILITY OF RELEVANT OBSERVERS PARTIALLY DETERMINES THE ACCEPTABILITY OF OBSERVATION REPORTS.
Role of Concepts in Inductive Arguments:
Aspect:
Observations are FRAMED IN TERMS OF CONCEPTS, and the ACCEPTABILITY OF REPORTS DEPENDS on the OBSERVER’S SKILL IN ADEQUATELY CLASSIFYING OBSERVED OBJECTS.
Assumption in Inductive Arguments:
Inductive arguments ASSUME THAT IRREGULARITIES PERSIST THROUH TIME AND/OR SPACE, JUSTIFYING EXTRAPOLATION FROM OBSERVED TO UNOBSERVED.
The philosophical dispute surrounds the justification and rationality of this assumption.
Importance of Regularities Assumption
– Regardless of the philosophical dispute,
IT IS CRUCIAL TO RECOGNISE THAT INDUCTIVE ARGUMENTS HEAVILY RELY ON THE ASSUMPTION THAT SOME REGULARITIES PERSIST THROUGH TIME AND/OR SPACE.
Term: Sampling and Support
- In inductive generalizations: observations about a section of a group or population lead to general conclusions about the entire population.
- The group we draw CONCLUSIONS about is the TARGET POPULATION, and the OBSERVED SECTION IS THE SAMPLE.
- A GOOD SAMPLE IS REPRESENTATIVE OF THE TARGET POPULATION.
Representative Sample:
A perfectly representative sample HAS THE SAME PERCENTAGE OF ITEMS A GIVEN CHARACTERISTIC AS THE TOTAL POPULATION.
IT ENSURES THE CONCLUSION DRAWN FROM THE SAMPLE APPLIES TO THE ENTIRE POPULATION.
Challenge in Representativeness
Question: What is the problem with defining a perfectly representative sample based on the percentage of items with a given characteristic?
The challenge lies in knowing WHETHER A SAMPLE IS REPRESENTATIVE WITHOUT PRIOR KNOWLEDGE OF THE ENTIRE TARGET POPULATION.
A criterial definition is needed to address this issue.
Methods for Representativeness:
RANDOMLY selected samples and STRATIFIED SAMPLES are methods to INCREASE LIKELYHOODS OF HAVING REPRESENTATIVE SAMPLES.
These methods provide criteria for identifying samples that accurately represent the target population.
Perfectly Representative Sample
(Criterial or Operational Definition)
- All members of a sample, S, are randomly selected to be perfectly representative of a population, P, regarding characteristic, C.
- Alternatively, a sample, S, is perfectly representative of a population, P, with respect to characteristic, C, when all members of the sample are selected using a suitably stratified sampling procedure.
Criterial Definition:
A criterial definition is necessary to determine whether a sample is representative.
Methods such as random selection and stratified sampling help establish criteria for identifying representative samples.
Term:
Perfectly Representative Sample (Essential Definition)
A sample, S, is perfectly representative of a population, P, concerning characteristic, C, when the percentage of S with C equals the percentage of P with C.
Random Samples (Technical Sense)
A sample is considered random when every member of the target population has an equal chance of being included.
It is crucial to distinguish the mathematical sense of ‘random’ from the everyday sense, ensuring the technical sense applies for a sample to be truly representative.
Everyday Sense of ‘Random’
Refers to something being perceived as NON-SYSTEMATIC or WITHOUT DISCERNIBLE PATTERN.
This differs from the technical sense of ‘random,’ which specifically denotes equal probability for each member in a sample.
Significance of Stratified Samples:
- Stratified samples are EFFECTIVE WHEN INFORMATION ABOUT THE TARGET POPULATION’S DEMOGRAPHIC, such as percentages of gender or income groups, iIS AVALIABLE.
By REPLICATING THESE PERCENTAGES IN THE SAMPLE, stratified sampling enhances the likelihood of accurately representing features under investigation.
Stratified Samples:
Specially constructed samples can be used when significant information about the target population is already known.
Stratified samples guarantee representativeness for specific characteristics by mimicking the known percentages of groups within the population.
Biased Samples:
Biased samples are chosen in a way that makes lack of representativeness highly likely.
An example would be drawing conclusions about all UWA students but only sampling those who attend philosophy lectures or frequent the tavern, introducing potential biases.
Challenges of Biased Samples
- It’s challenging to eliminate all biases from sampling.
- However, it’s essential to recognize that even biased samples CAN OFFER USEFUL INFORMATION.
- The PRIMARY GOAL IS TO IDENTIFY POSSIBLE BIASES AND WORK TOWARDS MINIMIZING THEM.
Example of Biased Sampling
Question: Provide an example illustrating biased sampling in the context of drawing conclusions about UWA students.
A biased sample occurs if one only selects UWA students who attend philosophy lectures or frequent the tavern.
This narrow selection may not represent the diverse preferences of the entire student population, leading to potential inaccuracies in conclusions.
Information from Biased Samples
Question: What should be remembered about the information obtained from biased samples?
It’s important to remember that even biased samples can provide helpful information.
However, the focus should be on detecting potential biases and taking steps to minimize them for more accurate conclusions.
Goal in Biased Sampling
Question: What should be the primary aim when dealing with biased samples?
The main goal in dealing with biased samples is to DETECT and ACKNOWLEDGE possible biases and make efforts to REDUCE THEM, ensuring MORE RELIABLE and REPRESENTATIVE FINDINGS.
“A singular induction is a generalisation.”
FALSE
“An inductive generalisation is an inference from a sample of a population to the wider population.”
TRUE
“An inference to the best explanation is not an example of an explanatory induction.”
FALSE
“The following definition of “perfectly representative” is a good operational or criterial definition: A sample, S, is perfectly representative of a population, P, with respect to characteristic, C, when the percentage of S that have C = the percentage of P that have C.”
FALSE
“The following definition of “perfectly representative” is a good essential definition: A sample, S, is perfectly representative of a population, P, with respect to characteristic, C, when the percentage of S that have C = the percentage of P that have C.”
TRUE
“The conclusion to an inductive argument is one which the premises make probable (to varying degrees).”
TRUE
“The conclusion to an inductive argument is a factual statement (a statement of what is claimed to be factual).”
TRUE
“Causal inductions are examples of explanatory inductions.”
TRUE
