Course outline
Module 1
Module 1: What Inductive Reasoning Is
What Inductive Reasoning Is
Knowledge check
Q1. Which of the following best describes inductive reasoning?
- A. Drawing a certain conclusion from universally accepted premises
- B. Drawing a probable conclusion from specific observations or evidence ✓
- C. Deriving a conclusion that must be true if the premises are true, using formal rules
- D. Testing a hypothesis by deducing predictions and checking them against data
Why: Inductive reasoning moves from specific observations to a general conclusion that is probable rather than guaranteed. Option A describes deductive reasoning, option C describes formal logic in a deductive sense, and option D describes the hypothetico-deductive method.
Q2. Every swan I have observed so far has been white. What is the most reasonable inductive conclusion?
- A. All swans are white. ✓
- B. Some swans are black.
- C. Swans are birds that cannot be any other color.
- D. There are no swans in the world except white ones.
Why: Inductive reasoning generalizes from observed cases to a broader claim, so 'all swans are white' is the natural inductive generalization. It is only probable, as the discovery of black swans in Australia later showed, but it is the conclusion the observations support.
Q3. Which argument is an example of inductive reasoning?
- A. All mammals are warm-blooded. Whales are mammals. Therefore, whales are warm-blooded.
- B. Every time I have studied the night before, I have passed the exam. Therefore, studying the night before will likely help me pass this exam. ✓
- C. If it rains, the ground gets wet. It rained. Therefore, the ground is wet.
- D. The number is even or odd. It is not even. Therefore, it is odd.
Why: The studying example moves from past specific cases to a probable prediction about the future, which is inductive. The other options are deductive: they follow valid forms (syllogism, modus ponens, disjunctive syllogism) where the conclusion must follow from the premises.
Q4. What is the main reason inductive conclusions are not certain?
- A. Inductive reasoning always contains a logical contradiction.
- B. The premises of an inductive argument can be true while the conclusion is false, because the future or unobserved cases may differ. ✓
- C. Inductive reasoning relies on mathematical formulas that are never exact.
- D. Inductive reasoning can only be applied to fictional situations.
Why: Inductive arguments are non-demonstrative: even strong evidence leaves open the possibility that unobserved cases behave differently. This is why no amount of white swans proves that all swans are white. The other options misstate the nature of induction.
Q5. A researcher surveys 1,000 randomly selected voters and finds that 60% prefer Candidate X. She concludes that approximately 60% of all voters in the country prefer Candidate X. What type of reasoning is this?
- A. Deductive reasoning, because the conclusion follows from the survey data
- B. Inductive reasoning, because a general conclusion about the whole population is drawn from a sample ✓
- C. Abductive reasoning, because the researcher is seeking the best explanation
- D. Circular reasoning, because the survey assumes its own result
Why: Generalizing from a sample to a whole population is a classic inductive inference. The conclusion is probable and carries a margin of error, which reflects the uncertainty typical of inductive reasoning. Option A is wrong because the conclusion goes beyond what is strictly entailed by the data.
Module 2
Module 2: Generalization From Samples
Generalization From Samples
Knowledge check
Q1. A researcher surveys 50 students from a single dormitory about their favorite campus dining hall and concludes that all students at the university prefer that dining hall. What is the most significant weakness in this generalization?
- A. The sample size is too large to be representative.
- B. The sample is not representative because it comes from only one dormitory. ✓
- C. Students always give honest answers, so the data is reliable.
- D. Dining hall preferences cannot be measured through surveys.
Why: A generalization about the whole university requires a sample that reflects the whole population. Surveying students from only one dormitory introduces sampling bias, since those students may share living situations or preferences that differ from the rest of the student body.
Q2. Which of the following samples would be most likely to support a strong inductive generalization about the average height of adults in a country?
- A. Heights of 10 professional basketball players
- B. Heights of 200 adults chosen randomly from across the country ✓
- C. Heights of every adult in one small village
- D. Heights of 30 adults who volunteered online
Why: A large, randomly selected sample drawn from across the whole population minimizes bias and captures the variety present in that population, making the generalization more reliable.
Q3. Every swan I have observed so far has been white, so I conclude that all swans are white. Which statement best describes this reasoning?
- A. It is a deductive argument with a valid conclusion.
- B. It is an inductive generalization that is strong but not conclusive, since a single black swan would refute it. ✓
- C. It is a sound argument because the premises are true.
- D. It is a causal argument about why swans are white.
Why: Inductive conclusions go beyond the premises and are never guaranteed. The observations support the generalization, but the discovery of black swans in Australia showed that such conclusions can be overturned by new evidence.
Q4. A poll asks only people who attend a political rally whether they support a candidate, and 90% say yes. Why should we be cautious about generalizing this result to the whole electorate?
- A. Ninety percent is too high to be a real number.
- B. Rally attendees are likely to already share similar views, creating a biased sample. ✓
- C. Polls are never useful for predicting anything.
- D. The sample must be exactly 50% to be valid.
Why: People who attend a rally for a candidate are a self-selected group that probably holds more favorable views than the general electorate. This selection bias means the sample does not represent the whole population, weakening any generalization.
Q5. Which factor most strengthens an inductive generalization, all else being equal?
- A. The observations were all made on a single day.
- B. The observed cases are diverse and cover different conditions within the population. ✓
- C. The researcher already believed the conclusion before collecting data.
- D. The sample consists of the researcher's close friends.
Why: Diversity among the observed cases makes it less likely that the pattern is an artifact of narrow conditions. Varied, representative observations provide stronger support for the generalization than homogeneous or convenient ones.
Module 3
Module 3: Strength and Weakness of Arguments
Strength and Weakness of Arguments
Knowledge check
Q1. Which of the following best describes an inductive argument?
- A. An argument in which the premises guarantee the truth of the conclusion
- B. An argument in which the premises provide probable support for the conclusion ✓
- C. An argument in which the conclusion contradicts one of the premises
- D. An argument that relies solely on mathematical proof
Why: Inductive arguments aim to make their conclusions probable, not certain. Their premises provide evidence that supports the conclusion to some degree, unlike deductive arguments where truth is guaranteed.
Q2. Consider this argument: 'Every swan I have observed is white. Therefore, all swans are white.' How would this argument be evaluated?
- A. As a strong argument because the premises are all true
- B. As a valid deductive argument
- C. As a weak inductive argument because the sample is limited and the conclusion goes beyond the evidence ✓
- D. As a sound argument because it has no counterexamples
Why: The conclusion generalizes from a limited sample to all swans. Even though the observations may be accurate, the sample is too small or unrepresentative to justify the universal conclusion, making the argument weak.
Q3. What is the main factor that determines the strength of an inductive argument?
- A. The number of premises it contains
- B. The degree to which the premises make the conclusion probable ✓
- C. Whether the conclusion is surprising or unexpected
- D. The length of the argument
Why: An inductive argument is strong when its premises, if true, make the conclusion likely to be true. The number of premises or the length of the argument does not determine strength on its own.
Q4. Which of the following would most strengthen the inductive argument: 'Most of the 500 randomly selected voters in this town prefer Candidate X. Therefore, most voters in the town prefer Candidate X'?
- A. Learning that the town has a large population
- B. Learning that the selected voters were chosen only from one neighborhood
- C. Learning that the sample was random and representative of the whole town ✓
- D. Learning that Candidate X has a memorable name
Why: A large, random, and representative sample increases the likelihood that the sample reflects the population as a whole, which strengthens the inductive inference. A biased sample, such as one neighborhood, would weaken it.
Q5. Which of the following is an example of a weak inductive argument?
- A. Most birds I have seen can fly, so the next bird I see will probably be able to fly
- B. I have tasted three apples from this tree and all were sweet, so apples from this tree are probably sweet ✓
- C. My friend, a doctor, recommends this medicine, so it is probably effective
- D. Every triangle has three sides, so this triangle has three sides
Why: Three apples is a very small sample, so the conclusion that all apples from the tree are sweet is weakly supported. The other options either rely on larger samples or are deductive, and the first and third are reasonably supported inductive arguments.
Module 4
Module 4: Base Rates and Representativeness
Base Rates and Representativeness
Knowledge check
Q1. A disease affects 1 in 1,000 people. A test for the disease is 95% accurate (correctly identifies sick people 95% of the time and healthy people 95% of the time). If a randomly chosen person tests positive, what is the most likely probability that they actually have the disease?
- A. 95%
- B. About 2% ✓
- C. About 50%
- D. About 10%
Why: Base rates matter here. Out of 1,000 people, about 1 has the disease and tests positive, while about 50 healthy people (5% of 999) also test positive falsely. So only about 1 of roughly 51 positives is truly sick, which is around 2%. The low base rate overwhelms the test's accuracy.
Q2. Linda is 31, outspoken, and was a philosophy major who cared deeply about social justice. Which is more probable?
A) Linda is a bank teller.
B) Linda is a bank teller and is active in the feminist movement.
Which option should be chosen based on probability logic?
- A. Option A, because a conjunction can never be more probable than one of its components ✓
- B. Option B, because the description fits Linda better
- C. Both are equally probable
- D. Neither can be assessed without more data
Why: This is the conjunction fallacy. The probability that Linda is a bank teller AND a feminist can never exceed the probability that she is a bank teller alone. Option B only feels more representative of her description, but representativeness should not override the rules of probability.
Q3. Of the 100 people in a study, 70 are engineers and 30 are lawyers. A person is picked at random and described as someone who enjoys building things and is very detail-oriented. Based on base rates alone, before considering the description, what is the probability this person is an engineer?
- A. 30%
- B. 50%
- C. 70% ✓
- D. 100%
Why: Before using the personality description, the base rate is simply the proportion of engineers in the sample, which is 70 out of 100. Good reasoning starts from this prior and then updates it with new evidence, rather than ignoring the base rate in favor of stereotypes.
Q4. A small town has 3 hospitals. Hospital A records 45 births per day, Hospital B records 15 births per day, and Hospital C records 15 births per day. Over a year, which hospital is most likely to have the most days where over 60% of births are boys?
- A. Hospital A, because it has the largest number of births
- B. Hospital B, because smaller samples are more variable ✓
- C. Hospital C, because it is in the middle
- D. All three hospitals will have the same number of such days
Why: Smaller samples fluctuate more from the true average (about 50% boys). Extreme proportions, like over 60% boys, are much more likely in a sample of 15 than in a sample of 45. So the smaller hospitals, B and C, will show more extreme days, and both are tied for the most. Between the options given, B is the best answer, reflecting the law of small numbers.
Q5. You learn that a coin has landed heads 6 times in a row. What is the probability the next flip is heads, assuming the coin is fair?
- A. Less than 50%, because tails is now due
- B. More than 50%, because heads is on a streak
- C. Exactly 50%, because each flip is independent ✓
- D. Exactly 0%, because the streak must end
Why: Each flip of a fair coin is independent, so past outcomes do not change the odds. The gambler's fallacy leads people to believe tails is 'due,' but the coin has no memory. Six heads in a row is unusual but does not alter the probability of the next flip, which remains 50%.
Module 5
Module 5: Analogical Reasoning
Analogical Reasoning
Knowledge check
Q1. Bird is to wing as fish is to:
- A. Water
- B. Fin ✓
- C. Scale
- D. Swim
Why: A wing is the organ a bird uses for locomotion, just as a fin is the organ a fish uses for locomotion. The relationship is between an animal and the body part it uses to move.
Q2. Hot is to cold as light is to:
- A. Bright
- B. Sun
- C. Dark ✓
- D. Lamp
Why: Hot and cold are opposites, so the analogy is based on antonymy. The opposite of light is dark.
Q3. Pen is to write as knife is to:
- A. Sharp
- B. Kitchen
- C. Steel
- D. Cut ✓
Why: A pen is used to write, so the relationship is between a tool and its primary function. A knife is used to cut.
Q4. Doctor is to hospital as teacher is to:
- A. Students
- B. Book
- C. School ✓
- D. Lesson
Why: A doctor typically works in a hospital, so the analogy pairs a professional with their usual workplace. A teacher typically works in a school.
Q5. Puppy is to dog as kitten is to:
- A. Milk
- B. Mouse
- C. Cat ✓
- D. Fur
Why: A puppy is a young dog, so the relationship is between a young animal and its adult form. A kitten grows into an adult cat.
Module 6
Module 6: Causal Inference From Patterns
Causal Inference From Patterns
Knowledge check
Q1. A shop owner notices that every time she plays upbeat music, her customers buy more. She concludes the music causes the extra sales. Which inductive concern is most relevant?
- A. The sample size is too large to be reliable
- B. A third factor, such as busier shopping hours, could explain both the music and the sales ✓
- C. Inductive reasoning cannot be applied to retail settings
- D. Upbeat music always has a causal effect on purchasing behavior by definition
Why: Correlation alone does not establish causation. A confounding variable, like busier hours when more people are shopping, may drive both the music choice and the higher sales.
Q2. Over several weeks, a gardener observes that plants watered in the morning grow taller than those watered at night. Which conclusion is best supported by this pattern?
- A. Morning watering always produces taller plants in every species
- B. Watering time has no effect on plant growth
- C. Morning watering may be associated with taller growth, but more controlled testing is needed to confirm a causal link ✓
- D. Night watering kills plants over time
Why: Observed patterns suggest a possible link, but inductive conclusions should be tentative. Controlled comparison, such as varying only watering time, would strengthen the causal claim.
Q3. Every swan a naturalist has ever seen in Europe has been white. She concludes that all swans are white. What is the main weakness of this inductive argument?
- A. It uses too few observations to be valid in any case
- B. Its conclusion goes beyond its evidence, since the observed sample is limited and unrepresentative of all swans ✓
- C. It confuses a cause with an effect
- D. It is a deductive argument with a false premise
Why: Inductive generalizations can be strong or weak depending on how representative and extensive the observations are. Black swans in Australia show that a limited, geographically narrow sample can mislead.
Q4. A researcher finds that cities with more ice cream shops have higher rates of drowning. Which is the most reasonable next step before inferring causation?
- A. Conclude that ice cream causes drowning since the pattern is consistent
- B. Ignore the data because ice cream and drowning are unrelated topics
- C. Check for a common cause, such as hot summer weather that increases both ice cream sales and swimming ✓
- D. Assume the pattern is random and stop investigating
Why: Hot weather plausibly increases both ice cream consumption and swimming, which raises drowning risk. Identifying potential common causes is a key step in causal inference from patterns.
Q5. After noticing that three consecutive coin flips landed heads, a person predicts the next flip will be tails because tails is 'due.' What inductive error is this?
- A. Strong generalization from a large sample
- B. Correct use of base rates
- C. The gambler's fallacy, treating past independent events as influencing future ones ✓
- D. A valid causal inference about the coin
Why: Each coin flip is independent with a 50 percent chance of each outcome. Believing past results make the opposite outcome more likely is the gambler's fallacy, a common mistake in pattern-based reasoning.
Module 7
Module 7: Common Inductive Fallacies
Common Inductive Fallacies
Knowledge check
Q1. A student tried three restaurants in town and all three gave slow service. She concludes that every restaurant in town has slow service. Which inductive fallacy is she committing?
- A. Hasty generalization ✓
- B. Appeal to authority
- C. Post hoc reasoning
- D. Slippery slope
Why: A hasty generalization draws a broad conclusion from a sample that is too small or unrepresentative. Three restaurants is not enough evidence to conclude anything about every restaurant in town.
Q2. A company surveyed only its own loyal customers and found that 95% are satisfied with its products. It concludes that 95% of all potential customers are satisfied. What is the main flaw?
- A. Biased sample ✓
- B. Circular reasoning
- C. False dilemma
- D. Ad hominem
Why: The sample is biased because it includes only existing loyal customers, who are likely to be more satisfied than the general population. A representative sample is needed to support a conclusion about all potential customers.
Q3. Every time Maria wears her lucky socks, her team wins. Maria concludes that the socks cause the wins. Which fallacy is this?
- A. Post hoc ergo propter hoc ✓
- B. Straw man
- C. Equivocation
- D. Appeal to popularity
Why: Post hoc ergo propter hoc means 'after this, therefore because of this.' Because the wins follow the socks, Maria assumes causation, but the correlation may be coincidental or explained by other factors.
Q4. Ten people in a row have told you a coin toss came up heads, so you conclude that the next toss will almost certainly be heads too. What mistake is this?
- A. Gambler's fallacy ✓
- B. Hasty generalization
- C. Red herring
- D. Begging the question
Why: Assuming that a streak of random outcomes makes the next outcome more or less likely is the gambler's fallacy. Each coin toss is independent, so the chance of heads on the next toss remains 50%.
Q5. A psychologist studied only college sophomores in one city and concluded that all human beings process memory in the same way. What is the primary weakness of this argument?
- A. Unrepresentative sample ✓
- B. Appeal to emotion
- C. Tu quoque
- D. Denying the antecedent
Why: The conclusion covers all humans, but the sample includes only a narrow group of college sophomores in one city. Such a sample may not represent people of other ages, cultures, or backgrounds, so the generalization is unsupported.